mirror of
https://github.com/ruby/ruby.git
synced 2022-11-09 12:17:21 -05:00
c5e08b764e
safe functions during garbage collection such as xfree(). On default, T_DATA objects are freed at same points as fianlizers. This approach protects issues such as reported by [ruby-dev:35578]. However, freeing T_DATA objects immediately helps heap usage. Most of T_DATA (in other words, most of dfree functions) are safe. However, we turned off RUBY_TYPED_FREE_IMMEDIATELY by default for safety. * cont.c: ditto. * dir.c: ditto. * encoding.c: ditto. * enumerator.c: ditto. * error.c: ditto. * file.c: ditto. * gc.c: ditto. * io.c: ditto. * iseq.c: ditto. * marshal.c: ditto. * parse.y: ditto. * proc.c: ditto. * process.c: ditto. * random.c: ditto. * thread.c: ditto. * time.c: ditto. * transcode.c: ditto. * variable.c: ditto. * vm.c: ditto. * vm_backtrace.c: ditto. * vm_trace.c: ditto. * ext/bigdecimal/bigdecimal.c: ditto. * ext/objspace/objspace.c: ditto. * ext/stringio/stringio.c: ditto. * ext/strscan/strscan.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@43466 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
14908 lines
498 KiB
Text
14908 lines
498 KiB
Text
Tue Oct 29 20:01:58 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* add RUBY_TYPED_FREE_IMMEDIATELY to data types which only use
|
|
safe functions during garbage collection such as xfree().
|
|
|
|
On default, T_DATA objects are freed at same points as fianlizers.
|
|
This approach protects issues such as reported by [ruby-dev:35578].
|
|
However, freeing T_DATA objects immediately helps heap usage.
|
|
|
|
Most of T_DATA (in other words, most of dfree functions) are safe.
|
|
However, we turned off RUBY_TYPED_FREE_IMMEDIATELY by default
|
|
for safety.
|
|
|
|
* cont.c: ditto.
|
|
|
|
* dir.c: ditto.
|
|
|
|
* encoding.c: ditto.
|
|
|
|
* enumerator.c: ditto.
|
|
|
|
* error.c: ditto.
|
|
|
|
* file.c: ditto.
|
|
|
|
* gc.c: ditto.
|
|
|
|
* io.c: ditto.
|
|
|
|
* iseq.c: ditto.
|
|
|
|
* marshal.c: ditto.
|
|
|
|
* parse.y: ditto.
|
|
|
|
* proc.c: ditto.
|
|
|
|
* process.c: ditto.
|
|
|
|
* random.c: ditto.
|
|
|
|
* thread.c: ditto.
|
|
|
|
* time.c: ditto.
|
|
|
|
* transcode.c: ditto.
|
|
|
|
* variable.c: ditto.
|
|
|
|
* vm.c: ditto.
|
|
|
|
* vm_backtrace.c: ditto.
|
|
|
|
* vm_trace.c: ditto.
|
|
|
|
* ext/bigdecimal/bigdecimal.c: ditto.
|
|
|
|
* ext/objspace/objspace.c: ditto.
|
|
|
|
* ext/stringio/stringio.c: ditto.
|
|
|
|
* ext/strscan/strscan.c: ditto.
|
|
|
|
Tue Oct 29 19:48:33 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: fix typo (FL_WB_PROTECT -> FL_WB_PROTECTED).
|
|
|
|
Tue Oct 29 18:45:08 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_trace.c (tp_free): remvoed because empty free function.
|
|
Use RUBY_TYPED_NEVER_FREE instead.
|
|
|
|
Tue Oct 29 18:37:33 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: introduce new flags for T_TYPEDDATA.
|
|
* RUBY_TYPED_FREE_IMMEDIATELY: free the data given by DATA_PTR()
|
|
with dfree function immediately. Otherwise (default), the data
|
|
freed at finalizaton point.
|
|
* RUBY_TYPED_WB_PROTECTED: make this object with FL_WB_PROTECT
|
|
(not shady).
|
|
|
|
* gc.c (obj_free): support RUBY_TYPED_FREE_IMMEDIATELY.
|
|
|
|
Tue Oct 29 16:49:03 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (vm_malloc_increase): decrease it more carefully.
|
|
|
|
Tue Oct 29 16:24:52 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (heap_page_resurrect): return a page in tomb heap even if
|
|
freelist is NULL.
|
|
|
|
Tue Oct 29 15:46:30 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ruby_atomic.h (ATOMIC_SIZE_CAS): new macro, compare and swap size_t.
|
|
|
|
Tue Oct 29 12:08:05 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/readline/readline.c (readline_getc): Consider
|
|
NULL as input.
|
|
|
|
Tue Oct 29 11:10:08 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* gc.c (gc_profile_total_time): fix off-by-one error in
|
|
GC::Profiler.total_time.
|
|
* test/ruby/test_gc.rb (class TestGc): test for above.
|
|
|
|
Tue Oct 29 09:53:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* insns.def, vm.c, vm_insnhelper.c, vm_insnhelper.h, vm_method.c: split
|
|
ruby_vm_global_state_version into two separate counters - one for the
|
|
global method state and one for the global constant state. This means
|
|
changes to constants do not affect method caches, and changes to
|
|
methods do not affect constant caches. In particular, this means
|
|
inclusions of modules containing constants no longer globally
|
|
invalidate the method cache.
|
|
|
|
* class.c, eval.c, include/ruby/intern.h, insns.def, vm.c, vm_method.c:
|
|
rename rb_clear_cache_by_class to rb_clear_method_cache_by_class
|
|
|
|
* class.c, include/ruby/intern.h, variable.c, vm_method.c: add
|
|
rb_clear_constant_cache
|
|
|
|
* compile.c, vm_core.h, vm_insnhelper.c: rename vmstat field in
|
|
rb_call_info_struct to method_state
|
|
|
|
* vm_method.c: rename vmstat field in struct cache_entry to method_state
|
|
|
|
Mon Oct 28 23:26:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/readline/test_readline.rb (teardown): Clear Readline.input and
|
|
Readline.output.
|
|
|
|
Mon Oct 28 21:35:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/-test-/file/depend, ext/-test-/postponed_job/depend,
|
|
ext/-test-/tracepoint/depend: New files for dependencies.
|
|
|
|
Mon Oct 28 15:32:18 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/openssl/depend (ossl.o): work around of dependency of
|
|
thread_native.h, which depends on headers by THREAD_MODEL.
|
|
[ruby-dev:47777]
|
|
|
|
* ext/openssl/extconf.rb: need THREAD_MODEL.
|
|
|
|
Mon Oct 28 14:57:01 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* load.c (ruby_init_ext): share feature names between frame name and
|
|
provided features.
|
|
|
|
Mon Oct 28 14:41:27 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el: Import ruby-electric.el 2.1 from
|
|
https://github.com/knu/ruby-electric.el.
|
|
|
|
* Hitting the newline-and-indent key within a comment fires
|
|
comment-indent-new-line.
|
|
|
|
* Introduce a new feature
|
|
`ruby-electric-autoindent-on-closing-char`.
|
|
|
|
* Fix fallback behavior of ruby-electric-space/return that
|
|
caused error with auto-complete.
|
|
|
|
Mon Oct 28 13:17:17 2013 Or Cohen <orc@fewbytes.com>
|
|
|
|
* error.c (name_err_to_s): remove no longer needed overriding, since
|
|
r30455 which made exc_to_s almost same. Fixes [GH-413].
|
|
|
|
Mon Oct 28 12:42:11 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* common.mk, ext/objspace/depend, ext/coverage/depend,
|
|
ext/-test-/debug/depend, ext/date/depend: Update dependencies.
|
|
|
|
Mon Oct 28 09:29:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* vm.c: vm_clear_all_cache is not necessary now we use a 64 bit counter
|
|
for global state version.
|
|
|
|
* vm_insnhelper.h: ruby_vm_global_state_version overflow is unnecessary
|
|
|
|
Mon Oct 28 07:47:32 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* vm_backtrace.c (rb_profile_frame_classpath): do not use rb_inspect
|
|
directly, since it might have a custom implementation or show ivars.
|
|
|
|
Mon Oct 28 04:10:41 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* vm_backtrace.c (rb_profile_frame_classpath): handle singleton
|
|
methods defined directly on an object.
|
|
* test/-ext-/debug/test_profile_frames.rb: test for above.
|
|
|
|
Mon Oct 28 00:52:36 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* struct.c (new_struct): fix warning message, class name and encoding.
|
|
|
|
Sun Oct 27 20:53:08 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/readline/readline.c: Include ruby/thread.h for
|
|
rb_thread_call_without_gvl2.
|
|
(readline_rl_instream, readline_rl_outstream): Record FILE
|
|
structures allocated by this extension.
|
|
(getc_body): New function extracted from readline_getc.
|
|
(getc_func): New function.
|
|
(readline_getc): Use rb_thread_call_without_gvl2 to invoke getc_func.
|
|
[ruby-dev:47033] [Bug #8749]
|
|
(clear_rl_instream, clear_rl_outstream): Close FILE structure
|
|
allocated by this extension reliably. [ruby-core:57951] [Bug #9040]
|
|
(readline_readline): Use clear_rl_instream and clear_rl_outstream.
|
|
(readline_s_set_input): Set readline_rl_instream.
|
|
(readline_s_set_output): Set readline_rl_outstream.
|
|
(Init_readline): Don't call readline_s_set_input because
|
|
readline_getc doesn't block other threads for any FILE structure now.
|
|
|
|
[ruby-dev:47033] [Bug #8749] reported by Nobuhiro IMAI.
|
|
[ruby-core:57951] [Bug #9040] reported by Eamonn Webster.
|
|
|
|
Sat Oct 26 19:31:28 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* gc.c: catch up recent changes to compile on GC_DEBUG,
|
|
RGENGC_CHECK_MODE.
|
|
|
|
Sat Oct 26 19:08:00 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* range.c (range_initialize_copy): disallow to modify after
|
|
initialized.
|
|
|
|
Sat Oct 26 17:48:54 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/open-uri.rb (meta_add_field): : Re-implemented.
|
|
[ruby-core:58017] [Bug #9051] patch by Eamonn Webster.
|
|
|
|
Sat Oct 26 14:35:09 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_profile_dump_on): use "Page" terminology.
|
|
|
|
Sat Oct 26 13:25:45 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_sweep, gc_heap_lazy_sweep): fix measurement code.
|
|
We only need one sweep time measurement without lazy sweep.
|
|
|
|
Sat Oct 26 11:59:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* addr2line.c: Include ELF header after system headers (especially
|
|
sys/types.h) to avoid compilation failure,
|
|
"usr/include/sh3/elf_machdep.h:4:2: error: #error Define _BYTE_ORDER!",
|
|
on NetBSD/sh3 (dreamcast, hpcsh, landisk, mmeye).
|
|
|
|
Sat Oct 26 11:35:22 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: tuning parameters.
|
|
|
|
* gc.c (GC_MALLOC_LIMIT): change default value to 16MB.
|
|
|
|
* gc.c (GC_MALLOC_LIMIT_GROWTH_FACTOR): change default value to 2.0.
|
|
|
|
* gc.c (gc_before_sweep): change decrease ratio of `malloc_limit'
|
|
from 1/4 to 1/10.
|
|
|
|
Sat Oct 26 11:30:07 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (vm_malloc_increase): do gc_rest_sweep() before GC.
|
|
gc_rest_sweep() can reduce malloc_increase, so try it before GC.
|
|
Otherwise, malloc_increase can be less than malloc_limit at
|
|
gc_before_sweep(). This means that re-calculation of malloc_limit
|
|
may be wrong value.
|
|
|
|
Sat Oct 26 06:35:41 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (gc_before_heap_sweep): Restructure code to mean clearly.
|
|
heap->freelist is connected to end of list.
|
|
|
|
Sat Oct 26 04:01:35 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_before_heap_sweep): fix freelist management.
|
|
After rb_gc_force_recycle() for a object belonging to heap->freelist,
|
|
`heap->using_page->freelist' is not null.
|
|
|
|
Thu Oct 24 21:57:24 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
|
|
|
|
* parse.y: Remove +(binary) and -(binary) special cases
|
|
[Feature #9048]
|
|
|
|
Thu Oct 24 12:45:53 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* object.c: [DOC] Document first argument also takes string for:
|
|
|
|
rb_mod_const_get, rb_mod_const_set, rb_mod_const_defined
|
|
|
|
Also added note about NameError exception for invalid constant name
|
|
|
|
Thu Oct 24 12:23:58 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread.c (rb_thread_terminate_all): add a comment why we need
|
|
state check and call terminate_i again.
|
|
|
|
Thu Oct 24 12:15:02 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread.c (rb_thread_terminate_all): add a comment why infinite
|
|
sleep is safe.
|
|
|
|
Thu Oct 24 07:41:42 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* gc.c: add new initial_growth_max tuning parameter.
|
|
[ruby-core:57928] [Bug #9035]
|
|
* gc.c (heap_set_increment): when initial_growth_max is set,
|
|
do not grow number of slots by more than growth_max at a time.
|
|
* gc.c (rb_gc_set_params): load optional new tuning value from
|
|
RUBY_HEAP_SLOTS_GROWTH_MAX environment variable.
|
|
* test/ruby/test_gc.rb (class TestGc): test for above.
|
|
|
|
Thu Oct 24 01:34:12 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/win32.h (rb_infinity_float): suppress overflow in
|
|
constant arithmetic warnings. [ruby-core:57981] [Bug #9044]
|
|
|
|
Thu Oct 24 00:11:24 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
|
|
|
|
* lib/ostruct.rb: raise NoMethodError with a #name and #args.
|
|
Raise RuntimeError when modifying frozen instances
|
|
instead of TypeError.
|
|
(OpenStruct#each_pair): Return an enumerator with size
|
|
(OpenStruct#delete): Use the converted argument.
|
|
Patches by Kenichi Kamiya. [Fixes GH-383]
|
|
|
|
* test/ostruct/test_ostruct.rb: Added tests for above.
|
|
|
|
Thu Oct 24 00:10:22 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
|
|
|
|
* array.c: Add Array#to_h [Feature #7292]
|
|
|
|
* enum.c: Add Enumerable#to_h
|
|
|
|
Wed Oct 23 23:48:28 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* gc.c: Rename free_min to min_free_slots and free_min_page to
|
|
max_free_slots. The algorithm for heap growth is:
|
|
if (swept_slots < min_free_slots) pages++
|
|
if (swept_slots > max_free_slots) pages--
|
|
|
|
Wed Oct 23 22:51:03 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/Makefile.sub (config.h): VC 2013 supports C99 mathematics
|
|
functions. [ruby-core:57981] [Bug #9044]
|
|
|
|
Wed Oct 23 19:13:18 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: move increment from heap to heap_pages.
|
|
Share `increment' information with heaps.
|
|
|
|
* gc.c: change ratio of heap_pages_free_min_page
|
|
to 0.80.
|
|
This change means slow down page freeing speed.
|
|
|
|
Wed Oct 23 17:52:03 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (heap_pages_free_unused_pages): cast to (int) for size_t
|
|
variable `i'.
|
|
|
|
Wed Oct 23 17:39:35 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: introduce tomb heap.
|
|
Tomb heap is where zombie objects and ghost (freed slot) lived in.
|
|
Separate from other heaps (now there is only eden heap) at sweeping
|
|
helps freeing pages more efficiently.
|
|
Before this patch, even if there is an empty page at former phase
|
|
of sweeping, we can't free it.
|
|
|
|
Algorithm:
|
|
(1) Sweeping all pages in a heap and move empty pages from the
|
|
heap to tomb_heap.
|
|
(2) Check all existing pages and free a page
|
|
if all slots of this page are empty and
|
|
there is enough empty slots (checking by swept_num)
|
|
|
|
To introduce this patch, there are several tuning of GC parameters.
|
|
|
|
Wed Oct 23 14:20:56 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_prof_sweep_timer_stop): catch up recent changes
|
|
to compile on GC_PROFILE_MORE_DETAIL=1.
|
|
|
|
Wed Oct 23 11:43:27 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* file.c: [DOC] fix rdoc format of File#expand_path from r43386
|
|
|
|
Tue Oct 22 21:58:28 2013 URABE Shyouhei <shyouhei@ruby-lang.org>
|
|
|
|
* vm_core.h (enum): avoid syntax error.
|
|
|
|
* method.h: ditto.
|
|
|
|
* internal.h: ditto.
|
|
|
|
Tue Oct 22 19:53:16 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (Init_heap): move logics from heap_pages_init() and remove
|
|
heap_pages_init().
|
|
|
|
Tue Oct 22 19:19:05 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: allow multiple heaps.
|
|
Now, objects are managed by page. And a set of pages is called heap.
|
|
This commit supports multiple heaps in the object space.
|
|
|
|
* Functions heap_* and rb_heap_t manages heap data structure.
|
|
* Functions heap_page_* and struct heap_page manage page data
|
|
structure.
|
|
* Functions heap_pages_* and struct rb_objspace_t::heap_pages
|
|
maintains all pages.
|
|
For example, pages are allocated from the heap_pages.
|
|
|
|
See https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/GC_design
|
|
and https://bugs.ruby-lang.org/attachments/4015/data-heap_structure_with_multiple_heaps.png
|
|
for more details.
|
|
|
|
Now, there is only one heap called `eden', which is a space for all
|
|
new generated objects.
|
|
|
|
Tue Oct 22 18:26:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/pp.rb (object_address_group): Use Kernel#to_s to obtain the class
|
|
name and object address.
|
|
This fix a problem caused by %p in C generates variable length
|
|
address.
|
|
Reported by ko1 via IRC.
|
|
|
|
Tue Oct 22 16:57:48 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* file.c (File#expand_path): [DOC] improve documentation of File#expand_path.
|
|
Based on patch by Prathamesh Sonpatki. [ruby-core:57734] [Bug #9002]
|
|
|
|
Tue Oct 22 15:59:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* dir.c (glob_helper): don't skip current directories if FNM_DOTMATCH
|
|
is given. [ruby-core:53108] [Bug #8006]
|
|
|
|
Tue Oct 22 14:53:11 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_trace.c: exterminate Zombies.
|
|
There is a bug that T_ZOMBIE objects are not collected.
|
|
Because there is a pass to miss finalizer postponed job
|
|
with multi-threading. This patch solve this issue.
|
|
|
|
* vm_trace.c (rb_postponed_job_register_one): set
|
|
RUBY_VM_SET_POSTPONED_JOB_INTERRUPT(th) if another same job
|
|
is registered.
|
|
There is a possibility to remain a postponed job without
|
|
interrupt flag.
|
|
|
|
* vm_trace.c (rb_postponed_job_register_one): check interrupt
|
|
carefully.
|
|
|
|
* vm_trace.c (rb_postponed_job_register_one): use additional space
|
|
to avoid buffer full.
|
|
|
|
* gc.c (gc_finalize_deferred_register): check failure.
|
|
|
|
* thread.c (rb_threadptr_execute_interrupts): check
|
|
`postponed_job_interrupt' immediately. There is a possibility
|
|
to miss this flag.
|
|
|
|
Tue Oct 22 12:11:16 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in: check if the given CFLAGS and LDFLAGS are working, and
|
|
bail out early if not.
|
|
|
|
Tue Oct 22 00:06:57 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* file.c (rb_file_exists_p): warn deprecated name. [Bug #9041]
|
|
|
|
Mon Oct 21 23:57:53 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* encoding.c (load_encoding): should preserve outer errinfo, so that
|
|
expected exception may not be lost. [ruby-core:57949] [Bug #9038]
|
|
|
|
Sun Oct 20 15:41:22 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* io.c (rb_io_reopen): create a new, temporary FD via rb_sysopen and
|
|
call rb_cloexec_dup2 on it to atomically replace the file fptr->fd
|
|
points to. This leaves no possible window where fptr->fd is invalid
|
|
to userspace (even for any threads running w/o GVL). based on the
|
|
patch by Eric Wong <normalperson@yhbt.net> at [ruby-core:57943].
|
|
[Bug #9036]
|
|
|
|
Sun Oct 20 15:29:05 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* error.c (rb_syserr_fail_path_in): new function split from
|
|
rb_sys_fail_path_in to raise SystemCallError without errno.
|
|
|
|
* internal.h (rb_syserr_fail_path): like rb_sys_fail_path but without
|
|
errno.
|
|
|
|
Sun Oct 20 13:58:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h (rb_obj_wb_unprotect, rb_obj_written),
|
|
(rb_obj_write): suppress unused-parameter warnings.
|
|
|
|
Sun Oct 20 10:32:48 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update RubyGems to master 0886307. This commit
|
|
improves documentation and should bring ruby above 75% documented on
|
|
rubyci.
|
|
|
|
Sun Oct 20 09:30:56 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master 3de7e0f. Changes:
|
|
|
|
Only attempt to build extensions for newly-installed gems. This
|
|
prevents compilation attempts at gem activation time for gems that
|
|
already have extensions built.
|
|
|
|
Fix crash in the dependency resolver for dependencies that cannot be
|
|
resolved.
|
|
|
|
* test/rubygems: ditto.
|
|
|
|
Sun Oct 20 05:24:29 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* variable.c (rb_class2name): should return real class name, not
|
|
singleton class or iclass.
|
|
|
|
Sun Oct 20 04:18:48 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* variable.c (rb_class2name): call rb_tmp_class_path() directly to
|
|
avoid extra rb_str_dup() from rb_class_name().
|
|
|
|
Sat Oct 19 19:59:02 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/file.c (code_page): use simple array instead of st_table.
|
|
|
|
* encoding.c (rb_locale_encindex): defer initialization of win32 code
|
|
page table until encoding db loaded.
|
|
|
|
Sat Oct 19 08:25:05 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: fix rb_objspace_t.
|
|
* make "struct heap" and move most of variables
|
|
in rb_objspace_t::heap.
|
|
* rename rb_objspace_t::heap::sorted to
|
|
rb_objspace_t::heap_sorted_pages
|
|
and make a macro heap_sorted_pages.
|
|
* rename rb_objspace_t::heap::range to
|
|
rb_objspace_t::heap_range and rename macros
|
|
lomem/himem to heap_lomem/heap_himem.
|
|
|
|
Sat Oct 19 07:14:40 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master 42543b6. Changes:
|
|
|
|
Fix `gem update` for gems with multiple platforms.
|
|
|
|
* test/rubygems: ditto.
|
|
|
|
Sat Oct 19 06:55:52 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master 0a3814b. Changes:
|
|
|
|
Fixed extension directory in Gem::Specification#require_paths.
|
|
|
|
Allow installation of gems when $HOME is nonexistent or unwritable.
|
|
|
|
Use proper API in InstallCommand.
|
|
|
|
Improve support for path option in gem dependency files.
|
|
|
|
Remove warnings.
|
|
|
|
* test/rubygems: ditto.
|
|
|
|
Fri Oct 18 15:23:34 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: change terminology of heap.
|
|
Change "slot" to "page". "Slot" is a space of RVALUE.
|
|
1. "Heap" consists of a set of "heap_page"s (pages).
|
|
2. Each "heap_page" has "heap_page_body".
|
|
3. "heap_page_body" has RVALUE (a.k.a. "slot") spaces.
|
|
4. "sorted" is a sorted array of "heap_page"s, sorted
|
|
by address of heap_page_body (for "is_pointer_to_heap").
|
|
|
|
See https://bugs.ruby-lang.org/attachments/4008/data-heap_structure.png.
|
|
|
|
Fri Oct 18 09:40:43 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master cee6788. Changes:
|
|
|
|
Fix test failure on vc10-x64 Server on rubyci.org due to attempting
|
|
to File.chmod where it is not supported.
|
|
|
|
Continuing work on improved gem dependencies file (Gemfile) support.
|
|
|
|
* test: ditto.
|
|
|
|
Fri Oct 18 06:02:49 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master f738c67. Changes:
|
|
|
|
Fixed test bug for ruby with ENABLE_SHARED = no
|
|
|
|
* test/rubygems: ditto.
|
|
|
|
Fri Oct 18 00:57:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/tsort.rb (TSort.tsort): Extracted from TSort#tsort.
|
|
(TSort.tsort_each): Extracted from TSort#tsort_each.
|
|
(TSort.strongly_connected_components): Extracted from
|
|
TSort#strongly_connected_components.
|
|
(TSort.each_strongly_connected_component): Extracted from
|
|
TSort#each_strongly_connected_component.
|
|
|
|
Thu Oct 17 18:50:08 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (CALC_EXACT_MALLOC_SIZE_CHECK_OLD_SIZE): introduced.
|
|
This macro enable checker compare with allocated memory and
|
|
declared old_size of sized_xfree and sized_xrealloc.
|
|
|
|
Thu Oct 17 18:45:41 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* string.c (STR_HEAP_SIZE): includes TERM_LEN(str).
|
|
|
|
* string.c (rb_str_memsize): use STR_HEAP_SIZE().
|
|
|
|
Thu Oct 17 17:43:00 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* vm_insnhelper.c (vm_call_method): set ci->me to 0 when the
|
|
original method of a refined method is undef to avoid SEGV.
|
|
|
|
* vm_method.c (rb_method_entry_without_refinements): return 0 when
|
|
the original method of a refined method is undef to avoid SEGV.
|
|
|
|
* test/ruby/test_refinement.rb: related test.
|
|
|
|
Thu Oct 17 17:38:36 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c, internal.h: rename ruby_xsizefree/realloc to
|
|
rb_sized_free/realloc.
|
|
|
|
* array.c: catch up these changes.
|
|
|
|
* string.c: ditto.
|
|
|
|
Thu Oct 17 17:32:51 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c, string.c: use ruby_xsizedfree() and ruby_xsizedrealloc().
|
|
|
|
* internal.h (SIZED_REALLOC_N): define a macro as REALLOC_N().
|
|
|
|
Thu Oct 17 17:11:17 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (console_emulator_p): check by comparison between
|
|
module handle of WriteConsoleW and kernel32.dll.
|
|
|
|
* configure.in, win32/Makefile.sub, win32/setup.mak: no longer need
|
|
psapi.lib.
|
|
|
|
Thu Oct 17 16:53:30 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c, internal.h: add new internal memory management functions.
|
|
* void *ruby_xsizedrealloc(void *ptr, size_t new_size, size_t old_size)
|
|
* void ruby_xsizedfree(void *x, size_t size)
|
|
These functions accept additional size parameter to calculate more
|
|
accurate malloc_increase parameter which control GC timing.
|
|
[Feature #8985]
|
|
|
|
Thu Oct 17 14:21:34 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/file.c (rb_file_expand_path_internal): fix memory leaks at
|
|
a non-absolute home exception.
|
|
|
|
Thu Oct 17 14:06:39 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/object_tracing.c (newobj_i): fix memory leak.
|
|
There is possibility to remain info due to missing FREEOBJ event.
|
|
FREEOBJ events are skipped while suppress_tracing state, for example,
|
|
during trace events are invoking.
|
|
|
|
Thu Oct 17 12:30:16 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/tsort.rb (TSort.each_strongly_connected_component_from):
|
|
Extracted from TSort#each_strongly_connected_component_from.
|
|
|
|
Thu Oct 17 11:07:06 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master 941c21a. Changes:
|
|
|
|
Restored method bundler wants to remove for compatibility.
|
|
|
|
Improvements to Gemfile compatibility.
|
|
|
|
* test/rubygems: ditto.
|
|
|
|
Thu Oct 17 08:08:11 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/object_tracing.c (newobj_i): add workaround.
|
|
some bugs hits this check.
|
|
|
|
* ext/objspace/object_tracing.c (object_allocations_reporter_i): cast as pointer.
|
|
|
|
Thu Oct 17 07:36:53 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master 2abce58. Changes:
|
|
|
|
Fixed documentation generation when sdoc and json are installed as
|
|
gems.
|
|
|
|
Added some missing documentation.
|
|
|
|
Thu Oct 17 07:10:26 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/curses/curses.c: [DOC] Cleaned up formatting consistency of rdoc
|
|
comments for Curses, including period spacing and column width.
|
|
|
|
This patch also fixed some typos. Thanks to @postmodern for the patch!
|
|
[Fixes GH-420] https://github.com/ruby/ruby/pull/420
|
|
|
|
Thu Oct 17 06:58:42 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/date/date_core.c: [DOC] plural grammar fixed by @scott113341
|
|
Contributed via documenting-ruby.org: documenting-ruby/ruby#16
|
|
https://github.com/documenting-ruby/ruby/pull/16
|
|
|
|
Thu Oct 17 05:52:31 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/io/nonblock/nonblock.c: [DOC] Document io/nonblock by reprah
|
|
[Fixes GH-418] https://github.com/ruby/ruby/pull/418 based on the
|
|
original discussion from documenting-ruby/ruby#18
|
|
|
|
Thu Oct 17 05:40:33 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (objspace_each_objects): do not skip empty RVALUEs.
|
|
|
|
Thu Oct 17 05:31:31 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* error.c (rb_bug_reporter_add): return simply 0 if failed.
|
|
Please check return value.
|
|
|
|
Thu Oct 17 05:17:33 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/object_tracing.c: add new method
|
|
ObjectSpace.trace_object_allocations_debug_start for GC debugging.
|
|
If you encounter the BUG "... is T_NONE" (and so on) on your
|
|
application, please try this method at the beginning of your app.
|
|
|
|
Wed Oct 16 22:35:27 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/io/nonblock/nonblock.c: use rb_cIO instead of VALUE
|
|
|
|
Wed Oct 16 17:45:13 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* bootstraptest/runner.rb: check nil before calling `signal?'
|
|
for a process status.
|
|
|
|
Wed Oct 16 17:37:17 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* error.c, internal.h (rb_bug_reporter_add): add a new C-API.
|
|
rb_bug_reporter_add() allows to register a function which
|
|
is called at rb_bug() called.
|
|
|
|
* ext/-test-/bug_reporter/bug_reporter.c: add a test for this C-API.
|
|
|
|
* ext/-test-/bug_reporter/extconf.rb: ditto.
|
|
|
|
* test/-ext-/bug_reporter/test_bug_reporter.rb: ditto.
|
|
|
|
Wed Oct 16 15:14:21 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* NEWS: add a line into NEWS for last commit.
|
|
|
|
Wed Oct 16 15:09:14 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/objspace.c: add a new method `reachable_objects_from_root'.
|
|
ObjectSpace.reachable_objects_from_root returns all objects referred
|
|
from root (called "root objects").
|
|
This feature is for deep object analysis.
|
|
|
|
* test/objspace/test_objspace.rb: add a test.
|
|
|
|
Wed Oct 16 15:00:21 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master b955554. Changes:
|
|
|
|
Fixed NameError for Gem::Ext due to re-entering file lookup in
|
|
RubyGems' overridden require. Bug by Koichi Sasada.
|
|
|
|
Fixed possible circular require warning in tests.
|
|
|
|
Used existing constant for `gem install -g` dependency file list.
|
|
|
|
* test/rubygems: ditto.
|
|
|
|
Wed Oct 16 09:42:42 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master 278d00d. Changes:
|
|
|
|
Fixes building extensions without a "clean" make rule
|
|
|
|
Adds gem dependency file autodetection to "gem install -g"
|
|
|
|
* test/rubygems: Tests for the above.
|
|
|
|
Wed Oct 16 09:12:23 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems master commit 2a74263. This fixes
|
|
several bugs in RubyGems 2.2.0.preview.1.
|
|
|
|
* test/rubygems: ditto.
|
|
|
|
Wed Oct 16 07:25:02 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* gc.c (gc_mark_roots): rename roots to be categories
|
|
instead of function names.
|
|
|
|
Tue Oct 15 19:18:13 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.h (rb_objspace_reachable_objects_from_root): added.
|
|
This API provides information which objects are root objects.
|
|
`category' shows what kind of root objects.
|
|
|
|
* gc.c (gc_mark_roots): separate from gc_marks_body().
|
|
|
|
Tue Oct 15 17:47:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c: Fix a typo. MacOS X doesn't have ENOTSUPP.
|
|
|
|
Mon Oct 14 12:32:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ruby.c (process_options): load statically linked extensions before
|
|
rubygems, because of ext/thread.
|
|
|
|
* ruby.c (process_options): use gem_prelude instead of requiring
|
|
rubygems directly when --enable=gems is given.
|
|
|
|
* Makefile.in (DEFAULT_PRELUDES): always use gem_prelude regardless of
|
|
--disable-rubygems.
|
|
|
|
Mon Oct 14 11:07:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (have_framework): should append framework options to
|
|
$LIBS, not $LDFLAGS. The former is propagated to exts.mk when
|
|
enable-static-linked-ext.
|
|
|
|
* lib/mkmf.rb (create_makefile): ranlib on static library, not DLLIB.
|
|
|
|
Sun Oct 13 23:53:40 2013 Andrew Grimm <andrew.j.grimm@gmail.com>
|
|
|
|
* vsnprintf.c: Fix spelling from compliment to complement.
|
|
Patch by @agrimm.
|
|
|
|
* include/ruby/ruby.h: ditto
|
|
|
|
Sun Oct 13 20:59:27 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm.c (Init_BareVM): initialize defined_module_hash here,
|
|
Init_top_self() is too late to register core classes/modules.
|
|
|
|
* compile.c (compile_array_): no hash to merge if it is empty.
|
|
|
|
* vm.c (m_core_hash_merge_kwd): just check keys if only one argument
|
|
is given, without merging.
|
|
|
|
Sat Oct 12 06:35:01 2013-10-11 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rake: Update to rake 10.1.0
|
|
* bin/rake: ditto.
|
|
* test/rake: ditto.
|
|
|
|
* NEWS: Update NEWS to include rake 10.1.0 and links to release notes.
|
|
|
|
Sat Oct 12 03:26:04 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* class.c, variable.c, gc.c (rb_class_tbl): removed.
|
|
|
|
* vm.c, vm_core.h (rb_vm_add_root_module): added to register as a
|
|
defined root module or class.
|
|
This guard helps mark miss from defined classes/modules they are
|
|
only referred from C's global variables in C-exts.
|
|
Basically, it is extension's bug.
|
|
Register to hash object VM has.
|
|
Marking a hash objects allows generational GC supports.
|
|
|
|
* gc.c (RGENGC_PRINT_TICK): disable (revert).
|
|
|
|
Sat Oct 12 03:24:49 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_method.c (rb_gc_mark_unlinked_live_method_entries):
|
|
revert last commit to introduce debug prints.
|
|
|
|
Fri Oct 11 21:05:19 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* internal.h, parse.y: use `full_mark' instead of `full_marking'.
|
|
|
|
Fri Oct 11 20:58:16 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: use terminology `full_mark' instead of `minor_gc'
|
|
in mark functions.
|
|
|
|
Fri Oct 11 20:46:09 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: use __GNUC__ instead of __GCC__.
|
|
|
|
Fri Oct 11 20:35:59 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c, parse.y: support generational Symbol related marking.
|
|
Each symbols has String objects respectively to represent
|
|
Symbols.
|
|
These objects are marked only when:
|
|
* full marking
|
|
* new symbols are added
|
|
This hack reduce symbols (related strings) marking time.
|
|
For example, on my Linux environment, the following code
|
|
"20_000_000.times{''}"
|
|
with 40k symbols (similar symbol number on Rails 3.2.14 app,
|
|
@jugyo tells me) boosts, from 7.3sec to 4.2sec.
|
|
|
|
* internal.h: change prototype of rb_gc_mark_symbols().
|
|
|
|
Fri Oct 11 19:27:22 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el: Import ruby-electric.el 2.0.1 which fixes
|
|
a bug and a flaw with auto-end introduced in the revamp.
|
|
|
|
* ruby-forward-sexp is inappropriate here because it moves the
|
|
cursor past the keyword.
|
|
|
|
* Fix a reversed looking-back check in
|
|
ruby-electric--block-beg-keyword-at-point-p.
|
|
|
|
* Do not add end again if space or return is hit repeatedly
|
|
after a block beginning keyword.
|
|
|
|
Fri Oct 11 18:12:47 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/gc_hook.c: prohibit reentrant.
|
|
|
|
Fri Oct 11 18:11:34 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_trace.c (rb_postponed_job_flush): fix bit operation.
|
|
|
|
Fri Oct 11 17:33:24 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el: Import ruby-electric.el 2.0 from
|
|
https://github.com/knu/ruby-electric.el which integrates changes
|
|
from another fork by @qoobaa.
|
|
|
|
* Allow ruby-electric-mode to be disabled by introducing a
|
|
dedicated key map. Electric key bindings are now defined in
|
|
ruby-electric-mode-map instead of overwriting ruby-mode-map.
|
|
|
|
* Add ruby-electric-mode-hook.
|
|
|
|
* Use a remap in binding ruby-electric-delete-backward-char.
|
|
|
|
* Totally revamp electric keywords and then introduce electric
|
|
return. Modifier keywords are now properly detected making
|
|
use of ruby-mode's indentation level calculator, and
|
|
|
|
* block-mid keywords (then, else, elsif, when, rescue and
|
|
ensure) also become electric with automatic reindentation.
|
|
|
|
* Add standardized comments for ELPA integration.
|
|
|
|
* Fix interaction with smartparens-mode by disabling its end
|
|
keyword completion, since ruby-electric has become more clever
|
|
at it.
|
|
|
|
* The custom variable `ruby-electric-keywords` is changed to
|
|
`ruby-electric-keywords-alist`, allowing user to fine-grained
|
|
configuration.
|
|
|
|
Fri Oct 11 16:53:28 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_trace.c (rb_postponed_job_flush): simplify.
|
|
|
|
Fri Oct 11 03:36:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* thread.c (rb_threadptr_execute_interrupts): flush postponed job only
|
|
once at last.
|
|
|
|
* vm_trace.c (rb_postponed_job_flush): defer calling postponed jobs
|
|
registered while flushing to get rid of infinite reentrance of
|
|
ObjectSpace.after_gc_start_hook. [ruby-dev:47400] [Bug #8492]
|
|
|
|
Thu Oct 10 23:04:00 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_or): remove unused variables.
|
|
|
|
Thu Oct 10 23:01:16 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_or): use rb_hash_keys().
|
|
|
|
Thu Oct 10 21:36:16 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_compact_bang): use ary_resize_smaller().
|
|
|
|
Thu Oct 10 17:25:28 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm.c (vm_exec): support :b_return event for "lambda{return}.call".
|
|
[Bug #8622]
|
|
|
|
* test/ruby/test_settracefunc.rb: add a test.
|
|
|
|
Thu Oct 10 13:52:37 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_trace.c (postponed_job): use preallocated buffer.
|
|
Pre-allocate MAX_POSTPONED_JOB (1024) sized buffer
|
|
and use it.
|
|
If rb_postponed_job_register() cause overflow, simply it
|
|
fails and returns 0.
|
|
And maybe rb_postponed_job_register() is signal safe.
|
|
|
|
* vm_core.h: change data structure.
|
|
|
|
Thu Oct 10 11:11:33 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm.c (Init_VM): hide also the singleton class of frozen-core, not
|
|
only frozen-core itself.
|
|
|
|
Thu Oct 10 06:02:08 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/ruby/test_rand.rb: fix r43224. local variable `e' is
|
|
no longer available.
|
|
|
|
Thu Oct 10 00:02:35 2013 Yusuke Endoh <mame@tsg.ne.jp>
|
|
|
|
* numeric.c (fix_aref): avoid a possible undefined behavior.
|
|
1L << 63 on 64-bit platform is undefined, at least, according to
|
|
ISO/IEC 9899 (C99) 6.5.7.
|
|
|
|
Wed Oct 9 23:57:02 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* object.c (id_for_attr): avoid inadvertent symbol creation.
|
|
|
|
Wed Oct 9 18:03:01 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_method.c (rb_attr): preserve encoding of the attribute ID in
|
|
error message.
|
|
|
|
Wed Oct 9 17:40:16 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_fstring): because of lazy sweep, str may be unmarked
|
|
already and swept at next time, so mark it for the time being.
|
|
[ruby-core:57756]
|
|
|
|
Wed Oct 9 13:53:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* compar.c (cmp_eq): fail if recursion. [ruby-core:57736] [Bug #9003]
|
|
|
|
* thread.c (rb_exec_recursive_paired_outer): new function which is
|
|
combination of paired and outer variants.
|
|
|
|
Wed Oct 9 09:18:14 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/debug.h,
|
|
vm_backtrace.c (rb_profile_frame_full_label): add new C API
|
|
rb_profile_frame_full_label() which returns label with
|
|
qualified method name.
|
|
Note that in future version of Ruby label() may return
|
|
same return value of full_label().
|
|
|
|
* ext/-test-/debug/profile_frames.c,
|
|
test/-ext-/debug/test_profile_frames.rb: fix a test for this change.
|
|
|
|
|
|
Wed Oct 9 00:55:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* load.c (load_lock): display backtrace to $stderr at circular
|
|
require.
|
|
|
|
* vm_backtrace.c (rb_backtrace_print_to): new function to print
|
|
backtrace to the given output.
|
|
|
|
Tue Oct 8 21:03:35 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_backtrace.c, include/ruby/debug.h: add new APIs
|
|
* VALUE rb_profile_frame_method_name(VALUE frame)
|
|
* VALUE rb_profile_frame_qualified_method_name(VALUE frame)
|
|
|
|
* iseq.c (rb_iseq_klass), internal.h: add new internal function
|
|
rb_iseq_method_name().
|
|
|
|
* ext/-test-/debug/profile_frames.c (profile_frames),
|
|
test/-ext-/debug/test_profile_frames.rb: add a test.
|
|
|
|
Tue Oct 8 16:11:11 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* array.c (rb_ary_uniq): use rb_hash_values(), as well as the case no
|
|
block is given.
|
|
|
|
* internal.h: define rb_hash_values() as internal API.
|
|
|
|
Tue Oct 8 13:53:21 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_uniq): use rb_hash_keys().
|
|
|
|
* internal.h: define rb_hash_keys() as internal API.
|
|
|
|
* hash.c (rb_hash_keys): ditto.
|
|
|
|
Tue Oct 8 10:56:39 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* cont.c: disable FIBER_USE_NATIVE on GNU/Hurd because it doesn't
|
|
support a combination getcontext() and threads. Patch by
|
|
Gabriele Giacone (1o5g4r8o@gmail.com). [Bug #8990][ruby-core:57685]
|
|
|
|
Tue Oct 8 05:58:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/time.rb (Time.strptime): Time.strptime('0', '%s') returns local
|
|
time Time object as Ruby 2.0 and before.
|
|
|
|
Tue Oct 8 05:40:37 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* .travis.yml: Rebuild Travis CI's "ruby-head" version on successful
|
|
build. Patch by Konstantin Haase. [Fixes GH-417]
|
|
https://github.com/ruby/ruby/pull/417
|
|
|
|
Tue Oct 8 04:28:25 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-mode.el: Use preceding-char/following-char
|
|
(returning 0 at BOF/EOF) instead of char-before/char-after
|
|
(returning nil at BOF/EOF) to avoid error from char-syntax when
|
|
at BOF/EOF.
|
|
|
|
Tue Oct 8 04:12:45 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-additional.el (ruby-mode-set-encoding): Add a missing
|
|
else clause to unbreak with `cp932`, etc.
|
|
|
|
* misc/ruby-mode.el (ruby-mode-set-encoding): Ditto.
|
|
|
|
Tue Oct 8 03:57:34 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-additional.el (ruby-mode-set-encoding): Use
|
|
`default-buffer-file-coding-system` if the :prefer-utf-8
|
|
property is not available.
|
|
|
|
* misc/ruby-mode.el (ruby-mode-set-encoding): Ditto.
|
|
|
|
* misc/ruby-additional.el (ruby-encoding-map): Override the
|
|
default value.
|
|
|
|
Tue Oct 8 03:19:19 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-additional.el (ruby-mode-set-encoding): Add support
|
|
for `prefer-utf-8` which was introduced in Emacs trunk.
|
|
|
|
* misc/ruby-additional.el (ruby-encoding-map): Add a mapping from
|
|
`japanese-cp932` to `cp932` to fix the problem where saving a
|
|
source file written in Shift_JIS twice would end up having
|
|
`coding: japanese-cp932` which Ruby could not recognize.
|
|
|
|
* misc/ruby-additional.el (ruby-mode-set-encoding): Add support
|
|
for encodings mapped to nil in `ruby-encoding-map`.
|
|
|
|
* misc/ruby-additional.el (ruby-encoding-map): Map `us-ascii` and
|
|
`utf-8` to nil by default, meaning they need not be explicitly
|
|
declared in magic comment.
|
|
|
|
* misc/ruby-additional.el (ruby-encoding-map): Add type
|
|
declaration for better customize UI.
|
|
|
|
* misc/ruby-mode.el: Ditto for the above.
|
|
|
|
Tue Oct 8 00:14:53 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-additional.el: Add a standard header and footer,
|
|
including (provide 'ruby-additional).
|
|
|
|
Mon Oct 7 22:52:45 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el (ruby-electric-space-can-be-expanded-p):
|
|
Return nil to avoid "end" insertion when in smartparens-mode
|
|
that is configured to insert "end" for the same keyword.
|
|
|
|
* misc/ruby-electric.el (ruby-electric-keywords): New custom
|
|
variable to replace `ruby-electric-simple-keywords-re` with.
|
|
|
|
Mon Oct 7 22:52:16 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-additional.el: Use preceding-char/following-char
|
|
(returning 0 at BOF/EOF) instead of char-before/char-after
|
|
(returning nil at BOF/EOF) to avoid error from char-syntax when
|
|
at BOF/EOF.
|
|
|
|
Mon Oct 7 22:45:20 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* cont.c (FIBER_USE_NATIVE): split long conditions.
|
|
|
|
Mon Oct 7 20:29:31 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/time.rb: [DOC] typo in Time.rb overview by @srt32 [Fixes GH-416]
|
|
https://github.com/ruby/ruby/pull/416
|
|
|
|
Mon Oct 7 20:07:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/time.rb (Time.strptime): Use :offset.
|
|
Patch by Felipe Contreras. [ruby-core:57694]
|
|
|
|
Mon Oct 7 16:47:27 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/-ext-/debug/test_profile_frames.rb: rename class C to
|
|
something long name because one test depends on absence of
|
|
class ::C.
|
|
|
|
Mon Oct 7 16:33:10 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/-test-/debug/profile_frames.c:
|
|
test/-ext-/debug/test_profile_frames.rb: add a test for new C-APIs.
|
|
|
|
Mon Oct 7 16:12:36 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/debug.h: add backtrace collecting APIs for profiler.
|
|
* int rb_profile_frames(int start, int limit, VALUE *buff, int *lines);
|
|
Collect information of frame information.
|
|
|
|
* VALUE rb_profile_frame_path(VALUE frame);
|
|
* VALUE rb_profile_frame_absolute_path(VALUE frame);
|
|
* VALUE rb_profile_frame_label(VALUE frame);
|
|
* VALUE rb_profile_frame_base_label(VALUE frame);
|
|
* VALUE rb_profile_frame_first_lineno(VALUE frame);
|
|
* VALUE rb_profile_frame_classpath(VALUE frame);
|
|
* VALUE rb_profile_frame_singleton_method_p(VALUE frame);
|
|
Get information about each frame.
|
|
|
|
These APIs are designed for profilers, for example, no object allocation,
|
|
and enough information for profilers.
|
|
In this version, this API collects only Ruby level frames.
|
|
This issue will be fixed after Ruby 2.1.
|
|
|
|
* vm_backtrace.c: implement above APIs.
|
|
|
|
* iseq.c (rb_iseq_klass): return local_iseq's class.
|
|
|
|
Mon Oct 7 14:26:01 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* proc.c: catch up last commit.
|
|
Type of return value of rb_iseq_first_lineno() is now VALUE.
|
|
|
|
* vm_insnhelper.c (argument_error): ditto.
|
|
|
|
* vm_method.c (rb_method_entry_make): ditto.
|
|
|
|
Mon Oct 7 14:07:45 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* iseq.c, internal.h: change to public (but internal) functions
|
|
* VALUE rb_iseq_path(VALUE iseqval);
|
|
* VALUE rb_iseq_absolute_path(VALUE iseqval);
|
|
* VALUE rb_iseq_label(VALUE iseqval);
|
|
* VALUE rb_iseq_base_label(VALUE iseqval);
|
|
* VALUE rb_iseq_first_lineno(VALUE iseqval);
|
|
And new (temporary) function:
|
|
* VALUE rb_iseq_klass(VALUE iseqval);
|
|
|
|
* iseq.c. vm_core.h (int rb_iseq_first_lineno): remove
|
|
function `int rb_iseq_first_lineno(const rb_iseq_t *iseq)'.
|
|
Use `VALUE rb_iseq_first_lineno(VALUE iseqval)' instead.
|
|
|
|
* proc.c. vm_insnhelper.c, vm_method.c: catch up this change.
|
|
|
|
Sun Oct 6 08:37:39 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/webrick.rb: [DOC] fix grammar in WEBrick overview [Fixes GH-413]
|
|
Based on patch by @chastell https://github.com/ruby/ruby/pull/413
|
|
|
|
Sat Oct 5 11:21:01 2013 Aaron Pfeifer <aaron.pfeifer@gmail.com>
|
|
|
|
* thread.c (terminate_atfork_i): fix locking mutexes not unlocked in
|
|
forks when not tracked in thread. [ruby-core:55102] [Bug #8433]
|
|
|
|
Fri Oct 4 19:54:09 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/dbm/dbm.c: [DOC] Fix wrong constant name in DBM by @edward
|
|
[Fixes GH-409] https://github.com/ruby/ruby/pull/409
|
|
|
|
Fri Oct 4 19:49:42 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* gc.c: rename heap.free_num as heap.swept_num to clarify meaning and
|
|
avoid confusion with objspace_free_num().
|
|
|
|
Fri Oct 4 19:02:01 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* gc.c (objspace_free_num): new method for available/free slots on
|
|
heap. [ruby-core:57633] [Bug #8983]
|
|
* gc.c (gc_stat): change heap_free_num definition to use new method.
|
|
* test/ruby/test_gc.rb: test for above.
|
|
|
|
Fri Oct 4 18:53:42 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* gc.c: add rb_objspace.limit to keep accurate count of total heap
|
|
slots [ruby-core:57633] [Bug #8983]
|
|
|
|
Fri Oct 4 09:32:33 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/csv.rb (CSV.foreach): support enumerator. based on a patch by
|
|
Hanmac (Hans Mackowiak) at [ruby-core:57643]. [ruby-core:57283]
|
|
[Feature #8929]
|
|
|
|
Thu Oct 3 18:20:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (console_emulator_p, constat_handle): disable built-in
|
|
console colorizing when console-emulator-like DLL is injected.
|
|
[Feature #8201]
|
|
|
|
Thu Oct 3 18:01:44 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: define gc_profile_record::allocated_size if
|
|
CALC_EXACT_MALLOC_SIZE is true.
|
|
|
|
Thu Oct 3 13:42:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* common.mk (yes-test-sample): use RUNRUBY instead of MINIRUBY to set
|
|
runtime library path and run the built ruby. [Bug #8971]
|
|
|
|
Thu Oct 3 00:17:15 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-additional.el: Properly quote the body. An unquoted
|
|
body given to eval-after-load is evaluated immediately!
|
|
|
|
Wed Oct 2 21:38:30 2013 Yusuke Endoh <mame@tsg.ne.jp>
|
|
|
|
* ext/socket/ifaddr.c (rsock_getifaddrs): fix possible memory leak.
|
|
When a system had no interface, this function used xmalloc for root
|
|
but did not return any reference to it. This patch fixes it by
|
|
immediately returning an empty array if no interface is found.
|
|
Coverity Scan found this bug.
|
|
|
|
Wed Oct 2 21:37:04 2013 Yusuke Endoh <mame@tsg.ne.jp>
|
|
|
|
* random.c (make_seed_value): a local array declaration was accessed
|
|
out of scope. Coverity Scan found this bug.
|
|
|
|
Wed Oct 2 18:52:40 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: relax GC condition due to malloc_limit.
|
|
|
|
* gc.c (GC_MALLOC_LIMIT_MAX): change default value
|
|
(256MB -> 512MB) and permit zero to ignore max value.
|
|
|
|
* gc.c (vm_malloc_increase, vm_xrealloc): do not cause GC on realloc.
|
|
|
|
* gc.c (gc_before_sweep): change debug messages.
|
|
|
|
Wed Oct 2 16:26:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* io.c (rb_io_close_read): duplex IO should wait its child process
|
|
even after close_read.
|
|
|
|
Wed Oct 2 15:39:13 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* vm_core.h: use __has_attribute() instead of __clang__major__ because
|
|
clang says "Note that marketing version numbers should not be used
|
|
to check for language features, as different vendors use different
|
|
numbering schemes. Instead, use the Feature Checking Macros."
|
|
http://clang.llvm.org/docs/LanguageExtensions.html
|
|
|
|
Wed Oct 2 14:19:57 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* io.c (rb_io_close_write): detach tied IO for writing before closing
|
|
to get rid of race condition. [ruby-list:49598]
|
|
|
|
* io.c (rb_io_close_read): keep fptr in write_io to be discarded, to
|
|
fix freed pointer access when it is in use by other threads, and get
|
|
rid of potential memory/fd leak.
|
|
|
|
Tue Oct 1 23:44:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* vm_core.h: use __attribute__((unused)) in UNINITIALIZED_VAR on clang
|
|
4.0+ instead of just on 4.2. Clang has supported the unused attribute
|
|
since before version 4, so this should be safe.
|
|
|
|
Tue Oct 1 22:03:48 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/tempfile.rb (Tempfile#unlink): finalizer is no longer needed
|
|
after unlinking. patched by by normalperson (Eric Wong) at
|
|
[ruby-core:56521] [Bug #8768]
|
|
|
|
Tue Oct 1 20:54:33 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* file.c (stat_new_0): constify.
|
|
|
|
* file.c (rb_stat_new): constify and export. based on a patch by
|
|
Hanmac (Hans Mackowiak) at [ruby-core:53225]. [Feature #8050]
|
|
|
|
Tue Oct 1 16:03:42 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h (ruby_safe_level_4_warning): needed by extension
|
|
libraries which check safe level 4. [ruby-dev:47517] [Bug #8652]
|
|
|
|
Mon Sep 30 23:14:36 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/objspace/objspace.c: [DOC] Cleaned up many rdoc formatting
|
|
issues and several duplicate grammar bugs.
|
|
|
|
Mon Sep 30 23:01:01 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/objspace/object_tracing.c: [DOC] Adjust rdoc formatting and fix
|
|
small grammar typo
|
|
|
|
Mon Sep 30 17:28:39 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/object_tracing.c: [DOC] add some notes for
|
|
ObjectSpace::trace_object_allocations.
|
|
|
|
Mon Sep 30 16:46:58 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/object_tracing.c: add new 3 methods to control tracing.
|
|
* ObjectSpace::trace_object_allocations_start
|
|
* ObjectSpace::trace_object_allocations_stop
|
|
* ObjectSpace::trace_object_allocations_clear
|
|
And some refactoring.
|
|
|
|
* test/objspace/test_objspace.rb: add a test for new methods.
|
|
|
|
* NEWS: add a description for new methods.
|
|
|
|
Mon Sep 30 11:18:04 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_gc_disable): do rest_sweep() before disable GC.
|
|
This fix may solve a failure of
|
|
TestTracepointObj#test_tracks_objspace_events
|
|
[test/-ext-/tracepoint/test_tracepoint.rb:43].
|
|
|
|
Mon Sep 30 10:40:20 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* vm_method.c (rb_undef): raise a NameError if the original method
|
|
of a refined method is not defined.
|
|
|
|
* vm_insnhelper.c (rb_method_entry_eq): added NULL check to avoid SEGV.
|
|
|
|
* test/ruby/test_refinement.rb: related test.
|
|
|
|
Sun Sep 29 23:45:42 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (rb_id_attrset, intern_str): allow junk attrset ID for
|
|
Struct.
|
|
|
|
* parse.y (rb_id_attrset): fix inconsistency with literals, allow
|
|
ID_ATTRSET and return it itself, but ID_JUNK cannot make ID_ATTRSET.
|
|
and raise a NameError instead of rb_bug() for invalid argument.
|
|
|
|
Sun Sep 29 18:45:05 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* vm_insnhelper.c (vm_callee_setup_arg_complex, vm_yield_setup_block_args):
|
|
clear keyword arguments to prevent GC bug which occurs
|
|
while marking VM stack.
|
|
[ruby-dev:47729] [Bug #8964]
|
|
|
|
* test/ruby/test_keyword.rb: tests for the above.
|
|
|
|
Sat Sep 28 23:25:56 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* math.c (math_log, math_log2, math_log10): fix for Bignum argument.
|
|
numbits should be add only when right shifted.
|
|
|
|
Sat Sep 28 14:30:29 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* test/dl/test_base.rb: {libc, libm} detection now handle GNU/Hurd
|
|
correctly. Patch by Gabriele Giacone (1o5g4r8o@gmail.com).
|
|
[Bug #8937][ruby-core:57311]
|
|
* test/fiddle/helper.rb: ditto.
|
|
|
|
Sat Sep 28 00:19:41 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* ext/curses/extconf.rb: check the size of chtype.
|
|
|
|
* ext/curses/curses.c (NUM2CH, CH2NUM): use proper macros for
|
|
the size of chtype.
|
|
|
|
[ruby-core:56090] [Bug #8659]
|
|
|
|
Fri Sep 27 18:33:23 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: add two GC tuning environment variables.
|
|
RUBY_GC_MALLOC_LIMIT_MAX and RUBY_GC_MALLOC_LIMIT_GROWTH_FACTOR.
|
|
See r43067 for details.
|
|
|
|
* gc.c (rb_gc_set_params): refactoring. And change verbose notation.
|
|
Mostly duplicated functions get_envparam_int/double is not cool.
|
|
Please rewrite it.
|
|
|
|
* test/ruby/test_gc.rb: fix a test for this change.
|
|
|
|
Fri Sep 27 17:44:41 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (GC_MALLOC_LIMIT): 8,000,000 -> 8 * 1,024 * 1,024.
|
|
|
|
Fri Sep 27 17:19:39 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_before_sweep): cast to size_t to suppress warnings.
|
|
|
|
Fri Sep 27 17:07:55 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: add some fine-grained profiling codes to tuning marking phase.
|
|
If you enable RGENGC_PRINT_TICK to 1, then profiling results by RDTSC
|
|
(on x86/amd64 environment) are printed at last.
|
|
Thanks Yoshii-san.
|
|
|
|
Fri Sep 27 16:32:27 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: simplify threshold of GC caused by malloc_increase.
|
|
Now, malloc_limit is increased/decreased by mysterious logic.
|
|
This fix simplify malloc_limit increase/decrease logic such as:
|
|
if (malloc_increase > malloc_limit) /* so many malloc */
|
|
malloc_limit += malloc_limit * (GC_MALLOC_LIMIT_FACTOR-1);
|
|
else
|
|
malloc_limit -= malloc_limit * (GC_MALLOC_LIMIT_FACTOR-1)/4;
|
|
Default value of GC_MALLOC_LIMIT_FACTOR is 1.8.
|
|
malloc_limit is bounded by GC_MALLOC_LIMIT_MAX (256MB by default).
|
|
This logic runs at gc_before_sweep(). So there are no effect from
|
|
caused by lazy sweep. And we can remove malloc_increase2.
|
|
|
|
* gc.c (HEAP_MIN_SLOTS, FREE_MIN, HEAP_GROWTH_FACTOR): rename to
|
|
GC_HEAP_MIN_SLOTS, GC_FREE_MIN, GC_HEAP_GROWTH_FACTOR respectively.
|
|
Check them by `#ifndef' so you can specify these values outside gc.c.
|
|
|
|
* gc.c (ruby_gc_params_t): add initial_malloc_limit_factor and
|
|
initial_malloc_limit_max.
|
|
|
|
* gc.c (vm_malloc_prepare, vm_xrealloc): use vm_malloc_increase to
|
|
add and check malloc_increase.
|
|
|
|
Fri Sep 27 01:05:00 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* re.c: [DOC] arguments of Regexp::union receive #to_regexp [Bug #8205]
|
|
|
|
Fri Sep 27 00:39:27 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* struct.c: [DOC] grammar of ArgumentError in Struct.new [Bug #8936]
|
|
Patch by Prathamesh Sonpatki
|
|
|
|
Thu Sep 26 22:11:56 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/bigdecimal/bigdecimal.c: [DOC] several fixes by @chastell
|
|
This includes fixing the capitalization of Infinity, return value of
|
|
example "BigDecimal.new('NaN') == 0.0", and code style in example.
|
|
[Fixes GH-398] https://github.com/ruby/ruby/pull/398
|
|
|
|
Thu Sep 26 22:08:11 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/observer.rb: [DOC] syntax improvement in example by @chastell
|
|
[Fixes GH-400] https://github.com/ruby/ruby/pull/400
|
|
|
|
Thu Sep 26 22:03:15 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/digest/digest.c: [DOC] typo in overview by @chastell
|
|
[Fixes GH-399] https://github.com/ruby/ruby/pull/399
|
|
|
|
Thu Sep 26 22:00:42 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/openssl/ossl.c: [DOC] typo in example by @zoranzaric
|
|
[Fixes GH-401] https://github.com/ruby/ruby/pull/401
|
|
|
|
Thu Sep 26 21:07:49 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el (ruby-electric-delete-backward-char): Add
|
|
support for smartparens-mode.
|
|
|
|
* misc/ruby-electric.el (ruby-electric-cua-replace-region-maybe)
|
|
(ruby-electric-cua-delete-region-maybe): New functions that
|
|
combine `ruby-electric-cua-*-region` with
|
|
`ruby-electric-cua-*-region-p`, using a slightly better way to
|
|
detect if it is in cua-mode.
|
|
|
|
Thu Sep 26 16:51:00 2013 Shota Fukumori <her@sorah.jp>
|
|
|
|
* insns.def (opt_regexpmatch2): Check String#=~ hasn't overridden
|
|
before calling rb_reg_match().
|
|
|
|
* test/ruby/test_string.rb: Test for above.
|
|
|
|
* vm.c (vm_init_redefined_flag): Add BOP flag for String#=~
|
|
|
|
[ruby-core:57385] [Bug #8953]
|
|
|
|
Thu Sep 26 16:43:42 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el: Avoid use of the interactive function
|
|
`self-insert-command` which fires `post-self-insert-hook` and
|
|
`post-command-hook`, to make the ruby-electric commands work
|
|
nicely with those minor modes that make use of them to do
|
|
similar input assistance, such as electric-pair-mode,
|
|
autopair-mode and smartparens-mode.
|
|
|
|
Thu Sep 26 16:24:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* insns.def (opt_regexpmatch1): check Regexp#=~ is not defined before
|
|
calling rb_reg_match()
|
|
|
|
* test/ruby/test_regexp.rb: add test
|
|
|
|
* vm.c (ruby_vm_redefined_flag): change type to short[]
|
|
|
|
* vm.c (vm_redefinition_check_flag): return REGEXP_REDEFINED_OP_FLAG if
|
|
klass == rb_cRegexp
|
|
|
|
* vm.c (vm_init_redefined_flag): setup BOP flag for Regexp#=~
|
|
|
|
* vm_insnhelper.h: add REGEXP_REDEFINED_OP_FLAG
|
|
|
|
[ruby-core:57385] [Bug #8953]
|
|
|
|
Thu Sep 26 14:46:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* gc.c (mark_locations_array): disable AddressSanitizer. based on a
|
|
patch by halfie (Ruby Guy) at [ruby-core:57372].
|
|
[ruby-core:56155] [Bug #8680]
|
|
|
|
Wed Sep 25 17:41:29 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* README.EXT, README.EXT.ja: remove description of RARRAY_PTR()
|
|
and add a caution of accessing internal data structure directly.
|
|
Also add a description of rb_ary_store().
|
|
[Bug #8399]
|
|
|
|
Wed Sep 25 17:12:08 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: rename RARRAY_RAWPTR() to RARRAY_CONST_PTR().
|
|
RARRAY_RAWPTR(ary) returns (const VALUE *) type pointer and
|
|
usecase of this macro is not acquire raw pointer, but acquire
|
|
read-only pointer. So we rename to better name.
|
|
RSTRUCT_RAWPTR() is also renamed to RSTRUCT_CONST_PTR()
|
|
(I expect that nobody use it).
|
|
|
|
* array.c, compile.c, cont.c, enumerator.c, gc.c, proc.c, random.c,
|
|
string.c, struct.c, thread.c, vm_eval.c, vm_insnhelper.c:
|
|
catch up this change.
|
|
|
|
Wed Sep 25 16:58:33 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* internal.h (rb_float_value, rb_float_new): move inline functions
|
|
from ruby/ruby.h.
|
|
|
|
* numeric.c (rb_float_value, rb_float_new): define external functions
|
|
for extension libraries.
|
|
|
|
Wed Sep 25 15:37:02 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/rdoc/test_rdoc_generator_darkfish.rb: add a guard for windows.
|
|
|
|
Wed Sep 25 09:53:11 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Fix CVE-2013-4363. Miscellaneous minor improvements.
|
|
|
|
* test/rubygems: Tests for the above.
|
|
|
|
Tue Sep 24 17:38:56 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_str_inspect): get rid of out-of-bound access.
|
|
|
|
* string.c (rb_str_inspect): when a UTF-16/32 string doesn't have a
|
|
BOM, inspect as a dummy encoding string.
|
|
|
|
Tue Sep 24 17:15:10 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enc/encdb.c (ENC_DUMMY_UNICODE): make BOM-encodings dummy.
|
|
|
|
* encoding.c (enc_autoload): keep dummy encodings dummy.
|
|
|
|
Tue Sep 24 16:41:15 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/win32/lib/win32/registry.rb (Win32::Registry#write): data size
|
|
is in bytes, not chars. terminators should be placed automatically.
|
|
|
|
Tue Sep 24 16:39:36 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/win32/lib/win32/registry.rb (Win32::Registry#each_value): encode
|
|
name.
|
|
|
|
* ext/win32/lib/win32/registry.rb (Win32::Registry#each_key): ditto.
|
|
|
|
* ext/win32/lib/win32/registry.rb (Win32::Registry#export_string):
|
|
encode to locale encoding if default internal is not set.
|
|
|
|
Tue Sep 24 16:35:09 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/win32/lib/win32/registry.rb (Win32::Registry::API#EnumKey):
|
|
size of the name is in WCHARs, not in bytes.
|
|
|
|
Tue Sep 24 14:07:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* gc.c (free_method_cache_entry_i): unused function
|
|
|
|
* gc.c (rb_free_mc_table): ditto
|
|
|
|
* internal.h (method_cache_entry_t): unused struct
|
|
|
|
* vm_method.c (verify_method_cache): remove unused variable
|
|
|
|
* vm_method.c (rb_method_entry): ditto
|
|
|
|
Tue Sep 24 14:01:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* class.c (class_alloc): remove mc_tbl
|
|
|
|
* gc.c (obj_free): ditto
|
|
|
|
* internal.h (struct rb_classext_struct): ditto
|
|
|
|
* method.h (rb_method_entry): remove ent param
|
|
|
|
* vm_method.c: restore the global method cache. Per class cache tables
|
|
turned out to be far too slow.
|
|
|
|
[ruby-core:57289] [Bug #8930]
|
|
|
|
Tue Sep 24 12:51:07 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/win32/lib/win32/registry.rb (Win32::Registry::API): need
|
|
Constants.
|
|
|
|
* ext/win32/lib/win32/registry.rb (Win32::Registry::API#EnumValue):
|
|
size of the name is in WCHARs, not in bytes.
|
|
|
|
Mon Sep 23 22:16:09 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enc/encdb.c, enc/utf_16_32.h (ENC_DUMMY_UNICODE): Unicode with BOM
|
|
must be based on big endian variants, so that actual encodings would
|
|
work. [ruby-core:57318] [Bug #8940]
|
|
|
|
Mon Sep 23 12:11:26 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (env_each_pair): do not call rb_assoc_new() if
|
|
it isn't needed.
|
|
|
|
Mon Sep 23 10:42:30 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* test/ruby/test_module.rb (TestModule#test_include_toplevel): test
|
|
for top level main.include. based on a part of the patch by
|
|
kyrylo at [GH-395].
|
|
|
|
Mon Sep 23 05:07:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/intern.h (rb_ary_cat): move from internal.h, since it
|
|
is described in README.EXT.
|
|
|
|
Sun Sep 22 20:55:20 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* vm_insnhelper.c (vm_make_proc_with_iseq): fix bug message.
|
|
This is follow up to changes in r42637.
|
|
|
|
Sun Sep 22 20:35:38 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* ext/-test-/tracepoint/tracepoint.c (Init_tracepoint): prevent from GC.
|
|
|
|
Sun Sep 22 19:00:28 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* benchmark/bm_app_answer.rb: revert r42990, benchmark scripts should
|
|
be self-contained and avoid dependencies, especially such small one.
|
|
See https://github.com/ruby/ruby/pull/393#issuecomment-24861301.
|
|
|
|
Sat Sep 21 20:11:06 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* process.c (rb_fork_internal): remove cloexec setting on pipes
|
|
created by rb_cloexec_pipe. patch by normalperson (Eric Wong) at
|
|
[ruby-core:56523]. [Bug #8769]
|
|
|
|
Sat Sep 21 01:04:25 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/benchmark.rb: [DOC] grammar of Benchmark#bm [Bug #8888]
|
|
Patch by Prathamesh Sonpatki
|
|
|
|
Sat Sep 21 00:50:02 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* enumerator.c: [DOC] Enumerator#each arguments documentation [GH-388]
|
|
Patch by @kachick https://github.com/ruby/ruby/pull/388
|
|
|
|
Sat Sep 21 00:49:16 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* enum.c: [DOC] Enumerable#to_a accepts arguments [GH-388]
|
|
Patch by @kachick https://github.com/ruby/ruby/pull/388
|
|
|
|
Sat Sep 21 00:47:44 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_str_conv_enc_opts): make sure to scan coderange to get
|
|
rid of unnecessary conversion.
|
|
|
|
Sat Sep 21 00:21:08 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/openssl/lib/openssl/ssl.rb: [DOC] Document OpenSSL::SSLServer
|
|
Based on a patch by Rafal Lisowski [Bug #8758]
|
|
|
|
Fri Sep 20 23:54:03 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/gserver.rb: [DOC] correct gserver.rb license [Bug #8913]
|
|
|
|
Fri Sep 20 23:48:34 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/psych/yaml/yaml.h: [DOC] merge upstream typo fix by @GreenGeorge
|
|
https://github.com/tenderlove/psych/pull/161
|
|
|
|
Fri Sep 20 23:37:40 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/securerandom.rb: [DOC] SecureRandom.hex length argument
|
|
[Fixes GH-394] Patch by @avdi https://github.com/ruby/ruby/pull/394
|
|
|
|
Fri Sep 20 23:34:48 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* benchmark/bm_app_answer.rb: removed duplicate code [Fixes GH-393]
|
|
Patch by @gouravtiwari https://github.com/ruby/ruby/pull/393
|
|
|
|
Fri Sep 20 23:24:08 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* common.mk (btest, btest-ruby, test-knownbug): add $(RUN_OPTS) to
|
|
ruby to be run, so that tests are runnable before making exts.
|
|
|
|
* common.mk (test-sample): ditto, and use $(MINIRUBY) as rubytest.rb
|
|
does not need extension libraries.
|
|
|
|
* tool/rubytest.rb: pass $(RUN_OPTS) to testing ruby using --run-opt.
|
|
|
|
Fri Sep 20 15:01:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (intern_str): sigil only names are junk, at least one
|
|
identifier character is needed. [ruby-dev:47723] [Bug #8928]
|
|
|
|
* parse.y (rb_enc_symname_type): fix out of bound access.
|
|
|
|
Fri Sep 20 14:14:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/-test-/printf/printf.c (printf_test_call): Fix an end of buffer
|
|
argument.
|
|
|
|
Thu Sep 19 16:59:02 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (lambda): adjust position to the beginning of the block.
|
|
|
|
Thu Sep 19 16:25:06 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vsnprintf.c (BSD_vfprintf): initialize cp so that size is 0 in the
|
|
commented case. fix an accidental bug at r16716.
|
|
|
|
Thu Sep 19 14:33:14 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* NEWS: add a news for r42974.
|
|
|
|
Thu Sep 19 14:12:02 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: make Symbol objects frozen.
|
|
[Feature #8906]
|
|
I want to freeze this good day, too.
|
|
|
|
* test/ruby/test_eval.rb: catch up this change.
|
|
|
|
* test/ruby/test_symbol.rb: add a test to check frozen symbols.
|
|
|
|
Thu Sep 19 09:11:33 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* NEWS: Update for RDoc 4.1.0.preview.1 and RubyGems 2.2.0.preview.1
|
|
|
|
Thu Sep 19 08:59:41 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rdoc/markdown/literals_1_9.rb: Fix trailing whitespace.
|
|
|
|
Previously kpeg (which generates this file) added trailing
|
|
whitespace, but this bug is now fixed.
|
|
|
|
* lib/rdoc/markdown.rb: ditto.
|
|
|
|
Thu Sep 19 08:33:14 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rdoc: Update to RDoc 4.1.0.preview.1
|
|
|
|
RDoc 4.1.0 contains a number of enhancements including a new default
|
|
style and accessibility support. You can see the changelog here:
|
|
|
|
https://github.com/rdoc/rdoc/blob/v4.1.0.preview.1/History.rdoc
|
|
|
|
* test/rdoc: ditto.
|
|
|
|
Thu Sep 19 07:16:26 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych.rb: updating Psych version
|
|
|
|
* ext/psych/psych.gemspec: ditto
|
|
|
|
Thu Sep 19 06:39:40 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/dependency_resolver.rb: Switch the iterative resolver
|
|
algorithm from recursive to iterative to avoid possible
|
|
SystemStackError.
|
|
|
|
Thu Sep 19 06:29:30 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems 2.2.0.preview.1
|
|
|
|
This brings several new features to RubyGems summarized here:
|
|
|
|
https://github.com/rubygems/rubygems/blob/v2.2.0.preview.1/History.txt
|
|
|
|
* test/rubygems: ditto.
|
|
|
|
Wed Sep 18 23:14:58 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* string.c (rb_str_enumerate_lines): make String#each_line and
|
|
#lines not raise invalid byte sequence error when it is called
|
|
with an argument. The patch also causes performance improvement.
|
|
[ruby-dev:47549] [Bug #8698]
|
|
|
|
* test/ruby/test_m17n_comb.rb (test_str_each_line): remove
|
|
assertions which check that String#each_line and #lines will
|
|
raise an error if the receiver includes invalid byte sequence.
|
|
|
|
Wed Sep 18 16:32:15 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* proc.c (mnew_from_me): allocate structs after allocated wrapper
|
|
object successfully, to get rid of potential memory leak.
|
|
|
|
Tue Sep 17 15:54:03 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/shell/command-processor.rb (Shell::CommandProcessor#find_system_command):
|
|
return executable file only, should ignore directories and
|
|
unexecutable files. [ruby-core:57235] [Bug #8918]
|
|
|
|
* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert_throw):
|
|
assertion for throw. MiniTest::Assertions#assert_throws discards
|
|
the caught value.
|
|
|
|
* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert_nothing_thrown):
|
|
returns the result of the given block.
|
|
|
|
Tue Sep 17 12:55:58 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* doc/regexp.rdoc: [DOC] Replace paragraphs in verbatim sections with
|
|
plain paragraphs to improve readability as ri and HTML.
|
|
|
|
Mon Sep 16 07:32:35 2013 Tadayoshi Funaba <tadf@dotrb.org>
|
|
|
|
* complex.c: removed meaningless lines.
|
|
* rational.c: ditto.
|
|
|
|
Mon Sep 16 00:44:23 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* ext/socket/mkconstants.rb: define MSG_FASTOPEN.
|
|
[ruby-core:57138] [Feature #8897]
|
|
|
|
Sun Sep 15 13:31:23 2013 Tadayoshi Funaba <tadf@dotrb.org>
|
|
|
|
* rational.c (nurat_div): reverted r28844, r28886 and r28887.
|
|
REASON: Nobuyoshi Nakada <nobu@ruby-lang.org>'s commits are buggy.
|
|
So Rational#/ may produce exact number with inexact number.
|
|
Moreover, without reducing.
|
|
REALLY NONSENSE COMMITS.
|
|
A bug report by me [ruby-dev:44710] is also caused by this behavior.
|
|
Kenta Murata <mrkn@mrkn.jp> patched it up.
|
|
But he did not fix the origin.
|
|
Today, the bug is still alive in ruby 1.9.3 and 2.0.0.
|
|
|
|
Sat Sep 14 06:08:10 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* dir.c (dir_s_glob): [DOC] Improve wording and layout.
|
|
|
|
* dir.c (file_s_fnmatch): ditto.
|
|
|
|
* dir.c (Init_Dir): [DOC] Document File::Constants::FNM_XXX
|
|
constants. (These won't show up in RDoc until a new RDoc is
|
|
imported.)
|
|
|
|
Thu Sep 12 14:58:58 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* lib/uri/generic.rb (URI::Generic.find_proxy): return nil if
|
|
http_proxy environment variable is empty string.
|
|
[ruby-core:57140] [Bug #8898]
|
|
|
|
Fri Sep 13 10:40:28 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems 2.1.3
|
|
|
|
Fixed installing platform gems
|
|
|
|
Restored concurrent requires
|
|
|
|
Fixed installing gems with extensions with --install-dir
|
|
|
|
Fixed `gem fetch -v` to install the latest version
|
|
|
|
Fixed installing gems with "./" in their files entries
|
|
|
|
* test/rubygems/test_gem_package.rb: Tests for the above.
|
|
|
|
* NEWS: Updated for RubyGems 2.1.3
|
|
|
|
Thu Sep 12 22:40:03 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (RUBY_CHECK_SIGNEDNESS): macro to check signedness of a
|
|
type.
|
|
|
|
* configure.in (size_t): must be unsigned.
|
|
[ruby-core:57149] [Feature #8890]
|
|
|
|
Thu Sep 12 22:37:08 2013 Anton Ovchinnikov <revolver112@gmail.com>
|
|
|
|
* ext/bigdecimal/bigdecimal.c, ext/digest/md5/md5.c,
|
|
ext/json/fbuffer/fbuffer.h, ext/json/generator/generator.c:
|
|
Eliminate less-than-zero checks for unsigned variables.
|
|
According to section 4.1.5 of C89 standard, size_t is an unsigned
|
|
type. These checks were found with 'cppcheck' static analysis tool.
|
|
[ruby-core:57117] [Feature #8890]
|
|
|
|
Thu Sep 12 21:35:46 2013 Naohisa Goto <ngotogenome@gmail.com>
|
|
|
|
* Makefile.in (libruby-static.a): change LDFLAGS order. LDFLAGS may
|
|
include library path that should be specified before LIBS.
|
|
[ruby-dev:47707] [Bug #8901]
|
|
|
|
Thu Sep 12 20:07:29 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vsnprintf.c (MAXEXP, MAXFRACT): calculate depending on constants in
|
|
float.h.
|
|
|
|
* vsnprintf.c (BSD_vfprintf): limit length for cvt() to get rid of
|
|
buffer overflow. [ruby-core:57023] [Bug #8864]
|
|
|
|
* vsnprintf.c (exponent): make expbuf size more precise.
|
|
|
|
Wed Sep 11 17:30:45 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (RUNRUBY): append -- only after runruby.rb, not
|
|
cross-compiling baseruby, so that $(RUN_OPT) can be command line
|
|
options. [ruby-dev:47703] [Bug #8893]
|
|
|
|
Wed Sep 11 07:55:17 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread.c (rb_mutex_unlock): Mutex#unlock no longer raise
|
|
an exception even if uses on trap. [Bug #8891]
|
|
|
|
Tue Sep 10 14:37:01 2013 Shota Fukumori <sorah@tubusu.net>
|
|
|
|
* vm_backtrace.c (vm_backtrace_to_ary): Ignore the second argument if
|
|
it is nil. [Bug #8884] [ruby-core:57094]
|
|
|
|
* test/ruby/test_backtrace.rb (test_caller_with_nil_length):
|
|
Test for above.
|
|
|
|
Tue Sep 10 12:39:17 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* class.c (method_entry_i): should exclude refined methods from
|
|
instance method list. [ruby-core:57080] [Bug #8881]
|
|
|
|
Tue Sep 10 12:05:04 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* io.c (rb_f_printf): [DOC] add missing parenthesis in rdoc.
|
|
|
|
Tue Sep 10 10:08:00 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* NEWS: Update RubyGems note.
|
|
|
|
Tue Sep 10 09:51:22 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems 2.1.0. Fixes CVE-2013-4287.
|
|
|
|
See http://rubygems.rubyforge.org/rubygems-update/CVE-2013-4287_txt.html
|
|
for CVE information.
|
|
|
|
See http://rubygems.rubyforge.org/rubygems-update/History_txt.html#label-2.1.0+%2F+2013-09-09
|
|
for release notes.
|
|
|
|
* test/rubygems: Tests for the above.
|
|
|
|
Mon Sep 9 21:31:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c: Remove spaces between SI prefix and unit to follow
|
|
SI brochure.
|
|
http://www.bipm.org/en/si/si_brochure/
|
|
https://www.nmij.jp/library/units/si/
|
|
|
|
* time.c: Ditto.
|
|
|
|
* ext/socket/ancdata.c: Ditto.
|
|
|
|
Mon Sep 9 16:55:59 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_method.c (rb_add_refined_method_entry): clear cache in the
|
|
refined class since refining a method entry is modifying the class.
|
|
[ruby-core:57079] [Bug #8880]
|
|
|
|
Mon Sep 9 09:14:58 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* tool/rbinstall.rb (Gem::Specification#initialize): default date to
|
|
RUBY_RELEASE_DATE. [ruby-core:57072] [Bug #8878]
|
|
|
|
* tool/rbinstall.rb (Gem::Specification#to_ruby): add date.
|
|
|
|
Sun Sep 8 16:01:54 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* rational.c (f_gcd): Relax the condition to use GMP.
|
|
|
|
Sun Sep 8 13:56:38 2013 Masaki Suketa <masaki.suketa@nifty.ne.jp>
|
|
|
|
* ext/win32ole/win32ole.c (folevariant_initialize): check type of
|
|
element of array.
|
|
|
|
* test/win32ole/test_win32ole_variant.rb (test_s_new_ary): ditto.
|
|
|
|
Sat Sep 7 21:33:10 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* math.c (math_log): Test the sign for bignums.
|
|
(math_log2): Ditto.
|
|
(math_log10): Ditto.
|
|
|
|
Sat Sep 7 20:25:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* math.c (math_log): Support bignums bigger than 2**1024.
|
|
(math_log2): Ditto.
|
|
(math_log10): Ditto.
|
|
|
|
Sat Sep 7 15:36:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* vm_eval.c (vm_call0): fix prototype, the id parameter should be of
|
|
type ID, not VALUE
|
|
|
|
* vm_insnhelper.c (check_match): the rb_funcall family of functions
|
|
does not care about refinements. We need to use
|
|
rb_method_entry_with_refinements instead to call === with
|
|
refinements. Thanks to Jon Conley for reporting this bug.
|
|
[ruby-core:57051] [Bug #8872]
|
|
|
|
* test/ruby/test_refinement.rb: add test
|
|
|
|
Sat Sep 7 13:49:40 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* variable.c (classname): the name of class that has
|
|
non class id should not be nil. This bug was introduced
|
|
in r36577.
|
|
|
|
* test/thread/test_cv.rb: test for change.
|
|
|
|
Sat Sep 7 13:29:22 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* lib/find.rb (Find.find): respect the encodings of arguments.
|
|
[ruby-dev:47530] [Feature #8657]
|
|
|
|
* test/test_find.rb: add tests.
|
|
|
|
Sat Sep 7 10:40:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/mkconstants.rb (TCP_FASTOPEN): Defined for TCP fast open.
|
|
[ruby-core:57048] [Feature #8871] patch by Masaki Matsushita.
|
|
|
|
Fri Sep 6 23:53:31 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* common.mk: use RUNRUBY instead of MINIRUBY because MINIRUBY can't
|
|
require extension libraries. The patch is from nobu
|
|
(Nobuyoshi Nakada).
|
|
|
|
* ext/thread/extconf.rb: for build ext/thread/thread.c.
|
|
|
|
* include/ruby/intern.h: ditto.
|
|
|
|
* thread.c: ditto.
|
|
|
|
* lib/thread.rb: removed and replaced by ext/thread/thread.c.
|
|
|
|
* ext/thread/thread.c: Queue, SizedQueue and ConditionVariable
|
|
implementations in C. This patch is based on patches from panaggio
|
|
(Ricardo Panaggio) and funny_falcon (Yura Sokolov) and ko1
|
|
(Koichi Sasada). [ruby-core:31513] [Feature #3620]
|
|
|
|
* test/thread/test_queue.rb (test_queue_thread_raise): add a test for
|
|
ensuring that killed thread should be removed from waiting threads.
|
|
It is based on a code by ko1 (Koichi Sasada). [ruby-core:45950]
|
|
|
|
Fri Sep 6 22:47:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Define ac_cv_func_clock_getres to yes for mingw*.
|
|
|
|
Fri Sep 6 21:04:10 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* rational.c: Include gmp.h if GMP is used.
|
|
(GMP_GCD_DIGITS): New macro.
|
|
(rb_gcd_gmp): New function.
|
|
(f_gcd_normal): Renamed from f_gcd.
|
|
(rb_gcd_normal): New function.
|
|
(f_gcd): Invoke rb_gcd_gmp or f_gcd_normal.
|
|
|
|
* internal.h (rb_gcd_normal): Declared.
|
|
(rb_gcd_gmp): Ditto.
|
|
|
|
* ext/-test-/rational: New directory.
|
|
|
|
* test/-ext-/rational: New directory.
|
|
|
|
Fri Sep 6 14:23:22 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (clock_getres): required as well as clock_gettime().
|
|
[ruby-dev:47699] [Bug #8869]
|
|
|
|
Fri Sep 6 11:45:27 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* transcode.c (rb_econv_append): new function to append a string data
|
|
with converting its encoding. split from rb_econv_substr_append.
|
|
|
|
Fri Sep 6 02:37:22 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych/visitors/yaml_tree.rb: use double quotes when
|
|
strings start with special characters.
|
|
https://github.com/tenderlove/psych/issues/157
|
|
|
|
* test/psych/test_string.rb: test for change.
|
|
|
|
Fri Sep 6 00:05:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* class.c (rewrite_cref_stack): remove recursion.
|
|
|
|
Thu Sep 5 18:05:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* string.c (fstring_cmp): take string encoding into account when
|
|
comparing fstrings [ruby-core:57037] [Bug #8866]
|
|
|
|
* test/ruby/test_string.rb: add test
|
|
|
|
Thu Sep 5 17:25:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_fstring, rb_str_free): use st_data_t instead of VALUE.
|
|
|
|
* string.c (rb_fstring): get rid of duplicating already frozen object.
|
|
|
|
* parse.y (str_suffix_gen): freeze in advance to reduce objects.
|
|
|
|
Thu Sep 5 14:01:22 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/optparse.rb: The Integer acceptable now allows binary and
|
|
hexadecimal numbers per the documentation. [ruby-trunk - Bug #8865]
|
|
|
|
DecimalInteger, OctalInteger, DecimalNumeric now validate their input
|
|
before converting to a number. [ruby-trunk - Bug #8865]
|
|
|
|
* test/optparse/test_acceptable.rb: Tests for the above, tests for all
|
|
numeric acceptables for existing behavior.
|
|
|
|
Thu Sep 5 13:49:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h: add RSTRING_FSTR flag
|
|
|
|
* internal.h: add rb_fstring() prototype
|
|
|
|
* parse.y (str_suffix_gen): deduplicate frozen string literals
|
|
|
|
* string.c (rb_fstring): deduplicate frozen string literals
|
|
|
|
* string.c (rb_str_free): delete fstrings from frozen_strings table when
|
|
they are GC'd
|
|
|
|
* string.c (Init_String): initialize frozen_strings table
|
|
|
|
* test/ruby/test_string.rb: test frozen strings are deduplicated
|
|
|
|
Thu Sep 5 12:48:00 2013 Kenta Murata <mrkn@cookpad.com>
|
|
|
|
* configure.in (with_gmp): set with_gmp no if it is empty.
|
|
|
|
Thu Sep 5 10:41:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* vm_insnhelper.c (vm_getivar): use class sequence to check class
|
|
identity, instead of pointer + vm state
|
|
|
|
* vm_insnhelper.c (vm_setivar): ditto
|
|
|
|
Thu Sep 5 08:20:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (GMP_DIV_DIGITS): New macro.
|
|
(bary_divmod_gmp): New function.
|
|
(rb_big_divrem_gmp): Ditto.
|
|
(bary_divmod_branch): Ditto.
|
|
(bary_divmod): Use bary_divmod_branch.
|
|
(bigdivrem): Ditto.
|
|
|
|
* internal.h (rb_big_divrem_gmp): Declared.
|
|
|
|
Thu Sep 5 06:22:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_divmod_normal): Reduce temporary array allocations.
|
|
|
|
Thu Sep 5 02:17:06 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_divrem_normal): Add GC guards.
|
|
|
|
Thu Sep 5 00:38:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_divrem_normal): New function.
|
|
|
|
* internal.h (rb_big_divrem_normal): Declared.
|
|
|
|
* ext/-test-/bignum/div.c: New file.
|
|
|
|
* test/-ext-/bignum/test_div.rb: New file.
|
|
|
|
Thu Sep 5 00:08:44 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem_normal): Removed.
|
|
(bary_divmod_normal): New function.
|
|
(bary_divmod): Use bary_divmod_normal.
|
|
(bigdivrem): Use bary_divmod_normal.
|
|
|
|
Wed Sep 4 23:02:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem): Useless declaration removed.
|
|
|
|
Wed Sep 4 22:56:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* numeric.c (NUM_STEP_GET_INF): split from NUM_STEP_SCAN_ARGS(), since
|
|
inf is not used in num_step_size().
|
|
|
|
Wed Sep 4 20:22:43 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem_normal): Add assertions.
|
|
|
|
Wed Sep 4 19:18:40 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* internal.h (vm_state_version_t): prefer LONG_LONG to uint64_t.
|
|
|
|
Wed Sep 4 16:28:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* internal.h (vm_state_version_t): use uint64_t when it is larger than
|
|
LONG_LONG, and fallback to unsigned long.
|
|
|
|
Wed Sep 4 15:37:05 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* enc/trans/utf8_mac-tbl.rb: fix r42789.
|
|
Fix conversion table and logic. [ruby-dev:47680]
|
|
|
|
Wed Sep 4 14:08:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* class.c, compile.c, eval.c, gc.h, insns.def, internal.h, method.h,
|
|
variable.c, vm.c, vm_core.c, vm_insnhelper.c, vm_insnhelper.h,
|
|
vm_method.c: Implement class hierarchy method cache invalidation.
|
|
|
|
[ruby-core:55053] [Feature #8426] [GH-387]
|
|
|
|
Wed Sep 4 11:13:40 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (str_gsub): use BEG(0) for whole matched position not
|
|
return value from rb_reg_search(), for \K matching.
|
|
[ruby-dev:47694] [Bug #8856]
|
|
|
|
Wed Sep 4 11:11:37 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (SOLIBS): LIBRUBY_SO also needs linking with gmp, to
|
|
run worker processes in test-all on non-ELF platforms.
|
|
|
|
Tue Sep 3 23:01:41 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/parser/test_tree.rb
|
|
(TestTreeParser::TestInvalid#test_unmatched_close_tag):
|
|
Compute expected value from test value.
|
|
|
|
Tue Sep 3 22:59:58 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/treeparser.rb (REXML::Parsers::TreeParser#parse):
|
|
Add source information to parse exception on no close tag error.
|
|
[Bug #8844] [ruby-dev:47672]
|
|
Patch by Ippei Obayashi. Thanks!!!
|
|
* test/rexml/parser/test_tree.rb: Add a test for the above case.
|
|
|
|
Tue Sep 3 22:57:57 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/parser/test_tree.rb: Fix test name to describe test
|
|
content.
|
|
|
|
Tue Sep 3 22:54:46 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/treeparser.rb (REXML::Parsers::TreeParser#parse):
|
|
Remove needless nested parse exception information.
|
|
[Bug #8844] [ruby-dev:47672]
|
|
Reported by Ippei Obayashi. Thanks!!!
|
|
* test/rexml/parser/test_tree.rb: Add a test for the above case.
|
|
|
|
Tue Sep 3 22:03:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_enc_str_new_cstr): new function to create a string from
|
|
the C-string pointer with the specified encoding.
|
|
|
|
Tue Sep 3 21:41:37 2013 Akira Matsuda <ronnie@dio.jp>
|
|
|
|
* eval.c (Init_eval): Make Module#include and Module#prepend public
|
|
[Feature #8846]
|
|
|
|
* test/ruby/test_module.rb (class TestModule): Test for above
|
|
|
|
Tue Sep 3 21:35:19 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* thread_pthread.c (sys/dyntune.h): for gettune().
|
|
|
|
* thread_pthread.c (hpux_attr_getstackaddr): fix missing *.
|
|
[ruby-core:56983] [Feature #8793]
|
|
|
|
Tue Sep 3 20:12:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (GMP_STR2BIG_DIGITS): New macro.
|
|
(str2big_gmp): New function.
|
|
(rb_cstr_to_inum): Use str2big_gmp for big bignums.
|
|
(rb_str2big_gmp): New function.
|
|
|
|
* internal.h (rb_str2big_gmp): Declared.
|
|
|
|
Tue Sep 3 19:44:40 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ext/win32/lib/win32/registry.rb (Win32::Registry#values): added.
|
|
[Feature #7763] [ruby-core:51783]
|
|
|
|
Tue Sep 3 18:26:00 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/inf-ruby.el (inf-ruby-keys, run-ruby): Add magic autoload
|
|
comments.
|
|
|
|
* misc/rdoc-mode.el (rdoc-mode): Ditto.
|
|
|
|
* misc/ruby-electric.el (ruby-electric-mode): Ditto.
|
|
|
|
* misc/ruby-style.el (ruby-style-c-mode): Ditto.
|
|
|
|
Tue Sep 3 17:06:15 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/ruby/test_rubyoptions.rb
|
|
(TestRubyOptions::SEGVTest::ExpectedStderr): the URL was changed at
|
|
r42800.
|
|
|
|
Tue Sep 3 14:48:25 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/thread.rb: [DOC] CV#wait typo by @avdi [Fixes GH-386]
|
|
https://github.com/ruby/ruby/pull/386
|
|
|
|
Tue Sep 3 14:37:53 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* error.c: [DOC] Update bug tracker url by @ScotterC [Fixes GH-390]
|
|
https://github.com/ruby/ruby/pull/390
|
|
|
|
Tue Sep 3 12:45:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_str2big_poweroftwo): New function.
|
|
(rb_str2big_normal): Ditto.
|
|
(rb_str2big_karatsuba): Ditto.
|
|
|
|
* internal.h (rb_str2big_poweroftwo): Declared.
|
|
(rb_str2big_normal): Ditto.
|
|
(rb_str2big_karatsuba): Ditto.
|
|
|
|
* ext/-test-/bignum/str2big.c: New file.
|
|
|
|
* test/-ext-/bignum/test_str2big.rb: New file.
|
|
|
|
* ext/-test-/bignum/depend: Add the dependency for str2big.c.
|
|
|
|
Tue Sep 3 12:09:08 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): Support times() based monotonic clock.
|
|
(rb_clock_getres): Ditto.
|
|
|
|
Tue Sep 3 12:03:02 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (str2big_scan_digits): Extracted from rb_cstr_to_inum.
|
|
|
|
Tue Sep 3 11:23:57 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/win32.c (rb_w32_select_with_thread): rounding up the fraction of
|
|
tv_usec instead of rounding down.
|
|
this change is an experiment to get rid of failures on vc10-x64 CI.
|
|
|
|
Tue Sep 3 11:00:28 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/win32.c (do_select): constify timeout.
|
|
|
|
* win32/win32.c (rb_w32_select_with_thread): constify 10ms wait and
|
|
0ms wait structs.
|
|
|
|
Tue Sep 3 10:03:42 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/openssl/test_pair.rb
|
|
(OpenSSL::TestPair#test_write_nonblock_no_exceptions): on some CIs
|
|
such as Debian 6.0, Ubuntu 10.04, CentOS and vc10-x64 (maybe depend
|
|
on OpenSSL version), writing to SSLSocket after SSL_ERROR_WANT_WRITE
|
|
causes SSL_ERROR_SSL "bad write retry".
|
|
|
|
Tue Sep 3 08:20:46 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* enc/trans/utf8_mac-tbl.rb: update conversion table to recent OS X.
|
|
Previous table is used on Mac OS X 10.1 or prior.
|
|
This table is used on 10.2 or later. [ruby-dev:47680]
|
|
|
|
Tue Sep 3 07:49:25 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* numeric.c (NUM_STEP_SCAN_ARGS): On second thought, keep
|
|
Numeric#step backward compatible in that it raises TypeError
|
|
when nil is given as second argument.
|
|
|
|
* test/ruby/test_float.rb (TestFloat#test_num2dbl): Revert.
|
|
|
|
* test/ruby/test_numeric.rb (TestNumeric#test_step): Fix test
|
|
cases for the above change.
|
|
|
|
Tue Sep 3 07:39:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bytes_2comp): Define it only for little endian
|
|
environment.
|
|
|
|
Tue Sep 3 07:31:29 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* numeric.c (NUM_STEP_SCAN_ARGS): Numeric#step should raise
|
|
TypeError if a non-numeric parameter is given.
|
|
|
|
* test/ruby/test_float.rb (TestFloat#test_num2dbl): Allow nil as
|
|
step, as with the keyword argument.
|
|
|
|
* test/ruby/test_numeric.rb (TestNumeric#test_step): Add tests for
|
|
nil as step or limit.
|
|
|
|
Tue Sep 3 07:28:49 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (bit_length): Add casts to fix compilation error with
|
|
clang 3.0 -Werror,-Wshorten-64-to-32.
|
|
[ruby-dev:47687] reported by SASADA Koichi.
|
|
|
|
Tue Sep 3 03:17:26 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_insnhelper.c (vm_search_super_method): use ci->argc instead of
|
|
ci->orig_argc. ci->argc can be changed by splat arguments.
|
|
[ruby-list:49575]
|
|
This fix should be applied to Ruby 2.0.0 series.
|
|
|
|
* test/ruby/test_super.rb: add a test for above.
|
|
|
|
Mon Sep 2 23:46:29 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* numeric.c (num_step): Default the limit argument to infinity and
|
|
allow it to be omitted. Keyword arguments (by: and to:) are
|
|
introduced for ease of use. [Feature #8838] [ruby-dev:47662]
|
|
[ruby-dev:42194]
|
|
|
|
* numeric.c (num_step): Optimize for infinite loop.
|
|
|
|
Mon Sep 2 23:46:10 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_str_options): use valid suffix word only, as well as
|
|
numeric literal, for the backward compatibility.
|
|
|
|
Mon Sep 2 22:55:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (ISDIGIT): Unused macro removed.
|
|
|
|
Mon Sep 2 22:49:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (str2big_poweroftwo): Extracted from rb_cstr_to_inum.
|
|
(str2big_normal): Ditto.
|
|
(str2big_karatsuba): Ditto.
|
|
|
|
Mon Sep 2 17:53:33 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (str_suffix_gen): String#b creates new string object, use
|
|
force_encoding instead.
|
|
|
|
Mon Sep 2 16:06:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* NEWS: Add note about frozen string literals
|
|
|
|
* compile.c (case_when_optimizable_literal): optimize NODE_LIT strings
|
|
in when clauses of case statements
|
|
|
|
* ext/ripper/eventids2.c: add tSTRING_SUFFIX
|
|
|
|
* parse.y: add 'f' suffix on string literals for frozen strings
|
|
|
|
* test/ripper/test_scanner_events.rb: add scanner tests
|
|
|
|
* test/ruby/test_string.rb: add frozen string tests
|
|
|
|
[Feature #8579] [ruby-core:55699]
|
|
|
|
Mon Sep 2 14:39:29 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* ruby.c (Process#setproctitle): [DOC] Fix and improve rdoc.
|
|
|
|
* ruby.c (Process#argv0): [DOC] Improve rdoc.
|
|
|
|
Mon Sep 2 14:15:00 2013 Kenta Murata <mrkn@cookpad.com>
|
|
|
|
* NEWS: fix description of number literal suffixes.
|
|
|
|
Mon Sep 2 14:01:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* test/rake/test_rake_rules.rb: add space after string literal to
|
|
prevent conflict with string options syntax "foo"opts
|
|
|
|
* test/rss/rss-assertions.rb: ditto
|
|
|
|
Mon Sep 2 12:28:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/ruby/test_bignum.rb (test_interrupt_during_to_s): Disable it
|
|
when GMP is used.
|
|
|
|
Mon Sep 2 07:02:10 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (Init_Bignum): Define Bignum::GMP_VERSION when GMP is used.
|
|
|
|
Mon Sep 2 01:46:14 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_generic): Reduce arguments.
|
|
(big2str_gmp): Ditto.
|
|
(rb_big2str1): Follow the above change.
|
|
|
|
Mon Sep 2 00:08:08 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (get_mach_timebase_info): Extracted from rb_clock_gettime.
|
|
(rb_clock_gettime): Use get_mach_timebase_info.
|
|
(rb_clock_getres): Support MACH_ABSOLUTE_TIME_BASED_CLOCK_MONOTONIC.
|
|
|
|
Sun Sep 1 23:30:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (GMP_BIG2STR_DIGITS): New constant.
|
|
(big2str_gmp): New function.
|
|
(rb_big2str1): Use big2str_gmp for big bignums.
|
|
|
|
* internal.h (rb_big2str_gmp): Declared.
|
|
|
|
* ext/-test-/bignum/big2str.c (big2str_gmp): New method.
|
|
|
|
Sun Sep 1 22:37:51 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul_gmp): Use mpz_init and mpz_clear instead of
|
|
mpz_inits and mpz_clears.
|
|
Older GMP don't have them.
|
|
|
|
Sun Sep 1 21:17:54 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/net/http/test_http.rb (test_bind_to_local_port): Choose an open
|
|
port more reliably.
|
|
|
|
Sun Sep 1 20:32:40 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_base_poweroftwo): Renamed from
|
|
big2str_base_powerof2.
|
|
(rb_big2str_poweroftwo): New function for test.
|
|
(big2str_generic): Extracted from rb_big2str1.
|
|
(rb_big2str_generic): New function for test.
|
|
|
|
* internal.h (rb_big2str_poweroftwo): Declared.
|
|
(rb_big2str_generic): Ditto.
|
|
|
|
* ext/-test-/bignum/big2str.c: New file.
|
|
|
|
* test/-ext-/bignum/test_big2str.rb: New file.
|
|
|
|
Sun Sep 1 15:21:21 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_2bdigits): Renamed from big2str_orig.
|
|
|
|
Sun Sep 1 13:02:24 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Remove BITSPERDIG >= INT_MAX test. The static assertion,
|
|
SIZEOF_BDIGITS <= sizeof(BDIGIT) is enough.
|
|
|
|
Sun Sep 1 11:38:26 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (maxpow_in_bdigit): Removed.
|
|
|
|
Sun Sep 1 10:30:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* numeric.c (rb_fix_bit_length): Moved from bignum.c.
|
|
|
|
Sun Sep 1 09:55:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (bit_length): Moved from bignum.c.
|
|
(nlz_int): Ditto.
|
|
(nlz_long): Ditto.
|
|
(nlz_long_long): Ditto.
|
|
(nlz_int128): Ditto.
|
|
|
|
Sun Sep 1 03:32:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bit_length): Renamed from bitsize.
|
|
|
|
Sun Sep 1 00:07:09 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_bit_length): New method.
|
|
(rb_fix_bit_length): Ditto.
|
|
[ruby-core:56247] [Feature #8700]
|
|
|
|
Sat Aug 31 22:18:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_getres): New method.
|
|
(timetick2dblnum_reciprocal): New function.
|
|
|
|
* configure.in: Check clock_getres.
|
|
|
|
[ruby-core:56780] [Feature #8809] accepted as a CRuby feature at
|
|
DevelopersMeeting20130831Japan
|
|
https://bugs.ruby-lang.org/projects/ruby/wiki/DevelopersMeeting20130831Japan
|
|
|
|
Sat Aug 31 21:02:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Use GMP to accelerate big Bignum multiplication.
|
|
(bary_mul_gmp): New function.
|
|
(bary_mul): Use bary_mul_gmp.
|
|
(bigsq): Use different threshold with GMP.
|
|
|
|
* configure.in: Detect GMP.
|
|
|
|
[ruby-core:56658] [Feature #8796]
|
|
|
|
Sat Aug 31 15:03:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* compile.c (NODE_MATCH3): pass CALL_INFO to opt_regexpmatch2
|
|
|
|
* insns.def (opt_regexpmatch2): use CALL_SIMPLE_METHOD to call =~ if
|
|
the receiver is not a T_STRING [Bug #8847] [ruby-core:56916]
|
|
|
|
Sat Aug 31 14:07:11 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/securerandom.rb (random_bytes): Use Process.clock_gettime.
|
|
|
|
Sat Aug 31 00:25:15 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/encoding.h (rb_{ascii8bit,utf8,usascii}_encindex): get
|
|
rid of conflict with macros defined in internal.h.
|
|
|
|
Fri Aug 30 22:37:57 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* thread_pthread.c (native_thread_init_stack): wait the creator thread
|
|
to fill machine stack info, if get_stack_of() is available.
|
|
|
|
* thread_pthread.c (native_thread_create): fill the created thread
|
|
stack info after starting, if get_stack_of() is available.
|
|
|
|
* thread_pthread.c (native_thread_create): define attr only if it is
|
|
used, and merge pthread_create() calls.
|
|
|
|
* thread_pthread.c (get_main_stack): separate function to get stack of
|
|
main thread.
|
|
|
|
Thu Aug 29 18:05:33 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* struct.c (rb_struct_define_without_accessor_under): added.
|
|
This function is similar to rb_define_class_under() against
|
|
rb_define_class().
|
|
|
|
* include/ruby/intern.h: add a declaration of this function.
|
|
|
|
Thu Aug 29 17:03:10 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_insnhelper.c (vm_call_method): a method entry refers the based
|
|
class/module, so should search superclass from the origin i-class
|
|
where the entry belongs to, to get rid of infinite loop when zsuper
|
|
in a prepended class/module. [ruby-core:54105] [Bug #8238]
|
|
|
|
Thu Aug 29 05:35:58 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* ext/zlib/zlib.c (zstream_run): Fix handling of deflate streams that
|
|
need a dictionary but are being decompressed by Zlib::Inflate.inflate
|
|
(which has no option to set a dictionary). Now Zlib::NeedDict is
|
|
raised instead of crashing. [ruby-trunk - Bug #8829]
|
|
* test/zlib/test_zlib.rb (TestZlibInflate): Test for the above.
|
|
|
|
Thu Aug 29 02:40:45 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych/scalar_scanner.rb: invalid floats should be
|
|
treated as strings.
|
|
https://github.com/tenderlove/psych/issues/156
|
|
|
|
* test/psych/test_string.rb: test for change
|
|
|
|
Wed Aug 28 17:20:07 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* thread_pthread.c (hpux_attr_getstackaddr): basic support for the
|
|
get_stack() under HP-UX. based on the patch by michal@rokos.cz
|
|
(Michal Rokos) at [ruby-core:56645]. [Feature #8793]
|
|
|
|
Wed Aug 28 11:24:20 2013 Michal Rokos <michal@rokos.cz>
|
|
|
|
* configure.in (sys/pstat.h): fix missing header check for
|
|
missing/setproctitle.c on HP-UX. [ruby-core:56644] [Bug #8792]
|
|
|
|
Wed Aug 28 04:54:33 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* ext/openssl/ossl_ssl.c (ossl_ssl_read): Replace duplicate
|
|
wait_writable with wait_readable.
|
|
|
|
Tue Aug 27 17:18:40 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/timeout.rb (Timeout#timeout): skip rescue clause only when no
|
|
exception class is given.
|
|
|
|
Tue Aug 27 17:02:58 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* io.c (copy_stream_body): should write in binary mode. based on a
|
|
patch by godfat (Lin Jen-Shin) at [ruby-core:56556].
|
|
[ruby-core:56518] [Bug #8767]
|
|
|
|
Tue Aug 27 17:02:33 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* io.c (copy_stream_body): move common open flags.
|
|
|
|
Tue Aug 27 16:56:50 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enumerator.c (enumerator_size): use rb_check_funcall() instead of
|
|
respond_to? and call.
|
|
|
|
* enumerator.c (enumerator_each): ensure that argument array size
|
|
does not overflow at appending.
|
|
|
|
Tue Aug 27 16:46:05 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* array.c (rb_ary_index, rb_ary_rindex): use optimized equality to
|
|
improve performance. [Feature #8820]
|
|
|
|
* vm_insnhelper.c (rb_equal_opt): optimized equality function.
|
|
|
|
Tue Aug 27 16:11:05 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_insnhelper.c (opt_eq_func): use RBASIC_CLASS() instead of HEAP_CLASS_OF().
|
|
|
|
* insns.def (opt_plus, opt_minus, opt_mult, opt_div, opt_mod, opt_lt),
|
|
(opt_gt, opt_ltlt, opt_aref, opt_aset, opt_length, opt_size),
|
|
(opt_empty_p, opt_succ): ditto.
|
|
|
|
Tue Aug 27 16:08:26 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_eval.c (rb_check_funcall, rb_check_funcall_with_hook): constify
|
|
argv.
|
|
|
|
Tue Aug 27 13:03:33 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/stringio/stringio.c (strio_read_nonblock): declare local
|
|
variables at the first of function.
|
|
|
|
Tue Aug 27 11:51:37 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
|
|
|
|
* enumerator.c: Allow Enumerator size argument to be any callable.
|
|
Patch by Avdi Grimm. [bug #8641] [ruby-core:56032] [fix GH-362]
|
|
|
|
* test/ruby/test_enumerator.rb: Test for above
|
|
|
|
Tue Aug 27 11:46:31 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_profile_clear): do rest_sweep() before clearing
|
|
profile.current_record.
|
|
|
|
Tue Aug 27 07:35:05 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* io.c (io_read_nonblock): support non-blocking reads without raising
|
|
exceptions. As in: `io.read_nonblock(size, exception: false)`
|
|
[ruby-core:38666] [Feature #5138]
|
|
* ext/openssl/ossl_ssl.c (ossl_ssl_read_internal): ditto
|
|
* ext/stringio/stringio.c (strio_sysread): ditto
|
|
* io.c (rb_io_write_nonblock): support non-blocking writes without
|
|
raising an exception.
|
|
* ext/openssl/ossl_ssl.c (ossl_ssl_write_internal): ditto
|
|
* test/openssl/test_pair.rb (class OpenSSL): tests
|
|
* test/ruby/test_io.rb (class TestIO): ditto
|
|
* test/socket/test_nonblock.rb (class TestSocketNonblock): ditto
|
|
* test/stringio/test_stringio.rb (class TestStringIO): ditto
|
|
|
|
Tue Aug 27 05:24:34 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Import RubyGems 2.1.0 Release Candidate
|
|
* test/rubygems: ditto.
|
|
|
|
Mon Aug 26 16:24:58 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_nextc): warn carriage return in middle of line.
|
|
[ruby-core:56240] [Feature #8699]
|
|
|
|
Mon Aug 26 15:27:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/timeout.rb (Timeout#timeout): should not be caught by rescue
|
|
clause. [Bug #8730]
|
|
|
|
Mon Aug 26 14:44:26 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (rb_ary_splice): use RARRAY_PTR_USE() without WB because
|
|
there are not new relations.
|
|
|
|
* enum.c (enum_sort_by): ditto.
|
|
|
|
* struct.c (setup_struct): use RARRAY_RAWPTR().
|
|
|
|
* vm_eval.c (yield_under): ditto.
|
|
|
|
* ext/pathname/pathname.c (path_entries): use RARRAY_AREF().
|
|
|
|
* ext/pathname/pathname.c (path_s_glob): ditto.
|
|
|
|
Mon Aug 26 13:11:10 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* array.c (ary_ensure_room_for_push): fix typo in r42658.
|
|
|
|
Mon Aug 26 12:37:10 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* template/sizes.c.tmpl: generate automatically by extracting
|
|
RUBY_CHECK_SIZEOF from configure.in.
|
|
|
|
Mon Aug 26 10:16:59 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* process.c (gcd_timetick_int): Renamed from gcd_timtick_int.
|
|
|
|
Sun Aug 25 21:02:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* sizes.c (Init_sizes): Define the size of clock_t.
|
|
|
|
Sun Aug 25 01:47:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (BARY_SHORT_MUL): Renamed from BARY_MUL1.
|
|
(bary_short_mul): Renamed from bary_mul1.
|
|
|
|
Sat Aug 24 10:35:09 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): The emulated clock names changed.
|
|
|
|
Fri Aug 23 22:22:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): Add a cast to fix compile error by
|
|
-Werror,-Wshorten-64-to-32.
|
|
|
|
Fri Aug 23 22:12:13 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* process.c (rb_intern): no symbol cache while initialization.
|
|
|
|
Fri Aug 23 22:07:45 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (clock_t): needs time.h.
|
|
|
|
Fri Aug 23 21:37:28 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (reduce_factors): New function.
|
|
(timetick2dblnum): Use reduce_factors.
|
|
(timetick2integer): Ditto.
|
|
(make_clock_result): Follow the above change.
|
|
(rb_clock_gettime): Ditto.
|
|
|
|
Fri Aug 23 21:00:55 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (timetick_int_t): Renamed from timetick_giga_count_t.
|
|
(gcd_timtick_int): Renamed from gcd_ul and make the arguments
|
|
timetick_giga_count_t.
|
|
(reduce_fraction): Make the arguments timetick_int_t.
|
|
(timetick2integer): Ditto.
|
|
(make_clock_result): Ditto.
|
|
(timetick2dblnum): Fix the return type.
|
|
(rb_clock_gettime): Use timetick_int_t.
|
|
|
|
Fri Aug 23 20:50:40 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (gcd_ul): New function.
|
|
(reduce_fraction): Ditto.
|
|
(reduce_fraction): Ditto.
|
|
(timetick2dblnum): Ditto.
|
|
(timetick2integer): Ditto.
|
|
(make_clock_result): Use timetick2dblnum and timetick2integer.
|
|
(rb_clock_gettime): Follow the make_clock_result change.
|
|
|
|
Fri Aug 23 18:39:04 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_make_shared): shared ary as shady. Need more effort to
|
|
make it normal object.
|
|
|
|
* array.c (rb_ary_modify): use RARRAY_PTR_USE() without WB because
|
|
there are not new relations.
|
|
|
|
* array.c (ary_ensure_room_for_unshift): use RARRAY_RAWPTR() because
|
|
there are not new relations.
|
|
|
|
Fri Aug 23 11:25:57 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c: introduce ARY_SHARED_OCCUPIED(shared).
|
|
|
|
Fri Aug 23 11:07:08 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/Makefile.sub (config.h): now SIZEOF_CLOCK_T is needed for
|
|
unsigned_clock_t.
|
|
|
|
Thu Aug 22 22:01:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): Strip "s" from unit names.
|
|
|
|
Thu Aug 22 20:14:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (unsigned_clock_t): Defined.
|
|
(rb_clock_gettime): Consider clock_t overflow for
|
|
ISO_C_CLOCK_BASED_CLOCK_PROCESS_CPUTIME_ID.
|
|
|
|
* configure.in: Check the size of clock_t.
|
|
|
|
Thu Aug 22 16:22:48 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* compile.c (build_postexe_iseq): fix to setup the local table.
|
|
|
|
Thu Aug 22 15:42:43 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* compile.c (rb_iseq_compile_node): accept NODE_IFUNC to support
|
|
custom compilation.
|
|
|
|
* compile.c (NODE_POSTEXE): compile to
|
|
"ONCE{ VMFrozenCore::core#set_postexe{...} }" with a new custom
|
|
compiler `build_postexe_iseq()'.
|
|
|
|
* vm.c (m_core_set_postexe): remove parameters (passed by a block).
|
|
|
|
Thu Aug 22 06:54:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): Change emulation symbols for
|
|
Process.clock_gettime.
|
|
|
|
Thu Aug 22 06:24:54 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (make_clock_result): Extracted from rb_clock_gettime.
|
|
|
|
Wed Aug 21 22:30:51 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): clock() based CLOCK_PROCESS_CPUTIME_ID
|
|
emulation implemented.
|
|
|
|
Wed Aug 21 21:02:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_proc_times): Use RB_GC_GUARD to guard objects from GC.
|
|
|
|
Wed Aug 21 20:33:01 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (get_clk_tck): Extracted from rb_proc_times.
|
|
(rb_clock_gettime): times() based CLOCK_PROCESS_CPUTIME_ID emulation
|
|
is implemented.
|
|
|
|
Wed Aug 21 19:31:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c: POSIX_GETTIMEOFDAY_CLOCK_REALTIME is renamed to
|
|
SUS_GETTIMEOFDAY_CLOCK_REALTIME.
|
|
|
|
Wed Aug 21 19:17:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): CLOCK_PROCESS_CPUTIME_ID emulation
|
|
using getrusage is implemented.
|
|
|
|
Wed Aug 21 17:34:27 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* gc.c (getrusage_time): Fallback clock_gettime to getrusage when
|
|
clock_gettime fails.
|
|
Reported by Eric Saxby. [ruby-core:56762] [Bug #8805]
|
|
|
|
Wed Aug 21 02:32:32 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* insns.def: fix regexp's once option behavior.
|
|
fix [ruby-trunk - Bug #6701]
|
|
|
|
* insns.def: remove `onceinlinecache' and introduce `once' instruction.
|
|
`once' doesn't use `setinlinecache' insn any more.
|
|
|
|
* vm_core.h: `union iseq_inline_storage_entry' to store once data.
|
|
|
|
* compile.c: catch up above changes.
|
|
|
|
* iseq.c: ditto.
|
|
|
|
* vm.c, vm_insnhelper.c: ditto. fix `m_core_set_postexe()' which
|
|
is depend on `onceinlinecache' insn.
|
|
|
|
* test/ruby/test_regexp.rb: add tests.
|
|
|
|
* iseq.c: ISEQ_MINOR_VERSION to 1 (should increment major?)
|
|
|
|
Wed Aug 21 02:30:15 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_gcdebug_print_obj_condition): add printing information.
|
|
|
|
Tue Aug 20 13:38:00 2013 Naohisa Goto <ngotogenome@gmail.com>
|
|
|
|
* test/gdbm/test_gdbm.rb: skip TestGDBM#test_s_open_lock on Solaris.
|
|
On Solaris (and platforms which do not have flock and have lockf),
|
|
with GDBM 1.10, gdbm_open(3) blocks when opening already locked
|
|
gdbm file. [Bug #8790] [ruby-dev:47631]
|
|
|
|
Tue Aug 20 02:32:52 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/test/: [DOC] Document Test::Unit, hide most submodules and
|
|
classes from rdoc. Since lib/test is only present as a compatibility
|
|
layer with the legacy test suite many test/unit users will be using
|
|
minitest or the test/unit gem instead. It is recommended to use one
|
|
of these alternatives for writing new tests.
|
|
|
|
This patch was based on a patch submitted by Steve Klabnik.
|
|
[ruby-core:56694] [Bug #8778]
|
|
|
|
Tue Aug 20 02:10:19 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rss/rss.rb: [DOC] Document for constants by Steve Klabnik
|
|
[ruby-core:56705] [Bug #8798]
|
|
|
|
Tue Aug 20 02:01:10 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rss/xmlparser.rb: [DOC] Hide legacy constant from rdoc
|
|
Patch by Steve Klabnik [ruby-core:56708] [Bug #8799]
|
|
|
|
Tue Aug 20 01:52:05 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/socket/unixserver.c: [DOC] Document #accept
|
|
* ext/socket/tcpserver.c: ditto
|
|
* ext/socket/udpsocket.c: [DOC] Fix indentation of documentation
|
|
* ext/socket/socket.c: ditto
|
|
Patches by David Rodr'iguez [ruby-core:56734] [Bug #8802]
|
|
|
|
Tue Aug 20 01:19:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Define ac_cv_func_clock_gettime to yes for mingw*.
|
|
|
|
Mon Aug 19 21:31:35 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/defines.h: Fix a compilation error with
|
|
i586-mingw32msvc-gcc of gcc-mingw32 package on Debian squeeze.
|
|
ruby/missing.h should be included before include/ruby/win32.h
|
|
because struct timespec, used in the clock_gettime declaration in
|
|
include/ruby/win32.h, is defined in ruby/missing.h instead of
|
|
system headers.
|
|
|
|
Mon Aug 19 20:55:12 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: fix around GC_DEBUG.
|
|
|
|
* gc.c (RVALUE::line): should be VALUE. On some environment
|
|
(such as mswin64), `int' introduces alignment mismatch.
|
|
|
|
* gc.c (newobj_of): add an assertion to check VALUE alignment.
|
|
|
|
* gc.c (aligned_malloc): `&' is low priority than `=='.
|
|
|
|
* gc.c: define GC_DEBUG everytime and use it as value 0 or 1.
|
|
|
|
Mon Aug 19 17:43:44 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/ruby/test_fiber.rb: collect garbage fibers immediately.
|
|
|
|
Mon Aug 19 17:41:49 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/profile_test_all.rb: add `failed?' information.
|
|
|
|
Mon Aug 19 17:00:53 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* process.c (retry_fork): retry with GC if ENOMEM occurred, to free
|
|
swap/kernel space.
|
|
|
|
Mon Aug 19 13:28:47 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* include/ruby/win32.h (CLOCK_MONOTONIC): typo.
|
|
|
|
* win32/win32.c: removed duplicated declarations.
|
|
|
|
Mon Aug 19 13:03:08 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (clock_gettime): should not overwrite cache variable
|
|
with different condition. otherwise -lrt is not linked and the link
|
|
fails, after reconfig.
|
|
|
|
Mon Aug 19 12:56:49 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (Init_process): Add constants: CLOCK_REALTIME_ALARM and
|
|
CLOCK_BOOTTIME_ALARM.
|
|
|
|
Sun Aug 18 20:17:41 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* variable.c, vm_method.c: remove dead code.
|
|
|
|
* test/ruby/test_fiber.rb, test/ruby/test_thread.rb:
|
|
change accordingly.
|
|
|
|
Sun Aug 18 19:32:26 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* error.c, file.c, gc.c, hash.c, thread.c, variable.c, vm_eval.c, bin/erb:
|
|
$SAFE=4 is obsolete.
|
|
|
|
Sun Aug 18 14:30:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): Rename POSIX_TIME_CLOCK_REALTIME to
|
|
ISO_C_TIME_CLOCK_REALTIME.
|
|
|
|
Sun Aug 18 14:22:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Revert r42604. It causes linking librt on systems
|
|
with newer glibc uselessly.
|
|
|
|
Sun Aug 18 13:18:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (Init_process): Add constants: CLOCK_REALTIME_COARSE,
|
|
CLOCK_MONOTONIC_COARSE and CLOCK_BOOTTIME.
|
|
|
|
Sun Aug 18 12:41:50 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (clock_gettime): need to check with -lrt prior to check
|
|
for the function only. otherwise -lrt is not linked and the link
|
|
fails, when ac_cv_func_clock_gettime is cached as yes.
|
|
|
|
Sun Aug 18 10:05:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big2str1): Make an expression more explicit.
|
|
|
|
Sun Aug 18 03:18:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big2str1): Use power_level instead of bitsize(xn).
|
|
|
|
Sun Aug 18 00:44:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (BIGDIVREM_EXTRA_WORDS): Redefine to 1.
|
|
(bigdivrem_num_extra_words): Removed.
|
|
(bigdivrem_normal): Simplified.
|
|
(big2str_karatsuba): Ditto.
|
|
|
|
Sat Aug 17 23:25:19 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* test/ruby/test_time.rb: use the in_timezone() helper
|
|
and define it at the top with other helpers.
|
|
|
|
Sat Aug 17 22:20:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* time.c (time_mload): ignore auxiliary data, offset and zone, if
|
|
invalid. [ruby-core:56648] [Bug #8795]
|
|
|
|
Sat Aug 17 20:11:49 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* process.c: [DOC] MACH_ABSOLUTE_TIME_CLOCK_MONOTONIC is an
|
|
available emulation for a monotonic clock on Darwin.
|
|
https://developer.apple.com/library/mac/qa/qa1398/_index.html
|
|
|
|
Fri Aug 16 18:12:05 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/profile_test_all.rb: fix typo.
|
|
|
|
Fri Aug 16 18:09:20 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/profile_test_all.rb: remove space characters from test names.
|
|
|
|
Fri Aug 16 17:32:02 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/profile_test_all.rb: refactoring memory profiling tool for
|
|
test-all.
|
|
Add profiling targets /proc/meminfo and /proc/self/status.
|
|
|
|
* test/runner.rb: accept other than 'true'.
|
|
|
|
Fri Aug 16 11:23:35 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* file.c (rb_file_size, rb_file_flock): improve performance of Windows.
|
|
|
|
* file.c (rb_file_truncate): removed unnecessary #ifdef.
|
|
|
|
* test/test_file.rb (TestFile#test_truncate_size): added an assertion
|
|
for File#size.
|
|
|
|
Fri Aug 16 10:07:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem_single1): Renamed from bigdivrem_single. Add
|
|
x_higher_bdigit argument.
|
|
(bigdivrem_single): Just call bigdivrem_single1.
|
|
(bigdivrem_restoring): Use bigdivrem_single1 to avoid memmove.
|
|
|
|
Fri Aug 16 09:17:00 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_small_rshift): Specify the higher BDIGIT instead of
|
|
sign bit.
|
|
(big_shift3): Follow the above change.
|
|
|
|
Fri Aug 16 02:20:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul_toom3): Reduce a branch.
|
|
|
|
Fri Aug 16 02:14:09 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* process.c (rb_clock_gettime): add CLOCK_MONOTONIC support on OS X.
|
|
http://developer.apple.com/library/mac/qa/qa1398/_index.html
|
|
[Feature #8658]
|
|
|
|
Fri Aug 16 01:37:43 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem_single): Use shift when y is a power of two.
|
|
|
|
Fri Aug 16 01:09:33 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem_restoring): Use bigdivrem_single if non-topmost
|
|
BDIGITs of y are zero.
|
|
|
|
Fri Aug 16 00:33:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big2str1): Truncate topmost zeros of x.
|
|
|
|
Fri Aug 16 00:00:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_divmod): Simplify an expression.
|
|
|
|
Thu Aug 15 23:26:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem_normal): Remove a local variable.
|
|
|
|
Thu Aug 15 23:08:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_karatsuba): Use bigdivrem_restoring directly to
|
|
reduce working buffer and memory copy.
|
|
(rb_big2str1): Allocate working buffer for big2str_karatsuba here.
|
|
|
|
Thu Aug 15 20:51:29 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* io.c, internal.h (rb_io_flush_raw): new function to select calling
|
|
fsync() (on Windows).
|
|
|
|
* io.c (rb_io_flush_raw): use above function.
|
|
|
|
* file.c (rb_file_truncate): use above function.
|
|
|
|
* test/ruby/test_file.rb (TestFile#test_truncate_size): test for
|
|
above changes.
|
|
|
|
Thu Aug 15 18:39:31 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/win32.c (clock_gettime): improve precision when freq is less
|
|
than and nearly equals 10**9.
|
|
|
|
Thu Aug 15 17:43:15 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_lazy_sweep): remove heap_increment() here because heap_inc
|
|
may be 0.
|
|
|
|
Thu Aug 15 16:59:56 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* io.c (rb_io_rewind): remove fsync() for Windows to improve the
|
|
performance.
|
|
|
|
Thu Aug 15 16:30:23 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/fileutils/test_fileutils.rb (TestFileUtils#test_rmdir):
|
|
FileUtils.rmdir ignores Errno::ENOTEMPTY, so, in such cases, this
|
|
assertion is nonsense.
|
|
|
|
Thu Aug 15 15:49:35 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): [DOC] FreeBSD 7.1 supports
|
|
CLOCK_THREAD_CPUTIME_ID.
|
|
http://www.freebsd.org/releases/7.1R/relnotes.html
|
|
|
|
Thu Aug 15 14:30:23 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* include/ruby/win32.h, win32/Makefile.sub, win32/win32.c
|
|
(clock_gettime): [experimental] emulates clock_gettime(2) of posix.
|
|
|
|
Thu Aug 15 02:32:40 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* hash.c (rb_hash_aset): [DOC] Document key dup patch by @kachick
|
|
[Fixes GH-382] https://github.com/ruby/ruby/pull/382
|
|
|
|
Wed Aug 14 14:28:39 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* proc.c (rb_mod_define_method): now they return the symbols of the
|
|
defined methods, not the methods/procs themselves.
|
|
[ruby-dev:42151] [Feature #3753]
|
|
|
|
* NEWS: documents about above change and def-expr (see r42337).
|
|
|
|
* test/ruby/test_module.rb: tests about above change.
|
|
|
|
Wed Aug 14 00:51:14 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem_restoring): xn argument removed.
|
|
(bigdivrem_normal): Follow the above change.
|
|
|
|
Wed Aug 14 00:18:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big_div_struct): Remove xn and j field. Add zn field.
|
|
(bigdivrem1): Follow the above change.
|
|
(bigdivrem_restoring): Ditto.
|
|
|
|
Tue Aug 13 23:38:17 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big_div_struct): ynzero field removed.
|
|
(bigdivrem1): Follow the above change.
|
|
(bigdivrem_restoring): Ditto.
|
|
|
|
Tue Aug 13 23:01:16 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem_restoring): Extracted from bigdivrem_normal.
|
|
|
|
Tue Aug 13 22:12:59 2013 Kenichi Kamiya <kachick1@gmail.com>
|
|
|
|
* random.c (rb_random_ulong_limited): coerce before check negative.
|
|
[Fixes GH-379]
|
|
|
|
Tue Aug 13 21:52:15 2013 Kenichi Kamiya <kachick1@gmail.com>
|
|
|
|
* object.c (Init_Object): undef Module#prepend_features on Class, as
|
|
well as Module#append_features. [Fixes GH-376]
|
|
|
|
* test_class.rb: Added test for above. And ensure type checking
|
|
on similar methods as module_function.
|
|
|
|
Tue Aug 13 08:52:18 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* doc/syntax/literals.rdoc: [DOC] String literal concat by @cknadler
|
|
[Fixes GH-380] https://github.com/ruby/ruby/pull/380
|
|
|
|
Mon Aug 12 23:07:21 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (gc_marks_test): inhibit gc for st's operation.
|
|
|
|
Mon Aug 12 15:59:50 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_whole_match_p): treat CR in middle of a line as a
|
|
mere whitespace.
|
|
|
|
Mon Aug 12 15:16:58 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* class.c (rb_prepend_module): make T_ICLASS object shady because
|
|
this T_ICLASS object seems to share method table with other class
|
|
objects. It was causes WB miss.
|
|
TODO: need to know the data structure.
|
|
|
|
* test/ruby/test_module.rb: add a test for WB miss.
|
|
|
|
Mon Aug 12 13:47:54 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* process.c: [DOC] RDoc formatting of Process.clock_gettime
|
|
|
|
Mon Aug 12 13:29:09 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/yaml/dbm.rb: [DOC] Document call-seq for YAML::DBM
|
|
|
|
Mon Aug 12 12:57:26 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/dbm/extconf.rb: [DOC] Hide from RDoc
|
|
Some libraries might want to document extconf.rb so RDoc treats it
|
|
like any other ruby program. However, DBM users shouldn't care about
|
|
these methods.
|
|
|
|
Mon Aug 12 12:53:39 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/dbm/dbm.c: [DOC] Reformat headings of DBM class
|
|
|
|
Mon Aug 12 12:46:31 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/yaml.rb, lib/yaml/: [DOC] Document YAML::DBM#key and add
|
|
references to similar methods with more detail. This patch brings
|
|
lib/yaml to 100% documentation coverage.
|
|
|
|
Mon Aug 12 02:51:32 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/readline/readline.c (readline_s_set_input): on OS X with editline,
|
|
Readline.readline doesn't work because readline_get doesn't use
|
|
rl_getc. The difference is introduced by r42402 [ruby-dev:47509]
|
|
[Bug #8644]. Before it rb_io_stdio_file set ifp->stdio_file.
|
|
Therefore add manually setting the value.
|
|
|
|
* ext/readline/readline.c (readline_s_set_output): ditto.
|
|
|
|
Sun Aug 11 23:27:00 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* file.c (rb_str_encode_ospath): OS path encoding on Mac OS X is also
|
|
fixed.
|
|
|
|
Sun Aug 11 22:57:24 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* test/ruby/test_require.rb (assert_require_nonascii_path): OS path
|
|
encoding on Windows is fixed, so encoding of __FILE__ should be it.
|
|
[ruby-core:56498] [Bug #8764]
|
|
|
|
Sun Aug 11 19:11:45 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/parser/test_sax2.rb: Expand abbreviated class name.
|
|
|
|
Sun Aug 11 19:06:03 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/sax2listener.rb (REXML::SAX2Listener#notationdecl): Fix
|
|
wrong number of arguments in the template listener.
|
|
[Bug #8731] [ruby-dev:47582]
|
|
Reported by Ippei Obayashi.
|
|
* test/rexml/parser/test_sax2.rb: Add tests for parsing notation
|
|
declarations with SAX2 API.
|
|
|
|
Sun Aug 11 18:44:04 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/sax2listener.rb (REXML::SAX2Listener#elementdecl): Fix wrong
|
|
examples. [Bug #8731] [ruby-dev:47582]
|
|
Reported by Ippei Obayashi.
|
|
|
|
Sun Aug 11 18:42:13 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/sax2parser.rb
|
|
(REXML::Parsers::SAX2Parser#handle_entitydecl): Extract.
|
|
|
|
Sun Aug 11 18:40:25 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/sax2parser.rb (REXML::Parsers::SAX2Parser#parse):
|
|
Fix wrong "%" position in parameter entity declaration event argument.
|
|
* test/rexml/parser/test_sax2.rb: Add tests for the above case.
|
|
|
|
Sun Aug 11 18:08:40 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/sax2parser.rb (REXML::Parsers::SAX2Parser#parse):
|
|
Support NDATA in external ID entity declaration.
|
|
* test/rexml/parser/test_sax2.rb: Add tests for the above case.
|
|
|
|
Sun Aug 11 18:07:39 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/baseparser.rb
|
|
(REXML::Parsers::BaseParser#pull_event): Support optional NDATA
|
|
in external ID entity declaration.
|
|
|
|
Sun Aug 11 17:54:07 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* NEWS (REXML::Parsers::SAX2Parser): Add about this change.
|
|
* lib/rexml/parsers/sax2parser.rb (REXML::Parsers::SAX2Parser#parse):
|
|
Fix wrong number of arguments. Document says "an array of the
|
|
entity declaration" but it passes two or more arguments.
|
|
This is a bug but it break backward compatibility.
|
|
Reported by Ippei Obayashi. [Bug #8731] [ruby-dev:47582]
|
|
* lib/rexml/sax2listener.rb (REXML::SAX2Listener#entitydecl): ditto.
|
|
The listener template accepted two arguments.
|
|
* test/rexml/parser/test_sax2.rb: Add tests for external ID case.
|
|
|
|
Sun Aug 11 17:41:41 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/parser/test_sax2.rb: Add SAX2 API test.
|
|
|
|
Sun Aug 11 15:10:40 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (rb_enc_symname_type): allow ID_ATTRSET for ID_INSTANCE,
|
|
ID_GLOBAL, ID_CLASS, ID_JUNK too. [Bug #8756]
|
|
|
|
Sun Aug 11 13:17:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* include/ruby/encoding.h: Reduce ENCODING_INLINE_MAX to 127 as this
|
|
should be sufficient to represent all the encodings Ruby supports.
|
|
|
|
Sun Aug 11 11:54:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (rb_clock_gettime): New method.
|
|
This is accepted in the meeting:
|
|
https://bugs.ruby-lang.org/projects/ruby/wiki/DevelopersMeeting20130809
|
|
This method is accepted as a CRuby feature.
|
|
I.e. Other Ruby implementations don't need to implement it.
|
|
[ruby-core:56087] [Feature #8658]
|
|
|
|
Sun Aug 11 10:40:48 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/time.rb: [DOC] Correcting rdoc visibility of time.rb constants
|
|
Reported by Tanaka Akira [ruby-core:56517]
|
|
|
|
Sun Aug 11 04:48:14 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* file.c (rb_str_normalize_ospath):
|
|
HFS Plus (Mac OS Extended) uses a variant of Normal Form D in which
|
|
U+2000 through U+2FFF, U+F900 through U+FAFF, and U+2F800 through
|
|
U+2FAFF are not decomposed (this avoids problems with round trip
|
|
conversions from old Mac text encodings).
|
|
http://developer.apple.com/library/mac/qa/qa1173/_index.html
|
|
Therefore fix r42457 to exclude the range.
|
|
|
|
Sun Aug 11 03:26:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bitsize): Fix a conditional expression.
|
|
|
|
Sun Aug 11 02:44:03 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/time.rb: [DOC] Document constants by @markijbema [Fixes GH-377]
|
|
https://github.com/ruby/ruby/pull/377
|
|
|
|
Sun Aug 11 01:28:52 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Revert r42458.
|
|
It removes the HAVE_CLOCK_GETTIME from config.h.
|
|
http://www.rubyist.net/~akr/chkbuild/debian/ruby-trunk/log/20130809T044800Z.diff.html.gz
|
|
|
|
Sat Aug 10 13:53:22 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (rb_id_attrset): allow other than ID_ATTRSET.
|
|
|
|
* parse.y (intern_str): ditto. try stem ID for ID_INSTANCE,
|
|
ID_GLOBAL, ID_CLASS, ID_JUNK too. [Bug #8756]
|
|
|
|
Sat Aug 10 12:49:50 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/baseparser.rb
|
|
(REXML::Parsers::BaseParser::CDATA_END): Use "\A" instead of "^".
|
|
It is not an used constant but I fix it. (Or should I remove it?)
|
|
|
|
Sat Aug 10 12:47:19 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser):
|
|
Fix wrong constant name. "]>" pattern match is the same but
|
|
it is used for "<!DOCTYPE" end mark not "<![CDATA[" end mark.
|
|
|
|
Sat Aug 10 12:43:15 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser):
|
|
Use "\A" instead of "^" in document type declaration patterns
|
|
because they are used as the head match in content not the head
|
|
match in line. They don't cause any problems in the current code
|
|
but it should be fixed.
|
|
|
|
Sat Aug 10 12:39:00 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/parse/test_document_type_declaration.rb: Add tests for
|
|
parsing document type declaration.
|
|
|
|
Sat Aug 10 12:00:45 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser::SYSTEM):
|
|
Fix loose "head" match regular expression. It doesn't cause any
|
|
problem in the current code but it should be fixed because readers
|
|
may confuse it.
|
|
Patch by Ippei Obayashi. Thanks!!!
|
|
|
|
Sat Aug 10 11:58:24 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/parse/test_notation_declaration.rb (#test_system_public):
|
|
Add a test for PUBLIC notation and SYSTEM notation order case.
|
|
|
|
Sat Aug 10 11:31:35 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser::PUBLIC):
|
|
Fix loose "head" match regular expression.
|
|
[Bug #8701] [ruby-dev:47551]
|
|
Patch by Ippei Obayashi. Thanks!!!
|
|
* test/rexml/parse/test_notation_declaration.rb (#test_system_public):
|
|
Add a test for the above case.
|
|
|
|
Sat Aug 10 09:20:21 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* NEWS: [DOC] typo in example reported by @moretea
|
|
https://github.com/ruby/ruby/commit/a39e724#commitcomment-3831489
|
|
|
|
Sat Aug 10 09:19:04 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* proc.c: [DOC] rdoc code formatting
|
|
|
|
Sat Aug 10 09:12:01 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (rb_id_attrset): check if the argument is valid type as an
|
|
attribute.
|
|
|
|
Sat Aug 10 05:44:08 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rss/trackback.rb: [DOC] Hide RSS::Trackback from rdoc
|
|
Patch by Steve Klabnik [Bug #8755] [ruby-core:56456]
|
|
|
|
Sat Aug 10 04:52:21 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big_div_struct): Use size_t.
|
|
(bigdivrem1): Ditto.
|
|
(bigdivrem_num_extra_words): Ditto.
|
|
(bigdivrem_single): Ditto.
|
|
(bigdivrem_normal): Ditto.
|
|
(bary_divmod): Ditto.
|
|
|
|
Fri Aug 9 23:47:15 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rss/rexmlparser.rb: Remove needless REXML version check.
|
|
Both RSS Parser and REXML are bundled in Ruby. RSS Parser can
|
|
always use the latest REXML. [Bug #8754] [ruby-core:56454]
|
|
Patch by Steve Klabnik. Thanks!!!
|
|
|
|
Fri Aug 9 22:51:10 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (XLDFLAGS, LIBRUBYARG_STATIC): CoreFoundation framework
|
|
option is now needed always, regardless enable-shared.
|
|
[ruby-core:56467] [Bug #8759]
|
|
|
|
Fri Aug 9 22:20:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ruby.c (load_file_internal): use rb_parser_compile_string_path and
|
|
rb_parser_compile_file_path, String path name versions. [Bug #8753]
|
|
|
|
Fri Aug 9 07:16:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* ext/io/console/console.c: delete redefinition of rb_cloexec_open.
|
|
drop support for 1.8 and 1.9 from the next release of io-console gem.
|
|
|
|
Fri Aug 9 19:13:54 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* NEWS: update about new methods for Binding.
|
|
|
|
Fri Aug 9 18:48:09 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* proc.c: add Binding#local_variable_get/set/defined?
|
|
to access local variables which a binding contains.
|
|
Most part of implementation by nobu.
|
|
|
|
* test/ruby/test_proc.rb: add a tests for above.
|
|
|
|
* vm.c, vm_core.h (rb_binding_add_dynavars): add a new function
|
|
to add a new environment to create space for new local variables.
|
|
|
|
Fri Aug 9 14:02:01 2013 SHIBATA Hiroshi <shibata.hiroshi@gmail.com>
|
|
|
|
* tool/make-snapshot: Fix order of priority for option parameter.
|
|
|
|
Fri Aug 9 12:06:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* file.c (rb_str_normalize_ospath): normalize to Normalization Form C
|
|
using CFString.
|
|
|
|
Fri Aug 9 10:53:57 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* time.c (get_timeval, get_new_timeval): use rb_obj_class()
|
|
instead of CLASS_OF() because CLASS_OF() may return
|
|
a singleton class.
|
|
|
|
Fri Aug 9 10:42:11 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* vm_insnhelper.c (vm_invoke_block): returning from lambda proc
|
|
now always exits from the Proc. [ruby-core:56193] [Feature #8693]
|
|
|
|
* NEWS, test/ruby/test_lambda.rb: ditto. Patch by nobu.
|
|
|
|
Fri Aug 9 00:10:32 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enumerator.c (lazy_zip_func): fix non-single argument. fix
|
|
out-of-bound access and pack multiple yielded values.
|
|
[ruby-core:56383] [Bug #8735]
|
|
|
|
Thu Aug 8 23:01:20 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* object.c (rb_mod_singleton_p): new method Module#singleton_class? to
|
|
return whether the receiver is a singleton class or not.
|
|
[ruby-core:51087] [Feature #7609]
|
|
|
|
Thu Aug 8 21:56:44 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* time.c (time_overflow_p): Avoid signed integer overflow.
|
|
(rb_time_new): Fix overflow condition.
|
|
|
|
Thu Aug 8 19:58:02 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* thread.c (rb_threadptr_pending_interrupt_check_mask):
|
|
use RARRAY_RAWPTR() instead of RARRAY_PTR() because
|
|
there is no new reference.
|
|
|
|
Thu Aug 8 19:56:52 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* string.c (rb_str_format_m): use RARRAY_RAWPTR() instead of
|
|
RARRAY_PTR() because there is no new reference.
|
|
|
|
Thu Aug 8 19:55:51 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: define USE_RGENGC_LOGGING_WB_UNPROTECT.
|
|
|
|
Thu Aug 8 16:44:25 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: add old macro name `RUBY_EVENT_SWITCH'.
|
|
This macro name is obsolete because it is renamed to
|
|
RUBY_INTERNAL_EVENT_SWITCH, but it has compatibility problem
|
|
using this macro name like ruby-prof.
|
|
I want to remove this macro after ruby 2.1.
|
|
|
|
Thu Aug 8 15:37:53 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/coverage/test_coverage.rb (TestCoverage#test_big_code): use `1'
|
|
instead of `p' to get rid of a side effect.
|
|
Kernel#p without any argument seems to do nothing, but flushes stdout.
|
|
and, if stdout is redirected to file, fsync() will be called on
|
|
Windows. so, when running test-all on Windows with redirection, such
|
|
as CI environment, this test took a lot of time.
|
|
|
|
Thu Aug 8 14:54:18 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* NEWS: add description of incompatibility introduced by r42396.
|
|
[ruby-core:56329] [Bug #8722]
|
|
|
|
Thu Aug 8 14:50:36 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* common.mk (mini): portable target to build miniruby
|
|
|
|
* common.mk (bisect): run git-bisect with miniruby
|
|
|
|
* common.mk (bisect-ruby): run git-bisect with ruby
|
|
|
|
* tool/bisect.sh: script for git-bisect
|
|
|
|
Thu Aug 8 12:11:43 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/webrick/test_httpresponse.rb (test_send_body_*_chunked): these
|
|
expectations assumes that the IOs are binmode. fixed test failures
|
|
introduced at r42427 on Windows.
|
|
|
|
Thu Aug 8 10:27:18 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* range.c (range_last): revert r42400. [Bug #8739]
|
|
|
|
Thu Aug 8 10:26:25 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* file.c (rb_str_normalize_ospath): extract and move from dir.c.
|
|
|
|
Thu Aug 8 05:59:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* test/openssl/test_ssl.rb: Fix test for CVE-2013-4073.
|
|
Patch by Antonio Terceiro. [Bug #8750] [ruby-core:56437]
|
|
|
|
Thu Aug 8 03:37:38 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/webrick/httpresponse.rb: Allow #body to be an IO-like object
|
|
that responds to #readpartial and #read.
|
|
[ruby-trunk - Feature #8155]
|
|
* NEWS: NEWS for above
|
|
* test/webrick/test_httpresponse.rb: Tests for above.
|
|
|
|
Wed Aug 7 23:06:26 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* ruby.c (Process.argv0): New method to return the original value
|
|
of $0. [Feature #8696]
|
|
|
|
Wed Aug 7 23:05:55 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* ruby.c (Process.setproctitle): New method to change the title of
|
|
the running process that is shown in ps(1). [Feature #8696]
|
|
|
|
Wed Aug 7 20:05:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_odd_p): Check the bignum length.
|
|
(rb_big_even_p): Ditto.
|
|
|
|
Wed Aug 7 19:29:26 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (dbl2big): A condition simplified.
|
|
|
|
Wed Aug 7 16:34:30 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/webrick/test_cgi.rb (TestWEBrickCGI#{start_cgi_server,test_cgi}):
|
|
mswin is not only mswin32 but also mswin64. [Bug #8746]
|
|
|
|
Wed Aug 7 16:19:12 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* cont.c (rb_fiber_start): use RARRAY_RAWPTR() instead of
|
|
RARRAY_PTR() because there is no new reference.
|
|
|
|
* proc.c (curry): ditto.
|
|
|
|
* proc.c (rb_proc_call): remove line break.
|
|
|
|
Wed Aug 7 13:20:12 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* random.c (random_load): use RARRAY_RAWPTR() instead of
|
|
RARRAY_PTR() because there is no new reference.
|
|
|
|
Wed Aug 7 12:58:23 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* thread.c (thread_start_func_2): use RARRAY_RAWPTR() instead of
|
|
RARRAY_PTR() because there is no new reference.
|
|
|
|
Wed Aug 7 09:00:24 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* string.c: [DOC] Description of rb_str_equal [Fixes GH-375]
|
|
Based on a patch by @markijbema
|
|
https://github.com/ruby/ruby/pull/375
|
|
|
|
Wed Aug 7 08:30:38 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/openssl/ossl_hmac.c: [DOC] Documentation for OpenSSL::HMAC
|
|
based on a patch by @repah documenting-ruby/ruby#14
|
|
https://github.com/documenting-ruby/ruby/pull/14
|
|
|
|
Wed Aug 7 07:46:23 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rss/utils.rb: [DOC] RSS::Utils by Steve Klabnik [Bug #8745]
|
|
|
|
Wed Aug 7 07:38:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (nlz16): Removed.
|
|
(nlz32): Ditto.
|
|
(nlz64): Ditto.
|
|
(nlz128): Ditto.
|
|
(nlz_int): New function.
|
|
(nlz_long): New function.
|
|
(nlz_long_long): New function.
|
|
(nlz_int128): New function.
|
|
(nlz): Follow above changes.
|
|
(bitsize): Follow above changes.
|
|
|
|
Tue Aug 6 22:38:15 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* time.c: [DOC] Typo in Time overview by @sparr [Fixes GH-374]
|
|
https://github.com/ruby/ruby/pull/374
|
|
|
|
Tue Aug 6 22:35:32 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rss/1.0.rb: [DOC] Document RSS10 by Steve Klabnik [Bug #8740]
|
|
|
|
Tue Aug 6 22:14:11 2013 Kouji Takao <kouji.takao@gmail.com>
|
|
|
|
* ext/readline/readline.c (readline_s_delete_text): remove
|
|
checking "$SAFE == 4".
|
|
|
|
* ext/readline/readline.c: fix rdoc, remove "Raises SecurityError"
|
|
and add "Raises NotImplementedError".
|
|
|
|
Tue Aug 6 22:04:38 2013 Kouji Takao <kouji.takao@gmail.com>
|
|
|
|
* ext/readline/readline.c, test/readline/test_readline.rb: fix
|
|
indent.
|
|
|
|
Tue Aug 6 21:59:56 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* range.c (range_last): return nil for empty range, or in the case the
|
|
predecessor is smaller than the begin. [Bug #8739]
|
|
|
|
Tue Aug 6 21:48:31 2013 Kouji Takao <kouji.takao@gmail.com>
|
|
|
|
* ext/readline/readline.c (readline_s_set_point, Init_readline):
|
|
add Readline.point=(pos). Patched by naruse. [ruby-dev:47535]
|
|
[Feature #8675]
|
|
|
|
Tue Aug 6 21:14:11 2013 Kouji Takao <kouji.takao@gmail.com>
|
|
|
|
* ext/readline/readline.c (Init_readline, readline_s_set_output)
|
|
(clear_rl_outstream, readline_s_set_input, clear_rl_instream)
|
|
(readline_readline): fix causing SEGV if closed IO object that is
|
|
set Readline.input or Readline.output. Patched by akr
|
|
[ruby-dev:47509] [Bug #8644]
|
|
|
|
Tue Aug 6 17:56:40 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_insnhelper.c (vm_push_frame): change type of stack_max to size_t.
|
|
|
|
Tue Aug 6 17:42:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* range.c (range_last): exclude the last number of the exclusive range
|
|
if the end is Numeric. [ruby-dev:47587] [Bug #8739]
|
|
|
|
Tue Aug 6 17:42:21 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (rb_w32_conv_from_wchar): converted string to CP_UTF8
|
|
should have UTF-8 encoding. otherwise no conversion takes place
|
|
later.
|
|
|
|
Tue Aug 6 17:21:38 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_insnhelper.c (vm_push_frame): fix stack overflow check codes.
|
|
Stack overflow check should be done *after* pushing a stack frame.
|
|
However, some stack overflow checking codes checked *before*
|
|
pushing a stack frame with iseq->stack_max.
|
|
To solve this problem, add a new parameter `stack_max' to specify
|
|
a possible consuming stack size.
|
|
|
|
* vm_core.h (CHECK_VM_STACK_OVERFLOW0): add to share the stack overflow
|
|
checking code.
|
|
|
|
* insns.def: catch up this change.
|
|
|
|
* vm.c, vm_eval.c: ditto.
|
|
|
|
* test/ruby/test_exception.rb: add a stack overflow test.
|
|
This code is reported by nobu.
|
|
|
|
Tue Aug 6 17:02:17 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (rb_w32_conv_from_wchar): use WideCharToMultiByte(),
|
|
as like as mbstr_to_wstr(), in the first step of the conversion from
|
|
WCHAR.
|
|
|
|
Tue Aug 6 16:14:32 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* vm_eval.c (eval_string_with_cref): copy cref to limit the scope of
|
|
refinements in the eval string. [ruby-core:56329] [Bug #8722]
|
|
|
|
* test/ruby/test_refinement.rb: related test.
|
|
|
|
Tue Aug 6 12:23:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_realloc): Use VALGRIND_MAKE_MEM_UNDEFINED to
|
|
declare undefined memory area.
|
|
(bignew_1): Ditto.
|
|
|
|
* internal.h (VALGRIND_MAKE_MEM_DEFINED): Moved from gc.c
|
|
(VALGRIND_MAKE_MEM_UNDEFINED): Ditto.
|
|
|
|
Tue Aug 6 01:40:37 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* process.c: [DOC] Document caveats of command form of Process.spawn
|
|
with regard to the shell and OS. Patched by Steve Klabnik [Bug #8550]
|
|
|
|
Tue Aug 6 01:28:35 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rss/0.9.rb: [DOC] Typo in example [Bug #8732]
|
|
|
|
Tue Aug 6 01:22:37 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rss/2.0.rb: [DOC] Document RSS::Rss by Steve Klabnik #8740
|
|
* lib/rss/atom.rb: [DOC] Typo in rdoc by Steve Klabnik
|
|
|
|
Mon Aug 5 23:47:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Rename local variables.
|
|
|
|
Mon Aug 5 22:23:59 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* vm_trace.c: [DOC] Fix TracePoint return values in examples
|
|
Based on a patch by @sho-h [Fixes GH-373]
|
|
https://github.com/ruby/ruby/pull/373
|
|
|
|
Mon Aug 5 17:38:15 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (rb_w32_write_console): use MultiByteToWideChar() for
|
|
the last step of conversion to WCHAR, to get rid of warnings from
|
|
rb_enc_find() in miniruby. [ruby-dev:47584] [Bug #8733]
|
|
|
|
* win32/win32.c (wstr_to_mbstr, mbstr_to_wstr): fix wrong trimming.
|
|
WideCharToMultiByte() and MultiByteToWideChar() do not count
|
|
NUL-terminator in the size for conversion result, unless the input
|
|
length is -1.
|
|
|
|
Mon Aug 5 11:51:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* include/ruby/encoding.h: document which user flags are used by
|
|
ENCODING_MASK for better greppability
|
|
|
|
Mon Aug 5 10:01:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* object.c (rb_class_inherited_p): allow iclasses to be tested for
|
|
inheritance. [Bug #8686] [ruby-core:56174]
|
|
|
|
* test/ruby/test_method.rb: add test
|
|
|
|
Mon Aug 5 06:13:48 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* enumerator.c: [DOC] Remove reference to Enumerator::Lazy#cycle
|
|
Patch by @kachick [Fixes GH-372]
|
|
https://github.com/ruby/ruby/pull/372
|
|
|
|
Mon Aug 5 03:57:16 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rss/0.9.rb: [DOC] Document RSS09 by Steve Klabnik [Bug #8732]
|
|
|
|
Mon Aug 5 03:35:11 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rexml/attribute.rb: [DOC] Update example for #namespace
|
|
Patch by Ippei Obayashi [Bug #8685] [ruby-core:56173]
|
|
|
|
Sun Aug 4 21:08:29 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_zip): performance implement by using
|
|
ALLOCA_N() to allocate tmp buffer.
|
|
|
|
Sun Aug 4 07:14:49 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* README.EXT, README.EXT.ja: Mention rb_integer_pack and
|
|
rb_integer_unpack.
|
|
|
|
Sun Aug 4 01:54:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (BARY_TRUNC): New macro.
|
|
(bary_cmp): Use BARY_TRUNC.
|
|
(bary_mul_toom3): Ditto.
|
|
(bary_divmod): Ditto.
|
|
(abs2twocomp): Ditto.
|
|
(bigfixize): Ditto.
|
|
(rb_cstr_to_inum): Ditto.
|
|
(big2str_karatsuba): Ditto.
|
|
(bigdivrem): Ditto.
|
|
|
|
Sun Aug 4 00:57:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_karatsuba): Don't allocate new temporary buffer
|
|
if the buffer is enough for current invocation.
|
|
|
|
Sun Aug 4 00:22:34 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary2bdigitdbl): New function.
|
|
(bdigitdbl2bary): Ditto.
|
|
(bary_mul_single): Use bdigitdbl2bary.
|
|
(power_cache_get_power): Ditto.
|
|
(bary_divmod): Use bary2bdigitdbl.
|
|
(big2str_orig): Ditto.
|
|
(bigdivrem): Ditto.
|
|
|
|
Sat Aug 3 22:47:11 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: The branch condition of selecting multiplication
|
|
algorithms should check smaller argument because Karatsuba and Toom3
|
|
is effective only if both arguments are big.
|
|
(bary_mul_toom3_branch): Compare the smaller argument to
|
|
TOOM3_MUL_DIGITS.
|
|
(bary_mul): Compare the smaller argument to KARATSUBA_MUL_DIGITS.
|
|
|
|
Sat Aug 3 22:23:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_orig): Receive the number to stringize as
|
|
BDIGIT array and size.
|
|
(big2str_karatsuba): Receive the number to stringize as BDIGIT array
|
|
and size. Use an temporary array of BDIGIT.
|
|
(rb_big2str1): Follow the above change.
|
|
|
|
Sat Aug 3 13:30:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (MAX_BASE36_POWER_TABLE_ENTRIES): Renamed from
|
|
MAX_BIG2STR_TABLE_ENTRIES.
|
|
(base36_power_cache): Renamed from big2str_power_cache.
|
|
(base36_numdigits_cache): Renamed from big2str_numdigits_cache.
|
|
|
|
Sat Aug 3 10:33:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_set_integer_literal): use rb_rational_raw1() for
|
|
integral rational because no reduction is needed with 1.
|
|
|
|
Sat Aug 3 09:46:07 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/etc/etc.c (setup_passwd, setup_group): set proper encodings to
|
|
string members.
|
|
|
|
Sat Aug 3 09:30:57 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* struct.c (rb_struct_define_under): new function to define Struct
|
|
under the given namespace, not under Struct. [Feature #8264]
|
|
|
|
* ext/etc/etc.c: use rb_struct_define_under.
|
|
|
|
Sat Aug 3 06:55:29 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* parse.y (value_expr_gen): now NODE_DEFN and NODE_DEFS are not void
|
|
value expressions. get rid of wrong warning with -w, and make to
|
|
pass tests with chkbuild. ref. [Feature #3753]
|
|
|
|
Sat Aug 3 04:23:48 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* doc/syntax/refinements.rdoc: Remove mention of instance_eval and
|
|
module_eval from scope section per:
|
|
http://twitter.com/shugomaeda/status/363219951336693761
|
|
|
|
Sat Aug 3 02:22:05 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_orig): Refactored.
|
|
|
|
Sat Aug 3 01:20:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigadd_core): Removed.
|
|
(bigadd): Use bary_add instead of bigadd_core.
|
|
|
|
Sat Aug 3 00:52:43 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big2str1): Simplify power_level calculation.
|
|
|
|
Sat Aug 3 00:34:20 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_zip): use rb_ary_new2() to create buffer
|
|
if rb_block_arity() > 1.
|
|
|
|
Sat Aug 3 00:12:00 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* NEWS: Add the description that IO#seek supports SEEK_DATA
|
|
and SEEK_HOLE.
|
|
|
|
Fri Aug 2 23:57:57 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* vm.c (m_core_define_method, m_core_define_singleton_method): now
|
|
the value of def-expr is the Symbol of the name of the method, not
|
|
nil.
|
|
ref. [ruby-dev:42151] [Feature #3753]
|
|
|
|
* test/ruby/test_syntax.rb (TestSyntax#test_value_of_def): test for
|
|
above changes.
|
|
|
|
Fri Aug 2 23:54:11 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_zip): performance improvement by avoiding
|
|
array creation if rb_block_arity() > 1.
|
|
|
|
Fri Aug 2 23:50:53 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (power_cache_get_power): Apply bigtrunc to the result of
|
|
bigsq.
|
|
(big2str_karatsuba): Fix number of leading zero characters.
|
|
|
|
Fri Aug 2 23:48:36 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_yylex): calculate denominator directly as powers of
|
|
ten, not parsing string.
|
|
|
|
* parse.y (parser_number_literal_suffix): return bit set of found
|
|
suffixes.
|
|
|
|
* parse.y (parser_set_number_literal, parser_set_integer_literal):
|
|
split from parser_number_literal_suffix to set yylval.
|
|
|
|
* parse.y (parser_yylex): parse rational number literal with decimal
|
|
point precisely.
|
|
|
|
* parse.y (simple_numeric): integrate numeric literals and simplify
|
|
numeric rules.
|
|
|
|
* ext/ripper/eventids2.c (ripper_init_eventids2): ripper support for
|
|
new literals, tRATIONAL and tIMAGINARY.
|
|
|
|
Fri Aug 2 18:33:28 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_karatsuba): Reduce power_level more than one at
|
|
recursion, if possible.
|
|
(rb_big2str1): Follow the above change.
|
|
|
|
Fri Aug 2 12:25:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul): Swap x and y for bary_mul1 if x is longer than y.
|
|
[ruby-dev:47565] [Bug #8719] Reported by Narihiro Nakamura.
|
|
|
|
Fri Aug 2 10:39:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* parse.y (negate_lit): add T_RATIONAL and T_COMPLEX to the switch
|
|
statement, and call rb_bug() if an unknown type is passed to
|
|
negate_lit(). [ruby-core:56316] [Bug #8717]
|
|
|
|
* bootstraptest/test_literal_suffix.rb (assert_equal): add test
|
|
|
|
Fri Aug 2 09:14:47 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* doc/syntax/refinements.rdoc: Improve description of where you may
|
|
activate refinements.
|
|
|
|
Fri Aug 2 07:45:55 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_orig): Remove len argument.
|
|
(big2str_karatsuba): Ditto.
|
|
(rb_big2str1): Follow above change.
|
|
|
|
Thu Aug 2 02:32:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* NEWS: Add the description of number literal suffixes.
|
|
|
|
Thu Aug 2 00:02:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* bootstraptest/test_literal_suffix.rb: add two test cases to
|
|
examine that "1if true" and "1rescue nil" are recognized as 1.
|
|
|
|
Thu Aug 1 23:45:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* rational.c (rb_flt_rationalize_with_prec): new public C function
|
|
to rationalize a Float instance with a precision.
|
|
|
|
* rational.c (rb_flt_rationalize): new public C function to
|
|
rationalize a Float instance. A precision is calculated from
|
|
the given float number.
|
|
|
|
* include/ruby/intern.h: Add rb_flt_rationalize_with_prec and
|
|
rb_flt_rationalize.
|
|
|
|
* parse.y: implement number literal suffixes, 'r' and 'i'.
|
|
[ruby-core:55096] [Feature #8430]
|
|
|
|
* bootstraptest/test_literal_suffix.rb: add tests for parser to scan
|
|
number literals with the above tsuffixes.
|
|
|
|
Thu Aug 1 23:55:08 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big2str1): Remove a local variable.
|
|
|
|
Thu Aug 1 23:33:01 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_cstr_to_inum): Use power_cache_get_power.
|
|
|
|
Thu Aug 1 21:02:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big2str1): Raise an error for too big number.
|
|
|
|
Thu Aug 1 20:46:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (power_cache_get_power): Hide cached Bignum objects.
|
|
|
|
Thu Aug 1 19:15:05 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big2str1): Remove non-trim mode.
|
|
(rb_big2str0): Non-trim mode implemented here.
|
|
(big2str_find_n1): Change the result type to long again.
|
|
(big2str_base_powerof2): Don't take arguments: len and trim.
|
|
(rb_big2str): Follow above change.
|
|
|
|
Thu Aug 1 12:37:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_alloc): New function to allocate the result string.
|
|
It is called after actual length is calculated.
|
|
(big2str_struct): Add fields: negative, result and ptr.
|
|
(big2str_orig): Write out the result via b2s->ptr.
|
|
(big2str_orig): Ditto.
|
|
(rb_big2str1): Don't allocate the result string at beginning.
|
|
|
|
Thu Aug 1 07:36:27 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_orig): Use temporary buffer when trim mode.
|
|
|
|
Thu Aug 1 06:28:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_orig): Simplified because RBIGNUM_LEN(x) <= 2 now.
|
|
(big2str_struct): Two fields added: hbase2, hbase2_numdigits.
|
|
(rb_big2str1): Initialize above fields.
|
|
|
|
Thu Aug 1 04:06:17 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/rdoc/options.rb (RDoc#finish): include root path in include
|
|
paths, to work in another directory than the source directory.
|
|
[ruby-core:56282] [Bug #8712]
|
|
|
|
* test/test_rdoc_markup_pre_process.rb (TestRDocMarkupPreProcess#setup):
|
|
fix input_file_name, as the test script is not pre-processed.
|
|
|
|
Thu Aug 1 01:45:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_karatsuba): Fix a condition of power_level.
|
|
|
|
Thu Aug 1 01:09:02 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (LOG2_KARATSUBA_BIG2STR_DIGITS): Removed.
|
|
(KARATSUBA_BIG2STR_DIGITS): Removed.
|
|
(big2str_numdigits_cache): New variable.
|
|
(power_cache_get_power): Merged with power_cache_get_power0.
|
|
This function returns maxpow_in_bdigit_dbl(base)**(2**power_level).
|
|
(rb_big2str1): use power_cache_get_power.
|
|
|
|
Wed Jul 31 23:59:28 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_find_n1): Change the return type to size_t.
|
|
(big2str_orig): Ditto.
|
|
(big2str_karatsuba): Ditto.
|
|
(rb_big2str1): Follow the above changes.
|
|
|
|
Wed Jul 31 23:19:06 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (power_cache_get_power): Change numdigits_ret to size_t *.
|
|
(big2str_orig): Change len argument to size_t.
|
|
(big2str_karatsuba): Ditto.
|
|
(rb_big2str1): Follow the above changes.
|
|
|
|
Wed Jul 31 22:59:47 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/parse/test_notation_declaration.rb: Change class
|
|
name to follow file name change.
|
|
|
|
Wed Jul 31 22:57:50 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_parsetest.rb: Rename to ...
|
|
* test/rexml/parse/test_notation_declaration.rb: ... this.
|
|
|
|
Wed Jul 31 22:54:39 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_mixin.rb: Remove duplicated tests.
|
|
|
|
Wed Jul 31 22:52:55 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_parsetest.rb: Fix typos in expected
|
|
value.
|
|
pubilc ->
|
|
public
|
|
^^
|
|
|
|
Wed Jul 31 22:50:51 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_parsetest.rb: Add tests that focus
|
|
system literal in external ID system notation declaration.
|
|
|
|
Wed Jul 31 22:36:21 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_cmp): Extracted from rb_big_cmp.
|
|
(power_cache_get_power): Change n1 argument (number of digits) to
|
|
power_level which is just passed to power_cache_get_power0.
|
|
(big2str_karatsuba): Ditto.
|
|
(rb_big2str1): Calculate the initial power_level.
|
|
|
|
Wed Jul 31 22:04:36 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_parsetest.rb: Fix a typo.
|
|
Extern ID ->
|
|
ExternalID
|
|
^^
|
|
|
|
Wed Jul 31 22:01:36 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_parsetest.rb: Add tests that focus
|
|
public ID in external ID notation declaration.
|
|
|
|
Wed Jul 31 22:01:24 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* parse.y: fix build error with bison-3.0.
|
|
|
|
Wed Jul 31 21:58:53 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_parsetest.rb: Split test patterns.
|
|
|
|
Wed Jul 31 21:42:33 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_parsetest.rb: Group tests.
|
|
|
|
Wed Jul 31 21:37:51 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_mixin.rb (TestNotationDecl#test_name):
|
|
Move to ...
|
|
* test/rexml/test_notationdecl_parsetest.rb
|
|
(TestNotationDecl#test_name): ... here.
|
|
|
|
Wed Jul 31 21:37:47 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
Wed Jul 31 21:31:49 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_parsetest.rb: Remove setup because it
|
|
doesn't share anything with other tests.
|
|
|
|
Wed Jul 31 21:24:55 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_attributes_mixin.rb: Remove a needless shebang.
|
|
* test/rexml/test_notationdecl_mixin.rb: ditto.
|
|
* test/rexml/test_doctype.rb: ditto.
|
|
* test/rexml/test_xml_declaration.rb: ditto.
|
|
* test/rexml/test_changing_encoding.rb: ditto.
|
|
|
|
Wed Jul 31 21:20:08 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_notationdecl_parsetest.rb: remove a needless shebang.
|
|
|
|
Wed Jul 31 20:11:01 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* string.c (rb_str_rindex): fix bug introduced in r42269.
|
|
"".rindex("") should return 0.
|
|
(str_rindex): ditto.
|
|
|
|
Wed Jul 31 19:55:33 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (MAX_BIG2STR_TABLE_ENTRIES): Use SIZEOF_SIZE_T.
|
|
(power_cache_get_power0): Add rb_bug call for too bit i argument.
|
|
(power_cache_get_power): Simplified.
|
|
|
|
Wed Jul 31 18:32:25 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/uri/common.rb (URI.decode_www_form_component): Use String#b.
|
|
|
|
Wed Jul 31 18:24:02 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* eval.c (rb_mod_refine, mod_using, top_using): don't show
|
|
warnings because Refinements are no longer experimental.
|
|
[ruby-core:55993] [Feature #8632]
|
|
|
|
* test/ruby/test_refinement.rb: related test.
|
|
|
|
* NEWS: fixes for the above change.
|
|
|
|
Wed Jul 31 17:55:55 2013 Shota Fukumori <her@sorah.jp>
|
|
|
|
* lib/uri/common.rb (URI.decode_www_form_component):
|
|
Don't raise error when str includes multibyte characters.
|
|
|
|
Wed Jul 31 17:45:39 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* string.c (rb_str_rindex): performance improvement by using
|
|
memrchr(3).
|
|
|
|
Wed Jul 31 16:43:30 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* string.c (rb_str_rindex): refactoring and avoid to call str_nth() if
|
|
pos == 0.
|
|
|
|
Wed Jul 31 14:41:36 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/set.rb: [DOC] Add a couple of notes on Hash as storage.
|
|
ref. [Feature #6589]
|
|
|
|
Wed Jul 31 14:38:52 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/set.rb: [DOC] Fix example result. Hash is now ordered.
|
|
|
|
Wed Jul 31 14:38:10 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/set.rb: [DOC] Use the term "sorted" instead of "ordered"
|
|
when mentioning SortSet.
|
|
|
|
Wed Jul 31 12:18:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_struct): New structure.
|
|
(big2str_orig): Use big2str_struct.
|
|
(big2str_karatsuba): Ditto.
|
|
(rb_big2str1): Ditto.
|
|
|
|
Wed Jul 31 12:02:16 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rubygems.rb: [DOC] typo in url patch by @Red54 [Fixes #369]
|
|
https://github.com/ruby/ruby/pull/369
|
|
|
|
Wed Jul 31 07:09:07 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Import RubyGems from master as of commit 523551c
|
|
* test/rubygems: ditto.
|
|
|
|
Tue Jul 30 22:21:54 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* test/ruby/test_hash.rb: add a test for enumeration order of Hash.
|
|
|
|
Tue Jul 30 18:52:27 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/set.rb (Set#intersect?, Set#disjoint?): Add new methods for
|
|
testing if two sets have any element in common.
|
|
[ruby-core:45641] [Feature #6588] Based on the code by marcandre.
|
|
|
|
Tue Jul 30 17:16:15 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* sprintf.c (ruby__sfvextra): add QUOTE flag to escape unprintable
|
|
characters.
|
|
|
|
Tue Jul 30 11:00:52 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/curses/extconf.rb: [DOC] nodoc to reduce Object pollution
|
|
|
|
Tue Jul 30 08:19:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* sizes.c (Init_sizes): Define sizes only if the type actually exists.
|
|
|
|
Mon Jul 29 22:55:26 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* sizes.c (Init_sizes): define RbConfig::SIZEOF. [Feature #8568]
|
|
|
|
Mon Jul 29 22:25:20 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/curses/curses.c: [DOC] Update location of samples
|
|
* samples/curses/*: Move Curses samples and refactor from mixin
|
|
The samples are included in rdoc for module and use of mixin is
|
|
confusing
|
|
|
|
Mon Jul 29 22:16:11 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (LOG2_KARATSUBA_BIG2STR_DIGITS): Renamed from
|
|
LOG2_KARATSUBA_DIGITS.
|
|
(KARATSUBA_BIG2STR_DIGITS): Renamed from KARATSUBA_DIGITS.
|
|
|
|
Mon Jul 29 22:04:45 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_compare_by_id): add function prototype.
|
|
|
|
Mon Jul 29 21:53:41 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_compare_by_id): don't call rb_hash_rehash()
|
|
if self.compare_by_identity? == true.
|
|
|
|
Mon Jul 29 21:29:48 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_assoc): performance improvement by replacing
|
|
compare function in RHASH(hash)->ntbl->type temporarily like r42224.
|
|
it falls back to rb_hash_foreach() if st_lookup() doesn't find the key.
|
|
|
|
* test/ruby/test_hash.rb: add a test for above.
|
|
|
|
Mon Jul 29 21:15:30 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* test/ruby/test_lazy_enumerator.rb
|
|
(TestLazyEnumerator#test_initialize): Make sure
|
|
Enumerator::Lazy#initialize raises error if the object is
|
|
frozen. The check was performed by rb_ivar_set() before
|
|
rb_check_frozen() was added to enumerator_init().
|
|
|
|
Mon Jul 29 21:06:42 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* enumerator.c (enumerator_init): Add a frozenness check to
|
|
prevent a frozen Enumerator object from being reinitialized with
|
|
a different enumerable object. This is the least we should do,
|
|
and more fixes will follow. [Fixes GH-368] Patch by Kenichi
|
|
Kamiya.
|
|
|
|
* enumerator.c (generator_init): Ditto.
|
|
|
|
Mon Jul 29 20:14:24 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_assoc): revert r42224. table->type->compare is
|
|
called only if hashes are matched.
|
|
|
|
* test/ruby/test_hash.rb: add a test to check using #== to compare.
|
|
|
|
Mon Jul 29 17:00:31 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (yycompile): store file name as String to keep the encoding.
|
|
|
|
* parse.y (rb_parser_compile_string_path, rb_parser_compile_file_path):
|
|
new functions to pass file name as a String.
|
|
|
|
* parse.y (gettable_gen): return a copy of the original file name, not
|
|
a copy in filesystem encoding.
|
|
|
|
* vm_eval.c (eval_string_with_cref): use Qundef instead of "(eval)".
|
|
|
|
Mon Jul 29 16:53:18 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* hash.c (rb_hash_initialize_copy): copy st_table type even if empty.
|
|
[ruby-core:56256] [Bug #8703]
|
|
|
|
Mon Jul 29 16:34:29 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* hash.c (rb_hash_initialize_copy): clear old table before copy new
|
|
table.
|
|
|
|
Mon Jul 29 16:34:09 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* hash.c (rb_hash_assoc): aggregate object can be initialized only
|
|
with link time constants.
|
|
|
|
Mon Jul 29 14:54:44 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_assoc): performance improvement by replacing
|
|
compare function in RHASH(hash)->ntbl->type temporarily.
|
|
|
|
Mon Jul 29 14:52:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (xsystem): expand environment variable in all macros not
|
|
expanded with RbConfig. [Bug #8702]
|
|
|
|
* test/mkmf/test_framework.rb (create_framework): replace all $@ not
|
|
only once.
|
|
|
|
Mon Jul 29 06:54:30 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (rb_w32_pipe): use enum for compile time constants,
|
|
instead of const int for debugging.
|
|
|
|
Mon Jul 29 00:11:49 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem): Specialized implementation added for
|
|
nx == 2 && ny == 2
|
|
|
|
Sun Jul 28 20:28:41 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* io.c (io_getpartial): use rb_str_locktmp_ensure().
|
|
[ruby-core:56121] [Bug #8669]
|
|
|
|
* io.c (rb_io_sysread): ditto.
|
|
|
|
* test/ruby/test_io.rb: add tests for above.
|
|
|
|
Sun Jul 28 20:10:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/extmk.rb (extmake): should make static libraries for extensions
|
|
to be statically linked. [Bug #7948]
|
|
|
|
Sun Jul 28 17:38:32 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* string.c: add internal API rb_str_locktmp_ensure().
|
|
|
|
* io.c (io_fread): use rb_str_locktmp_ensure().
|
|
[ruby-core:56121] [Bug #8669]
|
|
|
|
* test/ruby/test_io.rb: add a test for above.
|
|
|
|
Sun Jul 28 13:04:39 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* io.c (interpret_seek_whence): support SEEK_DATA and SEEK_HOLE.
|
|
These are whences for lseek(2) supported by Linux since version 3.1.
|
|
[ruby-core:56123] [Feature #8671]
|
|
|
|
* test/ruby/test_io.rb: Add tests for above.
|
|
|
|
Sun Jul 28 12:41:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (absint_numwords_generic): The char_bit variable changed
|
|
to static constant.
|
|
|
|
Sun Jul 28 12:03:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Constify bary_* functions.
|
|
|
|
Sun Jul 28 11:12:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/intern.h (rb_absint_size): Declaration moved from
|
|
internal.h to calculate required buffer size to pack integers.
|
|
(rb_absint_numwords): Ditto.
|
|
(rb_absint_singlebit_p): Ditto.
|
|
[ruby-core:42813] [Feature #6065]
|
|
|
|
Sun Jul 28 10:54:26 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (rb_w32_pipe): fix pipe name formatting. as "%x" may
|
|
not contain '0' at all, fill at fixed position instead.
|
|
|
|
Sun Jul 28 00:35:14 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_size): Return the bignum "bytewise" size.
|
|
[ruby-core:55578] [Feature #8553]
|
|
This is accepted by matz on DevelopersMeeting20130727Japan.
|
|
|
|
Sun Jul 28 00:07:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/intern.h (rb_integer_pack): Declaration moved from
|
|
internal.h.
|
|
(rb_integer_unpack): Ditto.
|
|
[ruby-core:42813] [Feature #6065]
|
|
|
|
Fri Jul 26 23:18:13 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* NEWS: Add a new feature that REXML::Parsers::StreamParser
|
|
supports "entity" event.
|
|
|
|
Fri Jul 26 23:14:31 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/parsers/streamparser.rb
|
|
(REXML::Parsers::StreamParser#parse): Add "entity" event support to
|
|
listener. [Bug #8689] [ruby-dev:47542]
|
|
Reported by Ippei Obayashi.
|
|
* test/rexml/test_stream.rb (StreamTester#entity): Add a test for
|
|
the above case.
|
|
|
|
Fri Jul 26 23:05:27 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_yylex): separate numeric literal from succeeding
|
|
token, and treat 'e' as floating point number only if followed by
|
|
exponent part.
|
|
|
|
Fri Jul 26 22:14:10 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_exec.h (CHECK_VM_STACK_OVERFLOW_FOR_INSN): surround with
|
|
do/while (0), and remove unnecessary casts.
|
|
|
|
Fri Jul 26 20:12:07 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* ext/syslog/lib/syslog/logger.rb (Syslog::Logger): Add facility
|
|
to Syslog::Logger. [Fixes GH-305] patch by Max Shytikov
|
|
https://github.com/ruby/ruby/pull/305
|
|
|
|
Fri Jul 26 19:25:17 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_exec.h, tool/instruction.rb: not an error, but a BUG if stack
|
|
overflow checking failed just before/after the beginning of an
|
|
instruction. It should be treated as a BUG.
|
|
Please tell us if your code cause BUG with this problem.
|
|
This check will removed soon (for performance).
|
|
|
|
Fri Jul 26 18:30:14 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_memcpy): cast to int to suppress a warning.
|
|
|
|
Fri Jul 26 18:21:58 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_memcpy): try to enable optimization.
|
|
At least on my environments, I don't see any errors
|
|
with many trials. Please tell us if you find any GC bugs.
|
|
|
|
Fri Jul 26 17:49:26 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/file.c (fix_string_encoding): fix target encoding. the
|
|
parameter `encoding' is not the target encoding but the original
|
|
encoding.
|
|
|
|
Fri Jul 26 14:05:19 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/fiddle/*: [DOC] More doc on dlopen and RTLD_DEFAULT from r42184
|
|
|
|
Fri Jul 26 13:08:53 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/fiddle/lib/fiddle.rb: [DOC] Document Fiddle.dlopen(nil)
|
|
* ext/fiddle/handle.c: [DOC] Document Fiddle::Handle.new(nil)
|
|
|
|
Fri Jul 26 13:04:15 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* load.c (rb_load_internal): use rb_load_file_str() to keep path
|
|
encoding.
|
|
|
|
* load.c (rb_require_safe): search in OS path encoding for Windows.
|
|
|
|
* ruby.c (rb_load_file_str): load file with keeping path encoding.
|
|
|
|
* win32/file.c (rb_file_load_ok): use WCHAR type API assuming incoming
|
|
path is encoded in UTF-8. [ruby-core:56136] [Bug #8676]
|
|
|
|
* file.c (rb_str_encode_ospath): simplify using rb_str_conv_enc().
|
|
|
|
* win32/file.c (fix_string_encoding): simplify with rb_str_conv_enc().
|
|
|
|
* win32/file.c (convert_mb_to_wchar): use bare pointer instead of
|
|
VALUE, and remove useless argument.
|
|
|
|
Fri Jul 26 11:42:07 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* rational.c (f_round_common): Rational is expected to be returned by
|
|
Rational#*, but mathn.rb breaks that assumption. [ruby-core:56177]
|
|
[Bug #8687]
|
|
|
|
Fri Jul 26 01:37:45 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h: check defined(USE_RGENGC_LOGGING_WB_UNPROTECT)
|
|
|
|
Fri Jul 26 01:21:41 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* file.c (rb_file_expand_path_internal): fix r42160; skip '~'.
|
|
|
|
Thu Jul 25 17:53:18 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* lib/net/http.rb (Net::HTTP#connect): disable Nagle's algorithm on
|
|
HTTP connection. [ruby-core:56158] [Feature #8681]
|
|
|
|
Thu Jul 25 17:49:42 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* re.c (rb_reg_to_s): convert closing parenthesis to the target encoding
|
|
if it is ASCII incompatible encoding. [ruby-core:56063] [Bug #8650]
|
|
|
|
Thu Jul 25 17:21:21 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* encoding.c (is_obj_encoding): new macro to check if obj is an
|
|
Encoding. obj can be any type while is_data_encoding expects T_DATA
|
|
only.
|
|
|
|
Thu Jul 25 17:17:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* file.c (rb_file_expand_path_internal): should clear coderange after
|
|
copying user name as binary data.
|
|
|
|
Thu Jul 25 16:17:55 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* encoding.c (check_encoding): Check T_DATA or not.
|
|
is_data_encoding(obj) assumes that `obj' is T_DATA.
|
|
|
|
Thu Jul 25 13:06:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* dir.c (dir_s_home): use rb_home_dir_of and rb_default_home_dir.
|
|
|
|
* file.c (rb_home_dir_of): split from rb_home_dir() for the home
|
|
directry of the given user, and the user name is a VALUE, not a bare
|
|
pointer. should raise if the user does not exist.
|
|
|
|
* file.c (rb_default_home_dir): split from rb_home_dir() for the home
|
|
directry of the current user.
|
|
|
|
Thu Jul 25 12:32:11 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/openssl/ossl.c: support additional three thread synchronization
|
|
functions. [ruby-trunk - Bug #8386]
|
|
|
|
Thu Jul 25 07:15:58 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Import RubyGems from master as of commit 4ff70cc
|
|
* test/rubygems: ditto.
|
|
|
|
Wed Jul 24 20:57:44 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* compile.c (iseq_set_arguments): use RARRAY_RAWPTR() instead of
|
|
RARRAY_PTR() because there is no new reference.
|
|
|
|
* compile.c (iseq_set_exception_table): ditto.
|
|
|
|
Wed Jul 24 19:49:54 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* lib/uri/generic.rb (find_proxy): raise BadURIError if the URI is
|
|
a relative URI. [Bug #8645]
|
|
|
|
Wed Jul 24 18:56:06 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_insnhelper.c (vm_expandarray): use RARRAY_RAWPTR() instead of
|
|
RARRAY_PTR() because there is no new reference.
|
|
|
|
* vm_insnhelper.c (vm_caller_setup_args): ditto.
|
|
|
|
* vm_insnhelper.c (vm_yield_setup_block_args): ditto.
|
|
|
|
Wed Jul 24 18:40:11 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c, gc.c: move ary_unprotect_logging() into
|
|
rb_gc_unprotect_logging() which is general version
|
|
|
|
* include/ruby/ruby.h: add USE_RGENGC_LOGGING_WB_UNPROTECT
|
|
to enable.
|
|
|
|
Wed Jul 24 17:37:50 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* file.c (rb_file_expand_path_internal): preserve the file name
|
|
encoding in an exception message.
|
|
|
|
Wed Jul 24 08:04:49 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/-ext-/tracepoint/test_tracepoint.rb: add GC on/off to count
|
|
GC events strictly.
|
|
|
|
Tue Jul 23 23:19:24 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/openssl/extconf.rb (CRYPTO_THREADID): check exist or not.
|
|
|
|
* ext/openssl/ossl.c (ossl_thread_id): use rb_nativethread_self()
|
|
implemented at r42137 to allow threads which doesn't associated with
|
|
Ruby thread to use openssl functions.
|
|
|
|
* ext/openssl/ossl.c (Init_ossl_locks): If CRYPTO_THREADID is defined
|
|
(OpenSSL 1.0.0 or later has it) use CRYPTO_THREADID_set_callback()
|
|
instead of CRYPTO_set_id_callback() because its argument is
|
|
unsigned long; it may cause id collision on mswin64
|
|
whose sizeof(unsigned long) < sizeof(void*).
|
|
http://www.openssl.org/docs/crypto/threads.html
|
|
|
|
* ext/openssl/ossl.c (ossl_threadid_func): defined for above.
|
|
|
|
Tue Jul 23 20:47:36 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Move functions.
|
|
|
|
Tue Jul 23 20:14:55 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_divmod): Add special cases for x < y easily detected
|
|
and nx == 2 && ny == 2.
|
|
|
|
Tue Jul 23 19:48:38 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* thread_(pthread|win32).h: rename rb_thread_cond_t to
|
|
rb_nativethread_cond_t.
|
|
|
|
* thread.c, thread_pthread.c, thread_win32.c, vm_core.h: catch up
|
|
renaming.
|
|
|
|
Tue Jul 23 19:44:32 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* thread_native.h: add rb_nativethread_self() which returns
|
|
current running native thread identifier.
|
|
|
|
* thread_[pthread|win32].c: implement rb_nativethread_self().
|
|
|
|
Tue Jul 23 19:34:11 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* thread_pthread.h, thread_win32.h: rename rb_thread_id_t to
|
|
rb_nativethread_id_t.
|
|
|
|
* thread_pthread.c, vm_core.h: use rb_nativethread_id_t.
|
|
|
|
Tue Jul 23 18:56:11 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/openssl/ossl.c: use system native (system provided)
|
|
thread locking APIs added by last commit.
|
|
This patch fixes [Bug #8386].
|
|
"rb_mutex_*" APIs control only "Ruby" threads.
|
|
Not for native threads.
|
|
|
|
Tue Jul 23 18:44:15 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* thread_native.h: added.
|
|
Move native thread related lines from vm_core.h.
|
|
And declare several functions "rb_nativethread_lock_*",
|
|
manipulate locking.
|
|
|
|
* common.mk: add thread_native.h.
|
|
|
|
* thread.c: add functions "rb_nativethread_lock_*".
|
|
|
|
* thread.c, thread_[pthread,win32].[ch]: rename rb_thread_lock_t
|
|
to rb_nativethread_lock_t to make it clear that this lock is for
|
|
native threads, not for ruby threads.
|
|
|
|
Tue Jul 23 16:14:57 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_before_sweep): fix spacing.
|
|
|
|
Tue Jul 23 15:57:11 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (heap_get_freeobj): clear slot->freelist here.
|
|
This means that this slot doesn't have any free objects.
|
|
And store this slot with objspace->heap.using_slot.
|
|
|
|
* gc.c (gc_before_sweep): restore objspace->freelist
|
|
into objspace->heap.using_slot->freelist.
|
|
This means that using_slot has free objects which are
|
|
pointed from objspace->freelist.
|
|
|
|
* gc.c (gc_slot_sweep): do not need to clear slot->freelist.
|
|
|
|
Tue Jul 23 09:34:49 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* sample/drb/README*.rdoc: [DOC] migrate DRb sample READMEs to rdoc
|
|
|
|
Tue Jul 23 09:28:05 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/drb/invokemethod.rb: [DOC] nodoc InvokeMethod18Mixin
|
|
|
|
Tue Jul 23 08:44:37 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* ext/openssl/ossl_asn1.c (asn1time_to_time): Implement YYMMDDhhmmZ
|
|
format for ASN.1 UTCTime. [ruby-trunk - Bug #8664]
|
|
* test/openssl/test_asn1.rb: Test for the above.
|
|
|
|
Tue Jul 23 08:11:32 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rexml/streamlistener.rb: [DOC] Fix examples in
|
|
REXML::StreamListener#entitydecl patch by Ippei Obayashi [Bug #8665]
|
|
|
|
Tue Jul 23 07:44:59 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Import RubyGems from master as of commit b165260
|
|
* test/rubygems: ditto.
|
|
|
|
Tue Jul 23 07:14:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mulsub_1xN): New function.
|
|
(bary_mul_toom3): Use bary_mulsub_1xN.
|
|
|
|
Tue Jul 23 03:32:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (KARATSUBA_BALANCED): New macro.
|
|
(TOOM3_BALANCED): Ditto.
|
|
(bary_mul_balance_with_mulfunc): Use KARATSUBA_BALANCED and
|
|
TOOM3_BALANCED.
|
|
(rb_big_mul_balance): Relax a condition.
|
|
(rb_big_mul_karatsuba): Use KARATSUBA_BALANCED.
|
|
(rb_big_mul_toom3): Use TOOM3_BALANCED.
|
|
(bary_mul_karatsuba_branch): Use KARATSUBA_BALANCED.
|
|
(bary_mul_toom3_branch): Use TOOM3_BALANCED.
|
|
|
|
Tue Jul 23 01:34:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem_mulsub): Extracted from bigdivrem1.
|
|
(bigdivrem1): Use bary_add.
|
|
|
|
Mon Jul 22 18:39:52 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* string.c (rb_str_enumerate_chars): specify array capa
|
|
with str_strlen().
|
|
|
|
* string.c (rb_str_enumerate_codepoints): ditto.
|
|
|
|
Mon Jul 22 18:01:33 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* string.c (rb_str_enumerate_chars): specify array capa.
|
|
|
|
Mon Jul 22 17:24:14 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* string.c (rb_str_each_char_size): performance improvement by
|
|
using rb_str_length().
|
|
|
|
Mon Jul 22 16:32:48 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_eval.c (eval_string_with_cref): check by Check_TypedStruct
|
|
instead of rb_obj_is_kind_of.
|
|
|
|
Mon Jul 22 13:19:22 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_resize_capa): use RARRAY_RAWPTR() because
|
|
this code creates no new references.
|
|
|
|
Mon Jul 22 12:58:18 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_memfill): added.
|
|
|
|
* array.c (rb_ary_initialize): use ary_memfill().
|
|
|
|
* array.c (rb_ary_fill): ditto.
|
|
|
|
* array.c (rb_ary_slice_bang): use RARRAY_RAWPTR() because
|
|
this code creates no new references.
|
|
|
|
Mon Jul 22 10:09:46 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_slot_sweep): need to add empty RVALUE as freeobj.
|
|
|
|
Mon Jul 22 09:48:31 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_eval.c (eval_string_with_cref): use the given file name unless
|
|
eval even if scope is given. additional fix for [Bug #8436].
|
|
based on the patch by srawlins at [ruby-core:56099] [Bug #8662].
|
|
|
|
Mon Jul 22 09:24:19 2013 Kouji Takao <kouji@takao7.net>
|
|
|
|
* ext/readline/readline.c (Init_readline): added
|
|
Readline.delete_text. [ruby-dev:45789] [Feature #6626]
|
|
* ext/readline/extconf.rb: check for rl_delete_text() in Readline library.
|
|
|
|
Thanks, Nobuyoshi Nakada, for the patch.
|
|
|
|
Mon Jul 22 03:15:54 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/date/date_parse.c (rfc2822_cb): check if wday is given, since it
|
|
can be omitted.
|
|
|
|
Mon Jul 22 00:15:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_sq_fast): Refine expressions.
|
|
|
|
Sun Jul 21 21:08:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul): Use simple multiplication if yl is small.
|
|
(rb_cstr_to_inum): Invoke bigsq instead of bigmul0.
|
|
(bigsq): Re-implemented.
|
|
(bigmul0): Invoke bigsq if two arguments are identical.
|
|
|
|
Sun Jul 21 09:58:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul_toom3): New function based on bigmul1_toom3.
|
|
(bary_mul_toom3_branch): Call bary_mul_toom3.
|
|
(rb_big_mul_toom3): Ditto.
|
|
(bigmul1_toom3): Removed.
|
|
(big_real_len): Ditto.
|
|
(big_split): Ditto.
|
|
(big_split3): Ditto.
|
|
|
|
Sun Jul 21 08:12:16 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* proc.c (proc_to_s): use PRIsVALUE to preserve the result encoding.
|
|
|
|
Sun Jul 21 03:36:18 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* hash.c (rb_hash_flatten): use NUM2INT to raise TypeError on 32bit
|
|
platform. it's introduced by r42039
|
|
|
|
Sun Jul 21 01:07:45 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* common.mk (help): Fix environment variable name and argument.
|
|
Actually it can also be a directory or any argument for
|
|
test/unit runner. [Fixes GH-363]
|
|
|
|
Sat Jul 20 22:44:50 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* common.mk: Document running a single test [Fixes GH-363]
|
|
Patch by Avdi Grimm https://github.com/ruby/ruby/pull/363
|
|
|
|
Sat Jul 20 22:39:56 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* sample/*: whitespace patch by Sergio Campama [Fixes GH-364]
|
|
https://github.com/ruby/ruby/pull/364
|
|
|
|
Sat Jul 20 22:33:13 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* doc/regexp.rdoc: [DOC] Fix typo in example [Fixes GH-365]
|
|
Patch by Juanito Fatas https://github.com/ruby/ruby/pull/365
|
|
|
|
Sat Jul 20 17:46:03 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* string.c (rb_str_succ): add missing case NEIGHBOR_WRAPPED.
|
|
r42078 caused buggy behavior like "\xFF".b -> "\x01\xFF".b
|
|
|
|
Sat Jul 20 15:22:38 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (rb_ary_resize): use simple memcpy because there are no new
|
|
references.
|
|
|
|
Sat Jul 20 15:02:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* safe.c (ruby_safe_level_4_warning): define for old extension
|
|
libraries. [Bug #8652]
|
|
|
|
Sat Jul 20 14:38:00 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_make_shared): make shared array shady.
|
|
Making non-shady shared array causes SEGV (see rubyci).
|
|
It seems a bug around shared array.
|
|
|
|
Sat Jul 20 12:14:07 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (enc_succ_char, enc_pred_char): consider wchar case.
|
|
[ruby-core:56071] [Bug #8653]
|
|
|
|
* string.c (rb_str_succ): do not replace with invalid char.
|
|
|
|
* encoding.c (rb_enc_code_to_mbclen): add new function which returns
|
|
mbclen from codepoint like as rb_enc_codelen() but 0 for invalid
|
|
char.
|
|
|
|
* include/ruby/encoding.h (rb_enc_code_to_mbclen): declaration and
|
|
shortcut macro.
|
|
|
|
Fri Jul 19 21:59:12 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: declare type_name() at the beginning of file.
|
|
|
|
Fri Jul 19 21:35:09 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c: reduce shady operations.
|
|
|
|
* array.c (rb_ary_modify, ary_make_partial, rb_ary_splice,
|
|
rb_ary_replace, rb_ary_eql, rb_ary_compact_bang):
|
|
use RARRAY_RAWPTR() instead of RARRAY_PTR().
|
|
|
|
* array.c (rb_ary_shift): use RARRAY_PTR_USE() without WB because
|
|
there are not new relations.
|
|
|
|
* array.c (ary_ensure_room_for_unshift): ditto.
|
|
|
|
* array.c (rb_ary_sort_bang): ditto.
|
|
|
|
* array.c (rb_ary_delete_at): ditto.
|
|
|
|
* array.c (rb_ary_reverse_m): use RARRAY_RAWPTR() because
|
|
there are not new relations.
|
|
|
|
Fri Jul 19 20:58:20 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c: reduce shade operations.
|
|
|
|
* array.c (rb_ary_modify): use RARRAY_RAWPTR().
|
|
|
|
* array.c (ary_make_substitution, rb_ary_s_create, ary_make_partial,
|
|
rb_ary_splice, rb_ary_resize, rb_ary_rotate_m, rb_ary_times):
|
|
use ary_memcpy().
|
|
|
|
Fri Jul 19 19:55:28 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_mem_clear): added. This operation doesn't need WB
|
|
because this operation creates a reference to Qnil.
|
|
|
|
* array.c (ary_make_shared, rb_ary_store, rb_ary_shift_m,
|
|
rb_ary_splice, rb_ary_resize, rb_ary_fill): use ary_mem_clear()
|
|
instead of rb_mem_clear().
|
|
|
|
* array.c (ary_make_shared): use RARRAY_RAWPTR() instead of RARRAY_PTR().
|
|
|
|
Fri Jul 19 19:18:51 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c: fix commit miss.
|
|
RGENGC_UNPROTECT_LOGGING should be 0.
|
|
|
|
Fri Jul 19 19:15:30 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (rb_ary_resurrect): use RARRAY_RAWPTR() because there is no
|
|
writing.
|
|
|
|
* array.c (rb_ary_new_from_values): use ary_memcpy().
|
|
|
|
Fri Jul 19 19:07:31 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_memcpy): add a function to copy VALUEs into ary
|
|
with write barrier. If ary is promoted, use write barrier correctly.
|
|
|
|
* array.c (rb_ary_cat, rb_ary_unshift_m, rb_ary_dup,
|
|
rb_ary_sort_bang, rb_ary_replace, rb_ary_plus): use ary_memcpy().
|
|
|
|
Fri Jul 19 15:32:57 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (rb_ary_store): use RARRAY_PTR_USE() instead of RARRAY_PTR().
|
|
Clearing memory space doesn't need WBs.
|
|
|
|
Fri Jul 19 15:19:37 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_ensure_room_for_push): use RARRAY_RAWPTR() instead of
|
|
RARRAY_PTR. In this code, there are no "write" operation.
|
|
|
|
* array.c (rb_ary_equal): ditto.
|
|
|
|
* array.c (recursive_equal): ditto.
|
|
|
|
Fri Jul 19 15:09:22 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c, internal.h (rb_gc_writebarrier_remember_promoted): add a new
|
|
function to remember an specified object. This api is only
|
|
experimental (strongly depend on WB/rgengc strategy).
|
|
|
|
Fri Jul 19 14:56:00 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_unprotect_logging): use (void *) for first parameter
|
|
because VALUE is not defined before including ruby/ruby.h.
|
|
|
|
Fri Jul 19 14:19:48 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* ext/pathname/pathname.c (path_inspect): use PRIsVALUE to preserve
|
|
the result encoding.
|
|
|
|
Fri Jul 19 12:35:41 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/socket/test_tcp.rb (test_initialize_failure): Use EADDRNOTAVAIL
|
|
to test an error message generated by bind() failure.
|
|
|
|
Fri Jul 19 11:27:38 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/racc/parser.rb: [DOC] Capitalize "Ruby" in documentation
|
|
Patch by Dave Worth https://github.com/ruby/ruby/pull/341
|
|
|
|
Fri Jul 19 11:26:28 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/psych/lib/psych*: [DOC] Capitalize "Ruby" in documentation
|
|
Patch by Dave Worth https://github.com/ruby/ruby/pull/341
|
|
|
|
Fri Jul 19 11:25:12 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rdoc/*: [DOC] Capitalize "Ruby" in documentation
|
|
Patch by Dave Worth https://github.com/ruby/ruby/pull/341
|
|
|
|
Fri Jul 19 11:23:55 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rubygems*: [DOC] Capitalize "Ruby" in documentation
|
|
Patch by Dave Worth https://github.com/ruby/ruby/pull/341
|
|
|
|
Fri Jul 19 11:16:54 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/set.rb (Set#to_set): Define Set#to_set so that aSet.to_set
|
|
returns self. [Fixes GH-359]
|
|
|
|
Fri Jul 19 11:10:23 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rake/*: [DOC] Capitalize "Ruby" in documentation
|
|
Patch by Dave Worth https://github.com/ruby/ruby/pull/341
|
|
|
|
Fri Jul 19 01:04:14 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/-test-/bignum/intpack.c: Renamed from ext/-test-/bignum/pack.c.
|
|
(Init_intpack): Renamed from Init_pack.
|
|
Reported by Naohisa Goto. [ruby-dev:47526] [Bug #8655]
|
|
|
|
Fri Jul 19 00:54:27 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* test/ruby/test_array.rb (test_count): add a test case for #count
|
|
with an argument. See Bug #8654.
|
|
|
|
Thu Jul 18 23:45:06 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_eql): compare RARRAY_PTR() for performance
|
|
improvement in case of that self and other are shared.
|
|
|
|
Thu Jul 18 22:46:42 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/cgi.rb: [DOC] Capitalize "Ruby" in documentation [Fixes GH-341]
|
|
Patch by Dave Worth https://github.com/ruby/ruby/pull/341
|
|
* lib/webrick.rb: ditto
|
|
* lib/scanf.rb: ditto
|
|
* lib/xmlrpc/config.rb: ditto
|
|
* lib/resolv.rb: ditto
|
|
* lib/e2mmap.rb: ditto
|
|
* lib/fileutils.rb: ditto
|
|
* lib/mkmf.rb: ditto
|
|
* lib/cgi/session.rb: ditto
|
|
* lib/yaml.rb: ditto
|
|
* lib/erb.rb: ditto
|
|
* lib/irb.rb: ditto
|
|
* lib/tracer.rb: ditto
|
|
* lib/net/http.rb: ditto
|
|
* ext/syslog/lib/syslog/logger.rb: ditto
|
|
* sample/pty/expect_sample.rb: ditto
|
|
|
|
Thu Jul 18 21:30:50 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_sq_fast): Specialize the last iteration of the
|
|
outer loop.
|
|
(bigfixize): A condition simplified.
|
|
|
|
Thu Jul 18 21:15:41 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_equal): compare RARRAY_PTR() for performance
|
|
improvement in case of that self and other are shared.
|
|
|
|
Thu Jul 18 20:44:51 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_fill): use memfill().
|
|
|
|
Thu Jul 18 20:35:14 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* array.c (rb_ary_count): check length to avoid SEGV
|
|
while iterating. Remove other pointer loop when arg is given.
|
|
|
|
* test/ruby/test_array.rb (test_count): add test for bug.
|
|
[ruby-core:56072] [Bug #8654]
|
|
|
|
Thu Jul 18 18:14:36 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* array.c (rb_ary_count): iterate items appropriately.
|
|
[Bug #8654]
|
|
|
|
Thu Jul 18 17:35:41 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_flatten): performance improvement by not using
|
|
rb_hash_to_a() to avoid array creation with rb_assoc_new().
|
|
|
|
Thu Jul 18 16:16:17 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c: add logging feature for RGenGC's write barrier unprotect
|
|
event.
|
|
|
|
Thu Jul 18 15:45:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h (RUBY_SAFE_LEVEL_CHECK): make only
|
|
rb_set_safe_level(4) an error always but make rb_secure(4) an error
|
|
only in the core. [ruby-dev:47517] [Bug #8652]
|
|
|
|
Thu Jul 18 15:42:01 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: fix spell miss.
|
|
|
|
Thu Jul 18 15:11:11 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h (ruby_safe_level_4): get rid of special
|
|
character. [ruby-dev:47512] [misc #8646]
|
|
|
|
Thu Jul 18 14:51:39 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (ary_alloc): slim setup process.
|
|
|
|
Thu Jul 18 14:37:57 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* string.c (str_alloc): no need to clear RString (already cleared).
|
|
|
|
Thu Jul 18 12:57:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (BDIGITS_ZERO): Defined.
|
|
(bary_pack): Use BDIGITS_ZERO.
|
|
(bary_unpack): Ditto.
|
|
(bary_mul_single): Ditto.
|
|
(bary_mul_normal): Ditto.
|
|
(bary_sq_fast): Ditto.
|
|
(bary_mul_balance_with_mulfunc): Ditto.
|
|
(bary_mul_precheck): Ditto.
|
|
(bary_mul_toom3_branch): Ditto.
|
|
(rb_cstr_to_inum): Ditto.
|
|
(big_shift3): Ditto.
|
|
(bigmul1_toom3): Ditto.
|
|
(bary_divmod): Ditto.
|
|
|
|
Thu Jul 18 06:30:02 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: rename gc related functions with prefix "gc_".
|
|
* before_gc_sweep() -> gc_before_sweep().
|
|
* after_gc_sweep() -> gc_after_sweep().
|
|
* lazy_sweep() -> gc_lazy_sweep().
|
|
* rest_sweep() -> gc_rest_sweep().
|
|
* slot_sweep() -> gc_slot_sweep().
|
|
|
|
* gc.c: rename a heap management function with prefix "heap_".
|
|
* get_freeobj() -> heap_get_freeobj().
|
|
|
|
* gc.c: rename markable_object_p() to is_markable_object().
|
|
|
|
Wed Jul 17 22:57:40 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (delete_if_i): use ST_DELETE.
|
|
|
|
Wed Jul 17 22:34:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: An static assertion for relation of SIZEOF_LONG and
|
|
SIZEOF_BDIGITS is added.
|
|
(bary_mul_precheck): Reduce comparisons.
|
|
(bary_mul): Invoke bary_sq_fast or bary_mul1 if the bignum size is
|
|
small.
|
|
(bigfixize): Resize the argument bignum here.
|
|
(bignorm): Don't call bigtrunc after bigfixize.
|
|
|
|
Wed Jul 17 22:13:26 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_replace): performance improvement by using
|
|
st_copy().
|
|
|
|
Wed Jul 17 17:19:54 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: rename heap management functions with prefix "heap_".
|
|
* allocate_sorted_array() -> heap_allocate_sorted_array().
|
|
* slot_add_freeobj() -> heap_slot_add_freeobj().
|
|
* assign_heap_slot() -> heap_assign_slot().
|
|
* add_heap_slots() -> heap_add_slots().
|
|
* init_heap() -> heap_init().
|
|
* set_heap_increment() -> heap_set_increment().
|
|
|
|
* gc.c (initial_expand_heap): inlined in rb_gc_set_params().
|
|
|
|
Wed Jul 17 17:12:23 2013 Matthew M. Boedicker <matthewm@boedicker.org>
|
|
|
|
* hash.c (env_fetch): Add key name to message on ENV.fetch KeyError,
|
|
as well as Hash#fetch. [ruby-core:56062] [Feature #8649]
|
|
|
|
Wed Jul 17 15:59:33 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: catch up last changes for debugging/checking mode.
|
|
|
|
Wed Jul 17 15:50:10 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_objspace_free): free slot itself.
|
|
|
|
* gc.c (objspace_each_objects): fix condition.
|
|
Use slot->body instead of slot.
|
|
|
|
* gc.c (count_objects): use "slot" variable.
|
|
|
|
Wed Jul 17 15:21:10 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (unlink_heap_slot): fix memory leak.
|
|
free slot itself at free_heap_slot().
|
|
|
|
Reproduce-able code is here:
|
|
N1 = 100_000; N2 = 1_000_000
|
|
N1.times{ary = []; N2.times{ary << ''}}
|
|
Maybe this problem is remaining in Ruby 2.0.0.
|
|
|
|
* gc.c (unlink_heap_slot): remove not working code.
|
|
|
|
Wed Jul 17 14:31:13 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: re-design the heap structure.
|
|
|
|
(1) The heap is consists of a set of slots.
|
|
(2) Each "slot" has a "slot_body".
|
|
slot::start and slot::limit specify RVALUE beginning address
|
|
and number of RVALUE in a "slot_body".
|
|
(3) "slot_body" contains a pointer to slot (slot_body::header::slot)
|
|
and an array of RVALUE.
|
|
(4) heap::sorted is an array of "slots", sorted by an address of
|
|
slot::body.
|
|
|
|
See https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/GC_design
|
|
for more details (figure).
|
|
|
|
* gc.c: Avoid "heaps" terminology. It is ambiguous.
|
|
|
|
Wed Jul 17 13:29:16 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: fix heaps_header and heaps_slot to reduce memory consumption.
|
|
(1) move heaps_header::start and limit to heaps_slot.
|
|
(2) remove heaps_header::end which can be calculated by start+limit.
|
|
|
|
* gc.c: catch up above change.
|
|
|
|
Wed Jul 17 12:30:05 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/st.h (st_strcasecmp): Macro defined for compatibility.
|
|
(st_strncasecmp): Ditto.
|
|
|
|
Wed Jul 17 11:57:45 2013 Takeyuki FUJIOKA <xibbar@ruby-lang.org>
|
|
|
|
* lib/cgi/util.rb (CGI::Util#escape, unescape): Avoid use of regexp
|
|
special global variable. [Feature #8648] Thanks to fotos.
|
|
|
|
Wed Jul 17 11:57:10 2013 Takeyuki FUJIOKA <xibbar@ruby-lang.org>
|
|
|
|
* lib/erb.rb (ERB::Util#url_encode): Avoid use of regexp special global
|
|
variable. [Feature #8648] Thanks to fotos.
|
|
|
|
Wed Jul 17 08:12:41 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* st.c (st_locale_insensitive_strcasecmp): Renamed from st_strcasecmp.
|
|
(st_locale_insensitive_strncasecmp): Renamed from st_strncasecmp.
|
|
|
|
* include/ruby/st.h: Follow above changes.
|
|
|
|
* include/ruby/ruby.h: Ditto.
|
|
|
|
Wed Jul 17 00:14:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigmul1_toom3): Use bigdivrem_single instead of bigdivrem.
|
|
(big_three): Removed.
|
|
(Init_Bignum): Don't initialize big_three.
|
|
|
|
Tue Jul 16 21:46:03 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* configure.in: revert r42008. strcasecmp() uses the current locale.
|
|
|
|
* include/ruby/ruby.h: ditto.
|
|
|
|
* st.c (st_strcasecmp): ditto.
|
|
|
|
Tue Jul 16 21:07:04 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* configure.in: check strcasecmp().
|
|
|
|
* include/ruby/ruby.h: use strcasecmp() as st_strcasecmp() if it
|
|
exists.
|
|
|
|
* st.c (st_strcasecmp): define the function only if strcasecmp()
|
|
doesn't exist.
|
|
|
|
Tue Jul 16 20:21:28 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigsq): Renamed from bigsqr.
|
|
|
|
Tue Jul 16 19:42:08 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (USHORT): Unused macro removed.
|
|
|
|
Tue Jul 16 19:18:51 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: slim a path of newobj_of().
|
|
|
|
* gc.c (objspace): add a new field objspace::freelist, which contains
|
|
available RVALUEs.
|
|
|
|
* gc.c (newobj_of): simply call new function `get_freeobj()'.
|
|
get_freeobj() returns objspace::freelist. If objspace::freelist
|
|
is not available, refill objspace::freelist with a slot pointed by
|
|
objspace::heap::free_slots.
|
|
|
|
* gc.c (before_gc_sweep): clear objspace::freelist.
|
|
|
|
* gc.c (slot_sweep): clear slot::freelist.
|
|
|
|
* gc.c (heaps_prepare_freeslot): renamed to heaps_prepare_freeslot.
|
|
|
|
* gc.c (unlink_free_heap_slot): remove unused function.
|
|
|
|
* gc.c (rb_free_const_table): remove unused function.
|
|
|
|
Tue Jul 16 19:05:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big_shift3): Big shift width is not a problem for right
|
|
shift.
|
|
|
|
Tue Jul 16 18:50:08 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* array.c (rb_ary_count): [DOC] fix typo. Array#count uses ==, not
|
|
===. a question at asakusa.rb ML.
|
|
|
|
Tue Jul 16 18:35:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul_karatsuba): Avoid duplicate calculation when
|
|
squaring.
|
|
(bary_mul_toom3_branch): Ditto.
|
|
|
|
Tue Jul 16 17:43:22 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (link_free_heap_slot): removed.
|
|
|
|
* gc.c (slot_sweep): use `heaps_add_freeslot' instead of
|
|
`link_free_heap_slot'.
|
|
|
|
* gc.c (assign_heap_slot): use local variable `slot' instead of
|
|
`heaps'.
|
|
|
|
Tue Jul 16 17:21:39 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (assign_heap_slot): refactoring variable names.
|
|
|
|
* gc.c (slot_add_freeobj): added.
|
|
|
|
* gc.c (heaps_add_freeslot): added.
|
|
|
|
* gc.c (finalize_list, rb_gc_force_recycle, slot_sweep): use
|
|
`slot_add_freeobj' instead of modifying linked list directly.
|
|
|
|
Tue Jul 16 16:30:58 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (lazy_sweep): refactoring.
|
|
|
|
Tue Jul 16 13:32:06 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* encoding.c (enc_set_index): since r41967, old terminator is dealt
|
|
with in str_fill_term(). should not consider it here because this
|
|
function is called before any encoding is set.
|
|
|
|
Tue Jul 16 11:12:03 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* proc.c (rb_block_arity): raise ArgumentError if no block given.
|
|
|
|
Tue Jul 16 08:15:22 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* ext/bigdecimal/lib/bigdecimal/util.rb: [DOC] document top-level
|
|
classes from BigDecimal utils native extensions
|
|
|
|
Tue Jul 16 03:23:03 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* numeric.c: [DOC] improve rdoc formatting for parameters and links
|
|
|
|
Mon Jul 15 14:40:00 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/intern.h (rb_big2str0): Deprecated.
|
|
|
|
* bignum.c (rb_big2str1): Renamed from rb_big2str0.
|
|
(rb_big2str0): Deprecated wrapper for rb_big2str1.
|
|
(rb_big2str): Invoke rb_big2str1 instead of rb_big2str0.
|
|
|
|
Mon Jul 15 14:13:02 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* struct.c (rb_struct_each_pair): use rb_yield_values(2, key, value)
|
|
instead of rb_yield(rb_assoc_new(key, value)) if rb_block_arity()
|
|
is greater than 1.
|
|
|
|
Mon Jul 15 13:46:26 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Add static assertions.
|
|
|
|
Mon Jul 15 13:36:02 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_each_pair): performance improvement by using
|
|
rb_block_arity().
|
|
|
|
Mon Jul 15 13:15:37 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* proc.c (rb_block_arity): create internal API rb_block_arity().
|
|
it returns arity of given block.
|
|
|
|
Mon Jul 15 13:07:27 2013 Yuki Yugui Sonoda <yugui@yugui.jp>
|
|
|
|
* lib/prime.rb (Prime::EratosthenesGenerator,
|
|
Prime::EratosthenesSieve): New implementation by
|
|
robertjlooby <robertjlooby AT gmail.com>.
|
|
|
|
* test/test_prime.rb: updated with new method name
|
|
|
|
Mon Jul 15 11:32:46 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* numeric.c (rb_cNumeric): [DOC] Added comment for Numeric to fix doc
|
|
|
|
Mon Jul 15 11:24:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (maxpow_in_bdigit_dbl): Useless #if removed.
|
|
|
|
Mon Jul 15 11:10:46 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* bignum.c (rb_big_coerce): [DOC] Add docs for Bignum#coerce
|
|
Based on patch by Juanito Fatas [Fixes GH-360]
|
|
https://github.com/ruby/ruby/pull/360
|
|
|
|
Mon Jul 15 10:56:01 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* thread.c (mutex_sleep): [DOC] Awake thread will reacquire lock
|
|
By Tim Abdulla [Fixes GH-342] https://github.com/ruby/ruby/pull/342
|
|
|
|
Mon Jul 15 10:45:09 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (nlz16): Use __builtin_clz if possible.
|
|
(nlz32): Use __builtin_clz or __builtin_clzl if possible.
|
|
(nlz64): Use __builtin_clzl or __builtin_clzll if possible.
|
|
(nlz128): Use __builtin_clzll if possible.
|
|
|
|
* configure.in: Check __builtin_clz, __builtin_clzl and
|
|
__builtin_clzll.
|
|
|
|
Mon Jul 15 09:39:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (power_cache_get_power): Use bitsize instead of ceil_log2.
|
|
(ones): Removed.
|
|
(next_pow2): Removed.
|
|
(floor_log2): Removed.
|
|
(ceil_log2): Removed.
|
|
|
|
* configure.in (__builtin_popcountl): Don't check.
|
|
|
|
Mon Jul 15 02:47:09 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* localeinit.c (rb_locale_charmap, Init_enc_set_filesystem_encoding):
|
|
move from encoding.c.
|
|
|
|
* miniinit.c (rb_locale_charmap, Init_enc_set_filesystem_encoding):
|
|
define miniruby specific functions only.
|
|
|
|
Mon Jul 15 02:32:58 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* encoding.c (rb_enc_init): no longer needs NO_PRESERVED_ENCODING.
|
|
|
|
* encoding.c (enc_inspect): defer loading autoloaded encoding.
|
|
|
|
* encoding.c (enc_check_encoding): use is_data_encoding() to check
|
|
type consistently.
|
|
|
|
* encoding.c (must_encoding): return rb_encoding* instead of encoding
|
|
index.
|
|
|
|
* encoding.c (enc_check_encoding): use is_data_encoding() to check
|
|
type consistently.
|
|
|
|
* encoding.c (must_encoding): return rb_encoding* instead of encoding
|
|
index.
|
|
|
|
Mon Jul 15 02:21:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (str_fill_term): consider old terminator length, and should
|
|
not use rb_enc_ascget since it depends on the current encoding which
|
|
may not be compatible with the new terminator. [Bug #8634]
|
|
|
|
* encoding.c (enc_inspect): use PRIsVALUE to preserve the result
|
|
encoding.
|
|
|
|
Sun Jul 14 23:21:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Check __builtin_popcountl, __builtin_bswap32 and
|
|
__builtin_bswap64.
|
|
|
|
* internal.h (swap32): Use the configure result for the condition to
|
|
use __builtin_bswap32.
|
|
(swap64): Use the configure result for the condition to use
|
|
__builtin_bswap64.
|
|
|
|
* bignum.c (ones): Use the configure result for the condition to use
|
|
__builtin_popcountl.
|
|
(bary_unpack_internal): Use appropriate types for swap argument.
|
|
|
|
Sun Jul 14 22:21:11 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_subb): Support xn < yn.
|
|
(bigsub_core): Removed.
|
|
(bigsub): Don't compare before subtraction. Just subtract and
|
|
get the two's complement if the subtraction causes a borrow.
|
|
|
|
Sun Jul 14 00:36:03 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (DIGSPERLONG): Unused macro removed.
|
|
(DIGSPERLL): Ditto.
|
|
|
|
Sun Jul 14 00:32:51 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_aref): Less scan when the number is negative.
|
|
|
|
Sun Jul 14 00:17:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big_shift): Avoid signed integer overflow.
|
|
|
|
Sun Jul 14 00:14:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul_precheck): Use bary_small_lshift or
|
|
bary_mul_normal if xl is 1.
|
|
|
|
Sat Jul 13 22:58:16 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big_shift3): New function.
|
|
big_lshift and big_rshift are merged.
|
|
(big_shift2): New function.
|
|
(big_lshift): Use big_shift3.
|
|
(big_rshift): Ditto.
|
|
(check_shiftdown): Removed.
|
|
(rb_big_lshift): Use big_shift2 and big_shift3.
|
|
(rb_big_rshift): Ditto.
|
|
(big_lshift): Removed.
|
|
(big_rshift): Ditto.
|
|
|
|
Sat Jul 13 15:51:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_small_lshift): Use size_t instead of long.
|
|
(bary_small_rshift): Ditto.
|
|
|
|
Sat Jul 13 15:33:33 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_small_lshift): Functions moved to remove
|
|
declaration.
|
|
(bary_small_rshift): Ditto.
|
|
|
|
Sat Jul 13 12:27:34 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* encoding.c (rb_enc_associate_index): fill new terminator length, not
|
|
old one.
|
|
|
|
Sat Jul 13 12:24:24 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/win32: move from ext/dl and ext/fiddle. since ext/extmk.rb
|
|
builds extensions in alphabetical order, compiled?('fiddle') under
|
|
ext/dl makes no sense.
|
|
|
|
Sat Jul 13 09:26:09 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (biglsh_bang): Removed.
|
|
(bigrsh_bang): Ditto.
|
|
(bigmul1_toom3): Use bary_small_lshift and bary_small_rshift.
|
|
|
|
Sat Jul 13 01:04:43 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/rubygems/psych_additions.rb: Ignore Psych docs here
|
|
|
|
Fri Jul 12 18:10:46 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ext/fiddle/win32/lib/win32/registry.rb
|
|
(Win32::Registry::API#make_wstr): same as r41922.
|
|
|
|
Fri Jul 12 16:28:37 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* encoding.c (rb_enc_associate_index): refill the terminator if it
|
|
becomes longer than before. [ruby-dev:47500] [Bug #8624]
|
|
|
|
* string.c (str_null_char, str_fill_term): get rid of out of bound
|
|
access.
|
|
|
|
* string.c (rb_str_fill_terminator): add a parameter for the length of
|
|
new terminator.
|
|
|
|
Fri Jul 12 11:26:25 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_reject_bang): do not call rb_hash_foreach() if RHash
|
|
has ntbl and it is empty.
|
|
|
|
Fri Jul 12 11:17:41 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (recursive_hash): use RHASH_SIZE() to check hash size.
|
|
|
|
Fri Jul 12 00:20:00 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_size): use RHASH_SIZE().
|
|
|
|
Fri Jul 12 00:08:24 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_values): set array capa to RHASH_SIZE().
|
|
|
|
Thu Jul 11 23:54:45 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* hash.c (rb_hash_keys): set array capa to RHASH_SIZE().
|
|
|
|
Thu Jul 11 21:30:17 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (rb_w32_pow): undef pow to get rid of infinite
|
|
recursive call. re-fix [Bug #8495]. [ruby-core:55923] [Bug #8621]
|
|
|
|
Thu Jul 11 20:18:13 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/dl/win32/lib/win32/registry.rb (Win32::Registry::API#make_wstr):
|
|
remove workaround to append WCHAR terminator.
|
|
|
|
* transcode.c (str_encode_associate): fill terminator after conversion.
|
|
|
|
* string.c (rb_enc_str_new, rb_str_set_len, rb_str_resize): fill
|
|
minimum length of the encoding as the terminator.
|
|
|
|
* string.c (str_buf_cat, rb_str_buf_append, rb_str_splice_0): ditto.
|
|
|
|
* string.c (str_make_independent_expand, rb_str_modify_expand): make
|
|
the capacity enough for multi-byte terminator.
|
|
|
|
* string.c (rb_string_value_cstr): fill minimum length of the encoding
|
|
as the terminator.
|
|
|
|
* string.c (rb_string_value_cstr): check null char in char, not in
|
|
byte.
|
|
|
|
Thu Jul 11 14:48:35 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* array.c: Replace confusing example for #reverse_each in overview
|
|
Patch by Earl St Sauver [Fixes documenting-ruby/ruby-12]
|
|
https://github.com/documenting-ruby/ruby/pull/12
|
|
|
|
Thu Jul 11 14:22:37 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* test/drb/ut_eq.rb: Use localhost for drb tests [Bug #7311]
|
|
Patch by Vit Ondruch [ruby-core:49101]
|
|
* test/drb/ut_array.rb: ditto
|
|
* test/drb/ut_array_drbssl.rb: ditto
|
|
|
|
Thu Jul 11 13:48:03 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* sprintf.c: Fix typo patch by @hynkle [Fixes GH-357]
|
|
https://github.com/ruby/ruby/pull/357
|
|
|
|
Thu Jul 11 13:00:34 2013 Zachary Scott <e@zzak.io>
|
|
|
|
* lib/securerandom.rb: Refactor conditions by Rafal Chmiel
|
|
[Fixes GH-326] https://github.com/ruby/ruby/pull/326
|
|
|
|
Thu Jul 11 12:04:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Don't use toom3 after once karatsuba is chosen.
|
|
(mulfunc_t): New type.
|
|
(bary_mul_toom3_start): Renamed from bary_mul.
|
|
(bary_mul_karatsuba_start): Renamed from bary_mul.
|
|
(bary_mul_balance_with_mulfunc): Renamed from bary_mul_balance and
|
|
new argument, mulfunc, is added.
|
|
(rb_big_mul_balance): Invoke bary_mul_balance_with_mulfunc with
|
|
bary_mul_toom3_start.
|
|
(bary_mul_karatsuba): Invoke bary_mul_karatsuba_start instead of
|
|
bary_mul.
|
|
(bary_mul_precheck): Extracted from bary_mul.
|
|
(bary_mul_karatsuba_branch): Extracted from bary_mul.
|
|
(bary_mul_karatsuba_start): New function to call bary_mul_precheck
|
|
and bary_mul_karatsuba_branch.
|
|
(bary_mul_toom3_branch): Extracted from bary_mul.
|
|
(bary_mul_toom3_start): New function to call bary_mul_precheck and
|
|
bary_mul_toom3_branch.
|
|
(bary_mul): Just call bary_mul_toom3_start.
|
|
Arguments for work memory are removed.
|
|
(rb_cstr_to_inum): Follow the bary_mul change.
|
|
(bigmul0): Ditto.
|
|
|
|
Thu Jul 11 10:46:38 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* tool/probes_to_wiki.rb: fix usage comment. use Enumerable#grep
|
|
which yields each elements to reduce unnecessary array.
|
|
|
|
Thu Jul 11 10:09:18 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* process.c (rb_daemon): daemon(3) is implemented with fork(2).
|
|
Therefore it needs rb_thread_atfork(). (and revert r41903)
|
|
|
|
Thu Jul 11 03:22:10 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* tool/probes_to_wiki.rb: adding a script to convert probes.d to wiki
|
|
format for easy wiki updates.
|
|
|
|
Thu Jul 11 00:54:07 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* man/ri.1: Incorrect use of .Dd macro [Bug #8620] by Tristan Hill
|
|
|
|
Thu Jul 11 00:48:29 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/delegate.rb: Add example for __setobj__ and __getobj__
|
|
[Bug #8615] Patch by Caleb Thompson
|
|
|
|
Wed Jul 10 23:29:22 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/logger.rb: Use :call-seq: for method signature rdoc
|
|
|
|
Wed Jul 10 23:23:18 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/logger.rb (#add): Remove incorrect rdoc for return value
|
|
[Bug #8567] Reported by Tim Pease.
|
|
|
|
Wed Jul 10 23:12:00 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_str_subpos): make public function.
|
|
|
|
Wed Jul 10 22:44:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Add a static assertion for RBIGNUM_EMBED_LEN_MAX.
|
|
|
|
Wed Jul 10 22:31:25 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* string.c (rb_str_index): cache single byte flag and some
|
|
cosmetic changes.
|
|
|
|
Wed Jul 10 22:03:27 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_2comp): Don't use bary_plus_one.
|
|
(bary_add_one): Replaced by the implementation of bary_plus_one.
|
|
|
|
Wed Jul 10 20:48:22 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* bignum.c (sizeof_bdigit_dbl): check sizeof(BDIGIT_DBL).
|
|
|
|
* internal.h (STATIC_ASSERT): move from enum.c.
|
|
|
|
Wed Jul 10 20:08:21 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (SIZEOF_BDIGIT_DBL): Add a ifdef guard for test.
|
|
|
|
Wed Jul 10 14:18:59 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* process.c (fork_daemon): kill the other threads all and abandon the
|
|
kept mutexes.
|
|
|
|
Wed Jul 10 11:35:36 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/net/http/test_http.rb (TestNetHTTP_v1_2#test_get,
|
|
TestNetHTTP_v1_2_chunked#test_get): shouldn't check
|
|
HttpResponse#decode_content if Zlib is not available.
|
|
ko1 complained via IRC.
|
|
|
|
Wed Jul 10 10:20:07 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* tool/rbinstall.rb: always require rubygems to stabilize rubygems
|
|
related status like whether Gem::Specification is defined or not.
|
|
|
|
* tool/rbinstall.rb (Gem::Specification.unresolved_deps): define stub.
|
|
|
|
Wed Jul 10 08:21:15 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Import RubyGems 2.1
|
|
* test/rubygems: Ditto.
|
|
|
|
Wed Jul 10 07:34:34 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb: Remove siteconf file after
|
|
building the gem.
|
|
* test/rubygems/test_gem_ext_ext_conf_builder.rb: Test for the above.
|
|
|
|
* lib/rubygems/psych_tree.rb (module Gem): Add backward compatibility
|
|
for r41148
|
|
|
|
* test/rubygems/test_gem_package.rb: Add backward compatibility for
|
|
double-slash elimination.
|
|
|
|
Wed Jul 10 06:22:27 2013 Tadayoshi Funaba <tadf@dotrb.org>
|
|
|
|
* ext/date/date_parse.c (date_zone_to_diff): [ruby-core:55831].
|
|
|
|
Wed Jul 10 00:41:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul): x*1 is x.
|
|
|
|
Tue Jul 9 22:24:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul1): No need to invoke MEMZERO at last.
|
|
(bary_mul_single): Invoke MEMZERO here.
|
|
|
|
Tue Jul 9 21:40:01 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/rexml/test_text.rb: Add missing tests for Text#<<.
|
|
Reported by nagachika. Thanks!!!
|
|
|
|
Tue Jul 9 18:02:38 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/fileutils.rb (FileUtils#chown_R): Do not skip traversal even
|
|
if user and group are both nil, to be consistent with #chown and
|
|
other commands.
|
|
|
|
Tue Jul 9 17:58:26 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* test/fileutils/test_fileutils.rb
|
|
(TestFileUtils#assert_output_lines): New utility assertion
|
|
method for testing verbose output.
|
|
|
|
Tue Jul 9 17:43:57 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/test_tracer.rb: catch up recent rubygems changes.
|
|
|
|
Tue Jul 9 16:58:30 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ext/{dl,fiddle}/win32/lib/win32/registry.rb: hope that the final
|
|
resolution to fix the failure of test-all. and includes Win64
|
|
support (fixed a potential bug).
|
|
|
|
Tue Jul 9 15:57:20 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* object.c: Fix rdoc for Kernel#<=>. [Fixes GH-352]
|
|
|
|
Tue Jul 9 15:53:51 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/fileutils.rb (FileUtils#mode_to_s): Define mode_to_s() also
|
|
as singleton method, or FileUtils.chmod fails in verbose mode.
|
|
|
|
Tue Jul 9 15:16:02 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* test/fileutils/fileasserts.rb
|
|
(Test::Unit::FileAssertions#assert_not_symlink): Add a missing
|
|
optional argument "message".
|
|
|
|
Tue Jul 9 15:03:24 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/fileutils.rb (FileUtils#chown, FileUtils#chown_R): If user
|
|
and group are both nil, print ":".
|
|
|
|
Tue Jul 9 12:47:08 2013 Masaki Matsushita <glass.saga@gmail.com>
|
|
|
|
* io.c (appendline): use READ_CHAR_PENDING_XXX macros and
|
|
RSTRING_END().
|
|
|
|
* io.c (rb_io_getline_1): rewrite nested if statement into one
|
|
statement.
|
|
|
|
Tue Jul 9 11:04:35 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ext/{dl,fiddle}/win32/lib/win32/registry.rb (Win32::Registry#check):
|
|
should report the position of the error.
|
|
|
|
* ext/{dl,fiddle}/win32/lib/win32/registry.rb
|
|
(Win32::Registry#QueryValue): workaround for test-all crash.
|
|
|
|
Tue Jul 9 10:27:56 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ext/{dl,fiddle}/win32/lib/win32/registry.rb
|
|
(Win32::Registry.expand_environ): use suitable encoding for the
|
|
string.
|
|
|
|
* ext/{dl,fiddle}/win32/lib/win32/registry.rb (Win32::Registry#read):
|
|
should return REG_SZ, REG_EXPAND_SZ and REG_MULTI_SZ values with
|
|
the expected encoding -- assumed as the same encoding of name.
|
|
|
|
Tue Jul 9 10:02:45 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ext/{dl,fiddle}/win32/lib/win32/registry.rb
|
|
(Win32::Registry::Error#initialize): use suitable encoding for the
|
|
string.
|
|
|
|
Tue Jul 9 09:46:53 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ext/dl/win32/lib/win32/registry.rb (Win32::Registry.expand_environ):
|
|
use suitable encoding for the string. fixed a test-all error of
|
|
r41838.
|
|
|
|
* ext/fiddle/win32/lib/win32/registry.rb: same changes of r41838 and
|
|
this revision of dl's win32/registry.rb.
|
|
|
|
Tue Jul 9 07:39:45 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems 2.0.4. See
|
|
https://github.com/rubygems/rubygems/blob/2.0/History.txt for changes
|
|
|
|
Tue Jul 9 01:47:16 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (biglsh_bang): Don't shift a BDIGIT with BITSPERDIG bits.
|
|
(bigrsh_bang): Ditto.
|
|
|
|
Tue Jul 9 01:17:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigrsh_bang): Fix bignum digits overrun.
|
|
|
|
Tue Jul 9 00:46:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (biglsh_bang): Fix bignum digits under-run.
|
|
|
|
Mon Jul 8 23:36:45 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/dl/win32/lib/win32/registry.rb (Error, API): use WCHAR
|
|
interfaces. c.f. [Bug #8508]
|
|
|
|
Mon Jul 8 23:13:11 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (rb_w32_pow): move from win32.h and disable strict
|
|
ANSI mode macro to let _controlfp() stuff defined.
|
|
[ruby-core:55312] [Bug #8495]
|
|
|
|
* numeric.c (finite): add declaration for strict ANSI.
|
|
[ruby-core:55312] [Bug #8495]
|
|
|
|
* thread_win32.c (w32_thread_start_func, thread_start_func_1),
|
|
(timer_thread_func): use __stdcall instead of _stdcall which is
|
|
unavailable in strict ANSI mode. [ruby-core:55312] [Bug #8495]
|
|
|
|
* win32/win32.c (gettimeofday): use __cdecl instead of _cdecl.
|
|
|
|
Mon Jul 8 22:41:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul): Arguments for work memory added.
|
|
(bary_mul_balance): Ditto.
|
|
(bary_mul_karatsuba): Ditto.
|
|
|
|
Mon Jul 8 22:03:30 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_sq_fast): New function for testing.
|
|
(rb_big_mul_toom3): Ditto.
|
|
|
|
* internal.h (rb_big_sq_fast): Declared.
|
|
(rb_big_mul_toom3): Ditto.
|
|
|
|
Mon Jul 8 21:59:34 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul_balance): Initialize a local variable to suppress
|
|
a warning.
|
|
|
|
Mon Jul 8 20:55:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul_balance): Reduce work memory.
|
|
|
|
Mon Jul 8 08:26:15 2013 Martin Bosslet <Martin.Bosslet@gmail.com>
|
|
|
|
* test/openssl/test_pkey_ec.rb: Skip tests for "Oakley" curves as
|
|
they are not suitable for ECDSA.
|
|
[ruby-core:54881] [Bug #8384]
|
|
|
|
Mon Jul 8 08:03:01 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul): Add a RB_GC_GUARD.
|
|
|
|
Sun Jul 7 23:56:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul_karatsuba): Unreachable code removed. Remove
|
|
several branches.
|
|
|
|
Sun Jul 7 22:59:06 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (rb_big_mul_normal): Declared.
|
|
(rb_big_mul_balance): Ditto.
|
|
(rb_big_mul_karatsuba): Ditto.
|
|
|
|
* bignum.c (rb_big_mul_normal): New function for tests.
|
|
(rb_big_mul_balance): Ditto.
|
|
(rb_big_mul_karatsuba): Ditto.
|
|
|
|
Sun Jul 7 19:21:30 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Reorder functions to decrease forward reference.
|
|
|
|
Sun Jul 7 14:41:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: (bigsub_core): Use bary_sub.
|
|
(bary_sub): Returns a borrow flag. Use bary_subb.
|
|
(bary_subb): New function for actually calculating subtraction with
|
|
borrow.
|
|
(bary_sub_one): New function.
|
|
(bigadd_core): Use bary_add.
|
|
(bary_add): Returns a carry flag. Use bary_addc.
|
|
(bary_addc): New function for actually calculating addition with
|
|
carry.
|
|
(bary_add_one): New function.
|
|
(bary_muladd_1xN): Extracted from bary_mul_normal.
|
|
(bigmul1_normal): Removed.
|
|
(bary_mul_karatsuba): New function.
|
|
(bary_mul1): Invoke rb_thread_check_ints after bary_mul_normal.
|
|
(bary_mul): Remove most and least significant zeros before actual
|
|
multiplication. Use bary_sq_fast, bary_mul_balance,
|
|
bary_mul_karatsuba and bigmul1_toom3 as bigmul0.
|
|
(bigmul1_balance): Removed.
|
|
(bigmul1_karatsuba): Removed.
|
|
(bigsqr_fast): Removed.
|
|
(bary_sparse_p): Extracted from big_sparse_p.
|
|
(big_sparse_p): Removed.
|
|
(bigmul0): Use bary_mul.
|
|
|
|
Sun Jul 7 11:54:33 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* NEWS: Add REXML::Text#<< related updates.
|
|
|
|
Sun Jul 7 11:49:19 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/text.rb (REXML::Text#<<): Support appending in not
|
|
"raw" mode. [Bug #8602] [ruby-dev:47482]
|
|
Reported by Ippei Obayashi. Thanks!!!
|
|
|
|
Sun Jul 7 11:43:13 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/text.rb (REXML::Text#<<): Support method chain use by "<<"
|
|
like other objects.
|
|
|
|
Sun Jul 7 11:34:18 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/text.rb (REXML::Text#clear_cache): Extract common
|
|
cache clear code.
|
|
|
|
Sun Jul 7 11:01:03 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (RUBY_DTRACE_POSTPROCESS): dtrace version SUN D 1.11
|
|
introduces a check in the dtrace compiler to ensure that probes
|
|
actually exist. If there are no probes, then the -G step will
|
|
fail. As this test is only being used to determine whether -G is
|
|
necessary (for instance, on OSX it is not), adding a real probe to
|
|
the conftest allows it to succeed on newer versions of dtrace.
|
|
Patch by Eric Saxby <sax AT livinginthepast.org> at
|
|
[ruby-core:55826]. [Fixes GH-351], [Bug #8606].
|
|
|
|
Sun Jul 7 10:07:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_sq_fast): Extracted from bigsqr_fast and
|
|
ensure not to access zds[2*xn].
|
|
(bigsqr_fast): Allocate the result bignum with 2*xn words.
|
|
|
|
Sat Jul 6 07:37:43 2013 Martin Bosslet <Martin.Bosslet@gmail.com>
|
|
|
|
* ext/openssl/ossl_pkey_ec.c: Ensure compatibility to builds of
|
|
OpenSSL with OPENSSL_NO_EC2M defined, but OPENSSL_NO_EC not
|
|
defined.
|
|
* test/openssl/test_pkey_ec.rb: Iterate over built-in curves
|
|
(and assert their non-emptiness!) instead of hard-coding them, as
|
|
this may cause problems with respect to the different availability
|
|
of individual curves in individual OpenSSL builds.
|
|
[ruby-core:54881] [Bug #8384]
|
|
|
|
Thanks to Vit Ondruch for providing the patch!
|
|
|
|
Sat Jul 6 07:12:39 2013 Martin Bosslet <Martin.Bosslet@gmail.com>
|
|
|
|
* test/openssl/test_x509crl.rb: Remove unused variable.
|
|
[ruby-core:53501] [Bug #8114]
|
|
|
|
Thanks, Vipul Amler, for pointing this out!
|
|
|
|
Sat Jul 6 06:37:10 2013 Martin Bosslet <Martin.Bosslet@gmail.com>
|
|
|
|
* ext/openssl/ossl.c: Provide CRYPTO_set_locking_callback() and
|
|
CRYPTO_set_id_callback() callback functions ossl_thread_id and
|
|
ossl_lock_callback to ensure the OpenSSL extension is usable in
|
|
multi-threaded environments.
|
|
[ruby-core:54900] [Bug #8386]
|
|
|
|
Thanks, Dirkjan Bussink, for the patch!
|
|
|
|
Sat Jul 6 06:06:16 2013 Martin Bosslet <Martin.Bosslet@gmail.com>
|
|
|
|
* lib/openssl/ssl.rb: Fix SSL client connection crash for SAN marked
|
|
critical.
|
|
The patch for CVE-2013-4073 caused SSL crash when a SSL server returns
|
|
the certificate that has critical SAN value. X509 extension could
|
|
include 2 or 3 elements in it:
|
|
|
|
[id, criticality, octet_string] if critical,
|
|
[id, octet_string] if not.
|
|
|
|
Making sure to pick the last element of X509 extension and use it as
|
|
SAN value.
|
|
[ruby-core:55685] [Bug #8575]
|
|
|
|
Thank you @nahi for providing the patch!
|
|
|
|
Sat Jul 6 04:49:38 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych/visitors/yaml_tree.rb: register time objects so
|
|
they are referenced as ids during output.
|
|
* test/psych/test_date_time.rb: corresponding test.
|
|
|
|
Fri Jul 5 20:46:39 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/ruby/test_unicode_escape.rb (TestUnicodeEscape#test_basic): this
|
|
assertion doesn't seems to be checking the unicode string on command
|
|
line, but seems to be checking how to treat the unicode string from
|
|
stdin. so, should escape '\' before 'u'. this fixes a test failure
|
|
on Windows.
|
|
|
|
Fri Jul 5 19:05:40 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/fileutils.rb (FileUtils#chown, FileUtils#chown_R): Fix the
|
|
wrong output message when user is nil, which should be "chown
|
|
:group file" instead of "chown group file".
|
|
|
|
Fri Jul 5 16:21:56 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* test/ruby/test_regexp.rb
|
|
(TestRegexp#test_options_in_look_behind)
|
|
(TestRegexp#assert_match_at): Add tests for another problem
|
|
fixed in Onigmo 5.13.5. Previously Onigmo did not allow option
|
|
enclosures in look-behind, which makes it impossible to
|
|
interpolate a regexp into another in the middle of a look-behind
|
|
pattern. cf. https://github.com/k-takata/Onigmo/pull/17
|
|
|
|
* test/ruby/test_regexp.rb
|
|
(TestRegexp#test_options_in_look_behind)
|
|
(TestRegexp#assert_match_at): Parse regexps in run time rather
|
|
than in compile time.
|
|
|
|
Fri Jul 5 12:14:40 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_notfound): after
|
|
r41710, the path of command uses backslash as the separator on
|
|
Windows.
|
|
|
|
Fri Jul 5 11:29:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/test/unit/assertions.rb (assert_raise_with_message): move from
|
|
test/fileutils/test_fileutils.rb. this is still experimental and
|
|
the interface may be changed.
|
|
|
|
Fri Jul 5 11:08:00 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/win32.c (w32_spawn): r41710 made that if the command starts with
|
|
a quote and includes slash, removed the top quote and NOT removed the
|
|
last quote.
|
|
this fixes test failures on test/ruby/test_process.rb and
|
|
test/webrick.
|
|
|
|
Fri Jul 5 09:53:15 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
* lib/mkmf.rb (CONFIG['CPPOUTFILE']): fix r41769; CONFIG['CPPOUTFILE']
|
|
may be nil.
|
|
|
|
Fri Jul 5 05:39:53 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (BARY_MUL1): Renamed from BARY_MUL.
|
|
(bary_mul1): Renamed from bary_mul.
|
|
(bary_mul): Renamed from bary_mul2.
|
|
|
|
Fri Jul 5 04:58:05 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul_balance): Extracted from bigmul1_balance and
|
|
use bary_mul2 and bary_add to decrease allocations.
|
|
|
|
Fri Jul 5 02:14:00 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/fileutils.rb (FileUtils#symbolic_modes_to_i): Fix the wrong
|
|
character class [+-=], which happened to match all desired
|
|
characters but also match undesired characters.
|
|
|
|
* lib/fileutils.rb (FileUtils.chmod{,_R}): Enhance the symbolic
|
|
mode parser to support the permission symbols u/g/o and multiple
|
|
actions as defined in SUS, so that chmod("g=o+w", file) works as
|
|
expected. Invalid symbolic modes are now rejected with
|
|
ArgumentError.
|
|
|
|
Fri Jul 5 00:25:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (have_framework): allow header file to check.
|
|
[ruby-core:55745] [Bug #8593]
|
|
|
|
Thu Jul 4 22:31:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* object.c (rb_obj_equal): Fixed an rb_obj_equal documentation typo
|
|
where "a" was used instead of "obj".
|
|
Fixes GH-349. Patch by @adnandoric
|
|
|
|
Thu Jul 4 20:39:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* tool/make-snapshot: Exit with EXIT_FAILURE when it fails.
|
|
|
|
Thu Jul 4 20:20:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (maxpow_in_bdigit_dbl): Use tables if available.
|
|
(maxpow_in_bdigit): Ditto.
|
|
(U16): New macro.
|
|
(U32): Ditto.
|
|
(U64): Ditto.
|
|
(U128): Ditto.
|
|
(maxpow16_exp): New table.
|
|
(maxpow16_num): New table.
|
|
(maxpow32_exp): New table.
|
|
(maxpow32_num): New table.
|
|
(maxpow64_exp): New table.
|
|
(maxpow64_num): New table.
|
|
(maxpow128_exp): New table.
|
|
(maxpow128_num): New table.
|
|
|
|
Thu Jul 4 18:25:25 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_cstr_to_inum): Avoid temporary buffer allocation except
|
|
very big base non-power-of-2 numbers.
|
|
|
|
Thu Jul 4 15:51:56 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* string.c (rb_str_succ): use ONIGENC_MBCLEN_CHARFOUND_P correctly.
|
|
|
|
* string.c (rb_str_dump): ditto.
|
|
|
|
Thu Jul 4 10:04:11 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* regcomp.c (): Merge Onigmo 5.13.5 23b523076d6f1161.
|
|
|
|
* [bug] (thanks Akinori MUSHA and Ippei Obayashi)
|
|
Fix a renumbering bug in condition regexp with a named
|
|
capture. [Bug #8583]
|
|
* [spec] (thanks Akinori MUSHA)
|
|
Allow ENCLOSE_OPTION in look-behind.
|
|
|
|
Thu Jul 4 00:36:03 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* internal.h (SIGNED_INTEGER_MAX): suppress warning C4146 on VC6.
|
|
seems a logical ORed expression becomes unsigned.
|
|
|
|
Thu Jul 4 00:13:01 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ruby_atomic.h (rb_w32_atomic_cas): call InterlockedCompareExchange
|
|
directly.
|
|
|
|
* ruby_atomic.h (ATOMIC_CAS): fix missing function call.
|
|
|
|
Wed Jul 3 23:47:35 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ruby_atomic.h (ATOMIC_CAS): suppress C4022 and C4047 warnings in
|
|
VC6. only InterlockedCompareExchange is declared using PVOID.
|
|
|
|
Wed Jul 3 22:29:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (ruby_digit36_to_number_table): Declared.
|
|
|
|
* util.c (ruby_digit36_to_number_table): Moved from scan_digits.
|
|
|
|
* bignum.c (conv_digit): Use ruby_digit36_to_number_table.
|
|
|
|
* pack.c (hex2num): Ditto.
|
|
|
|
Wed Jul 3 18:12:56 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (install_dirs): revert DESTDIR prefix by r39841, since
|
|
it is fixed by r41648. [ruby-core:55760] [Bug #8115]
|
|
|
|
Wed Jul 3 14:15:25 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* dir.c (do_stat): use rb_w32_ustati64() in win32.c to get rid of
|
|
mysterious behavior of FindFirstFile() Windows API which treat "<"
|
|
and ">" like as wildcard characters. [ruby-core:55764] [Bug #8597]
|
|
|
|
Wed Jul 3 12:06:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (maxpow_in_bdigit): Renamed from calc_hbase and return
|
|
maxpow.
|
|
|
|
Tue Jul 2 23:47:50 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (roomof): Cast to long.
|
|
(rb_ull2big): Fix bignew arguments.
|
|
|
|
Tue Jul 2 21:17:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_cstr_to_inum): Merge two temporary buffers.
|
|
|
|
Tue Jul 2 20:25:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_cstr_to_inum): Use BDIGIT_DBL to collect adjacent digits.
|
|
(BDIGIT_DBL_MAX): New macro.
|
|
(maxpow_in_bdigit_dbl): New function.
|
|
|
|
Tue Jul 2 17:23:33 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* doc/syntax/refinements.rdoc: add description of Module#using and
|
|
refinement inheritance by module inclusion.
|
|
|
|
Tue Jul 2 17:22:44 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* internal.h: add EUC-JP and Windows-31J.
|
|
|
|
* re.c (rb_char_to_option_kcode): use built-in encoding indexes in
|
|
internal.h.
|
|
|
|
* internal.h: add UTF8-MAC.
|
|
|
|
* dir.c (rb_utf8mac_encoding): use built-in encoding indexes in
|
|
internal.h.
|
|
|
|
* internal.h: add UTF-{16,32} dummy encodings.
|
|
|
|
* string.c (rb_str_inspect, str_scrub0): use built-in encoding indexes
|
|
in internal.h.
|
|
|
|
* internal.h: add UTF-{16,32}{BE,LE}.
|
|
|
|
* io.c (io_strip_bom): use built-in encoding indexes in internal.h.
|
|
|
|
* internal.h (rb_{ascii8bit,utf8,usascii}_encindex): use built-in
|
|
encoding indexes for optimization.
|
|
|
|
* encoding.c (enc_inspect, rb_locale_encindex),
|
|
(enc_set_filesystem_encoding, rb_filesystem_encindex): use built-in
|
|
encoding indexes directly.
|
|
|
|
* encoding.c (rb_enc_set_index, rb_enc_associate_index): validate
|
|
argument encoding index.
|
|
|
|
* include/ruby/encoding.h (ENCODING_SET): use rb_enc_set_index()
|
|
instead of setting inlined bits directly.
|
|
|
|
* encoding.c (rb_enc_init): register preserved indexes.
|
|
|
|
* internal.h (ruby_preserved_encindex): move from encoding.c.
|
|
|
|
Tue Jul 2 11:14:36 2013 Shota Fukumori <sorah@cookpad.com>
|
|
|
|
* lib/mkmf.rb (try_config): Fix to not replace $LDFLAGS with $libs
|
|
(1.9.3 behavior) [ruby-core:55752] [Bug #8595]
|
|
|
|
Tue Jul 2 00:39:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/ipsocket.c (init_inetsock_internal): Don't try mismatched
|
|
address family if already failed.
|
|
|
|
Mon Jul 1 23:07:38 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* template/encdb.h.tmpl: define encoding index macros to use the index
|
|
statically from C source.
|
|
|
|
Mon Jul 1 22:57:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul2): New function.
|
|
(rb_cstr_to_inum): Use a better algorithm to compose the result
|
|
if input length is very long.
|
|
|
|
Mon Jul 1 20:22:00 2013 Kenta Murata <mrkn@cookpad.com>
|
|
|
|
* ext/bigdecimal/bigdecimal.h (RB_UNUSED_VAR, UNREACHABLE):
|
|
import macros from ruby.h for 1.9.3.
|
|
[Bug #8588] [ruby-core:55730]
|
|
|
|
* ext/bigdecimal/bigdecimal.gemspec: Bump version to 1.2.1.
|
|
|
|
Mon Jul 1 20:03:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/ipsocket.c (init_inetsock_internal): Use an address
|
|
family for local address which is different to the remote
|
|
address if no other choice.
|
|
|
|
Mon Jul 1 15:05:00 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/csv.rb (CSV#<<): use StringIO#set_encoding instead of creating
|
|
new StringIO instance with String#force_encoding, forcing encoding
|
|
discards the cached coderange bits and can make further operations
|
|
very slow. [ruby-core:55714] [Bug #8585]
|
|
|
|
* ext/stringio/stringio.c (strio_write): keep coderange of
|
|
ptr->string.
|
|
|
|
* string.c (rb_enc_cr_str_buf_cat, rb_str_append): consider an empty
|
|
string 7bit-clean and should not discard cached coderange of string
|
|
to be appended.
|
|
|
|
Mon Jul 1 12:56:41 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* eval.c (rb_using_module): activate refinements in the ancestors of
|
|
the argument module to support refinement inheritance by
|
|
Module#include. [ruby-core:55671] [Feature #8571]
|
|
|
|
* test/ruby/test_refinement.rb: related test.
|
|
|
|
Mon Jul 1 12:02:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_cstr_to_inum): Skip leading zeros.
|
|
|
|
Mon Jul 1 00:59:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (nlz16): New function.
|
|
(nlz32): Ditto.
|
|
(nlz64): Ditto.
|
|
(nlz128): Ditto.
|
|
(nlz): Redefined using an above function.
|
|
(bitsize): New macro.
|
|
(rb_cstr_to_inum): Use bitsize instead of nlz.
|
|
|
|
Sun Jun 30 22:40:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* lib/prime.rb: Corrected a few comments. Patch by @Nullset14.
|
|
Fixes GH-346.
|
|
|
|
Sun Jun 30 21:53:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_cstr_to_inum): Use rb_integer_unpack if base is a power
|
|
of 2.
|
|
|
|
Sun Jun 30 10:59:23 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (join_argv): use backslash instead of slash in program
|
|
path, otherwise cannot invoke "./c\u{1ee7}a.exe" for some reason.
|
|
[ruby-core:24309] [Bug #1771]
|
|
|
|
* io.c (spawnv, spawn): use UTF-8 spawn family. [Bug #1771]
|
|
|
|
* process.c (proc_exec_sh, proc_spawn_cmd, proc_spawn_sh): ditto.
|
|
|
|
* win32/win32.c (translate_char, join_argv, has_redirection): make
|
|
codepage aware.
|
|
|
|
* win32/win32.c (rb_w32_udln_find_exe_r, rb_w32_udln_find_file_r):
|
|
codepage independent versions.
|
|
|
|
* win32/win32.c (w32_spawn): extract codepage aware code from
|
|
rb_w32_spawn().
|
|
|
|
* win32/win32.c (rb_w32_uspawn): add UTF-8 version function.
|
|
|
|
* win32/win32.c (w32_aspawn_flags): extract codepage aware code from
|
|
rb_w32_aspawn_flags().
|
|
|
|
* win32/win32.c (rb_w32_uaspawn_flags, rb_w32_uaspawn_flags): add
|
|
UTF-8 version functions.
|
|
|
|
* win32/win32.c (w32_getenv): extract codepage aware code from
|
|
rb_w32_ugetenv() and rb_w32_getenv().
|
|
|
|
* win32/win32.c (w32_stati64): extract codepage aware code from
|
|
rb_w32_ustati64() and rb_w32_stati64().
|
|
|
|
* dln.h (DLN_FIND_EXTRA_ARG, DLN_FIND_EXTRA_ARG_DECL): allow extra
|
|
arguments to dln_find_{exe,file}_r().
|
|
|
|
* dln_find.c (dln_find_exe_r, dln_find_file_r): add extract arguments.
|
|
|
|
* process.c (EXPORT_STR, EXPORT_DUP): convert to default process
|
|
encoding if defined.
|
|
|
|
* process.c (check_exec_env_i): convert environment variables too.
|
|
|
|
* process.c (rb_exec_fillarg): convert program path and arguments too.
|
|
|
|
Sun Jun 30 01:57:08 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big_rshift): Use abs2twocomp and twocomp2abs_bang.
|
|
|
|
Sun Jun 30 00:14:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (RBIGNUM_SET_NEGATIVE_SIGN): New macro.
|
|
(RBIGNUM_SET_POSITIVE_SIGN): Ditto.
|
|
(rb_big_neg): Inline get2comp to avoid double negation.
|
|
|
|
Sat Jun 29 23:26:41 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_neg): Extracted from bary_2comp.
|
|
(bary_plus_one): Extracted from bary_2comp.
|
|
(bary_2comp): Use bary_neg and bary_plus_one.
|
|
(big_extend_carry): Extracted from get2comp.
|
|
(get2comp): Use big_extend_carry.
|
|
(rb_integer_unpack): Use big_extend_carry.
|
|
(rb_big_neg): Use bary_neg.
|
|
|
|
Sat Jun 29 22:31:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_2comp): Simplified.
|
|
|
|
Sat Jun 29 09:33:53 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigor_int): Return -1 if y == -1.
|
|
|
|
Sat Jun 29 09:07:16 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigor_int): Use RB_GC_GUARD.
|
|
(bigxor_int): Take xn and hibitsx arguments. Use twocomp2abs_bang.
|
|
(rb_big_xor): Use abs2twocomp and twocomp2abs_bang.
|
|
|
|
Sat Jun 29 08:19:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigand_int): Don't apply bitwise and for BDIGIT and long.
|
|
(bigor_int): Take xn and hibitsx arguments. Use twocomp2abs_bang.
|
|
(rb_big_or): Use abs2twocomp and twocomp2abs_bang.
|
|
|
|
Fri Jun 29 01:08:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* numeric.c (fix_mul): remove FIT_SQRT_LONG test as it was causing
|
|
fix_mul to return an incorrect result for -2147483648*-2147483648
|
|
on 64 bit platforms
|
|
|
|
* test/ruby/test_integer_comb.rb (class TestIntegerComb): add test case
|
|
|
|
Fri Jun 28 12:26:53 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_and): Allocate new bignum with same size to shorter
|
|
argument if it's high bits are zero.
|
|
|
|
Fri Jun 28 12:14:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/ipsocket.c (init_inetsock_internal): Don't use local
|
|
addresses which address family is different to remote address.
|
|
|
|
Fri Jun 28 08:06:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigand_int): Add arguments, xn and hibitsx.
|
|
Use twocomp2abs_bang.
|
|
|
|
Thu Jun 27 23:58:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (abs2twocomp_bang): Removed.
|
|
(abs2twocomp): Take n_ret argument to return actual length.
|
|
(rb_big_and): Follow above change.
|
|
|
|
Thu Jun 27 22:52:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (get2comp): Use bary_2comp.
|
|
(abs2twocomp_bang): New function.
|
|
(abs2twocomp): New function.
|
|
(twocomp2abs_bang): New function.
|
|
(rb_big_and): Use abs2twocomp and twocomp2abs_bang.
|
|
|
|
Thu Jun 27 20:03:13 2013 CHIKANAGA Tomoyuki <nagachika@ruby-lang.org>
|
|
|
|
* ext/openssl/lib/openssl/ssl.rb (verify_certificate_identity): fix
|
|
hostname verification. Patched by nahi.
|
|
|
|
* test/openssl/test_ssl.rb (test_verify_certificate_identity): test for
|
|
above.
|
|
|
|
|
|
Thu Jun 27 00:23:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_pow): Retry if y is a Bignum and it is
|
|
representable as a Fixnum.
|
|
Use rb_absint_numwords.
|
|
|
|
Wed Jun 26 23:53:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* ext/bigdecimal/bigdecimal.c (BigDecimal_save_rounding_mode): fix typo.
|
|
Fixes GH-343. Patch by @jgarber.
|
|
|
|
Wed Jun 26 23:22:21 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enumerator.c (rb_enumeratorize_with_size): use strict definition
|
|
rb_enumerator_size_func.
|
|
|
|
Wed Jun 26 23:11:14 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* gc.c (is_before_sweep): Add a missing space before a parenthesis.
|
|
* gc.c (rb_gc_force_recycle): Add a missing space around a parenthesis.
|
|
|
|
Wed Jun 26 22:44:00 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/intern.h (rb_enumeratorize_with_size): cast for
|
|
backward compatibility.
|
|
|
|
* include/ruby/intern.h (rb_enumerator_size_func): define strict
|
|
function declaration for rb_enumeratorize_with_size().
|
|
|
|
Wed Jun 26 21:01:22 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* test/ruby/test_io.rb (TestIO#test_write_32bit_boundary): skip if
|
|
writing a file is slow.
|
|
[ruby-core:55541] [Bug #8519]
|
|
|
|
Wed Jun 26 16:42:11 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb: should use expanded values for header directories
|
|
unless extmk. patch by vo.x (Vit Ondruch) at [ruby-core:55653]
|
|
[Bug #8115], rhbz#921650.
|
|
|
|
Wed Jun 26 12:48:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigxor_int): Fix a buffer over read.
|
|
|
|
Wed Jun 26 12:13:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigand_int): Consider negative values.
|
|
(bigor_int): The allocated bignum should have enough size
|
|
to store long.
|
|
This fixes (bignum fits in a BDIGIT) | (fixnum bigger than BDIGIT)
|
|
on platforms which SIZEOF_BDIGITS < SIZEOF_LONG,
|
|
such as LP64 with 32bit BDIGIT (no int128).
|
|
|
|
Wed Jun 26 12:08:51 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/socket/test_udp.rb: Close sockets explicitly.
|
|
Don't use fixed port number.
|
|
|
|
Wed Jun 26 07:27:17 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigand_int): Fix a buffer over read.
|
|
|
|
Wed Jun 26 06:48:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigadd_int): Fix a buffer over read.
|
|
|
|
Wed Jun 26 01:18:13 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (is_before_sweep): Add new helper function that check the object
|
|
is before sweep?
|
|
* gc.c (rb_gc_force_recycle): Have to clear mark bit if object's slot
|
|
already ready to minor sweep.
|
|
|
|
Wed Jun 26 01:17:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigsub_int): Fix a buffer over read.
|
|
|
|
Tue Jun 25 22:45:43 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_absint_singlebit_p): Use POW2_P.
|
|
(bary_pack): Ditto.
|
|
(rb_big2str0): Ditto.
|
|
(POW2_P): Moved to top.
|
|
|
|
Tue Jun 25 22:28:07 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/rubygems/ext/builder.rb (Gem::Ext::Builder.make): Pass
|
|
DESTDIR via command line to override what's in MAKEFLAGS. This
|
|
fixes an installation problem under a package building
|
|
environment where DESTDIR is specified in the (parent) command
|
|
line. [Fixes GH-327]
|
|
|
|
Tue Jun 25 21:43:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2dbl): Use (BDIGIT)1 instead of 1UL.
|
|
(bary_mul_normal): Remove a useless cast.
|
|
|
|
Tue Jun 25 21:26:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* ext/bigdecimal/bigdecimal.c (BigMath_s_exp): Fix for the cases when
|
|
the argument x is not a BigDecimal.
|
|
This change is based on the patch made by Heesob Park and Garth Snyder.
|
|
[Bug #6862] [ruby-core:47145]
|
|
[Fixes GH-332] https://github.com/ruby/ruby/pull/332
|
|
|
|
Tue Jun 25 20:36:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2ulong): "check" argument removed.
|
|
(rb_big2ulong): Follow above change.
|
|
(rb_big2long): Ditto.
|
|
(rb_big_rshift): Ditto.
|
|
(rb_big_aref): Ditto.
|
|
|
|
Tue Jun 25 20:08:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big2ulong_pack): Use rb_integer_pack.
|
|
(rb_big_aref): Call big2ulong with TRUE for "check" argument.
|
|
It should be non-effective.
|
|
|
|
Tue Jun 25 19:07:33 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (LSHIFTX): Revert r41611.
|
|
The redundant expression suppresses a warning, C4293, by Visual
|
|
Studio.
|
|
http://ruby-mswin.cloudapp.net/vc10-x64/ruby-trunk/log/20130625T072854Z.log.html.gz#miniruby
|
|
|
|
Tue Jun 25 19:03:00 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2ulong): Add a cast.
|
|
(big2ull): Add a specialized code for SIZEOF_LONG_LONG <=
|
|
SIZEOF_BDIGITS.
|
|
|
|
Tue Jun 25 12:42:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (integer_unpack_single_bdigit): Use "1 + ~u" instead of
|
|
"-u" to suppress warning (C4146) by Visual Studio.
|
|
Reported by ko1 via IRC.
|
|
|
|
Tue Jun 25 12:28:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2ulong): Add code specialized for SIZEOF_LONG <=
|
|
SIZEOF_BDIGITS.
|
|
This prevents shift width warning from "num <<= BITSPERDIG".
|
|
|
|
Tue Jun 25 12:23:30 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: fix oldgen/remembered_shady counting algorithm.
|
|
|
|
* gc.c (rgengc_check_shady): increment
|
|
`objspace->rgengc.remembered_shady_object_count' here.
|
|
|
|
* gc.c (rgengc_remember): return FALSE if obj is already remembered.
|
|
|
|
* gc.c (rgengc_rememberset_mark): make it void.
|
|
|
|
* gc.c (gc_mark_children): fix to double counting oldgen_object_count
|
|
at minor GC.
|
|
|
|
Tue Jun 25 12:07:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (MSB): Removed.
|
|
(BDIGIT_MSB): Defined using BIGRAD_HALF.
|
|
(bary_2comp): Apply BIGLO after possible over flow of BDIGIT.
|
|
(get2comp): Ditto.
|
|
(bary_unpack_internal): Use BDIGIT_MSB.
|
|
Apply BIGLO after possible over flow of BDIGIT.
|
|
(rb_integer_unpack): Use BDIGIT_MSB.
|
|
(calc_hbase): Use BDIGMAX.
|
|
(big2dbl): Use BDIGMAX.
|
|
Apply BIGLO after possible over flow of BDIGIT.
|
|
(rb_big_neg): Apply BIGLO after possible over flow of BDIGIT.
|
|
(biglsh_bang): Ditto.
|
|
(bigrsh_bang): Ditto.
|
|
(bary_divmod): Use BDIGIT_MSB.
|
|
(bigdivrem): Ditto.
|
|
(bigxor_int): Apply BIGLO after possible over flow of BDIGIT.
|
|
|
|
* marshal.c (shortlen): Use SIZEOF_BDIGITS instead of sizeof(BDIGIT).
|
|
|
|
* ext/openssl/ossl_bn.c (ossl_bn_initialize): Use SIZEOF_BDIGITS
|
|
instead of sizeof(BDIGIT).
|
|
|
|
Tue Jun 25 11:40:08 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* bignum.c (big2ulong): suppress shorten-64-to-32 warning. BDIGIT can
|
|
be bigger than long now.
|
|
|
|
* bignum.c (LSHIFTX): remove redundant never-true expression.
|
|
|
|
Tue Jun 25 00:55:54 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (typedef struct rb_objspace): Change members for monitor objects.
|
|
* gc.c (gc_marks_test): Check all WriteBarrier Errors and track them in obj-tree.
|
|
* gc.c (rgengc_check_shady): Ditto.
|
|
* gc.c (gc_marks): Move 2 function calls to gc_marks_test for test initialize.
|
|
|
|
Mon Jun 24 23:30:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (integer_unpack_single_bdigit): Refine code to filling
|
|
higher bits and use BIGLO.
|
|
|
|
Mon Jun 24 22:26:31 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* test/rinda/test_rinda.rb (RingIPv6#prepare_ipv6):
|
|
ifindex() function may not be implemented on Windows. We use another
|
|
check for the case.
|
|
|
|
Mon Jun 24 22:11:37 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* test/gdbm/test_gdbm.rb (TestGDBM#test_s_open_nolock):
|
|
skip a failing test on Windows because flock() implementation is
|
|
different from Unix.
|
|
|
|
Mon Jun 24 22:06:14 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* test/rubygems/test_gem_installer.rb (test_install_extension_flat):
|
|
use ruby in build directory in case ruby is not installed.
|
|
[ruby-core:53265] [Bug #8058]
|
|
|
|
Mon Jun 24 22:04:02 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* ext/dl/cfunc.c (rb_dlcfunc_call): fix conversion from Bignum to
|
|
pointer. sizeof(DLSTACK_TYPE) is larger than sizeof(long) on
|
|
Windows x64 and higher bits over sizeof(long) of DLSTACK_TYPE was
|
|
zero even if a pointer value was over 32 bits which causes SEGV on
|
|
DL::TestCPtr#test_to_ptr_io. Adding a cast solves the bug.
|
|
|
|
Mon Jun 24 22:04:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* eval_error.c (warn_printf): use rb_vsprintf instead so ruby specific
|
|
extensions like PRIsVALUE can be used in format strings
|
|
* eval_error.c (error_print): use warn_print_str (alias for
|
|
rb_write_error_str) to print a string value instead of using
|
|
RSTRING_PTR and RSTRING_LEN manually
|
|
* eval.c (setup_exception): use PRIsVALUE instead of %s and RSTRING_PTR
|
|
|
|
Mon Jun 24 20:31:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* compile.c (make_name_for_block): use PRIsVALUE in format string
|
|
instead of %s and RSTRING_PTR to protect objects from being garbage
|
|
collected too soon
|
|
* encoding.c (str_to_encindex): ditto
|
|
* hash.c (rb_hash_fetch_m): ditto
|
|
* io.c (rb_io_reopen): ditto
|
|
* parse.y (reg_fragment_check_gen): ditto
|
|
* parse.y (reg_compile_gen): ditto
|
|
* parse.y (ripper_assert_Qundef): ditto
|
|
* re.c (rb_reg_raise): ditto
|
|
* ruby.c (set_option_encoding_once): ditto
|
|
* vm_eval.c (rb_throw_obj): ditto
|
|
|
|
Mon Jun 24 07:57:18 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (after_gc_sweep): Have to record malloc info before reset.
|
|
* gc.c (gc_prof_timer_start): Pick out part of new record creation as gc_prof_setup_new_record.
|
|
* gc.c (gc_prof_set_malloc_info): Move point of recording allocation size to front of mark.
|
|
|
|
Mon Jun 24 02:53:09 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* array.c: Return value in Array overview example found by @PragTob
|
|
[Fixes GH-336] https://github.com/ruby/ruby/pull/336
|
|
|
|
Mon Jun 24 02:45:51 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* array.c (rb_ary_zip): typo by @PragTob [Fixes GH-337]
|
|
https://github.com/ruby/ruby/pull/337
|
|
|
|
Mon Jun 24 02:42:01 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* win32/README.win32: grammar typo by @blankenshipz [Fixes GH-334]
|
|
https://github.com/ruby/ruby/pull/334
|
|
|
|
Mon Jun 24 00:59:35 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (BIGUP): Use LSHIFTX and avoid cast to consider the type
|
|
of x is bigger than BDIGIT_DBL.
|
|
(big2ulong): Use unsigned long to store the result.
|
|
(big2ull): Use unsigned LONG_LONG to store the result.
|
|
(bigand_int): Use long for num to avoid data loss.
|
|
(bigor_int): Ditto.
|
|
(bigxor_int): Ditto.
|
|
|
|
Sun Jun 23 23:05:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/defines.h (BDIGIT): Define it only if it is not defined
|
|
yet. This eases tests and debug.
|
|
(SIZEOF_BDIGITS): Ditto.
|
|
(BDIGIT_DBL): Ditto.
|
|
(BDIGIT_DBL_SIGNED): Ditto.
|
|
(PRI_BDIGIT_PREFIX): Ditto.
|
|
(PRI_BDIGIT_DBL_PREFIX): Ditto.
|
|
(PRIdBDIGIT): Define it only if PRI_BDIGIT_PREFIX is defined.
|
|
(PRIiBDIGIT): Ditto.
|
|
(PRIoBDIGIT): Ditto.
|
|
(PRIuBDIGIT): Ditto.
|
|
(PRIxBDIGIT): Ditto.
|
|
(PRIXBDIGIT): Ditto.
|
|
(PRIdBDIGIT_DBL): Ditto.
|
|
(PRIiBDIGIT_DBL): Ditto.
|
|
(PRIoBDIGIT_DBL): Ditto.
|
|
(PRIuBDIGIT_DBL): Ditto.
|
|
(PRIxBDIGIT_DBL): Ditto.
|
|
(PRIXBDIGIT_DBL): Ditto.
|
|
|
|
* include/ruby/ruby.h (RBIGNUM_EMBED_LEN_MAX): Define it only if it is
|
|
not defined yet.
|
|
|
|
Sun Jun 23 17:29:51 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (integer_unpack_single_bdigit): Use a cast.
|
|
|
|
Sun Jun 23 15:38:07 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* bootstraptest/test_thread.rb: rescue resource limitation errors.
|
|
|
|
Sun Jun 23 08:19:27 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (integer_unpack_single_bdigit): Extracted from
|
|
bary_unpack_internal.
|
|
|
|
Sun Jun 23 07:41:52 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_unpack_internal): Suppress warnings (C4146) on Visual Studio.
|
|
Reported by ko1 via IRC.
|
|
|
|
Sun Jun 23 06:49:28 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h, gc.c: rename macros and functions:
|
|
OBJ_WB_GIVEUP() -> OBJ_WB_UNPROTECT(),
|
|
rb_obj_wb_giveup() -> rb_obj_wb_unprotect(),
|
|
rb_gc_giveup_promoted_writebarrier() ->
|
|
rb_gc_writebarrier_unprotect_promoted(),
|
|
|
|
* class.c, eval.c, hash.c: use OBJ_WB_UNPROTECT().
|
|
|
|
Sun Jun 23 05:41:32 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* class.c (rb_include_class_new), eval.c (rb_using_refinement):
|
|
make classes/modules (who share method table) shady.
|
|
If module `a' and `b' shares method table m_tbl and new method
|
|
with iseq is added, then write barrier is applied only `a' or `b'.
|
|
To avoid this issue, shade such classes/modules.
|
|
|
|
* vm_method.c (rb_method_entry_make): add write barriers.
|
|
|
|
Sun Jun 23 01:27:54 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bytes_zero_p): Removed.
|
|
(bary_pack): Don't call bytes_zero_p.
|
|
|
|
Sun Jun 23 00:51:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bytes_zero_p): Extracted from bary_pack.
|
|
(bary_pack): Use bytes_zero_p.
|
|
|
|
Sun Jun 23 00:16:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (MSB): New macro.
|
|
(bary_unpack_internal): Use MSB.
|
|
(bary_divmod): Ditto.
|
|
(bigdivrem): Ditto.
|
|
|
|
Sat Jun 22 23:45:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_swap): New function.
|
|
(bary_pack): Use bary_swap.
|
|
(bary_unpack_internal): Ditto.
|
|
|
|
Sat Jun 22 23:18:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bytes_2comp): Renamed from quad_buf_complement.
|
|
(bary_pack): Use bytes_2comp.
|
|
(rb_quad_pack): Use rb_integer_pack.
|
|
(rb_quad_unpack): Use rb_integer_unpack.
|
|
|
|
Sat Jun 22 21:46:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_integer_unpack): Don't allocate a Bignum if possible.
|
|
|
|
Sat Jun 22 21:03:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* pack.c (pack_unpack): Remove specialized unpackers for integers.
|
|
|
|
Sat Jun 22 20:36:50 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_unpack_internal): Specialized unpacker implemented.
|
|
(bary_unpack): Support INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION.
|
|
(rb_integer_unpack): Support INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION.
|
|
|
|
Sat Jun 22 18:53:10 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_pack): Support
|
|
INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION flag.
|
|
Fix byte order and word order handling in code specialized for
|
|
wordsize % SIZEOF_BDIGITS == 0.
|
|
|
|
* internal.h (INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION): Defined.
|
|
|
|
Sat Jun 22 15:41:25 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rgengc_check_shady): add new WB miss checking
|
|
on RGENGC_CHECK_MODE >= 2.
|
|
|
|
(1) Save bitmaps before marking
|
|
(2) Run full marking
|
|
(3) On each traceable object,
|
|
(a) object was not oldgen (== newly or shady object) &&
|
|
(b) parent object was oldgen &&
|
|
(c) parent object was not remembered &&
|
|
(d) object was not remembered
|
|
then, it should be WB miss.
|
|
|
|
This idea of this checker is by Masaya Tarui <tarui@ruby-lang.org>.
|
|
|
|
Sat Jun 22 15:25:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* ext/etc/etc.c (setup_passwd): revert r41560, unnecessary
|
|
|
|
Sat Jun 22 14:39:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* ext/etc/etc.c (Init_etc): omit 'passwd' from definition of Etc::Passwd
|
|
if HAVE_STRUCT_PASSWD_PW_PASSWD is not defined to prevent mismatch of
|
|
fields and values in setup_passwd
|
|
|
|
Sat Jun 22 14:35:40 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/dl/cfunc.c (rb_dlcfunc_call): Use rb_big_pack instead of
|
|
rb_big2ulong_pack and rb_big2ull.
|
|
|
|
* include/ruby/intern.h (rb_big2ulong_pack): Deprecated.
|
|
|
|
Sat Jun 22 14:31:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* ext/etc/etc.c (setup_passwd): pass 0 as VALUE to rb_struct_new to
|
|
prevent segfault if the compiler passes it as a 32 bit integer on
|
|
a 64 bit ruby
|
|
|
|
Sat Jun 22 13:47:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_pack): MEMZERO can be used even if nails is not zero.
|
|
|
|
Sat Jun 22 13:43:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* ext/etc/etc.c (etc_getpwnam): use PRIsVALUE in format string instead
|
|
of %s and RSTRING_PTR
|
|
|
|
* ext/etc/etc.c (etc_getgrnam): ditto
|
|
|
|
Sat Jun 22 13:07:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (CLEAR_LOWBITS): Rewritten without RSHIFTX.
|
|
(RSHIFTX): Removed.
|
|
|
|
Sat Jun 22 10:38:03 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* pack.c (num2i32): Removed.
|
|
(pack_pack): Don't use num2i32.
|
|
|
|
Sat Jun 22 09:55:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (LSHIFTX): Defined to suppress a warning.
|
|
(RSHIFTX): Ditto.
|
|
(CLEAR_LOWBITS): Use LSHIFTX and RSHIFTX.
|
|
(FILL_LOWBITS): Use LSHIFTX.
|
|
Reported by ko1 via IRC.
|
|
|
|
Sat Jun 22 09:11:33 2013 Ryan Davis <ryand-ruby@zenspider.com>
|
|
|
|
* lib/minitest/*: Imported minitest 4.7.5 (r8724)
|
|
* test/minitest/*: ditto
|
|
|
|
Sat Jun 22 07:20:30 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_prof_set_heap_info, after_gc_sweep): call
|
|
gc_prof_set_heap_info() just after sweeping to calculate
|
|
live object number correctly.
|
|
(live object number = total generated number (before marking) -
|
|
total freed number (after sweeping))
|
|
|
|
* gc.c (gc_marks): record `oldgen_object_count' into current profile`
|
|
record directly.
|
|
|
|
* gc.c (rgengc_rememberset_mark): same for remembered_normal_objects
|
|
and remembered_shady_objects.
|
|
|
|
Sat Jun 22 06:46:04 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_objspace::profile): rename rb_objspace::profile::record to
|
|
records (because it points a set of records) and add a field
|
|
rb_objspace::profile::current_record to point a current profiling
|
|
record.
|
|
|
|
* gc.c: use above fields.
|
|
|
|
Sat Jun 22 06:05:36 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_gc_giveup_promoted_writebarrier): remove `rest_sweep()'
|
|
because all of remembered objects are called for gc_mark_children().
|
|
|
|
Sat Jun 22 05:08:03 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rgengc_rememberset_mark): call gc_mark_children() for
|
|
remembered objects directly instead of pushing on the mark stack.
|
|
|
|
Sat Jun 22 04:48:53 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h (OBJ_WRITE): cast to (VALUE *) for second
|
|
parameter `slot'. You don't need to write a cast (VALUE *) any more.
|
|
|
|
* class.c, compile.c, hash.c, iseq.c, proc.c, re.c, variable.c,
|
|
vm.c, vm_method.c: remove cast expressions for OBJ_WRITE().
|
|
|
|
Sat Jun 22 04:37:08 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (slot_sweep_body): rename to slot_sweep().
|
|
No need to separate major/minor GC.
|
|
|
|
* gc.c (gc_setup_mark_bits): remove gc_clear_mark_bits() and unify to
|
|
this function.
|
|
|
|
Sat Jun 22 04:20:21 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (check_bitmap_consistency): add to check flag and bitmap consistency.
|
|
Use this function in several places.
|
|
|
|
Sat Jun 22 02:18:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_pack): Specialized packers implemented.
|
|
(HOST_BIGENDIAN_P): New macro.
|
|
(ALIGNOF): New macro.
|
|
(CLEAR_LOWBITS): New macro.
|
|
(FILL_LOWBITS): New macro.
|
|
(swap_bdigit): New macro.
|
|
(bary_2comp): Returns an int.
|
|
|
|
* internal.h (swap16): Moved from pack.c
|
|
(swap32): Ditto.
|
|
(swap64): Ditto.
|
|
|
|
Fri Jun 21 21:29:49 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (typedef enum): Introduce flags of major gc reason.
|
|
* gc.c (garbage_collect_body): Ditto.
|
|
* gc.c (gc_profile_flags): Ditto.
|
|
* gc.c (gc_profile_dump_on): Ditto.
|
|
|
|
Fri Jun 21 21:11:53 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (allocate_sorted_heaps): remove unused variable `add'.
|
|
|
|
Fri Jun 21 20:50:32 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: constify RArray::as::ary and RArray::heap::ptr.
|
|
Use RARRAY_ASET() or RARRAY_PTR_USE() to modify Array objects.
|
|
|
|
* array.c, gc.c: catch up above changes.
|
|
|
|
Fri Jun 21 20:32:13 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_eval.c (eval_string_with_cref): fix WB miss.
|
|
|
|
Fri Jun 21 20:15:49 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: support write barrier protection for T_STRUCT.
|
|
Introduce the following C APIs:
|
|
* RSTRUCT_RAWPTR(st) returns pointer (do WB on your risk).
|
|
The type of returned pointer is (const VALUE *).
|
|
* RSTRUCT_GET(st, idx) returns idx-th value of struct.
|
|
* RSTRUCT_SET(st, idx, v) set idx-th value by v with WB.
|
|
And
|
|
* RSTRUCT_PTR(st) returns pointer with shady operation.
|
|
The type of returned pointer is (VALUE *).
|
|
|
|
* struct.c, re.c, gc.c, marshal.c: rewrite with above APIs.
|
|
|
|
Fri Jun 21 19:38:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (BDIGMAX): Use BIGRAD.
|
|
(BIGLO): Use BDIGMAX.
|
|
(bigdivrem1): Ditto.
|
|
(bigor_int): Ditto.
|
|
(rb_big_or): Ditto.
|
|
|
|
Fri Jun 21 19:18:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* pack.c (pack_pack): Move the implementation for 'c' directive after
|
|
pack_integer label.
|
|
|
|
Fri Jun 21 19:11:56 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h, re.c: support write barrier for T_REGEXP.
|
|
|
|
Note: T_MATCH object is also easy to support write barriers.
|
|
However, most of T_MATCH objects are short-lived objects.
|
|
So I skipped to support non-shady T_MATCH.
|
|
|
|
Fri Jun 21 18:56:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigsub_int): Use bdigit_roomof.
|
|
(bigadd_int): Ditto.
|
|
(bigand_int): Ditto.
|
|
(bigor_int): Ditto.
|
|
(bigxor_int): Ditto.
|
|
|
|
Fri Jun 21 17:56:25 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/gc/gcbench.rb: fix summary of benchmark result notation.
|
|
|
|
Fri Jun 21 16:38:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* ext/openssl/ossl_x509attr.c: change OSSL_X509ATTR_IS_SINGLE and
|
|
OSSL_X509ATTR_SET_SINGLE macros to use ->value.set rather than
|
|
->set to fix compile failure
|
|
|
|
Fri Jun 21 15:26:45 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_sweep): profile sweep time correctly when LAZY_SWEEP is
|
|
disabled.
|
|
|
|
* gc.c (gc_marks_test): store oldgen count and shady count
|
|
before test marking and restore them after marking.
|
|
|
|
Fri Jun 21 15:07:42 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: enable lazy sweep (commit miss).
|
|
|
|
Fri Jun 21 14:31:29 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* hash.c (ruby_setenv): refine error message so include the variable
|
|
name.
|
|
|
|
Fri Jun 21 14:15:08 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: fix to use total_allocated_object_num and heaps_used
|
|
at the GC time for profiler.
|
|
|
|
Fri Jun 21 12:35:35 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: RGENGC_CHECK_MODE should be 0.
|
|
|
|
Fri Jun 21 11:18:25 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_marks_body): fix to get `th' in this function.
|
|
|
|
Fri Jun 21 10:21:44 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (heaps_header/heaps_slot): embed bitmaps into heaps_slot.
|
|
no need to maintain allocation/free bitmaps.
|
|
|
|
Fri Jun 21 09:22:16 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (slot_sweep_body): add counters at a time.
|
|
|
|
* gc.c (gc_profile_dump_on): fix line break position.
|
|
|
|
Fri Jun 21 08:14:00 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c: refactoring bitmaps. introduce bits_t type and some Consts.
|
|
|
|
Fri Jun 21 08:04:32 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: fix to support USE_RGENGC == 0 (disable RGenGC).
|
|
If USE_RGENGC==0, it caused compilation error.
|
|
|
|
Fri Jun 21 08:08:11 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (lazy_sweep): Use is_lazy_sweeping()
|
|
* gc.c (rest_sweep): Ditto.
|
|
* gc.c (gc_prepare_free_objects): Ditto.
|
|
|
|
Fri Jun 21 07:34:47 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_profile_record::oldgen_objects): added.
|
|
|
|
* gc.c (gc_profile_dump_on): print the following information:
|
|
* Living object counts
|
|
* Free object counts
|
|
If RGENGC_PROFILE > 0 then
|
|
* Oldgen object counts
|
|
* Remembered normal object counts
|
|
* Remembered shady object counts
|
|
|
|
Fri Jun 21 06:43:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_ull2big): Refactored.
|
|
(rb_uint2big): Useless code removed.
|
|
|
|
Fri Jun 21 05:37:39 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_prof_sweep_timer_stop): accumulate sweep time only when
|
|
record->gc_time > 0.
|
|
|
|
Fri Jun 21 00:37:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/bigdecimal: Workaround fix for bigdecimal test failures caused
|
|
by [ruby-dev:47413] [Feature #8509]
|
|
|
|
* ext/bigdecimal/bigdecimal.h (BDIGIT): Make it independent from the
|
|
definition for bignum.c.
|
|
(SIZEOF_BDIGITS): Ditto.
|
|
(BDIGIT_DBL): Ditto.
|
|
(BDIGIT_DBL_SIGNED): Ditto.
|
|
(PRI_BDIGIT_PREFIX): Undefine the definition.
|
|
(PRI_BDIGIT_DBL_PREFIX): Ditto.
|
|
|
|
* ext/bigdecimal/bigdecimal.c (RBIGNUM_ZERO_P): Use rb_bigzero_p.
|
|
(bigzero_p): Removed.
|
|
(is_even): Use rb_big_pack.
|
|
|
|
Thu Jun 20 22:52:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigmul1_toom3): Don't call bignorm twice.
|
|
|
|
Thu Jun 20 22:49:27 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bignorm): Don't call bigtrunc if the result is a fixnum.
|
|
|
|
Thu Jun 20 22:29:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_uint2big): Refactored.
|
|
|
|
Thu Jun 20 22:24:41 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (dump_bignum): Use SIZEOF_BDIGITS.
|
|
|
|
Thu Jun 20 22:22:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2ulong): Change the return type to unsigned long.
|
|
(rb_big2ulong_pack): Follow the above change.
|
|
(rb_big2long): Ditto.
|
|
(rb_big_lshift): Ditto.
|
|
(rb_big_rshift): Ditto.
|
|
(rb_big_aref): Ditto.
|
|
|
|
Thu Jun 20 22:02:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_unpack_internal): Return -2 when negative overflow.
|
|
(bary_unpack): Set the overflowed bit if an extra BDIGIT exists.
|
|
(rb_integer_unpack): Set the overflowed bit.
|
|
|
|
Thu Jun 20 21:17:19 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rgengc_rememberset_mark): record
|
|
(1) normal objects count in remember set
|
|
(2) shady objects count in remember set
|
|
each GC timing.
|
|
|
|
* gc.c (gc_profile_record_get): enable to access above information
|
|
and REMOVING_OBJECTS, EMPTY_OBJECTS.
|
|
|
|
Thu Jun 20 18:29:26 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/gc/gcbench.rb: Do not use GC::Profiler::disable because
|
|
GC::Profiler::disable prohibit to access profiling data. It should
|
|
be spec bug.
|
|
|
|
Skip GC::Profiler::report if RUBY_VERSION < '2.0.0'
|
|
|
|
Thu Jun 20 17:59:08 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/gc/gcbench.rb: stop GC::Profiler before output results.
|
|
Generating GC::Profiler result under profiling causes infinite loop.
|
|
|
|
Thu Jun 20 17:24:24 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/gc/gcbench.rb: don't use __dir__ to make compatible
|
|
with ruby 1.9.3.
|
|
|
|
Thu Jun 20 16:57:19 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/bm_app_aobench.rb: use attr_accessor/reader instead of
|
|
defining methods.
|
|
|
|
Thu Jun 20 16:46:46 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/bm_app_aobench.rb: added.
|
|
|
|
* benchmark/gc/aobench.rb: added.
|
|
|
|
Thu Jun 20 16:28:33 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/bm_so_binary_trees.rb: disable `puts' method
|
|
and change iteration parameter to increase execution time.
|
|
|
|
* benchmark/gc/binarytree.rb: added.
|
|
|
|
Thu Jun 20 16:06:37 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/gc/pentomino.rb: added.
|
|
Simply load pentomino puzzle in the benchmark/ directory.
|
|
|
|
Thu Jun 20 15:32:56 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/gc/redblack.rb: import red black tree benchmark from
|
|
https://github.com/jruby/rubybench/blob/master/time/bench_red_black.rb
|
|
|
|
* benchmark/gc/ring.rb: add a benchmark. This benchmark create many
|
|
old objects.
|
|
|
|
Thu Jun 20 15:14:00 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* benchmark/gc: create a directory to store GC related benchmark.
|
|
|
|
* benchmark/gc/gcbench.rb: moved from tool/gcbench.rb.
|
|
|
|
* benchmark/gc/hash(1|2).rb: ditto.
|
|
|
|
* benchmark/gc/rdoc.rb: ditto.
|
|
|
|
* benchmark/gc/null.rb: added.
|
|
|
|
* common.mk: fix rule.
|
|
|
|
Thu Jun 20 14:09:54 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* tool/hashbench1.rb: fix parameter too. Increase temporary objects.
|
|
|
|
Thu Jun 20 14:01:35 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* tool/hashbench1.rb: fix parameters.
|
|
|
|
Thu Jun 20 14:00:34 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* common.mk: remove dependency from ruby.
|
|
|
|
Thu Jun 20 13:14:06 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* error.c (rb_check_backtrace): evaluate RARRAY_AREF only once.
|
|
the first argument of RB_TYPE_P is expanded twice for non-immediate
|
|
types.
|
|
|
|
Thu Jun 20 08:09:29 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* tool/gcbench.rb: Summary in one line.
|
|
|
|
* common.mk: separate gcbench-hash to gcbench-hash1 and gcbench-hash2.
|
|
|
|
Thu Jun 20 08:07:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (BIGSIZE): New macro.
|
|
(bigfixize): Use BIGSIZE.
|
|
(big2ulong): Ditto.
|
|
(check_shiftdown): Ditto.
|
|
(rb_big_aref): Ditto.
|
|
|
|
Thu Jun 20 07:46:48 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (rb_gc_writebarrier): give up rescan A and register B directly
|
|
if A has huge number of children.
|
|
|
|
Thu Jun 20 07:30:35 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* common.mk: add new rules `gcbench-rdoc', `gcbench-hash'.
|
|
|
|
* tool/gcbench.rb: separate GC bench framework and process.
|
|
|
|
* tool/hashbench1.rb, tool/hashbench2.rb: add two types GC bench.
|
|
hashbench1: many temporal objects (GC by newobj)
|
|
hashbench2: hash size becomes bigger and bigger (GC by malloc)
|
|
Two benches are executed by `gcbench-hash' rule.
|
|
|
|
* tool/rdocbench.rb: separated.
|
|
|
|
Thu Jun 20 06:25:39 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* tool/rdocbench.rb: add summary.
|
|
|
|
Thu Jun 20 06:18:01 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_profile_total_time): check objspace->profile.next_index > 0.
|
|
|
|
Thu Jun 20 05:47:41 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_prof_sweep_timer_start): fix merge miss.
|
|
|
|
* gc.c (GC_PROFILE_MORE_DETAIL): set it 0.
|
|
|
|
Thu Jun 20 05:38:56 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: Accumulate sweep time to GC time.
|
|
Now [GC time] is [mark time] + [sweep time] + [misc].
|
|
([GC time] >= [mark time] + [sweep time])
|
|
|
|
* gc.c (gc_prof_sweep_slot_timer_start/stop): rename to
|
|
gc_prof_sweep_timer_start/stop and locate at lazy_sweep().
|
|
|
|
* gc.c (elapsed_time_from): add a utility function.
|
|
|
|
Thu Jun 20 05:08:53 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_marks): fix wrong option. FALSE means major/full GC.
|
|
It should be TRUE (minor marking).
|
|
|
|
Thu Jun 20 02:44:45 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (waitpid): should not return 0 but wait until exit
|
|
unless WNOHANG is given. waiting huge process may return while
|
|
active, for some reason.
|
|
|
|
Thu Jun 20 01:34:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bdigit_roomof): Use SIZEOF_BDIGITS.
|
|
(bigfixize): Refine an ifdef condition.
|
|
(rb_absint_size): Use bdigit_roomof.
|
|
(rb_absint_singlebit_p): Ditto.
|
|
(rb_integer_pack): Ditto.
|
|
(integer_pack_fill_dd): Use BITSPERDIG.
|
|
(integer_unpack_push_bits): Use BITSPERDIG, BIGLO and BIGDN.
|
|
|
|
Thu Jun 20 01:07:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* gc.c (MARKED_IN_BITMAP, FL_TEST2): return boolean value since always
|
|
used as boolean value.
|
|
|
|
* gc.c (MARK_IN_BITMAP, CLEAR_IN_BITMAP): evaluate bits once.
|
|
|
|
Thu Jun 20 00:05:07 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (RVALUE_PROMOTED): fix type.
|
|
|
|
Wed Jun 19 23:39:01 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_marks_test): rewrite checking code.
|
|
When RGENGC_CHECK_MODE >= 2, all minor marking, run normal minor
|
|
marking *and* major/full marking. After that, compare the results
|
|
and shows BUG if a object living with major/full marking but dead
|
|
with minor marking.
|
|
After detecting bugs, print references information.
|
|
(RGENGC_CHECK_MODE == 2, show references to dead object)
|
|
(RGENGC_CHECK_MODE == 3, show all references)
|
|
|
|
Wed Jun 19 23:51:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigfixize): Use rb_absint_size.
|
|
(check_shiftdown): Ditto.
|
|
(big2ulong): Use bdigit_roomof.
|
|
|
|
Wed Jun 19 23:32:23 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (RVALUE_PROMOTED): check consistency between oldgen flag and
|
|
oldgen bitmap if RGENGC_CHECK_MODE > 0.
|
|
|
|
Wed Jun 19 23:29:29 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_gc_force_recycle): clear oldgen bitmap, too.
|
|
|
|
Wed Jun 19 21:02:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_uint2big): Consider environments BDIGIT is bigger than
|
|
long.
|
|
(big2ulong): Ditto.
|
|
(rb_big_aref): Ditto.
|
|
(rb_big_pack): Just call rb_integer_pack.
|
|
(rb_big_unpack): Just call rb_integer_unpack.
|
|
|
|
Wed Jun 19 20:51:21 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* gc.c (gc_stress_get): GC.stress can be Fixnum.
|
|
|
|
Wed Jun 19 19:31:30 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (DIGSPERLONG): Don't define if BDIGIT is bigger than long.
|
|
(DIGSPERLL): Don't define if BDIGIT is bigger than LONG_LONG
|
|
(rb_absint_size): Consider environments BDIGIT is bigger than long.
|
|
Use BIGLO and BIGDN.
|
|
(rb_absint_singlebit_p): Ditto.
|
|
(rb_integer_pack): Ditto.
|
|
(bigsub_int): Consider environments BDIGIT is bigger than long.
|
|
Use SIZEOF_BDIGITS instead of sizeof(BDIGIT).
|
|
(bigadd_int): Ditto.
|
|
(bigand_int): Ditto.
|
|
(bigor_int): Ditto.
|
|
(bigxor_int): Ditto.
|
|
|
|
Wed Jun 19 15:14:30 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h (struct rb_data_type_struct), gc.c: add
|
|
rb_data_type_struct::flags. Now, this flags is passed
|
|
at T_DATA object creation. You can specify FL_WB_PROTECTED
|
|
on this flag.
|
|
|
|
* iseq.c: making non-shady iseq objects.
|
|
|
|
* class.c, compile.c, proc.c, vm.c: add WB for iseq objects.
|
|
|
|
* vm_core.h, iseq.h: constify fields to detect WB insertion.
|
|
|
|
Wed Jun 19 15:11:13 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* gc.c (gc_mark_children): show more info for broken object.
|
|
|
|
Wed Jun 19 14:04:41 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* test/ruby/envutil.rb (EnvUtil#rubybin): remove unnecessary
|
|
unless expression.
|
|
|
|
Wed Jun 19 07:47:48 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (garbage_collect_body): use FIX2INT for ruby_gc_stress.
|
|
|
|
Wed Jun 19 07:44:31 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_objspace::gc_stress): int -> VALUE to store Fixnum object.
|
|
|
|
Wed Jun 19 07:25:35 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (make_deferred): clear flags to T_ZOMBIE.
|
|
|
|
* gc.c (slot_sweep_body): fix indent.
|
|
|
|
Wed Jun 19 07:18:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_aref): Apply BIGLO to ~xds[i] for environment which
|
|
BDIGIT is 16bit.
|
|
|
|
Wed Jun 19 07:09:26 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rgengc_remember): fix output level.
|
|
|
|
* gc.c (rgengc_rememberset_mark): fix to output clear count.
|
|
(shady_object_count + clear_count = count of remembered objects)
|
|
|
|
Wed Jun 19 07:06:21 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rgengc_remember): check T_NONE and T_ZOMBIE
|
|
if RGENGC_CHECK_MODE > 0.
|
|
|
|
Wed Jun 19 07:02:19 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (RGENGC_CHECK_MODE): add new check mode `3'.
|
|
In this mode, show all references if there is
|
|
a miss-corrected object.
|
|
|
|
Wed Jun 19 06:31:08 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_stress_set): add special option of GC.stress.
|
|
`GC.stress=(flag)' accepts integer to control behavior of GC.
|
|
See code for details. Of course, this feature is only for MRI.
|
|
|
|
You can debug RGenGC (WB) using `GC.stress = 1'.
|
|
Using this option, do minor marking at all possible places.
|
|
|
|
GC::STRESS_MINOR_MARK = 1 and GC::STRESS_LAZY_SWEEP = 2
|
|
seem good to add.
|
|
|
|
Wed Jun 19 06:29:31 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm.c (kwmerge_i): add WB.
|
|
|
|
Wed Jun 19 06:26:49 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* hash.c: `st_update()' also has same issue of last fix.
|
|
write barriers at callback function are too early.
|
|
All write barriers are executed after `st_update()'
|
|
|
|
Wed Jun 19 04:33:22 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* variable.c (rb_const_set): fix WB miss.
|
|
|
|
WBs had located before creating reference between a klass
|
|
and constant value. It causes GC bug.
|
|
|
|
# pseudo code:
|
|
WB(klass, value); # WB and remember klass
|
|
st_insert(klass->const_table, const_id, value);
|
|
|
|
`st_insert()' can cause GC before inserting `value' and
|
|
forget `klass' from the remember set. After that, relationship
|
|
between `klass' and `value' are created with constant table.
|
|
Now, `value' can be young (shady) object and `klass' can be old
|
|
object, without remembering `klass' object.
|
|
At the next GC, old `klass' object will be skipped and
|
|
young (shady) `value' will be miss-collected. -> GC bug
|
|
|
|
Lesson: The place of a WB is important.
|
|
|
|
Tue Jun 18 22:01:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* vm_insnhelper.c (vm_call_method): ensure methods of type
|
|
VM_METHOD_TYPE_ATTR_SET are called with 1 argument
|
|
|
|
* test/ruby/test_module.rb
|
|
(TestModule#test_attr_writer_with_no_arguments): add test
|
|
[ruby-core:55543] [Bug #8540]
|
|
|
|
Tue Jun 18 22:36:23 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (gc_profile_record_flag): fix typo.
|
|
|
|
Tue Jun 18 22:08:53 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/objspace/object_tracing.c: Return for ::allocation_generation
|
|
|
|
Tue Jun 18 22:04:35 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/objspace/object_tracing.c: Document object_tracing methods.
|
|
|
|
Tue Jun 18 21:58:17 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* gc.c: Rename rb_mObSpace -> rb_mObjSpace
|
|
|
|
Tue Jun 18 20:55:05 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/objspace/objspace.c: Document ObjectSpace::InternalObjectWrapper.
|
|
|
|
Tue Jun 18 20:39:04 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/objspace/object_tracing.c: Teach rdoc object_tracing.c [Bug #8537]
|
|
|
|
Tue Jun 18 20:29:47 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/.document: add object_tracing.c to document file
|
|
|
|
Tue Jun 18 20:20:27 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/objspace/objspace.c: rdoc on require to overview from r41355
|
|
|
|
Tue Jun 18 18:39:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Check __int128.
|
|
|
|
* include/ruby/defines.h (BDIGIT_DBL): Use uint128_t if it is available.
|
|
(BDIGIT): Use uint64_t if uint128_t is available.
|
|
(SIZEOF_BDIGITS): Defined for above case.
|
|
(BDIGIT_DBL_SIGNED): Ditto.
|
|
(PRI_BDIGIT_PREFIX): Ditto.
|
|
|
|
* include/ruby/ruby.h (PRI_64_PREFIX): Defined.
|
|
|
|
* bignum.c (rb_big_pow): Don't use BITSPERDIG for the condition which
|
|
rb_big_pow returns Float or Bignum.
|
|
|
|
[ruby-dev:47413] [Feature #8509]
|
|
|
|
Tue Jun 18 16:43:44 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_heredoc_restore): clear lex_strterm always to get
|
|
rid of marking recycled node. this bug is revealed by r41372 with
|
|
GC.stress=true.
|
|
|
|
Tue Jun 18 12:53:25 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (nlz): Cast the result explicitly.
|
|
(big2dbl): Don't assign BDIGIT values to int variable.
|
|
|
|
Tue Jun 18 12:25:16 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_xor): Non-effective code removed.
|
|
|
|
Tue Jun 18 11:26:05 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_stat): add `generated_normal_object_count_types' for
|
|
RGENGC_PROFILE >= 2.
|
|
|
|
Tue Jun 18 11:02:18 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_mark_maybe): check to skip T_NONE.
|
|
|
|
* gc.c (markable_object_p): do not need to check (flags == 0) here.
|
|
|
|
Tue Jun 18 10:17:37 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* variable.c (rb_autoload): fix WB miss.
|
|
|
|
Tue Jun 18 04:20:18 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_mark_children): don't need to care about T_ZOMBIE here.
|
|
|
|
Mon Jun 17 22:16:02 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* test/ruby/test_proc.rb (TestProc#test_block_given_method_to_proc):
|
|
run test for r41359.
|
|
|
|
Mon Jun 17 21:42:18 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* include/ruby/ruby.h, vm_eval.c (rb_funcall_with_block):
|
|
new function to invoke a method with a block passed
|
|
as an argument.
|
|
|
|
* string.c (sym_call): use the above function to avoid
|
|
a block sharing. [ruby-dev:47438] [Bug #8531]
|
|
|
|
* vm_insnhelper.c (vm_yield_with_cfunc): don't set block
|
|
in the frame.
|
|
|
|
* test/ruby/test_symbol.rb (TestSymbol#test_block_given_to_proc):
|
|
run related tests.
|
|
|
|
Mon Jun 17 21:33:27 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* include/ruby/intern.h, proc.c (rb_method_call_with_block):
|
|
new function to invoke a Method object with a block passed
|
|
as an argument.
|
|
|
|
* proc.c (bmcall): use the above function to avoid a block sharing.
|
|
[ruby-core:54626] [Bug #8341]
|
|
|
|
* test/ruby/test_proc.rb (TestProc#test_block_persist_between_calls):
|
|
run related tests.
|
|
|
|
Mon Jun 17 20:53:21 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* loadpath.c (RUBY_REVISION): Defined to suppress revision.h
|
|
inclusion actually. r41352 removes the dependency.
|
|
|
|
Mon Jun 17 18:15:57 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* ext/objspace/objspace.c: let rdoc know about objspace methods.
|
|
Specify 'objspace' should be required. See #8537.
|
|
|
|
Mon Jun 17 17:44:31 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* gc.c (ObjectSpace): is a module not a class.
|
|
|
|
* ext/objspace/objspace.c: try to include overview in rdoc,
|
|
see #8537.
|
|
|
|
Mon Jun 17 17:38:24 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* gc.c: fix example of ObjectSpace.define_finalizer in overview
|
|
|
|
Mon Jun 17 16:59:53 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/tk/tkutil/tkutil.c: use rb_sprintf(), rb_id2str(), and
|
|
rb_intern_str() instead of rb_intern() and RSTRING_PTR() with
|
|
RB_GC_GUARD(), to prevent temporary objects from GC.
|
|
[ruby-core:39000] [Bug #5199]
|
|
|
|
Mon Jun 17 14:27:54 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* vm_backtrace.c: Update rdoc for Backtrace#label with @_ko1
|
|
|
|
Mon Jun 17 13:04:01 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* tool/ifchange (until): Fix the condition, although harmless in
|
|
this case.
|
|
|
|
Mon Jun 17 11:50:29 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_mark_maybe): added. check `is_pointer_to_heap()' and
|
|
type is not T_ZOMBIE.
|
|
|
|
* gc.c: use `gc_mark_maybe()'. T_ZOMBIE objects should not be pushed
|
|
to the mark stack.
|
|
|
|
Mon Jun 17 07:56:24 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_small_lshift): Renamed from bdigs_small_lshift.
|
|
(bary_small_rshift): Renamed from bdigs_small_rshift.
|
|
|
|
Mon Jun 17 07:38:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (absint_numwords_bytes): Removed.
|
|
(rb_absint_numwords): Don't call absint_numwords_bytes.
|
|
|
|
Sun Jun 16 23:14:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (BARY_ADD): New macro.
|
|
(BARY_SUB): Ditto.
|
|
(BARY_MUL): Ditto.
|
|
(BARY_DIVMOD): Ditto.
|
|
(BARY_ZERO_P): Ditto.
|
|
(absint_numwords_generic): Use these macros.
|
|
|
|
Sun Jun 16 21:41:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_2comp): Extracted from get2comp.
|
|
(integer_unpack_num_bdigits): Extracted from
|
|
rb_integer_unpack_internal.
|
|
(bary_unpack_internal): Renamed from bary_unpack and support
|
|
INTEGER_PACK_2COMP.
|
|
(bary_unpack): New function to validate arguments and invoke
|
|
bary_unpack_internal.
|
|
(rb_integer_unpack_internal): Removed.
|
|
(rb_integer_unpack): Invoke bary_unpack_internal.
|
|
(rb_integer_unpack_2comp): Removed.
|
|
|
|
* internal.h (rb_integer_unpack_2comp): Removed.
|
|
|
|
* pack.c: Follow the above change.
|
|
|
|
Sun Jun 16 18:41:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (INTEGER_PACK_2COMP): Defined.
|
|
(rb_integer_pack_2comp): Removed.
|
|
|
|
* bignum.c (bary_pack): Support INTEGER_PACK_2COMP.
|
|
(rb_integer_pack): Invoke bary_pack directly.
|
|
(rb_integer_pack_2comp): Removed.
|
|
(rb_integer_pack_internal): Ditto.
|
|
(absint_numwords_generic): Follow the above change.
|
|
|
|
* pack.c (pack_pack): Ditto.
|
|
|
|
* sprintf.c (rb_str_format): Ditto.
|
|
|
|
Sun Jun 16 17:48:14 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (absint_numwords_generic): rb_funcall invocations removed.
|
|
|
|
Sun Jun 16 16:04:38 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* tool/config_files.rb: use URI.read to allow it runs with Ruby 1.8.5.
|
|
|
|
Sun Jun 16 14:32:25 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_pack) Extracted from rb_integer_pack_internal.
|
|
(absint_numwords_generic): Use bary_pack.
|
|
|
|
Sun Jun 16 11:01:57 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* NEWS (XMLRPC::Client#http): Add.
|
|
[ruby-core:55197] [Feature #8461]
|
|
|
|
Sun Jun 16 10:38:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_add): New function.
|
|
(bary_zero_p): Extracted from bigzero_p.
|
|
(absint_numwords_generic): Use bary_zero_p and bary_add.
|
|
(bary_mul): Fix an argument for bary_mul_single.
|
|
(bary_divmod): Use size_t for arguments.
|
|
|
|
Sun Jun 16 08:55:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem): Use a BDIGIT variable to store the return
|
|
value of bigdivrem_single.
|
|
|
|
Sun Jun 16 08:43:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_divmod): New function.
|
|
(absint_numwords_generic): Use bary_divmod.
|
|
(bigdivrem_num_extra_words): Extracted from bigdivrem.
|
|
(bigdivrem_single): Ditto.
|
|
(bigdivrem_normal): Ditto.
|
|
(BIGDIVREM_EXTRA_WORDS): Defined.
|
|
|
|
Sun Jun 16 05:51:51 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c: Fixup around GC by MALLOC.
|
|
Add allocate size to malloc_increase before GC
|
|
for updating limit in after_gc_sweep.
|
|
Reset malloc_increase into garbage_collect()
|
|
for preventing GC again soon.
|
|
|
|
Sun Jun 16 05:15:36 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c: Add some columns to more detail profile.
|
|
new columns: Allocated size, Prepare Time, Removing Objects, Empty Objects
|
|
|
|
Sun Jun 16 02:04:40 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (gc_prof_timer_stop): Merge function codes of GC_PROFILE_MORE_DETAIL and !GC_PROFILE_MORE_DETAIL.
|
|
* gc.c (gc_prof_mark_timer_start): Ditto.
|
|
* gc.c (gc_prof_mark_timer_stop): Ditto.
|
|
* gc.c (gc_prof_sweep_slot_timer_start): Ditto.
|
|
* gc.c (gc_prof_sweep_slot_timer_stop): Ditto.
|
|
* gc.c (gc_prof_set_malloc_info): Ditto.
|
|
* gc.c (gc_prof_set_heap_info): Ditto.
|
|
|
|
Sat Jun 15 23:50:24 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_sub): New function.
|
|
(absint_numwords_generic): Use bary_sub.
|
|
(bigsub_core): Skip unnecessary copy.
|
|
|
|
Sat Jun 15 22:05:30 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_mul): New function.
|
|
(absint_numwords_generic): Use bary_mul.
|
|
(bary_mul_single): Extracted from bigmul1_single.
|
|
(bary_mul_normal): Extracted from bigmul1_normal.
|
|
|
|
Sat Jun 15 20:13:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bary_unpack): Extracted from rb_integer_unpack_internal.
|
|
(absint_numwords_generic): Use bary_unpack.
|
|
(roomof): Defined.
|
|
(bdigit_roomof): Defined.
|
|
(BARY_ARGS): Defined.
|
|
(bary_unpack): Declared.
|
|
|
|
Sat Jun 15 19:35:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (absint_numwords_bytes): Make it static.
|
|
(absint_numwords_small): Ditto.
|
|
(absint_numwords_generic): Ditto.
|
|
|
|
Sat Jun 15 17:14:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigmul1_normal): Shrink the result Bignum length.
|
|
|
|
Sat Jun 15 10:19:42 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/bigdecimal/bigdecimal.c: Update overview formatting of headers
|
|
|
|
Sat Jun 15 10:19:06 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/bigdecimal/bigdecimal.gemspec: Update authors
|
|
|
|
Sat Jun 15 10:02:26 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bdigs_small_rshift): Extracted from big_rshift.
|
|
(bigdivrem): Use bdigs_small_rshift.
|
|
|
|
Sat Jun 15 08:37:28 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_eval.c (eval_string_with_cref): propagate absolute path from the
|
|
binding if it is given explicitly. patch by Gat (Dawid Janczak) at
|
|
[ruby-core:55123]. [Bug #8436]
|
|
|
|
Sat Jun 15 02:40:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bdigs_small_lshift): Extracted from big_lshift.
|
|
(bigdivrem): Use bdigs_small_lshift.
|
|
|
|
Fri Jun 14 20:47:41 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem): Reduce number of digits before bignew() for div.
|
|
|
|
Fri Jun 14 20:12:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem): Use bignew when ny == 1.
|
|
|
|
Fri Jun 14 18:52:51 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* compile.c (rb_iseq_compile_node): fix location of a `trace'
|
|
instruction (b_return event).
|
|
[ruby-core:55305] [ruby-trunk - Bug #8489]
|
|
(need a backport to 2.0.0?)
|
|
|
|
* test/ruby/test_settracefunc.rb: add a test.
|
|
|
|
Fri Jun 14 18:18:07 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* class.c, include/ruby/ruby.h: add write barriers for T_CLASS,
|
|
T_MODULE, T_ICLASS.
|
|
|
|
* constant.h: constify rb_const_entry_t::value and file to detect
|
|
assignment.
|
|
|
|
* variable.c, internal.h (rb_st_insert_id_and_value, rb_st_copy):
|
|
added. update table with write barrier.
|
|
|
|
* method.h: constify some variables to detect assignment.
|
|
|
|
* object.c (init_copy): add WBs.
|
|
|
|
* variable.c: ditto.
|
|
|
|
* vm_method.c (rb_add_method): ditto.
|
|
|
|
Fri Jun 14 14:33:47 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* NEWS: add a note for Module#using.
|
|
|
|
Fri Jun 14 13:40:27 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* .travis.yml (before_script): update config files.
|
|
|
|
* common.mk ($(srcdir)/tool/config.{guess,sub}): use get-config_files.
|
|
|
|
* tool/config_files.rb: split get-config_files.
|
|
|
|
* common.mk (update-config_files): rule to download config files.
|
|
|
|
* tool/config.guess, tool/config.sub: remove and download from the
|
|
upstream.
|
|
|
|
* tool/config_files.rb: download config files from GNU.
|
|
|
|
Fri Jun 14 12:21:20 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h (RUBY_SAFE_LEVEL_CHECK): suppress warnings
|
|
"left-hand operand of comma expression has no effect", on gcc 4.4.
|
|
|
|
Fri Jun 14 09:48:48 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* NEWS: add notes for $SAFE.
|
|
|
|
* doc/security.rdoc: remove the description of $SAFE=4.
|
|
|
|
Fri Jun 14 00:14:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem): Zero test condition simplified.
|
|
|
|
Thu Jun 13 23:43:11 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/bigdecimal/*: improve documentation, nodoc samples with @mrkn
|
|
|
|
Thu Jun 13 23:02:14 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/xmlrpc/client.rb (XMLRPC::Client#http): Add reader for raw
|
|
Net::HTTP. [ruby-core:55197] [Feature #8461]
|
|
Reported by Herwin Weststrate. Thanks!!!
|
|
|
|
Thu Jun 13 22:44:52 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/xmlrpc/client.rb (XMLRPC::Client#parse_set_cookies): Support
|
|
multiple names in a response. [ruby-core:41711] [Bug #5774]
|
|
Reported by Roman Riha. Thanks!!!
|
|
* test/xmlrpc/test_client.rb (XMLRPC::ClientTest#test_cookie_override):
|
|
Add a test of the above case.
|
|
|
|
Thu Jun 13 22:35:50 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/xmlrpc/client.rb (XMLRPC::Client#parse_set_cookies): Use
|
|
guard style.
|
|
|
|
Thu Jun 13 22:12:32 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/fileutils.rb (FileUtils#rmdir): fix traversal loop, not trying
|
|
remove same directory only.
|
|
|
|
Thu Jun 13 21:30:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (opt-dir), tool/ifchange: get rid of "alternate value"
|
|
expansion for legacy sh. [ruby-dev:47420] [Bug #8524]
|
|
|
|
Thu Jun 13 21:24:09 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem): Refactored to use ALLOCV_N for temporary
|
|
buffers.
|
|
|
|
Thu Jun 13 18:54:11 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* bignum.c (integer_unpack_num_bdigits_generic): reorder terms (but not
|
|
changed the intention of the expression) because VC++ reports a
|
|
warning for it. reported by ko1 via IRC.
|
|
|
|
Thu Jun 13 18:53:14 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/ruby/test_thread.rb (test_thread_local_security): Don't create
|
|
an unused thread.
|
|
|
|
Thu Jun 13 18:34:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bigdivrem): Use nlz.
|
|
|
|
Thu Jun 13 14:51:06 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h (RUBY_SAFE_LEVEL_CHECK): check constant safe
|
|
level at compile time.
|
|
|
|
Thu Jun 13 14:39:08 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* test/-ext-/test_printf.rb, test/rss/test_parser.rb,
|
|
test/ruby/test_array.rb, test/ruby/test_hash.rb,
|
|
test/ruby/test_m17n.rb, test/ruby/test_marshal.rb,
|
|
test/ruby/test_object.rb, test/ruby/test_string.rb: don't use
|
|
untrusted?, untrust, and trust to avoid warnings in case $VERBOSE is
|
|
true.
|
|
|
|
Thu Jun 13 10:47:16 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* bootstraptest/test_autoload.rb, bootstraptest/test_method.rb:
|
|
remove tests for $SAFE=4.
|
|
|
|
* lib/pp.rb: use taint instead of untrust to avoid warnings when
|
|
$VERBOSE is set to true.
|
|
|
|
Thu Jun 13 06:12:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (integer_unpack_num_bdigits_small): Fix a compile error on
|
|
clang -Werror,-Wshorten-64-to-32
|
|
Reported by Eric Hodel. [ruby-core:55467] [Bug #8522]
|
|
|
|
Thu Jun 13 05:32:13 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* ext/socket/extconf.rb: Enable RFC 3542 IPV6 socket options for OS X
|
|
10.7+. [ruby-trunk - Bug #8517]
|
|
|
|
Thu Jun 13 00:17:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_integer_unpack_2comp): New function.
|
|
(rb_integer_unpack_internal): Extracted from rb_integer_unpack and
|
|
nlp_bits_ret argument added.
|
|
(integer_unpack_num_bdigits_small): nlp_bits_ret argument added to
|
|
return number of leading padding bits.
|
|
(integer_unpack_num_bdigits_generic): Ditto.
|
|
|
|
* internal.h (rb_integer_unpack_2comp): Declared.
|
|
|
|
* pack.c (pack_unpack): Use rb_integer_unpack_2comp and
|
|
rb_integer_unpack.
|
|
|
|
Wed Jun 12 23:27:03 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* eval.c (mod_using): new method Module#using, which activates
|
|
refinements of the specified module only in the current class or
|
|
module definition. [ruby-core:55273] [Feature #8481]
|
|
|
|
* test/ruby/test_refinement.rb: related test.
|
|
|
|
Wed Jun 12 22:58:48 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* safe.c (rb_set_safe_level, safe_setter): raise an ArgumentError
|
|
when $SAFE is set to 4. $SAFE=4 is now obsolete.
|
|
[ruby-core:55222] [Feature #8468]
|
|
|
|
* object.c (rb_obj_untrusted, rb_obj_untrust, rb_obj_trust):
|
|
Kernel#untrusted?, untrust, and trust are now deprecated.
|
|
Their behavior is same as tainted?, taint, and untaint,
|
|
respectively.
|
|
|
|
* include/ruby/ruby.h (OBJ_UNTRUSTED, OBJ_UNTRUST): OBJ_UNTRUSTED()
|
|
and OBJ_UNTRUST() are aliases of OBJ_TAINTED() and OBJ_TAINT(),
|
|
respectively.
|
|
|
|
* array.c, class.c, debug.c, dir.c, encoding.c, error.c, eval.c,
|
|
ext/curses/curses.c, ext/dbm/dbm.c, ext/dl/cfunc.c,
|
|
ext/dl/cptr.c, ext/dl/dl.c, ext/etc/etc.c, ext/fiddle/fiddle.c,
|
|
ext/fiddle/pointer.c, ext/gdbm/gdbm.c, ext/readline/readline.c,
|
|
ext/sdbm/init.c, ext/socket/ancdata.c, ext/socket/basicsocket.c,
|
|
ext/socket/socket.c, ext/socket/udpsocket.c,
|
|
ext/stringio/stringio.c, ext/syslog/syslog.c, ext/tk/tcltklib.c,
|
|
ext/win32ole/win32ole.c, file.c, gc.c, hash.c, io.c, iseq.c,
|
|
load.c, marshal.c, object.c, proc.c, process.c, random.c, re.c,
|
|
safe.c, string.c, thread.c, transcode.c, variable.c,
|
|
vm_insnhelper.c, vm_method.c, vm_trace.c: remove code for
|
|
$SAFE=4.
|
|
|
|
* test/dl/test_dl2.rb, test/erb/test_erb.rb,
|
|
test/readline/test_readline.rb,
|
|
test/readline/test_readline_history.rb, test/ruby/test_alias.rb,
|
|
test/ruby/test_array.rb, test/ruby/test_dir.rb,
|
|
test/ruby/test_encoding.rb, test/ruby/test_env.rb,
|
|
test/ruby/test_eval.rb, test/ruby/test_exception.rb,
|
|
test/ruby/test_file_exhaustive.rb, test/ruby/test_hash.rb,
|
|
test/ruby/test_io.rb, test/ruby/test_method.rb,
|
|
test/ruby/test_module.rb, test/ruby/test_object.rb,
|
|
test/ruby/test_pack.rb, test/ruby/test_rand.rb,
|
|
test/ruby/test_regexp.rb, test/ruby/test_settracefunc.rb,
|
|
test/ruby/test_struct.rb, test/ruby/test_thread.rb,
|
|
test/ruby/test_time.rb: remove tests for $SAFE=4.
|
|
|
|
Wed Jun 12 22:18:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (integer_unpack_num_bdigits_generic): Rewritten without
|
|
rb_funcall.
|
|
(integer_unpack_num_bdigits_bytes): Removed.
|
|
(rb_integer_unpack): integer_unpack_num_bdigits_bytes invocation
|
|
removed.
|
|
|
|
Wed Jun 12 20:18:03 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/xmlrpc/client.rb (XMLRPC::Client#parse_set_cookies): Extract.
|
|
|
|
Wed Jun 12 18:19:41 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (validate_integer_pack_format): supported_flags argument
|
|
added and validate given flags.
|
|
(rb_integer_pack_internal): Specify supported_flags.
|
|
(rb_integer_unpack): Ditto.
|
|
|
|
Wed Jun 12 16:41:38 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* array.c (rb_ary_sort_bang): remove duplicated assertions.
|
|
ARY_HEAP_PTR() implies ary not to be embedded. [ruby-dev:47419]
|
|
[Bug #8518]
|
|
|
|
Wed Jun 12 12:44:45 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* io.c (io_getc): fix 7bit coderange condition, check if ascii read
|
|
data instead of read length. [ruby-core:55444] [Bug #8516]
|
|
|
|
Wed Jun 12 12:35:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* pack.c (pack_pack): Use rb_integer_pack_2comp.
|
|
|
|
Wed Jun 12 12:07:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* sprintf.c (rb_str_format): Fix a dynamic format string.
|
|
|
|
Wed Jun 12 12:04:09 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* array.c (rb_ary_uniq_bang): must not be modified once frozen even in
|
|
a callback method.
|
|
|
|
Wed Jun 12 12:03:43 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* array.c (rb_ary_sort_bang): must not be modified once frozen even in
|
|
a callback method.
|
|
|
|
Wed Jun 12 12:00:15 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* array.c (FL_SET_EMBED): shared object is frozen even when get
|
|
unshared.
|
|
|
|
* array.c (rb_ary_modify): ARY_SET_CAPA needs unshared array.
|
|
|
|
Wed Jun 12 07:32:01 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (rand_int): Use rb_big_uminus.
|
|
|
|
Wed Jun 12 07:12:54 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* struct.c: Improve documentation: replace "instance variable" with
|
|
"member", recommend the use of a block to customize structs, note
|
|
that member accessors are created, general cleanup.
|
|
|
|
Wed Jun 12 06:35:01 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (INTEGER_PACK_NEGATIVE): Defined.
|
|
(rb_integer_unpack): sign argument removed.
|
|
|
|
* bignum.c (rb_integer_unpack): sign argument removed.
|
|
Non-negative integers generated by default.
|
|
INTEGER_PACK_NEGATIVE flag is used to generate non-positive integers.
|
|
|
|
* pack.c (pack_unpack): Follow the above change.
|
|
|
|
* random.c (int_pair_to_real_inclusive): Ditto.
|
|
(make_seed_value): Ditto.
|
|
(mt_state): Ditto.
|
|
(limited_big_rand): Ditto.
|
|
|
|
* marshal.c (r_object0): Ditto.
|
|
|
|
Wed Jun 12 00:07:46 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/xmlrpc/test_client.rb (XMLRPC::ClientTest#test_cookie_simple):
|
|
Add a test for the extracted method.
|
|
|
|
Tue Jun 11 23:56:24 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/xmlrpc/test_client.rb (XMLRPC::ClientTest::Fake::HTTP#started):
|
|
Add a missing empty line.
|
|
|
|
Tue Jun 11 23:37:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (validate_integer_pack_format): Don't require a word order
|
|
flag if numwords is 1 or less.
|
|
(absint_numwords_generic): Don't specify a word order for
|
|
rb_integer_pack.
|
|
|
|
* hash.c (rb_hash): Ditto.
|
|
|
|
* time.c (v2w_bignum): Ditto.
|
|
|
|
Tue Jun 11 23:01:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (validate_integer_pack_format): Refine error messages.
|
|
|
|
Tue Jun 11 22:25:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (validate_integer_pack_format): numwords argument added.
|
|
Move a varidation from rb_integer_pack_internal and rb_integer_unpack.
|
|
(rb_integer_pack_internal): Follow above change.
|
|
(rb_integer_unpack): Ditto.
|
|
|
|
Tue Jun 11 20:52:43 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_integer_pack_internal): Renamed from rb_integer_pack
|
|
and overflow_2comp argument added.
|
|
(rb_integer_pack): Just call rb_integer_pack_internal.
|
|
(rb_integer_pack_2comp): New function.
|
|
|
|
* internal.h (rb_integer_pack_2comp): Declared.
|
|
|
|
* sprintf.c (rb_str_format): Use rb_integer_pack and
|
|
rb_integer_pack_2comp to format binary/octal/hexadecimal integers.
|
|
(ruby_digitmap): Declared.
|
|
(remove_sign_bits): Removed.
|
|
(BITSPERDIG): Ditto.
|
|
(EXTENDSIGN): Ditto.
|
|
|
|
Tue Jun 11 16:15:03 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* array.c (ary_shrink_capa): shrink the capacity so it fits just with
|
|
the length.
|
|
|
|
* array.c (ary_make_shared): release never used elements from frozen
|
|
array to be shared. [ruby-dev:47416] [Bug #8510]
|
|
|
|
Tue Jun 11 12:49:01 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* doc/re.rdoc: Rename to doc/regexp.rdoc
|
|
* re.c: Update rdoc include for rename of file
|
|
|
|
Tue Jun 11 07:13:13 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* eval_error.c (error_print): keep that errat is non-shady object.
|
|
and guard errat from GC.
|
|
|
|
Tue Jun 11 05:04:25 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* ext/racc/cparse/cparse.c: use rb_ary_entry() and
|
|
rb_ary_subseq() instead of RARRAY_PTR.
|
|
Based on a patch by Dirkjan Bussink. See Bug #8399.
|
|
|
|
Mon Jun 10 23:51:51 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* array.c (rb_ary_new_from_values): fix a typo. pointed out by
|
|
nagachika.
|
|
http://d.hatena.ne.jp/nagachika/20130610/ruby_trunk_changes_41199_41220
|
|
|
|
Mon Jun 10 21:51:03 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* ext/socket/raddrinfo.c (nogvl_getaddrinfo): Fix indent.
|
|
|
|
Mon Jun 10 21:49:43 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* ext/socket/raddrinfo.c (nogvl_getaddrinfo): Add missing return
|
|
value assignment.
|
|
|
|
Mon Jun 10 20:58:11 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/socket/raddrinfo.c (nogvl_getaddrinfo): work around for Ubuntu
|
|
13.04's getaddrinfo issue with mdns4. [ruby-list:49420]
|
|
|
|
Mon Jun 10 19:34:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_integer_pack): Returns sign instead of words.
|
|
(absint_numwords_generic): Follow the above change.
|
|
(big2str_base_powerof2): Follow the above change.
|
|
|
|
* internal.h: Ditto.
|
|
|
|
* hash.c (rb_hash): Ditto.
|
|
|
|
* pack.c (pack_pack): Ditto.
|
|
|
|
* random.c (int_pair_to_real_inclusive): Ditto.
|
|
(rand_init): Ditto.
|
|
(random_load): Ditto.
|
|
(limited_big_rand): Ditto.
|
|
|
|
* time.c (v2w_bignum): Ditto.
|
|
|
|
Mon Jun 10 17:20:01 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rgengc_remember): permit promoted object.
|
|
(rb_gc_writebarrier -> remember)
|
|
|
|
Mon Jun 10 17:14:01 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (RVALUE_PROMOTE): fix parameter name (`x' to `obj')
|
|
and make it inline function (like RVALUE_PROMOTE).
|
|
|
|
Mon Jun 10 16:22:50 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* array.c (rb_ary_new_from_values): add assertion
|
|
(ary should be young object).
|
|
|
|
Mon Jun 10 16:05:59 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (wmap_mark): check allocation of `w->obj2wmap'.
|
|
(no-allocation `w->obj2wmap' will be NULL pointer reference)
|
|
|
|
Mon Jun 10 15:36:00 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* eval_error.c (error_print): use checking functions instead of
|
|
catching exceptions.
|
|
|
|
* eval_error.c (error_print): restore errinfo for the case new
|
|
exception raised while printing the message. [ruby-core:55365]
|
|
[Bug #8501]
|
|
|
|
* eval_error.c (error_print): reduce calling setjmp.
|
|
|
|
Mon Jun 10 12:10:06 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (integer_unpack_num_bdigits_small: Extracted from
|
|
rb_integer_unpack.
|
|
(integer_unpack_num_bdigits_generic): Ditto.
|
|
(integer_unpack_num_bdigits_bytes): New function.
|
|
(rb_integer_unpack): Use above functions.
|
|
Return a Bignum for INTEGER_PACK_FORCE_BIGNUM even when the result
|
|
is zero.
|
|
|
|
Mon Jun 10 05:38:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (absint_numwords_small): New function.
|
|
(absint_numwords_generic): Use absint_numwords_small if possible.
|
|
|
|
Mon Jun 10 01:07:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (absint_numwords_bytes): New function.
|
|
(absint_numwords_generic): Extracted from rb_absint_numwords.
|
|
(rb_absint_numwords): Use absint_numwords_bytes if possible.
|
|
|
|
Sun Jun 9 21:33:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_absint_numwords): Return (size_t)-1 when overflow.
|
|
Refine variable names.
|
|
(rb_absint_size): Refine variable names.
|
|
|
|
* internal.h (rb_absint_size): Refine an argument name.
|
|
(rb_absint_numwords): Ditto.
|
|
|
|
Sun Jun 9 16:51:41 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_absint_numwords): Renamed from rb_absint_size_in_word.
|
|
|
|
* internal.h (rb_absint_numwords): Follow the above change.
|
|
|
|
* pack.c (pack_pack): Ditto.
|
|
|
|
* random.c (rand_init): Ditto.
|
|
(limited_big_rand): Ditto.
|
|
|
|
Sun Jun 9 14:41:05 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_integer_pack): numwords_allocated argument removed.
|
|
|
|
* internal.h (rb_integer_pack): Follow the above change.
|
|
|
|
* hash.c (rb_hash): Ditto.
|
|
|
|
* time.c (v2w_bignum): Ditto.
|
|
|
|
* pack.c (pack_pack): Ditto.
|
|
|
|
* random.c (int_pair_to_real_inclusive): Ditto.
|
|
(rand_init): Ditto.
|
|
(random_load): Ditto.
|
|
(limited_big_rand): Ditto.
|
|
|
|
Sun Jun 9 09:34:44 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big2str_base_powerof2): New function.
|
|
(rb_big2str0): Use big2str_base_powerof2 if base is 2, 4, 8, 16 or 32.
|
|
|
|
Sun Jun 9 00:59:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* hash.c (rb_hash): Use rb_integer_pack to obtain least significant
|
|
long integer.
|
|
|
|
Sat Jun 8 23:56:00 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* numeric.c (rb_num_to_uint): Use rb_absint_size instead of
|
|
RBIGNUM_LEN.
|
|
|
|
Sat Jun 8 22:53:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* marshal.c (r_object0): Use rb_integer_unpack.
|
|
|
|
Sat Jun 8 22:18:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* time.c (v2w): Use rb_absint_size instead of RBIGNUM_LEN.
|
|
|
|
Sat Jun 8 21:47:33 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* time.c (v2w_bignum): Simplified using rb_integer_pack.
|
|
(rb_big_abs_find_maxbit): Removed.
|
|
|
|
Sat Jun 8 21:03:40 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_absint_singlebit_p): New function.
|
|
|
|
* internal.h (rb_absint_singlebit_p): Declared.
|
|
|
|
* time.c (v2w_bignum): Use rb_absint_singlebit_p instead of
|
|
rb_big_abs_find_minbit.
|
|
(rb_big_abs_find_minbit): Removed.
|
|
|
|
Sat Jun 8 20:24:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* time.c (rb_big_abs_find_maxbit): Use rb_absint_size.
|
|
(bdigit_find_maxbit): Removed.
|
|
|
|
Sat Jun 8 19:47:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* class.c (include_modules_at): invalidate method cache if included
|
|
module contains constants
|
|
|
|
* test/ruby/test_module.rb: add test
|
|
|
|
Sat Jun 8 19:31:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* random.c (limited_big_rand): declare rnd, lim and mask as uint32_t
|
|
to avoid 64 bit to 32 bit shorten warnings.
|
|
|
|
Sat Jun 8 19:23:53 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* win32/Makefile.sub: r41163 changed win32/win32.c and configure.in
|
|
but it didn't treat about mswin32/mswin64, so fix it.
|
|
NOTE: this needs a review by usa whether additional condition is
|
|
required or not.
|
|
|
|
Sat Jun 8 19:06:26 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c: Unused RBignum internal accessing macros removed.
|
|
|
|
Sat Jun 8 19:04:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (limited_big_rand): The argument, limit, is changed to
|
|
VALUE. Use rb_integer_pack and rb_integer_unpack.
|
|
|
|
Sat Jun 8 17:15:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (make_seed_value): Fix the length given for
|
|
rb_integer_unpack.
|
|
|
|
Sat Jun 8 16:38:02 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_integer_unpack): Don't use rb_funcall if possible.
|
|
|
|
* random.c: Use uint32_t for elements of seed.
|
|
(make_seed_value): Use rb_integer_unpack.
|
|
|
|
Sat Jun 8 15:58:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (rand_init): Add a cast to fix clang compile error:
|
|
random.c:410:32: error: implicit conversion loses integer precision:
|
|
'size_t' (aka 'unsigned long') to 'int' [-Werror,-Wshorten-64-to-32]
|
|
This cast doesn't cause a problem because len is not bigger than
|
|
MT_MAX_STATE.
|
|
|
|
Sat Jun 8 15:30:03 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (rand_init): Use rb_integer_pack.
|
|
(roomof): Removed.
|
|
|
|
Sat Jun 8 14:58:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (INTEGER_PACK_FORCE_BIGNUM): New flag constant.
|
|
|
|
* bignum.c (rb_integer_unpack): Support INTEGER_PACK_FORCE_BIGNUM.
|
|
|
|
* random.c (int_pair_to_real_inclusive): Use
|
|
INTEGER_PACK_FORCE_BIGNUM to use rb_big_mul instead of rb_funcall.
|
|
|
|
Sat Jun 8 14:17:01 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in: check for NET_LUID. header macro varies across
|
|
compiler versions.
|
|
|
|
* win32/win32.c: use configured macro.
|
|
|
|
Sat Jun 8 11:59:55 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (int_pair_to_real_inclusive): Use rb_funcall instead of
|
|
rb_big_mul because rb_integer_unpack can return a Fixnum.
|
|
|
|
Sat Jun 8 11:17:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (int_pair_to_real_inclusive): Use rb_integer_pack.
|
|
|
|
Sat Jun 8 09:49:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (int_pair_to_real_inclusive): Use rb_integer_unpack.
|
|
|
|
Sat Jun 8 08:12:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (random_load): Use rb_integer_pack.
|
|
|
|
Sat Jun 8 06:15:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (numberof): Removed.
|
|
|
|
Sat Jun 8 06:00:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c: include internal.h.
|
|
(mt_state): Use rb_integer_unpack.
|
|
|
|
Sat Jun 8 00:55:51 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (integer_pack_loop_setup): word_num_nailbytes_ret argument
|
|
removed.
|
|
(rb_integer_pack): Follow the above change.
|
|
(rb_integer_unpack): Follow the above change.
|
|
|
|
Sat Jun 8 00:37:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (validate_integer_pack_format): Renamed from
|
|
validate_integer_format.
|
|
(integer_pack_loop_setup): Renamed from integer_format_loop_setup.
|
|
(integer_pack_fill_dd): Renamed from int_export_fill_dd.
|
|
(integer_pack_take_lowbits): Renamed from int_export_take_lowbits.
|
|
(integer_unpack_push_bits): Renamed from int_import_push_bits.
|
|
|
|
Fri Jun 7 23:58:06 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_integer_pack): Arguments changed. Use flags to
|
|
specify word order and byte order.
|
|
(rb_integer_unpack): Ditto.
|
|
(validate_integer_format): Follow the above change.
|
|
(integer_format_loop_setup): Ditto.
|
|
|
|
* pack.c: Ditto.
|
|
|
|
* internal.h: Ditto.
|
|
(INTEGER_PACK_MSWORD_FIRST): Defined.
|
|
(INTEGER_PACK_LSWORD_FIRST): Ditto.
|
|
(INTEGER_PACK_MSBYTE_FIRST): Ditto.
|
|
(INTEGER_PACK_LSBYTE_FIRST): Ditto.
|
|
(INTEGER_PACK_NATIVE_BYTE_ORDER): Ditto.
|
|
(INTEGER_PACK_LITTLE_ENDIAN): Ditto.
|
|
(INTEGER_PACK_BIG_ENDIAN): Ditto.
|
|
|
|
Fri Jun 7 22:10:50 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* lib/rubygems/specification.rb (Gem::Specification#to_yaml):
|
|
use Gem::NoAliasYAMLTree.create instead of Gem::NoAliasYAMLTree.new
|
|
to suppress deprecated warnings.
|
|
|
|
Fri Jun 7 21:39:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_integer_pack): Renamed from rb_int_export.
|
|
(rb_integer_unpack): Renamed from rb_int_import.
|
|
|
|
* internal.h, pack.c: Follow the above change.
|
|
|
|
Fri Jun 7 21:05:26 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (integer_format_loop_setup): Extracted from rb_int_export
|
|
and rb_int_import.
|
|
|
|
Fri Jun 7 19:48:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (validate_integer_format): Extracted from rb_int_export and
|
|
rb_int_import.
|
|
|
|
Fri Jun 7 19:23:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_absint_size): Use numberof.
|
|
(rb_int_export): Ditto.
|
|
|
|
Fri Jun 7 18:58:56 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (numberof): Gathered from various files.
|
|
|
|
* array.c, math.c, thread_pthread.c, iseq.c, enum.c, string.c, io.c,
|
|
load.c, compile.c, struct.c, eval.c, gc.c, parse.y, process.c,
|
|
error.c, ruby.c: Remove the definitions of numberof.
|
|
|
|
Fri Jun 7 18:24:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_absint_size): Declare a variable, i, just before used
|
|
to suppress a warning.
|
|
(rb_int_export): Ditto.
|
|
|
|
Fri Jun 7 17:41:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* bignum.c (rb_absint_size): explicit cast to BDIGIT to avoid implicit
|
|
64 bit to 32 bit shortening warning
|
|
* bignum.c (rb_int_export): ditto
|
|
* bignum.c (int_import_push_bits): ditto
|
|
|
|
Fri Jun 7 17:31:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* internal.h (RCLASS_SUPER): use descriptive variable name
|
|
* internal.h (RCLASS_SET_SUPER): ditto
|
|
|
|
Fri Jun 7 13:25:27 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/json/fbuffer/fbuffer.h (fbuffer_append_str): change the place of
|
|
RB_GC_GUARD. it should be after the object is used.
|
|
|
|
Fri Jun 7 13:22:43 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* gc.c (before_gc_sweep): noinline can also avoid the segv instead of
|
|
-O0 of r41084. this way is expected less slow.
|
|
|
|
Fri Jun 7 11:45:42 2013 Kenta Murata <mrkn@cookpad.com>
|
|
|
|
* rational.c (numeric_quo): move num_quo in numeric.c to numeric_quo
|
|
in rational.c to refer canonicalization state for mathn support.
|
|
[ruby-core:41575] [Bug #5736]
|
|
|
|
* numeric.c (num_quo): ditto.
|
|
|
|
* test/test_mathn.rb: add a test for the change at r41109.
|
|
|
|
Fri Jun 7 11:41:42 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in: revert r41106. size_t may not be unsigned
|
|
|
|
* bignum.c (rb_absint_size_in_word, rb_int_export, rb_int_import): use
|
|
NUM2SIZET() and SIZET2NUM() already defined in ruby/ruby.h.
|
|
|
|
Fri Jun 7 11:28:37 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c: use oldgen bitmap as initial mark bitmap when major gc.
|
|
so can skip oldgen bitmap check around mark & sweep.
|
|
* gc.c (slot_sweep_body): change scan algorithm for performance:
|
|
from object's pointer base to bitmap one.
|
|
|
|
Fri Jun 7 11:25:56 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c: introduce oldgen bitmap for preparing performance tuning.
|
|
|
|
Fri Jun 7 11:20:57 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (MARKED_IN_BITMAP, MARK_IN_BITMAP, CLEAR_IN_BITMAP): bring
|
|
bitmap macros in one place, and introduce BITMAP_BIT.
|
|
|
|
Fri Jun 7 11:18:35 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* array.c (ary_new): change order of allocation in order
|
|
to remove FL_OLDGEN operation.
|
|
|
|
Fri Jun 7 11:16:28 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* tool/rdocbench.rb: add gc total time information.
|
|
|
|
Fri Jun 7 10:12:01 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: remove "Sunny" terminology.
|
|
"Sunny" doesn't mean antonym of "Shady" (questionable, doubtful, etc).
|
|
Instead of "Sunny", use "non-shady" or "normal".
|
|
|
|
Fri Jun 7 09:29:33 2013 Kenta Murata <mrkn@cookpad.com>
|
|
|
|
* bignum.c (rb_int_import): explicitly casting BDIGIT_DBL to BDIGIT
|
|
to prevent warning.
|
|
|
|
Fri Jun 7 07:29:33 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (rb_int_export): countp argument is split into
|
|
wordcount_allocated and wordcount.
|
|
|
|
* bignum.c (rb_int_export): Follow the above change.
|
|
|
|
* pack.c (pack_pack): Ditto.
|
|
|
|
Fri Jun 7 07:17:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* NEWS: describe a compatibility issue of Numeric#quo
|
|
introduced at r41109.
|
|
|
|
Fri Jun 7 07:15:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* NEWS: fix style.
|
|
|
|
Fri Jun 7 06:48:17 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* numeric.c: remove unused ID id_to_r introduced in r41109.
|
|
|
|
Fri Jun 7 06:15:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_int_import): New function.
|
|
(int_import_push_bits): Ditto.
|
|
|
|
* internal.h (rb_int_import): Declared.
|
|
|
|
* pack.c (pack_unpack): Use rb_int_import for BER compressed integer.
|
|
|
|
Thu Jun 6 22:24:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* numeric.c (num_quo): Use to_r method to convert the receiver to
|
|
rational. [ruby-core:41575] [Bug #5736]
|
|
|
|
* test/ruby/test_numeric.rb: add a test for the above change.
|
|
|
|
Thu Jun 6 20:40:17 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Invoke RUBY_REPLACE_TYPE for size_t.
|
|
Don't invoke RUBY_CHECK_PRINTF_PREFIX for size_t to avoid conflict
|
|
with RUBY_REPLACE_TYPE.
|
|
|
|
* internal.h (rb_absint_size): Declared.
|
|
(rb_absint_size_in_word): Ditto.
|
|
(rb_int_export): Ditto.
|
|
|
|
* bignum.c (rb_absint_size): New function.
|
|
(rb_absint_size_in_word): Ditto.
|
|
(int_export_fill_dd): Ditto.
|
|
(int_export_take_lowbits): Ditto.
|
|
(rb_int_export): Ditto.
|
|
|
|
* pack.c (pack_pack): Use rb_int_export for BER compressed integer.
|
|
|
|
Thu Jun 6 19:31:33 2013 Tadayoshi Funaba <tadf@dotrb.org>
|
|
|
|
* ext/date/date_core.c: fixed coding error [ruby-core:55337].
|
|
reported by Riley Lynch.
|
|
|
|
Thu Jun 6 14:16:37 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* ext/objspace/object_tracing.c: rename allocation_info to
|
|
lookup_allocation_info. At times I confused "struct
|
|
allocation_info" with "function allocation_info".
|
|
|
|
Thu Jun 6 13:57:06 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* ext/objspace/object_tracing.c: allocation_info function isn't
|
|
called by any other file.
|
|
|
|
Thu Jun 6 09:41:00 2013 Kenta Murata <mrkn@cookpad.com>
|
|
|
|
* numeric.c (num_quo): should return a Float for a Float argument.
|
|
[ruby-dev:44710] [Bug #5515]
|
|
|
|
* test/ruby/test_fixnum.rb: Add an assertion for the above change.
|
|
|
|
* test/ruby/test_bignum.rb: ditto.
|
|
|
|
Thu Jun 6 00:59:44 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (gc_mark): get rid of pushing useless objects.
|
|
* gc.c (rgengc_rememberset_mark): bypass gc_mark() in order to push
|
|
sunny old object at minor gc.
|
|
* gc.c (gc_mark_children): move sunny old check to gc_mark().
|
|
* gc.c (rgengc_check_shady): remove DEMOTE that already unnecessary.
|
|
* gc.c (rb_gc_writebarrier): ditto.
|
|
|
|
change sunny old check point in order to save mark stack and
|
|
remove unnatural rest_sweep & demote.
|
|
|
|
Thu Jun 6 00:52:42 2013 Masaya Tarui <tarui@ruby-lang.org>
|
|
|
|
* gc.c (rgengc_rememberset_mark): change scan algorithm for performance:
|
|
from object's pointer base to bitmap one.
|
|
|
|
Thu Jun 6 00:30:04 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* win32/win32.c (NET_LUID): define it on MinGW32.
|
|
mingw-w64 has NET_LUID but mingw32 (mingw.org) still doesn't have
|
|
NET_LUID. reported by taco on IRC
|
|
|
|
Thu Jun 6 00:05:08 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* string.c (String#b): Allow code range scan to happen later so
|
|
ascii_only? on a result string returns the correct value.
|
|
[ruby-core:55315] [Bug #8496]
|
|
|
|
Wed Jun 5 22:40:42 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* lib/net/imap.rb (capability_response): should ignore trailing
|
|
spaces. Thanks, Peter Kovacs. [ruby-core:55024] [Bug #8415]
|
|
|
|
* test/net/imap/test_imap_response_parser.rb: related test.
|
|
|
|
Wed Jun 5 21:17:08 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (big_fdiv): Use nlz() instead of bdigbitsize().
|
|
(bdigbitsize): Removed.
|
|
|
|
Wed Jun 5 20:32:00 2013 Kenta Murata <mrkn@cookpad.com>
|
|
|
|
* include/ruby/ruby.h: fix alignment in comment.
|
|
|
|
Wed Jun 5 20:05:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* random.c (int_pair_to_real_inclusive): Add a cast to BDIGIT.
|
|
(random_load): Fix shift width for fixnums.
|
|
Re-implement bignum extraction without ifdefs.
|
|
|
|
Wed Jun 5 15:26:10 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* gc.c (before_gc_sweep): don't optimize it to avoid segv on Ubuntu
|
|
10.04 gcc 4.4.
|
|
http://u32.rubyci.org/~chkbuild/ruby-trunk/log/20130527T190301Z.diff.html.gz
|
|
|
|
Wed Jun 5 09:46:46 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* test/fileutils/test_fileutils.rb (TestFileUtils#test_mkdir): add
|
|
EACCES for Windows.
|
|
|
|
Wed Jun 5 08:13:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_big_pow): Don't need to multiply SIZEOF_BDIGITS.
|
|
Use nlz instead of bitlength_bdigit.
|
|
(bitlength_bdigit): Removed.
|
|
|
|
Wed Jun 5 07:14:18 2013 Tadayoshi Funaba <tadf@dotrb.org>
|
|
|
|
* ext/date/date_core.c (d_lite_cmp, d_lite_equal): simplified.
|
|
|
|
Wed Jun 5 07:07:01 2013 Tadayoshi Funaba <tadf@dotrb.org>
|
|
|
|
* ext/date/date_core.c: fixed a bug [ruby-core:55295]. reported
|
|
by Riley Lynch.
|
|
|
|
Wed Jun 5 06:44:08 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems: Update to RubyGems 2.0.3
|
|
|
|
* test/rubygems: Tests for the above.
|
|
|
|
* NEWS: Added RubyGems 2.0.3 note.
|
|
|
|
Wed Jun 5 06:35:15 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* doc/marshal.rdoc: Add description of Marshal format.
|
|
|
|
Wed Jun 5 01:16:09 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* array.c (Array#+): fix documentation example.
|
|
Patch by Logan Serman. [Fixes GH-324]
|
|
|
|
Wed Jun 5 00:21:54 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* lib/irb/lc/ja/help-message: update help messages.
|
|
following r41028. [ruby-dev:46707] [Feature #7510]
|
|
|
|
Wed Jun 5 00:09:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* marshal.c (r_object0): Generalize a round up expression.
|
|
Use BDIGIT instead of int.
|
|
|
|
Tue Jun 4 23:44:02 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* object.c (rb_Hash): fix docs. patched by Stefan Sch"ussler.
|
|
[ruby-core:55299] [Bug #8487]
|
|
|
|
Tue Jun 4 23:16:49 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* lib/irb/completion.rb: Use %w literal construction for long lists.
|
|
Patch by Dave Goodchild. [Fixes GH-299]
|
|
|
|
Tue Jun 4 23:08:42 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* ext/objspace/objspace.c: improve wording and remove duplicated comment.
|
|
Based on a patch by Dave Goodchild. [Fixes GH-299]
|
|
|
|
Tue Jun 4 18:41:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bitlength_bdigit): Fix an off-by-one error.
|
|
|
|
Tue Jun 4 15:30:00 2013 Kenta Murata <mrkn@cookpad.com>
|
|
|
|
* ext/bigdecimal/lib/bigdecimal/util.rb (Float#to_d): fix the number
|
|
of figures. Patch by Vipul A M <vipulnsward@gmail.com>.
|
|
https://github.com/ruby/ruby/pull/323 fix GH-323
|
|
|
|
* test/bigdecimal/test_bigdecimal_util.rb: fix for the above change.
|
|
|
|
Tue Jun 4 00:44:27 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* test/fileutils/test_fileutils.rb (TestFileUtils#test_mkdir): add
|
|
EEXIST for Linux. (suggested by nurse)
|
|
|
|
Mon Jun 3 23:58:19 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* lib/fileutils.rb (FileUtils.rmdir): use remove_tailing_slash.
|
|
* test/fileutils/test_fileutils.rb: test for above.
|
|
|
|
Mon Jun 3 23:47:55 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (bitlength_bdigit): New function.
|
|
(rb_big_pow): Use bitlength_bdigit instead of ffs.
|
|
|
|
Mon Jun 3 23:11:19 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* lib/fileutils.rb: fix behavior when mkdir/mkdir_p accepted "/".
|
|
* test/fileutils/test_fileutils.rb: add test for above change.
|
|
Patched by Mitsunori Komatsu. [GH-319]
|
|
|
|
Mon Jun 3 19:02:20 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* dir.c (is_hfs): use the file descriptor instead of a path.
|
|
|
|
Mon Jun 3 07:15:17 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: removes AC_CHECK_FUNCS(readdir_r). readdir_r()
|
|
is only used from dir.c and it doesn't need readdir_r().
|
|
* configure.in (SIZEOF_STRUCT_DIRENT_TOO_SMALL): removed. It is
|
|
only used for readdir_r.
|
|
* dir.c: removes NAME_MAX_FOR_STRUCT_DIRENT. It is not right way
|
|
to detect maximum length of path len. POSIX require to use
|
|
fpathconf(). IOW, it might have lead to make a vulnerability
|
|
using stack smashing. Moreover, readdir() works enough for our
|
|
usage.
|
|
* dir.c (READDIR): removes an implementation which uses
|
|
readdir_r() and parenthesize in a macro body correctly.
|
|
* dir.c (dir_read): removes IF_HAVE_READDIR_R(DEFINE_STRUCT_DIRENT
|
|
entry), it is used only for readdir_r().
|
|
* dir.c (dir_each): ditto.
|
|
* dir.c (glob_helper): ditto.
|
|
|
|
* dir.c (READDIR): removes entry and dp argument.
|
|
* dir.c (dir_read): adjust for the above change.
|
|
* dir.c (dir_each): ditto.
|
|
* dir.c (glob_helper): ditto.
|
|
|
|
Mon Jun 3 03:40:29 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* vm_insnhelper.c (vm_yield_setup_block_args): partially revert r41019.
|
|
The code is not useless.
|
|
|
|
Mon Jun 3 01:25:25 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* test/socket/test_sockopt.rb: change test name. follow r41037.
|
|
|
|
Mon Jun 3 01:08:43 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* test/rinda/test_rinda.rb: rename functions introduced in r41009.
|
|
|
|
Sun Jun 2 23:33:42 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* enc/trans/japanese_euc.trans, test/ruby/test_transcode.rb,
|
|
tool/transcode-tblgen.rb: change EUC-JP-2004 to EUC-JIS-2004.
|
|
This is follow up to changes in r41024.
|
|
|
|
Sun Jun 2 22:44:42 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/socket/option.c: rename functions introduced in r41009
|
|
s/ip/ipv4/g because they are ipv4 functions.
|
|
(there's a policy that the name "ip" is for methods which supports
|
|
both ipv4 and ipv6)
|
|
|
|
Sun Jun 2 16:15:29 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* dln_find.c (dln_find_exe, dln_find_file): remove deprecated
|
|
non-reentrant functions.
|
|
|
|
Sun Jun 2 15:04:35 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/cgi/util.rb, lib/erb.rb: Use String#b [Feature #8394] by znz
|
|
|
|
Sun Jun 2 14:10:21 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/irb/lc/help-message: Apply english updates for irb --help #7510
|
|
|
|
Sun Jun 2 12:03:58 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* range.c: Fix rdoc on Range#bsearch [Bug #8242] [ruby-core:54143]
|
|
|
|
Sun Jun 2 02:08:37 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* enc/euc_jp.c: fix typo: the name of EUC-JIS-2004.
|
|
|
|
Sat Jun 1 23:17:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* vm_eval.c (rb_mod_module_eval): mention in docs that arguments passed
|
|
to the method are passed to the block
|
|
|
|
Sat Jun 1 17:58:13 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/set.rb (Set#freeze, taint, untaint): Save a "self" by
|
|
utilizing super returning self, and add tests while at it.
|
|
|
|
Sat Jun 1 17:24:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* compile.c (iseq_set_arguments): not a simple single argument if any
|
|
keyword arguments exist. [ruby-core:55203] [Bug #8463]
|
|
|
|
* vm_insnhelper.c (vm_yield_setup_block_args): split single parameter
|
|
if any keyword arguments exist, and then extract keyword arguments.
|
|
[ruby-core:55203] [Bug #8463]
|
|
|
|
Sat Jun 1 11:16:22 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* error.c (rb_exc_new_cstr): rename from rb_exc_new2.
|
|
|
|
* error.c (rb_exc_new_str): rename from rb_exc_new3.
|
|
|
|
Sat Jun 1 10:13:17 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_str_new[2-5], rb_{tainted,usascii}_str_new2),
|
|
(rb_str_buf_new2): remove old interfaces.
|
|
|
|
Sat Jun 1 08:00:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/zlib/zlib.c (gzfile_read, gzfile_read_all, gzfile_getc),
|
|
(gzreader_gets): check EOF. [ruby-core:55220] [Bug #8467]
|
|
|
|
Sat Jun 1 07:32:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Use BDIGIT type for hbase.
|
|
|
|
Sat Jun 1 02:37:35 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/socket/option.c (sockopt_s_byte): constructor of the sockopt
|
|
whose value's is byte.
|
|
|
|
* ext/socket/option.c (sockopt_byte): getter for above.
|
|
|
|
* ext/socket/option.c (inspect_byte): inspect for above.
|
|
|
|
* ext/socket/option.c (sockopt_s_ip_multicast_loop): constructor of
|
|
the sockopt whose optname is IP_MULTICAST_LOOP.
|
|
|
|
* ext/socket/option.c (sockopt_ip_multicast_loop): getter for above.
|
|
|
|
* ext/socket/option.c (sockopt_s_ip_multicast_ttl): constructor of
|
|
the sockopt whose optname is IP_MULTICAST_TTL.
|
|
|
|
* ext/socket/option.c (sockopt_ip_multicast_ttl): getter for above.
|
|
|
|
* ext/socket/option.c (sockopt_inspect): use above.
|
|
|
|
Sat Jun 01 01:50:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* ext/bigdecimal/bigdecimal.c (BigDecimal_power): use rb_dbl2big
|
|
to convert a double value to a Bignum.
|
|
|
|
Sat Jun 1 00:19:50 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (calc_hbase): Make hbase the maximum power of base
|
|
representable in BDIGIT.
|
|
|
|
Fri May 31 23:56:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (calc_hbase): Extracted from rb_big2str0.
|
|
|
|
Fri May 31 23:22:24 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c: Don't hard code SIZEOF_BDIGITS for log_base(hbase).
|
|
(big2str_orig): hbase_numdigits argument added.
|
|
(big2str_karatsuba): Ditto.
|
|
(rb_big2str0): Calculate hbase_numdigits.
|
|
|
|
Fri May 31 17:57:21 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* process.c: Improve Process::exec documentation
|
|
|
|
Fri May 31 17:26:42 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_eval.c (rb_funcallv): add better names of rb_funcall2.
|
|
|
|
* vm_eval.c (rb_funcallv_public): ditto for rb_funcall3.
|
|
|
|
Fri May 31 17:04:45 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* array.c (rb_ary_new_capa): add better names of rb_ary_new2.
|
|
|
|
* array.c (rb_ary_new_from_args): ditto for rb_ary_new3.
|
|
|
|
* array.c (rb_ary_new_from_values): ditto for rb_ary_new4.
|
|
|
|
Fri May 31 16:35:44 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (HAVE_ATTRIBUTE_FUNCTION_ALIAS): define to tell if
|
|
alias attribute is available.
|
|
|
|
Fri May 31 16:03:23 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* object.c, proc.c: s/call_seq/call-seq in rdoc. [Fixes GH-322]
|
|
|
|
Fri May 31 15:56:36 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/openssl/ossl_ssl.c: Add missing paren in rdoc [Fixes GH-321]
|
|
|
|
Fri May 31 11:58:24 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_method.c (set_visibility): extract from rb_mod_public(),
|
|
rb_mod_protected() and rb_mod_private().
|
|
|
|
Thu May 30 19:47:42 2013 Yusuke Endoh <mame@tsg.ne.jp>
|
|
|
|
* vm_insnhelper.c (vm_callee_setup_keyword_arg,
|
|
vm_callee_setup_arg_complex): consider a hash argument for keyword
|
|
only when the number of arguments is more than the expected
|
|
mandatory parameters. [ruby-core:53199] [ruby-trunk - Bug #8040]
|
|
|
|
* test/ruby/test_keyword.rb: update a test for above.
|
|
|
|
Thu May 30 17:55:04 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* process.c: RDoc on Process.spawn
|
|
|
|
Thu May 30 00:08:14 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_profile_enable): rest_sweep() to finish last GC.
|
|
Profiling record is allocated at first of marking phase.
|
|
Enable at lazy sweeping may cause an error (SEGV).
|
|
|
|
Wed May 29 10:33:27 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* hash.c: fix WB bug.
|
|
(1) Hash's key also needs WB.
|
|
(2) callback parameter *key and *value of st_update() is not a
|
|
storage of st_table itself (only local variable). So that
|
|
OBJ_WRITE() is not suitable, especially for `!existing'.
|
|
OBJ_WRITTEN() is used instead of OBJ_WRITE().
|
|
|
|
Tue May 28 12:31:21 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/object_tracing.c: fix a bug reported at
|
|
"[ruby-core:55182] [ruby-trunk - Bug #8456][Open] Sugfault in Ruby Head"
|
|
Care about the case TracePoint#path #=> `nil'.
|
|
|
|
* ext/objspace/object_tracing.c: add two new methods:
|
|
* ObjectSpace.allocation_class_path(o)
|
|
* ObjectSpace.allocation_method_id(o)
|
|
They are not useful for Object.new because they are always
|
|
"Class" and :new.
|
|
To trace more useful information, we need to maintain call-tree
|
|
using call/return hooks, which is implemented by
|
|
ll-prof <http://sunagae.net/wiki/doku.php?id=software:llprof>
|
|
|
|
* test/objspace/test_objspace.rb: add a test.
|
|
|
|
Tue May 28 11:30:02 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/extmk.rb (extmake): leave makefiles untouched if the content is
|
|
not changed, to get rid of unnecessary re-linking.
|
|
|
|
Tue May 28 03:11:02 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/gc_hook.c, ext/objspace/objspace.c: add new methods to
|
|
hook GC invocation.
|
|
* ObjectSpace.after_gc_start_hook=(proc)
|
|
* ObjectSpace.after_gc_end_hook=(proc)
|
|
|
|
Note that hooks are not kicked immediately. Procs are kicked
|
|
at postponed_job.
|
|
|
|
This feature is a sample of new internal event and
|
|
rb_postponed_job API.
|
|
|
|
Tue May 28 02:56:15 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_stat): remove wrong rest_sweep().
|
|
|
|
Tue May 28 02:44:23 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (garbage_collect_body): fix GC_ENABLE_LAZY_SWEEP condition.
|
|
|
|
* gc.c (GC_NOTIFY): move debug print location and use stderr instead
|
|
of stdout.
|
|
|
|
Tue May 28 02:07:21 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_trace.c (rb_postponed_job_register_one): fix iteration bug.
|
|
|
|
* ext/-test-/postponed_job/postponed_job.c,
|
|
test/-ext-/postponed_job/test_postponed_job.rb: add a test.
|
|
|
|
Tue May 28 00:34:23 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h, gc.c: add new internal event
|
|
RUBY_INTERNAL_EVENT_GC_END. This event invokes at the end of
|
|
after_sweep().
|
|
Time chart with lazy sweep is:
|
|
(1) Kick RUBY_INTERNAL_EVENT_GC_START
|
|
(2) [gc_marks()]
|
|
(3) [lazy_sweep()]
|
|
(4) [... run Ruby program (mutator) with lazy_sweep() ...]
|
|
(5) [after_sweep()]
|
|
(6) Kick RUBY_INTERNAL_EVENT_GC_END
|
|
(7) [... run Ruby program (mutator), and go to (1) ...]
|
|
Time chart without lazy sweep (GC.start, etc) is:
|
|
(1) Kick RUBY_INTERNAL_EVENT_GC_START
|
|
(2) [gc_marks()]
|
|
(3) [gc_sweep()]
|
|
(4) [after_sweep()]
|
|
(5) Kick RUBY_INTERNAL_EVENT_GC_END
|
|
(6) [... run Ruby program (mutator), and go to (1) ...]
|
|
|
|
* ext/-test-/tracepoint/tracepoint.c,
|
|
test/-ext-/tracepoint/test_tracepoint.rb: modify a test.
|
|
|
|
Tue May 28 00:18:57 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_trace.c (rb_postponed_job_flush): remove a wrong comment.
|
|
|
|
Mon May 27 22:09:33 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/ruby.h (RHASH_SIZE): Add a cast to suppress a
|
|
warning, comparison between signed and unsigned integer
|
|
expressions [-Wsign-compare], on ILP32.
|
|
|
|
Mon May 27 19:25:47 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: rename RUBY_INTERNAL_EVENT_FREE to
|
|
RUBY_INTERNAL_EVENT_FREEOBJ.
|
|
|
|
* ext/-test-/tracepoint/tracepoint.c,
|
|
ext/objspace/object_tracing.c,
|
|
gc.c, vm_trace.c: catch up this change.
|
|
|
|
Mon May 27 18:57:28 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/objspace/objspace.c: support ObjectSpace.trace_object_allocations.
|
|
Read the following test to know HOWTO.
|
|
This feature is a sample of RUBY_INTERNAL_EVENT.
|
|
|
|
* test/objspace/test_objspace.rb: add a test.
|
|
|
|
* ext/objspace/object_tracing.c: ditto.
|
|
|
|
* gc.c (rb_gc_count): add. This function returns GC count.
|
|
|
|
* internal.h: add decl. of rb_gc_count(). Same as `GC.count'.
|
|
|
|
Mon May 27 17:33:28 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* tool/rbinstall.rb (install_recursive): add maxdepth option.
|
|
|
|
* tool/rbinstall.rb (bin-comm): limit depth of bindir and reject empty
|
|
files. [ruby-core:55101] [Bug #8432]
|
|
|
|
Mon May 27 16:16:18 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_trace.c (rb_postponed_job_flush, rb_postponed_job_register): use
|
|
ruby_xmalloc/xfree. It is safe during GC.
|
|
|
|
Mon May 27 09:24:03 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* test/-ext-/postponed_job/test_postponed_job.rb: fix typo and class name.
|
|
|
|
Mon May 27 09:05:17 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h, gc.c, vm_trace.c: add internal events.
|
|
* RUBY_INTERNAL_EVENT_NEWOBJ: object created.
|
|
* RUBY_INTERNAL_EVENT_FREE: object freed.
|
|
* RUBY_INTERNAL_EVENT_GC_START: GC started.
|
|
And rename `RUBY_EVENT_SWITCH' to `RUBY_INTERNAL_EVENT_SWITCH'.
|
|
|
|
Internal events can not invoke any Ruby program because the tracing
|
|
timing may be critical (under huge restriction).
|
|
These events can be hooked only by C-extensions.
|
|
We recommend to use rb_postponed_job_register() API to call Ruby
|
|
program safely.
|
|
|
|
This change is mostly written by Aman Gupta (tmm1).
|
|
https://bugs.ruby-lang.org/issues/8107#note-12
|
|
[Feature #8107]
|
|
|
|
* include/ruby/debug.h, vm_trace.c: added two new APIs.
|
|
* rb_tracearg_event_flag() returns rb_event_flag_t of this event.
|
|
* rb_tracearg_object() returns created/freed object.
|
|
|
|
* ext/-test-/tracepoint/extconf.rb,
|
|
ext/-test-/tracepoint/tracepoint.c,
|
|
test/-ext-/tracepoint/test_tracepoint.rb: add a test.
|
|
|
|
Mon May 27 08:38:21 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* ext/-test-/postponed_job/postponed_job.c: fix `init' function name.
|
|
|
|
Mon May 27 06:22:41 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/debug.h, vm_trace.c: add rb_postponed_job API.
|
|
Postponed jobs are registered with this API. Registered jobs
|
|
are invoked at `ruby-running-safe-point' as soon as possible.
|
|
This timing is completely same as finalizer timing.
|
|
|
|
There are two APIs:
|
|
* rb_postponed_job_register(flags, func, data): register a
|
|
postponed job with data. flags are reserved.
|
|
* rb_postponed_job_register_one(flags, func, data): same as
|
|
`rb_postponed_job_register', but only one `func' job is
|
|
registered (skip if `func' is already registered).
|
|
|
|
This change is mostly written by Aman Gupta (tmm1).
|
|
https://bugs.ruby-lang.org/issues/8107#note-15
|
|
[Feature #8107]
|
|
|
|
* gc.c: use postponed job API for finalizer.
|
|
|
|
* common.mk: add dependency from vm_trace.c to debug.h.
|
|
|
|
* ext/-test-/postponed_job/extconf.rb, postponed_job.c,
|
|
test/-ext-/postponed_job/test_postponed_job.rb: add a test.
|
|
|
|
* thread.c: implement postponed API.
|
|
|
|
* vm_core.h: ditto.
|
|
|
|
Mon May 27 02:26:02 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_stat): collect promote_operation_count and
|
|
types (RGENGC_PROFILE >= 2).
|
|
|
|
Mon May 27 01:40:58 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_stat): collect shade_operation_count,
|
|
remembered_sunny_object_count and remembered_shady_object_count
|
|
for each types when RGENGC_PROFILE >= 2.
|
|
They are informative for optimization.
|
|
|
|
Mon May 27 01:15:22 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* hash.c (rb_hash_tbl_raw), internal.h: added.
|
|
Returns st_table without shading hash.
|
|
|
|
* array.c: use rb_hash_tbl_raw() for read-only purpose.
|
|
|
|
* compile.c (iseq_compile_each): ditto.
|
|
|
|
* gc.c (count_objects): ditto.
|
|
|
|
* insns.def: ditto.
|
|
|
|
* process.c: ditto.
|
|
|
|
* thread.c (clear_coverage): ditto.
|
|
|
|
* vm_insnhelper.c: ditto.
|
|
|
|
Mon May 27 00:31:09 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* tool/make-snapshot: use ENV["AUTOCONF"] instead of directly using
|
|
literal "autoconf".
|
|
|
|
Sun May 26 21:31:46 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* hash.c, include/ruby/ruby.h: support WB protected hash.
|
|
* constify RHash::ifnone and make new macro RHASH_SET_IFNONE().
|
|
* insert write barrier for st_update().
|
|
|
|
* include/ruby/intern.h: declare rb_hash_set_ifnone(hash, ifnone).
|
|
|
|
* marshal.c (r_object0): use RHASH_SET_IFNONE().
|
|
|
|
* ext/openssl/ossl_x509name.c (Init_ossl_x509name): ditto.
|
|
|
|
Sat May 25 23:22:38 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* test/fiddle/test_c_struct_entry.rb,
|
|
test/fiddle/test_c_union_entity.rb,
|
|
test/fiddle/test_cparser.rb, test/fiddle/test_func.rb,
|
|
test/fiddle/test_handle.rb, test/fiddle/test_import.rb,
|
|
test/fiddle/test_pointer.rb: don't run test if the system
|
|
don't support fiddle.
|
|
|
|
Sat May 25 21:29:34 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/pty/pty.c (get_device_once): FreeBSD 10-current and 9-stable
|
|
added O_CLOEXEC support to posix_openpt, so assume FreeBSD 9.2 or
|
|
later supports it.
|
|
http://www.freebsd.org/cgi/query-pr.cgi?pr=162374
|
|
|
|
Sat May 25 18:46:23 2013 Yusuke Endoh <mame@tsg.ne.jp>
|
|
|
|
* proc.c (rb_method_entry_min_max_arity): fix missing break in switch.
|
|
This was introduced in r38236, which is not intentional apparently.
|
|
This has caused no actual harm because VM_METHOD_TYPE_OPTIMIZED is
|
|
not used except for OPTIMIZED_METHOD_TYPE_SEND, but may do in
|
|
future. Coverity Scan found this inadequacy.
|
|
|
|
Sat May 25 18:08:06 2013 Yusuke Endoh <mame@tsg.ne.jp>
|
|
|
|
* dir.c (bracket): fix copy-paste error. When the first and last
|
|
characters of fnmatch range have different length, fnmatch may
|
|
have wrongly matched a path that does not really match.
|
|
Coverity Scan found this bug.
|
|
|
|
Sat May 25 17:06:25 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (after_gc_sweep): reduce full GC timing.
|
|
|
|
Sat May 25 11:28:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* variable.c (set_const_visibility): return without clearing method
|
|
cache if no arguments.
|
|
|
|
* vm_method.c (set_method_visibility): ditto.
|
|
|
|
Sat May 25 11:27:32 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_method.c (set_method_visibility): quote unprintable method name.
|
|
|
|
Sat May 25 11:24:24 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* eval.c (rb_frame_callee): returns the called name of the current
|
|
frame, not the previous frame.
|
|
|
|
* eval.c (prev_frame_callee, prev_frame_func): rename and make static,
|
|
as these are used by rb_f_method_name() and rb_f_callee_name() only.
|
|
|
|
* variable.c (set_const_visibility): use the called name.
|
|
|
|
Sat May 25 08:58:23 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_str_quote_unprintable): check if argument is a string.
|
|
|
|
Fri May 24 19:32:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* variable.c (set_const_visibility): use rb_frame_this_func() instead
|
|
of rb_frame_callee() for getting the name of the called method
|
|
|
|
* test/ruby/test_module.rb: add test for private_constant with no args
|
|
|
|
Fri May 24 18:53:10 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: do major/full GC when:
|
|
* number of oldgen object is bigger than twice of
|
|
number of oldgen object at last full GC.
|
|
* number of remembered shady object is bigger than twice of
|
|
number of remembered shady object at last full GC.
|
|
* number of oldgen object and remembered shady object is bigger
|
|
than half of total object space.
|
|
(please fix my English!)
|
|
|
|
Fri May 24 17:07:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
|
|
|
|
* intern.h: remove dangling rb_class_init_copy declaration
|
|
[ruby-core:55120] [Bug #8434]
|
|
|
|
Fri May 24 16:31:23 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/strscan/strscan.c (strscan_aref): raise error if given
|
|
name reference is not found.
|
|
|
|
Fri May 24 15:48:18 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (after_gc_sweep, garbage_collect_body): do major GC (full GC)
|
|
before extending heaps.
|
|
TODO: do major GC when there are many old (promoted) objects.
|
|
|
|
* gc.c (after_gc_sweep): remove TODO comments.
|
|
|
|
Fri May 24 11:04:00 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (LIBRUBY_RPATHFLAGS): do not append -L option with
|
|
runtime library directory if cross compiling, but only -R option.
|
|
runtime path makes no sense on the host system. [ruby-dev:47363]
|
|
[Bug #8443]
|
|
|
|
Fri May 24 02:57:17 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* object.c (rb_obj_clone): should not propagate OLDGEN status.
|
|
This propagation had caused WB miss for class.
|
|
|
|
Thu May 23 17:35:30 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* load.c (loaded_feature_path): fix invalid read by index underflow.
|
|
the beginning of name is also a boundary as well as just after '/'.
|
|
|
|
Thu May 23 17:21:22 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* gc.c (gc_profile_dump_on): revert r40898. ok to show the record
|
|
accumulating while lazy_sweep().
|
|
|
|
Wed May 22 16:50:18 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* gc.c (gc_profile_dump_on): use size_t to get rid of overflow and
|
|
show the header when next_index > 0, instead of next_index != 1.
|
|
|
|
Wed May 22 15:18:59 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (setup_overlapped): check the error code in addition
|
|
to the result of SetFilePointer() to determine if an error occurred,
|
|
because INVALID_SET_FILE_POINTER is a valid value.
|
|
[ruby-core:55098] [Bug #8431]
|
|
|
|
* win32/win32.c (setup_overlapped, finish_overlapped): extract from
|
|
rb_w32_read() and rb_w32_write().
|
|
|
|
Wed May 22 14:19:56 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_prepare_free_objects, rest_sweep, lazy_sweep): fix position
|
|
of `during_gc' setting.
|
|
|
|
Wed May 22 07:36:08 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (garbage_collect): all GC is start from garbage_collect()
|
|
(or garbage_collect_body()). `garbage_collect()' accept additional
|
|
two parameters `full_mark' and `immediate_sweep'.
|
|
If `full_mark' is TRUE, then force it full gc (major gc), otherwise,
|
|
it depends on status of object space. Now, it will be minor gc.
|
|
If `immediate_sweep' is TRUE, then disable lazy sweep.
|
|
To allocate free memory, `full_mark' and `immediate_sweep' should be
|
|
TRUE. Otherwise, they should be FALSE.
|
|
|
|
* gc.c (gc_prepare_free_objects): use `garbage_collect_body()'.
|
|
|
|
* gc.c (slot_sweep, before_gc_sweep, after_gc_sweep): add logging code.
|
|
|
|
Tue May 21 22:47:06 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/strscan/strscan.c (strscan_aref): support named captures.
|
|
patched by Konstantin Haase [ruby-core:54664] [Feature #8343]
|
|
|
|
Tue May 21 21:48:44 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* test/ruby/test_dir_m17n.rb (TestDir_M17N#test_entries_compose):
|
|
Use #each instead of #map just for iteration.
|
|
|
|
Tue May 21 19:57:22 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* ext/digest/lib/digest.rb (Digest::Class.file): Take optional
|
|
arguments that are passed to the constructor of the digest
|
|
class.
|
|
|
|
Tue May 21 17:21:12 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: remove gc_profile_record::is_marked. always true.
|
|
|
|
Tue May 21 17:13:40 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: fix to collect additional information for GC::Profiler.
|
|
* major/minor GC
|
|
* trigger reason of GC
|
|
|
|
* gc.c (gc_profile_dump_on): change reporting format with
|
|
added information.
|
|
|
|
* gc.c (gc_profile_record_get): return added information by
|
|
:GC_FLAGS => array.
|
|
|
|
Tue May 21 16:45:31 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: GC::Profiler's sweeping time is accumulated all slot
|
|
sweeping time. At lazy GC, GC::Profiler makes new record entry
|
|
for each lazy_sweep(). In this change, accumulating all
|
|
slot_sweep() time.
|
|
And change indentation.
|
|
|
|
Tue May 21 16:29:09 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* common.mk (rdoc-bench): add a benchmark rule
|
|
using RDoc. Generate all rdoc related files
|
|
(same as `make rdoc') in temporary directory
|
|
and remove them. Execution time, GC::Profiler
|
|
and results of GC.stat are printed.
|
|
|
|
* tool/rdocbench.rb: added for `rdoc-bench'.
|
|
|
|
Tue May 21 16:25:05 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (gc_profile_dump_on): `count' should be (int) because it
|
|
can be negative number.
|
|
And use pointer for `record' (don't copy).
|
|
|
|
Tue May 21 03:11:18 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* dir.c (dir_each): compose HFS file names from
|
|
UTF8-MAC. [ruby-core:48745] [Bug #7267]
|
|
|
|
Tue May 21 03:08:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* test/ruby/envutil.rb (assert_separately): require envutil in the
|
|
child process too.
|
|
|
|
Tue May 21 03:07:26 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_str_conv_enc_opts): should infect.
|
|
|
|
Mon May 20 22:24:45 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/set.rb (Set#delete_if, Set#keep_if): Avoid blockless call of
|
|
proc, which is not portable to JRuby. Replace &method() with
|
|
faster and simpler literal blocks while at it.
|
|
|
|
Mon May 20 22:00:31 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/e2mmap.rb: Format of E2MM documentation
|
|
|
|
Mon May 20 21:41:15 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/extmk.rb: nodoc this file
|
|
|
|
Mon May 20 20:43:32 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/cmath.rb: Remove duplicate RDoc heading from overview
|
|
|
|
Mon May 20 20:36:19 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/securerandom.rb: Update position of overview for RDoc
|
|
|
|
Mon May 20 19:33:55 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* math.c: improve and fix documentation of sin, tan and log
|
|
|
|
Mon May 20 19:31:49 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* lib/logger.rb (Logger::Application): show namespace in documentation
|
|
|
|
Mon May 20 11:50:12 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/pp.rb: Revert part of r40834 and nodoc PP::ObjectMixin
|
|
[ruby-core:55068]
|
|
|
|
Mon May 20 10:40:21 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/webrick/htmlutils.rb (WEBrick::HTMLUtils#escape): replace HTML
|
|
meta chars even in non-ascii string. [Bug #8425] [ruby-core:55052]
|
|
|
|
* lib/webrick/httputils.rb (WEBrick::HTTPUtils#{_escape,_unescape}):
|
|
fix %-escape encodings. [Bug #8425] [ruby-core:55052]
|
|
|
|
* lib/webrick/httpservlet/filehandler.rb (set_dir_list): revert r20152
|
|
partially and fix misuse of bytesize and regexp repetition operator.
|
|
|
|
Mon May 20 08:03:51 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/profiler.rb: Document Profiler__ methods
|
|
|
|
Mon May 20 08:02:13 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/tempfile.rb: nodoc Tempfile#inspect
|
|
|
|
Mon May 20 07:48:24 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/stringio/stringio.c: Correct position of method rdoc
|
|
|
|
Mon May 20 07:27:41 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* math.c: RDoc formatting of Math core docs with domains and codomains
|
|
Patch by @eLobato [Fixes GH-309]
|
|
|
|
Mon May 20 05:58:12 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/bigdecimal/bigdecimal.c: Formatting for BigMath [Fixes GH-306]
|
|
Based on a patch by @eLobato.
|
|
* ext/bigdecimal/lib/bigdecimal/math.rb: ditto
|
|
|
|
Mon May 20 04:56:59 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/forwardable.rb: Forwardable examples in overview were broken
|
|
Based on patch by @joem [Fixes GH-303] [Bug #8392]
|
|
|
|
Mon May 20 03:35:26 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/optparse.rb: nodoc OptionParser::Version and SPLAT_PROC
|
|
|
|
Mon May 20 03:16:52 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/pp.rb: Document PP::ObjectMixin [Fixes GH-312]
|
|
|
|
Sun May 19 23:52:22 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* test/webrick/test_htmlutils.rb: add test for WEBrick::HTMLUtils.
|
|
|
|
Sun May 19 23:12:07 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* encoding.c: document fix, change default script encoding.
|
|
patched by @windwiny [Fixes GH-310]
|
|
|
|
Sun May 19 17:29:07 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/set.rb (Set#delete_if, Set#keep_if): Add comments.
|
|
|
|
Sun May 19 11:37:36 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* ext/fiddle/extconf.rb: ignore rc version of libffi to fix build failure.
|
|
|
|
Sun May 19 10:38:50 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el (ruby-electric-delete-backward-char): Use
|
|
delete-char instead of delete-backward-char, which is an
|
|
interactive function.
|
|
|
|
Sun May 19 03:59:29 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* string.c (str_scrub0): added for refactoring.
|
|
|
|
Sun May 19 03:48:26 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* lib/uri/common.rb (URI.decode_www_form): scrub string if decoded
|
|
bytes are invalid for the encoding.
|
|
|
|
Sun May 19 02:46:32 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/set.rb (Set#delete_if, Set#keep_if): Make Set#delete_if and
|
|
Set#keep_if more space and time efficient by avoiding to_a.
|
|
|
|
Sun May 19 02:33:09 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el (ruby-electric-setup-keymap): Make
|
|
backquotes electric as well. It was listed in
|
|
ruby-electric-expand-delimiters-list but not activated.
|
|
|
|
* misc/ruby-electric.el (ruby-electric-delete-backward-char):
|
|
Introduce electric DEL that deletes what the previous electric
|
|
command has input.
|
|
|
|
* misc/ruby-electric.el (ruby-electric-matching-char): Make
|
|
electric quotes work again at the end of buffer.
|
|
|
|
Sun May 19 01:39:50 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (setjmp-type): check if setjmpex() is really available.
|
|
workaround for i686-w64-mingw32 which declares it but lacks its
|
|
definition.
|
|
|
|
* include/ruby/defines.h: include setjmpex.h only if also setjmpex()
|
|
is available.
|
|
|
|
Sat May 18 23:57:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (setjmp-type): use setjmpex() on w64-mingw32 to get rid
|
|
of -Wclobbered warnings.
|
|
|
|
* include/ruby/defines.h: include setjmpex.h here becase setjmp.h is
|
|
included from win32.h via intrin.h, winnt.h, and so on.
|
|
|
|
Sat May 18 20:28:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/mkconstants.rb (INTEGER2NUM): Make less comparisons.
|
|
|
|
Sat May 18 20:15:28 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* string.c (str_scrub_bang): add String#scrub!. [Feature #8414]
|
|
|
|
Sat May 18 16:59:52 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/mkconstants.rb (INTEGER2NUM): Renamed from INTEGER2VALUE.
|
|
|
|
Sat May 18 16:57:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/mkconstants.rb (INTEGER2VALUE): Suppress a warning:
|
|
comparison between signed and unsigned integer expressions
|
|
|
|
Sat May 18 16:38:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* compile.c (iseq_compile_each): forward anonymous and first keyword
|
|
rest argument one. [ruby-core:55033] [Bug #8416].
|
|
|
|
Sat May 18 15:49:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_core.h (rb_vm_tag): move jmpbuf between tag and prev so ensure to
|
|
be accessible.
|
|
|
|
Sat May 18 11:05:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enumerator.c (inspect_enumerator): use VALUE instead of mere char*
|
|
by using rb_sprintf() and rb_id2str().
|
|
|
|
* enumerator.c (append_method): extract from inspect_enumerator().
|
|
|
|
Sat May 18 09:00:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/mkconstants.rb (INTEGER2VALUE): Use LONG2FIX if possible.
|
|
|
|
Sat May 18 00:38:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/mkconstants.rb: Convert integer constants bigger than int
|
|
correctly.
|
|
|
|
Fri May 17 22:02:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/ifaddr.c: Use unsigned LONG_LONG to represent flags
|
|
because SunOS 5.11 (OpenIndiana) defines ifa_flags as uint64_t.
|
|
|
|
Fri May 17 21:47:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* cont.c: Typo in constant MAX_MACHINE_STACK_CACHE from '..MAHINE..'
|
|
patch by @schmurfy [Fixes GH-307]
|
|
|
|
Fri May 17 19:18:24 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el (ruby-electric-matching-char): Do not put
|
|
a closing quote when the quote typed does not start a string, as
|
|
in $', ?\' or ?\".
|
|
|
|
Fri May 17 18:06:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Consider error messages to find out version option of
|
|
C compiler.
|
|
The C compiler of Sun Studio C emits "Warning: Option -qversion
|
|
passed to ld, if ld is invoked, ignored otherwise" and exit
|
|
successfully.
|
|
|
|
Fri May 17 17:34:48 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* gc.c (rb_gc_guarded_ptr): unoptimize on other compilers than gcc and
|
|
msvc.
|
|
|
|
Fri May 17 11:06:48 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* eval_intern.h (TH_PUSH_TAG): ensure jmpbuf to be accessible before
|
|
pushing tag to get rid of unaccessible tag by stack overflow.
|
|
|
|
Thu May 16 17:15:32 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* vm_eval.c (rb_catch_obj): add volatile to tag to prevent crash
|
|
experimentally.
|
|
http://www.rubyist.net/~akr/chkbuild/debian/ruby-trunk/log/20130515T133500Z.log.html.gz
|
|
|
|
Thu May 16 16:19:50 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/Makefile.sub (verconf.in): no longer used.
|
|
|
|
* win32/Makefile.sub (config.status): fix typo.
|
|
|
|
* configure.in, template/verconf.h.in (RUBY_EXEC_PREFIX): fix for
|
|
default prefix.
|
|
|
|
Thu May 16 13:12:27 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* template/verconf.h.in: generate verconf.h from the template and
|
|
rbconfig.rb.
|
|
|
|
Thu May 16 05:47:18 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych/visitors/yaml_tree.rb: fix syntax error.
|
|
Thanks @spastorino! [ruby-core:55011]
|
|
|
|
Thu May 16 03:05:45 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_node_newnode): use newobj_of() instead of rb_newobj().
|
|
|
|
Thu May 16 02:03:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/depend: Add a dependency for ifaddr.o.
|
|
|
|
Thu May 16 01:44:45 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* common.mk (verconf.h): $< cannot be used in explicit rules with
|
|
nmake.
|
|
|
|
* win32/Makefile.sub (CONFIG_H): create verconf.in instead of
|
|
verconf.h.
|
|
|
|
Thu May 16 01:25:07 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych/visitors/yaml_tree.rb: only emit warnings when
|
|
-w is enabled.
|
|
|
|
Wed May 15 18:58:17 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (newobj): rename to `newobj_of' and accept additional
|
|
three parameters v1, v2, v3. newobj_of() do OBJSETUP() and
|
|
fill values with v1, v2, v3.
|
|
|
|
* gc.c (rb_data_object_alloc, rb_data_typed_object_alloc):
|
|
use newobj_of().
|
|
|
|
Wed May 15 17:55:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (RUBY_PLATFORM): move to config.h as needed by
|
|
version.c.
|
|
|
|
Wed May 15 17:04:11 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: add an additional RGENGC_PROFILE mode (2).
|
|
Profiling result can be check by GC.stat.
|
|
|
|
* gc.c (type_name): separate from obj_type_name().
|
|
|
|
Wed May 15 16:58:24 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in: save configured load path values into verconf.in.
|
|
|
|
* common.mk (verconf.h): create from verconf.in with shvar_to_cpp.rb.
|
|
|
|
* tool/shvar_to_cpp.rb: turn shell variables into C macros.
|
|
[Bug #7959]
|
|
|
|
* loadpath.c: split load path staffs from version.c.
|
|
|
|
* dmyloadpath.c: miniruby has no builtin load paths, so verconf.h is
|
|
not needed.
|
|
|
|
Wed May 15 03:56:09 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych/visitors/yaml_tree.rb: adding backwards
|
|
compatible YAMLTree.new method
|
|
|
|
Wed May 15 02:22:16 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych.rb: Adding Psych.safe_load for loading a user
|
|
defined, restricted subset of Ruby object types.
|
|
* ext/psych/lib/psych/class_loader.rb: A class loader for
|
|
encapsulating the logic for which objects are allowed to be
|
|
deserialized.
|
|
* ext/psych/lib/psych/deprecated.rb: Changes to use the class loader
|
|
* ext/psych/lib/psych/exception.rb: ditto
|
|
* ext/psych/lib/psych/json/stream.rb: ditto
|
|
* ext/psych/lib/psych/nodes/node.rb: ditto
|
|
* ext/psych/lib/psych/scalar_scanner.rb: ditto
|
|
* ext/psych/lib/psych/stream.rb: ditto
|
|
* ext/psych/lib/psych/streaming.rb: ditto
|
|
* ext/psych/lib/psych/visitors/json_tree.rb: ditto
|
|
* ext/psych/lib/psych/visitors/to_ruby.rb: ditto
|
|
* ext/psych/lib/psych/visitors/yaml_tree.rb: ditto
|
|
* ext/psych/psych_to_ruby.c: ditto
|
|
* test/psych/helper.rb: ditto
|
|
* test/psych/test_safe_load.rb: tests for restricted subset.
|
|
* test/psych/test_scalar_scanner.rb: ditto
|
|
* test/psych/visitors/test_to_ruby.rb: ditto
|
|
* test/psych/visitors/test_yaml_tree.rb: ditto
|
|
|
|
Wed May 15 02:06:35 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* test/psych/helper.rb: envutil is not available outside Ruby, so
|
|
port the functions from envutil to the test helper.
|
|
|
|
* test/psych/test_deprecated.rb: ditto
|
|
|
|
* test/psych/test_encoding.rb: ditto
|
|
|
|
Wed May 15 00:42:54 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* signal.c: need to include unistd.h for write(2).
|
|
unistd.h is now included via ruby/defines.h, but should explicitly
|
|
include here. (suggested by kosaki)
|
|
|
|
Tue May 14 23:43:05 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/.document: Add ifaddr.c.
|
|
|
|
Tue May 14 23:24:31 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/socket/extconf.rb: check for if_nametoindex() for
|
|
i686-w64-mingw32, and check for declarations of if_indextoname() and
|
|
if_nametoindex().
|
|
|
|
* ext/socket/ifaddr.c (ifaddr_ifindex): not-implement unless
|
|
if_nametoindex() is available.
|
|
|
|
* ext/socket/rubysocket.h: declare if_indextoname() and
|
|
if_nametoindex() if available but not declared.
|
|
|
|
Tue May 14 19:58:17 2013 CHIKANAGA Tomoyuki <nagachika@ruby-lang.org>
|
|
|
|
* ext/dl/lib/dl/func.rb (DL::Function#call): check tainted when
|
|
$SAFE > 0.
|
|
* ext/fiddle/function.c (function_call): check tainted when $SAFE > 0.
|
|
* test/fiddle/test_func.rb (module Fiddle): add test for above.
|
|
|
|
|
|
Tue May 14 14:51:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/win32.h (INTPTR_MAX, INTPTR_MIN, UINTPTR_MAX): split
|
|
from intptr_t and uintptr_t, since VC9 defines the latter only in
|
|
crtdefs.h.
|
|
|
|
Tue May 14 12:21:28 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/win32.c (NET_LUID): mingw may have NET_LUID and not defined
|
|
_IFDEF_.
|
|
|
|
Tue May 14 03:33:17 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* string.c (rb_str_new_frozen): remove debug print.
|
|
|
|
Tue May 14 03:22:51 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: enable to generate write barrier protected
|
|
arrays (T_ARRAY).
|
|
|
|
Tue May 14 03:21:42 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: enable to generate write barrier protected
|
|
strings (T_STRING).
|
|
|
|
Tue May 14 03:19:59 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: enable to generate write barrier protected
|
|
objects (T_OBJECT).
|
|
|
|
Tue May 14 03:17:15 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: enable to generate write barrier protected
|
|
objects for numeric types (Float, Complex, Rational, Bignum).
|
|
|
|
Tue May 14 03:10:59 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: enable RGENGC (USE_RGENGC)
|
|
but no type creates write protected (sunny) objects
|
|
(RGENGC_WB_PROTECTED_* == 0).
|
|
|
|
Tue May 14 02:47:30 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: support RGENGC. [ruby-trunk - Feature #8339]
|
|
See this ticket about RGENGC.
|
|
|
|
* gc.c: Add several flags:
|
|
* RGENGC_DEBUG: if >0, then prints debug information.
|
|
* RGENGC_CHECK_MODE: if >0, add assertions.
|
|
* RGENGC_PROFILE: if >0, add profiling features.
|
|
check GC.stat and GC::Profiler.
|
|
|
|
* include/ruby/ruby.h: disable RGENGC by default (USE_RGENGC == 0).
|
|
|
|
* array.c: add write barriers for T_ARRAY and generate sunny objects.
|
|
|
|
* include/ruby/ruby.h (RARRAY_PTR_USE): added. Use this macro if
|
|
you want to access raw pointers. If you modify the contents which
|
|
pointer pointed, then you need to care write barrier.
|
|
|
|
* bignum.c, marshal.c, random.c: generate T_BIGNUM sunny objects.
|
|
|
|
* complex.c, include/ruby/ruby.h: add write barriers for T_COMPLEX
|
|
and generate sunny objects.
|
|
|
|
* rational.c (nurat_s_new_internal), include/ruby/ruby.h: add write
|
|
barriers for T_RATIONAL and generate sunny objects.
|
|
|
|
* internal.h: add write barriers for RBasic::klass.
|
|
|
|
* numeric.c (rb_float_new_in_heap): generate sunny T_FLOAT objects.
|
|
|
|
* object.c (rb_class_allocate_instance), range.c:
|
|
generate sunny T_OBJECT objects.
|
|
|
|
* string.c: add write barriers for T_STRING and generate sunny objects.
|
|
|
|
* variable.c: add write barriers for ivars.
|
|
|
|
* vm_insnhelper.c (vm_setivar): ditto.
|
|
|
|
* include/ruby/ruby.h, debug.c: use two flags
|
|
FL_WB_PROTECTED and FL_OLDGEN.
|
|
|
|
* node.h (NODE_FL_CREF_PUSHED_BY_EVAL, NODE_FL_CREF_OMOD_SHARED):
|
|
move flag bits.
|
|
|
|
Tue May 14 01:54:48 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: remove rb_objspace_t::marked_num.
|
|
We can use `objspace_live_num()' instead of removed `marked_num'
|
|
if it is after `after_gc_sweep()' function call.
|
|
|
|
* gc.c (after_gc_sweep): use objspace_live_num() instead of removed
|
|
rb_objspace_t::marked_num.
|
|
|
|
* gc.c (gc_mark_ptr, gc_marks): remove rb_objspace_t::marked_num code.
|
|
|
|
* gc.c (gc_prepare_free_objects): do not call set_heaps_increment()
|
|
with checking objspace->heap.marked_num. At this point, we only
|
|
need to check availability of free-cell.
|
|
|
|
* gc.c (lazy_sweep): call after_gc_sweep() if there are no sweep_able entry.
|
|
|
|
* gc.c (rest_sweep, gc_prepare_free_objects): remove after_gc_sweep() call.
|
|
|
|
Tue May 14 01:50:41 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: disable GC_PROFILE_MORE_DETAIL (fix last commit).
|
|
|
|
* gc.c (gc_prof_set_malloc_info): fix "objspace->heap.live_num" to
|
|
"objspace_live_num(objspace)". There is no such member variable.
|
|
|
|
Tue May 14 01:25:55 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c: refactoring GC::Profiler.
|
|
|
|
* gc.c (gc_prof_sweep_timer_start/stop): removed because
|
|
they doesn't support lazy sweep.
|
|
|
|
* gc.c (gc_prof_sweep_slot_timer_start/stop): added.
|
|
redefine `sweeping time' to accumulated time of all of
|
|
slot_sweep().
|
|
|
|
* gc.c (rb_objspace_t::profile::count): renamed to
|
|
rb_objspace_t::profile::next_index. `counter' seems ambiguous.
|
|
increment it when next record is acquired.
|
|
|
|
Tue May 14 00:48:55 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: constify RRational::(num,den) and
|
|
RComplex::(real,imag).
|
|
Add macro to set these values:
|
|
* RRATIONAL_SET_NUM()
|
|
* RRATIONAL_SET_DEN()
|
|
* RCOMPLEX_SET_REAL()
|
|
* RCOMPLEX_SET_IMAG()
|
|
This change is a part of RGENGC branch [ruby-trunk - Feature #8339].
|
|
|
|
TODO: API design. RRATIONAL_SET(rat,num,den) is enough?
|
|
TODO: Setting constify variable with cast has same issue of r40691.
|
|
|
|
* complex.c, rational.c: use above macros.
|
|
|
|
Mon May 13 21:49:17 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Check socketpair again.
|
|
It is required on Unix.
|
|
|
|
Mon May 13 21:20:32 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/win32.c (getipaddrs): use alternative interface name if
|
|
available, because if_nametoindex() requires them.
|
|
|
|
Mon May 13 20:23:24 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/win32.c, include/ruby/win32.h (getipaddrs): [experimental]
|
|
emulate getipaddrs(3) on Unix.
|
|
|
|
* win32/Makefile.sub, configure.in (LIBS): need iphlpapi.lib for above
|
|
function.
|
|
|
|
* include/ruby/win32.h (socketpair): rb_w32_socketpair() doesn't
|
|
substitute for any function, so use non-prefixed name.
|
|
|
|
* ext/socket/extconf.rb (socketpair); follow above change.
|
|
|
|
Mon May 13 20:11:06 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* iseq.c (prepare_iseq_build): remove additional line break.
|
|
|
|
Mon May 13 19:29:54 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: constify RBasic::klass and add
|
|
RBASIC_CLASS(obj) macro which returns a class of `obj'.
|
|
This change is a part of RGENGC branch [ruby-trunk - Feature #8339].
|
|
|
|
* object.c: add new function rb_obj_reveal().
|
|
This function reveal internal (hidden) object by rb_obj_hide().
|
|
Note that do not change class before and after hiding.
|
|
Only permitted example is:
|
|
klass = RBASIC_CLASS(obj);
|
|
rb_obj_hide(obj);
|
|
....
|
|
rb_obj_reveal(obj, klass);
|
|
|
|
TODO: API design. rb_obj_reveal() should be replaced with others.
|
|
|
|
TODO: modify constified variables using cast may be harmful for
|
|
compiler's analysis and optimization.
|
|
Any idea to prohibit inserting RBasic::klass directly?
|
|
If rename RBasic::klass and force to use RBASIC_CLASS(obj),
|
|
then all codes such as `RBASIC(obj)->klass' will be
|
|
compilation error. Is it acceptable? (We have similar
|
|
experience at Ruby 1.9,
|
|
for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)".
|
|
|
|
* internal.h: add some macros.
|
|
* RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal
|
|
object.
|
|
* RBASIC_SET_CLASS(obj, cls) set RBasic::klass.
|
|
* RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS
|
|
without write barrier (planned).
|
|
* RCLASS_SET_SUPER(a, b) set super class of a.
|
|
|
|
* array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c,
|
|
file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c,
|
|
parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c,
|
|
string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c:
|
|
Use above macros and functions to access RBasic::klass.
|
|
|
|
* ext/coverage/coverage.c, ext/readline/readline.c,
|
|
ext/socket/ancdata.c, ext/socket/init.c,
|
|
* ext/zlib/zlib.c: ditto.
|
|
|
|
Mon May 13 18:44:14 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* *.c, parse.y, insns.def: use RARRAY_AREF/ASET macro
|
|
instead of using RARRAY_PTR().
|
|
|
|
Mon May 13 16:53:53 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* include/ruby/ruby.h: add new utility macros to access
|
|
Array's element.
|
|
* RARRAY_AREF(a, i) returns i-th element of an array `a'
|
|
* RARRAY_ASET(a, i, v) set i-th element of `a' to `v'
|
|
This change is a part of RGENGC branch [ruby-trunk - Feature #8339].
|
|
|
|
Mon May 13 15:31:10 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* object.c (rb_obj_setup): added.
|
|
|
|
* include/ruby/ruby.h (OBJSETUP): use rb_obj_setup() instead of
|
|
a macro.
|
|
|
|
Mon May 13 15:24:16 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_data_object_alloc): check klass only if klass is not 0.
|
|
klass==0 means internal object.
|
|
|
|
Mon May 13 14:57:28 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* gc.c (rb_data_object_alloc, rb_data_typed_object_alloc):
|
|
use NEWOBJ_OF() instead of NEWOBJ().
|
|
|
|
Mon May 13 14:51:59 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* proc.c (rb_obj_singleton_method): new method Kernel#singleton_method
|
|
which returns a Method object of the singleton method.
|
|
non-singleton method causes NameError, but not aliased or zsuper
|
|
method, right now.
|
|
[ruby-core:54914] [Feature #8391]
|
|
|
|
* vm_method.c (rb_method_entry_at): return the method entry for id at
|
|
klass, without ancestors.
|
|
|
|
* class.c (rb_singleton_class_get): get the singleton class if exists,
|
|
or nil.
|
|
|
|
Mon May 13 10:20:59 2013 Yuki Yugui Sonoda <yugui@google.com>
|
|
|
|
* ext/openssl/ossl_ssl.c: Disabled OpenSSL::SSL::SSLSocket if
|
|
defined(OPENSSL_NO_SOCK).
|
|
|
|
This fixes a linkage error on platforms which do not have socket.
|
|
OpenSSL itself is still useful as a set of cryptographic functions
|
|
even on such platforms.
|
|
|
|
Mon May 13 10:30:04 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* hash.c: Hash[] and {} are not equivalent by @eam [Fixes GH-301]
|
|
|
|
Mon May 13 10:04:22 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* random.c: Document Random::DEFAULT by @eLobato [Fixes GH-304]
|
|
|
|
Sun May 12 21:12:42 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h (OFFT2NUM): RUBY_REPLACE_TYPE also defines macro
|
|
to convert int type to VALUE if found.
|
|
|
|
Wed May 8 13:46:52 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* include/ruby/intern.h (rb_iv_set, rb_iv_get): removed. Because
|
|
ruby.h has a declaration for that.
|
|
|
|
Wed May 8 13:49:06 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* include/ruby/intern.h (rb_uint2big, rb_int2big, rb_uint2inum)
|
|
(rb_int2inum, rb_ll2inum, rb_ull2inum): removed because ruby.h
|
|
has a declaration for these.
|
|
|
|
Sun May 12 17:52:23 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: removes 'ac_cv_func_fseeko=yes' form MinGW
|
|
specific definitions.
|
|
|
|
Sun May 12 17:25:46 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* file.c (rb_file_s_truncate): use correct type. chsize takes
|
|
a long.
|
|
|
|
Sun May 12 17:18:46 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* process.c: move '#define HAVE_SPAWNV 1' to win32/Makefile.sub.
|
|
* win32/Makefile.sub: see above.
|
|
|
|
Sun May 12 17:13:32 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: removes AC_CHECK_FUNCS(setitimer) because it's
|
|
unused.
|
|
|
|
Sun May 12 17:08:16 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: removes AC_CHECK_FUNCS(pause) because it's unused.
|
|
|
|
Sun May 12 17:05:18 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* signal.c (rb_f_kill): fixes typo. s/HAS_KILLPG/HAVE_KILLPG/.
|
|
|
|
Sun May 12 17:03:27 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: abort if gettimeofday doesn't exist.
|
|
|
|
Sun May 12 16:31:27 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: adds RUBY_REPLACE_TYPE(off_t) for creating
|
|
NUM2OFFT.
|
|
* file.c (rb_file_truncate): use correct type. chsize() take
|
|
a long.
|
|
* include/ruby/ruby.h (NUM2OFFT): use a definition created by
|
|
a configure script by default.
|
|
|
|
Sun May 12 16:03:41 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: removes AC_CHECK_FUNC(fseeko, fseeko64, ftello,
|
|
ftello64). They are not used from anywhere.
|
|
|
|
* win32/win32.c (fseeko): removes.
|
|
* win32/win32.c (rb_w32_ftello): removes.
|
|
* include/ruby/win32.h: removes declarations of rb_w32_ftello and
|
|
rb_w32_fseeko.
|
|
* win32/Makefile.sub: removes '#define HAVE_FTELLO 1'.
|
|
|
|
Sun May 12 15:51:47 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: remove AC_CHECK_FUNC(close). It is not used from
|
|
anywhere.
|
|
|
|
Sun May 12 15:50:45 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: adds comments for setjmp check.
|
|
|
|
Sun May 12 15:38:09 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: move clock_gettime() check into regular place.
|
|
|
|
Wed May 8 13:45:53 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: add getenv() declaration check.
|
|
* dln_find.c: add HAVE_DECL_GETENV test.
|
|
|
|
Sun May 12 15:33:18 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: sorts AC_CHECK_FUNCS()s as alphabetical order.
|
|
|
|
Wed May 8 13:41:57 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* bignum.c: remove redundant decl for big_lshift() big_rshift().
|
|
|
|
Sun May 12 16:06:43 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/socket/rubysocket.h (rsock_inspect_sockaddr): as r40646
|
|
check HAVE_TYPE_STRUCT_SOCKADDR_DL.
|
|
|
|
Sat May 11 23:01:58 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/socket/rubysocket.h (HAVE_TYPE_STRUCT_SOCKADDR_DL):
|
|
MSVC has struct sockaddr_dl, but its content is broken.
|
|
http://ruby-mswin.cloudapp.net/vc10-x64/ruby-trunk/log/20130511T103938Z.log.html.gz
|
|
|
|
Sat May 11 22:07:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/rinda/test_rinda.rb: Socket.getifaddrs may returns an interface
|
|
which #addr method returns nil for venet0 in OpenVZ.
|
|
|
|
Sat May 11 21:56:34 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/raddrinfo.c (rsock_inspect_sockaddr): Add casts to
|
|
suppress warnings.
|
|
|
|
Sat May 11 17:28:51 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket: New method, Socket.getifaddrs, implemented.
|
|
[ruby-core:54777] [Feature #8368]
|
|
|
|
Sat May 11 00:47:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* gc.h (SET_MACHINE_STACK_END): Add !defined(_ILP32) to a defining
|
|
condition to avoid compilation error on x32.
|
|
https://sites.google.com/site/x32abi/
|
|
|
|
Fri May 10 23:56:34 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_peek_variable_name): treat invalid global, class,
|
|
and instance variable names as mere strings rather than errors.
|
|
[ruby-core:54885] [Bug #8375]
|
|
|
|
Fri May 10 20:22:40 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Move library checks into "Checks for libraries." part.
|
|
|
|
Fri May 10 19:32:01 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Reformat arguments of AC_CHECK_HEADERS and
|
|
AC_CHECK_FUNCS to track modifications easily.
|
|
|
|
Fri May 10 12:01:36 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Don't link librt if clock_gettime is available in
|
|
the main C library.
|
|
glibc 2.17 moves clock_* from librt to the main C library.
|
|
http://sourceware.org/ml/libc-announce/2012/msg00001.html
|
|
|
|
Thu May 9 22:00:35 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/ancdata.c (bsock_sendmsg_internal): controls_num should
|
|
not be negative.
|
|
|
|
Thu May 9 21:09:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* file.c, ext/etc/etc.c, ext/socket/unixsocket.c,
|
|
ext/openssl/ossl.h, ext/openssl/openssl_missing.c: Use
|
|
HAVE_AGGREGATE_MEMBER instead of HAVE_ST_MEMBER.
|
|
|
|
Thu May 9 20:43:41 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/ancdata.c (bsock_sendmsg_internal): Always set
|
|
controls_num to raise NotImplementedError appropriately.
|
|
(bsock_recvmsg_internal): Raise NotImplementedError if
|
|
:scm_rights=>true is given on platforms which don't have
|
|
4.4BSD style control message.
|
|
|
|
Thu May 9 12:06:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/rubysocket.h, ext/socket/unixsocket.c,
|
|
ext/socket/ancdata.c: Use HAVE_STRUCT_MSGHDR_MSG_CONTROL instead
|
|
of HAVE_ST_MSG_CONTROL.
|
|
|
|
Thu May 9 11:30:02 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* string.c: Add call-seq alias for String#=== [Bug #8381]
|
|
|
|
Thu May 9 11:14:18 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* doc/contributing.rdoc: Add guide for contributing to CRuby
|
|
|
|
Thu May 9 04:55:49 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Check socket library again. shutdown() is used in
|
|
io.c.
|
|
|
|
Thu May 9 01:52:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Don't check socketpair. socketpair is not used in
|
|
ruby command itself.
|
|
|
|
Thu May 9 01:05:41 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* class.c (rb_mod_included_modules): should not include non-modules.
|
|
[ruby-core:53158] [Bug #8025]
|
|
|
|
Wed May 8 22:46:59 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* class.c (rb_mod_included_modules): should not include the original
|
|
module itself. [ruby-core:53158] [Bug #8025]
|
|
|
|
Wed May 8 17:43:55 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* io.c (rb_io_ext_int_to_encs): ignore internal encoding if external
|
|
encoding is ASCII-8BIT. [Bug #8342]
|
|
|
|
Wed May 8 13:49:38 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/json/generator/generator.c (isArrayOrObject): cast char to
|
|
unsigned char. [Bug #8378]
|
|
|
|
Wed May 8 13:46:10 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/json/generator/depend: fix dependencies [Bug #8379]
|
|
|
|
* ext/json/parser/depend: ditto.
|
|
|
|
Wed May 8 13:07:17 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_yylex): fail if $, @, @@ are not followed by a valid
|
|
name character. [ruby-core:54846] [Bug #8375].
|
|
|
|
Wed May 8 13:06:31 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h (ISGRAPH): add missing macro.
|
|
|
|
Wed May 8 06:42:56 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/socket/socket.c (socket_s_ip_address_list): fix wrongly filled
|
|
sin6_scope_id on KAME introduced by r40593 for OpenIndiana.
|
|
KAME uses fe80:<scope_id>::<interface id> for link-local address
|
|
internally.
|
|
Setting sin6_scope_id causes it leaked.
|
|
see also comments of sockaddr_obj().
|
|
|
|
Tue May 7 22:12:34 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/readline/readline.c (insert_ignore_escape): Add a cast to
|
|
unsigned char * before dereference.
|
|
This suppress a warning on Cygwin.
|
|
|
|
Tue May 7 12:15:24 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/ancdata.c (bsock_recvmsg_internal): Add a cast to
|
|
suppress warning.
|
|
Bionic defines socklen_t as int.
|
|
Bionic defines msg_controllen as unsigned int (__kernel_size_t)
|
|
instead of socklen_t as POSIX.
|
|
|
|
Tue May 7 12:12:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/ancdata.c (ancillary_inspect): Don't call
|
|
anc_inspect_ipv6_pktinfo if !HAVE_TYPE_STRUCT_IN6_PKTINFO.
|
|
anc_inspect_ipv6_pktinfo is not defined in the case.
|
|
|
|
Tue May 7 12:10:52 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/socket.c (socket_s_ip_address_list): Cast EXTRA_SPACE as
|
|
int. This suppress a warning.
|
|
|
|
Tue May 7 12:09:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Set close_fds false for Cygwin.
|
|
Cygwin doesn't support fd passing.
|
|
This enables socket extension library cross-compilable by default.
|
|
|
|
Tue May 7 12:07:35 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* pack.c (swap32): Don't redefine it if it is already defined.
|
|
Bionic defines it.
|
|
(swap64): Ditto.
|
|
|
|
Mon May 6 20:50:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/socket.c (socket_s_ip_address_list): Fill sin6_scope_id
|
|
if getifaddrs() returns an IPv6 link local address which
|
|
sin6_scope_id is zero, such as on OpenIndiana SunOS 5.11.
|
|
|
|
Sun May 5 18:56:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* insns.def (defined): use vm_search_superclass() like as normal super
|
|
call. based on a patch <https://gist.github.com/wanabe/5520026> by
|
|
wanabe.
|
|
|
|
* vm_insnhelper.c (vm_search_superclass): return error but not raise
|
|
exceptions.
|
|
|
|
* vm_insnhelper.c (vm_search_super_method): check the result of
|
|
vm_search_superclass and raise exceptions on error.
|
|
|
|
Sun May 5 16:29:41 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* insns.def (defined): get method entry from the method top level
|
|
frame, not block frame. [ruby-core:54769] [Bug #8367]
|
|
|
|
Sun May 5 13:28:54 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* template/ruby.pc.in (Cflags): use rubyarchhdrdir for multiarch.
|
|
[Bug #7874]
|
|
|
|
Sat May 4 07:20:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* doc/security.rdoc: Add note about reporting security vulns
|
|
|
|
Sat May 4 04:13:27 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* include/ruby/defines.h (RUBY_ATTR_ALLOC_SIZE): New for
|
|
attribute((alloc_size(params))).
|
|
|
|
* include/ruby/defines.h (xmalloc, xmalloc2, xcalloc)
|
|
(xrealloc, xrealloc2): Annotated by RUBY_ATTR_ALLOC_SIZE.
|
|
* include/ruby/ruby.h (rb_alloc_tmp_buffer): ditto.
|
|
|
|
Fri May 3 19:32:13 2013 Takeyuki FUJIOKA <xibbar@ruby-lang.org>
|
|
|
|
* lib/cgi/util.rb: All class methods modulized.
|
|
We can use these methods like a function when "include CGI::Util".
|
|
[Feature #8354]
|
|
|
|
Fri May 3 14:09:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Make default_ipv6 true for Cygwin.
|
|
Cygwin supports IPv6 since Cygwin 1.7.1 (2009-12).
|
|
http://cygwin.com/ml/cygwin-announce/2009-12/msg00027.html
|
|
|
|
Fri May 3 13:35:26 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/socket/{getaddrinfo,getnameinfo}.c: define socklen_t if not
|
|
defined, e.g., older VC.
|
|
|
|
Fri May 3 13:29:11 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/win32.h (INTPTR_MAX, INTPTR_MIN, UINTPTR_MAX): also
|
|
should be defined when defining intptr_t and uintptr_t.
|
|
bigdecimal.c requires the former two now.
|
|
|
|
Fri May 3 13:22:12 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (poll_child_status): fix build error on older mingw.
|
|
|
|
Fri May 3 00:15:58 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* common.mk: remove timestamps in distclean-ext realclean-ext.
|
|
|
|
Thu May 2 23:23:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* object.c (rb_obj_is_kind_of): skip prepending modules.
|
|
[ruby-core:54742] [Bug #8357]
|
|
|
|
* object.c (rb_class_inherited_p): ditto.
|
|
[ruby-core:54736] [Bug #8357]
|
|
|
|
Thu May 2 22:11:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* bin/irb: remove dead code from sample/irb.rb.
|
|
|
|
Thu May 2 17:32:45 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* marshal.c (copy_ivar_i): get rid of overwriting already copied
|
|
instance variables. c.f. [Bug #8276]
|
|
|
|
Thu May 2 16:55:43 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* thread.c (id_locals): use cached ID.
|
|
|
|
* vm.c (ruby_thread_init): ditto.
|
|
|
|
* defs/id.def: add more predefined IDs used in core.
|
|
|
|
Thu May 2 13:42:42 2013 Ryan Davis <ryand-ruby@zenspider.com>
|
|
|
|
* lib/minitest/*: Imported minitest 4.7.4 (r8483)
|
|
* test/minitest/*: ditto
|
|
|
|
Thu May 2 11:32:22 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/win32.c (poll_child_status): [experimental] set the cause of
|
|
a child's death to status if its exitcode seems to be an error.
|
|
|
|
* test/ruby/test_process.rb (TestProcess#test_no_curdir): maybe now
|
|
we can test it.
|
|
|
|
* test/ruby/test_thread.rb (TestThread#test_thread_timer_and_interrupt):
|
|
ditto.
|
|
|
|
Thu May 2 11:24:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/yaml.rb: nodoc EngineManager, add History doc #8344
|
|
|
|
Wed May 1 21:11:17 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* time.c (localtime_with_gmtoff_zone): musl libc may return NULL for
|
|
tm_zone.
|
|
|
|
Wed May 1 18:59:36 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* enum.c (Enumerable#chunk): fix grammar of error message
|
|
for symbols beginning with an underscore [Bug #8351]
|
|
|
|
Wed May 1 16:47:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/curses/extconf.rb (curses_version): try once for each tests, a
|
|
function or a variable. fallback to variable for old SVR4.
|
|
|
|
Wed May 1 16:17:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/extmk.rb (extmake): extensions not to be installed should not
|
|
make static libraries, but make dynamic libraries always.
|
|
|
|
Wed May 1 12:20:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/rake/version.rb: Fix RDoc warning with :include: [Bug #8347]
|
|
|
|
Wed May 1 11:40:25 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* defs/id.def (predefined): add "idProc".
|
|
|
|
* eval.c (frame_func_id): use predefined IDs.
|
|
|
|
* proc.c (mnew, mproc, mlambda): use predefined IDs.
|
|
|
|
* vm.c (rb_vm_control_frame_id_and_class): ditto.
|
|
|
|
* vm.c (Init_VM): ditto.
|
|
|
|
Tue Apr 30 23:18:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/benchmark.rb: Update Benchmark results on newer CPU
|
|
|
|
Tue Apr 30 12:31:40 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* proc.c (mproc, mlambda): use frozen core methods instead of plain
|
|
global methods, so that methods cannot be overridden.
|
|
[ruby-core:54687] [Bug #8345]
|
|
|
|
* vm.c (Init_VM): define proc and lambda on the frozen core object.
|
|
|
|
* include/ruby/intern.h (rb_block_lambda): add declaration instead of
|
|
deprecated rb_f_lambda.
|
|
|
|
Mon Apr 29 17:02:30 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/nkf/nkf-utf8/nkf.h: Bionic libc doesn't have locale.
|
|
[Feature #8338]
|
|
|
|
|
|
Mon Apr 29 06:58:30 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/openssl/ossl_bn.c (ossl_bn_initialize): no need of alloca for
|
|
small fixed size array.
|
|
|
|
* ext/openssl/ossl_bn.c (ossl_bn_initialize): check overflow first,
|
|
and use alloca for small size input.
|
|
|
|
Mon Apr 29 00:40:13 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* lib/yaml.rb: Clarify documentation about YAML being always Psych.
|
|
Give a tip about using Syck. See #8344.
|
|
|
|
Sun Apr 28 23:34:01 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* lib/yaml.rb: Use another trick to define the YAML module.
|
|
https://twitter.com/n0kada/status/328342207511801856
|
|
|
|
Sun Apr 28 23:19:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/pp.rb: Update PP module overview by @geopet
|
|
|
|
Sun Apr 28 22:04:37 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* ext/openssl/ossl_bn.c (ossl_bn_initialize): fix buffer overflow on
|
|
x64 Windows and memory leak when initializing with integer.
|
|
[ruby-core:54615] [Bug #8337]
|
|
|
|
Sun Apr 28 12:38:04 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* README.EXT: correct method name to be used. [Bug #7982]
|
|
|
|
* README.EXT.ja: add notes too.
|
|
|
|
Sun Apr 28 10:35:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* object.c: With feedback from Steve Klabnik, reverted a change to
|
|
#untrusted? and #tainted?. Also adjusted grammar for $SAFE levels
|
|
|
|
Sun Apr 28 10:10:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/yaml.rb: Disable setting YAML const twice [ruby-core:54642]
|
|
|
|
Sun Apr 28 09:50:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* object.c: Documentation for taint and trust [Bug #8162]
|
|
|
|
Sun Apr 28 09:40:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* README.EXT: Copy note from r40505 for rb_sprintf() [Bug #7982]
|
|
|
|
Sun Apr 28 08:28:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/curses/curses.c: Update Curses::Window example for nicer output
|
|
Patch by Michal Suchanek [Bug #8121] [ruby-core:53520]
|
|
|
|
Sun Apr 28 08:10:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* README.EXT: Update note from r40504, by Jeremy Evans [Bug #7982]
|
|
|
|
Sun Apr 28 08:02:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* README.EXT: Add note to warn use of %i in Exceptions [Bug #7982]
|
|
|
|
Sun Apr 28 02:41:05 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Fix a typo. Should check endgrent() instead of
|
|
endgrnam().
|
|
|
|
Sun Apr 28 00:35:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c (obj2gid): Don't call endgrent() if not exist.
|
|
Bionic (Android's libc) don't have endgrent().
|
|
|
|
* configure.in: Check endgrnam function.
|
|
|
|
Sat Apr 27 23:53:00 2013 Charlie Somerville <charlie@charliesomerville.com>
|
|
|
|
* lib/yaml.rb: add security warning to YAML documentation
|
|
|
|
Sat Apr 27 23:25:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/yaml.rb: Documentation for YAML module [Bug #8213]
|
|
|
|
Sat Apr 27 20:19:21 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* thread_pthread.c (ruby_init_stack): Add STACK_GROW_DIR_DETECTION.
|
|
This fixes a compilation failure while cross-compiling for Tensilica
|
|
Xtensa Processor.
|
|
|
|
Sat Apr 27 19:32:44 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* thread.c: fix typos and documentation
|
|
|
|
Sat Apr 27 19:04:55 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* sparc.c: Use __asm__ instead of asm for gcc.
|
|
gcc doesn't provide asm keyword if -ansi option is given.
|
|
http://gcc.gnu.org/onlinedocs/gcc/Alternate-Keywords.html
|
|
|
|
Sat Apr 27 17:22:50 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Redundant test removed.
|
|
|
|
Sat Apr 27 16:00:10 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb (test_recvmsg_with_msg_peek_creates_fds):
|
|
Extracted.
|
|
|
|
Sat Apr 27 15:50:40 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (SIGNED_INTEGER_TYPE_P): New macro.
|
|
(SIGNED_INTEGER_MAX): Ditto.
|
|
(SIGNED_INTEGER_MIN): Ditto.
|
|
(UNSIGNED_INTEGER_MAX): Ditto.
|
|
(TIMET_MAX): Use SIGNED_INTEGER_MAX and UNSIGNED_INTEGER_MAX.
|
|
(TIMET_MIN): Use SIGNED_INTEGER_MIN.
|
|
|
|
* thread.c (TIMEVAL_SEC_MAX): Use SIGNED_INTEGER_MAX.
|
|
(TIMEVAL_SEC_MIN): Use SIGNED_INTEGER_MIN.
|
|
|
|
Sat Apr 27 10:52:52 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* thread.c (TIMEVAL_SEC_MAX, TIMEVAL_SEC_MIN): Consider environments,
|
|
sizeof(time_t) is smaller than sizeof(tv_sec), such as
|
|
OpenBSD 5.2 (amd64).
|
|
|
|
Fri Apr 26 23:34:59 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/text.rb (REXML::Text.normalize): Fix a bug that all
|
|
entity filters are ignored. [ruby-dev:47278] [Bug #8302]
|
|
Patch by Ippei Obayashi. Thanks!!!
|
|
* test/rexml/test_entity.rb (EntityTester#test_entity_filter): Add
|
|
a test of the above change.
|
|
|
|
Fri Apr 26 22:53:55 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/element.rb (REXML::Attributes#to_a): Support
|
|
namespaced attributes. [ruby-dev:47277] [Bug #8301]
|
|
Patch by Ippei Obayashi. Thanks!!!
|
|
* test/rexml/test_attributes.rb
|
|
(AttributesTester#test_to_a_with_namespaces): Add a test of the
|
|
above change.
|
|
|
|
Fri Apr 26 21:48:29 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rss/atom.rb (RSS::Atom::Entry): Fix indent of document comment.
|
|
|
|
Fri Apr 26 21:21:17 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rss/maker.rb (RSS::Maker): Fix indent of document comment.
|
|
|
|
Fri Apr 26 18:41:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Use a block of enable_config() for
|
|
--{enable,disable}-close-fds-by-recvmsg-with-peek configure option
|
|
|
|
Fri Apr 26 18:08:08 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* dir.c (dir_set_pos): Fix a compilation error when seekdir() is not
|
|
exist.
|
|
|
|
Fri Apr 26 17:41:17 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* thread_pthread.c (ruby_init_stack): Add STACK_GROW_DIR_DETECTION.
|
|
This fixes a compilation failure while cross-compiling for ARM.
|
|
|
|
Fri Apr 26 14:35:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/rss/atom.rb: Documentation for RSS::Atom based on a patch by
|
|
Michael Denomy
|
|
* lib/rss/maker.rb: Documentation for RSS::Maker also by @mdenomy
|
|
|
|
Fri Apr 26 12:41:22 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/curses/extconf.rb: Test linkability of curses_version at first.
|
|
|
|
* ext/socket/extconf.rb: Test the behavior of fd passing with MSG_PEEK
|
|
only if recvmsg(), msg_control member, AF_UNIX and SCM_RIGHTS are
|
|
available.
|
|
|
|
Fri Apr 26 00:07:52 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* lib/rinda/ring.rb (Rinda::RingServer#initialize): accept array
|
|
arguments of address to specify multicast interface.
|
|
|
|
* lib/rinda/ring.rb (Rinda::RingServer#make_socket): add optional
|
|
arguments for multicast interface.
|
|
|
|
* test/rinda/test_rinda.rb
|
|
(TestRingFinger#test_ring_server_ipv4_multicast,
|
|
TestRingFinger#test_ring_server_ipv6_multicast): add tests for
|
|
above change.
|
|
|
|
* test/rinda/test_rinda.rb
|
|
(TestRingServer#test_make_socket_ipv4_multicast,
|
|
TestRingServer#test_make_socket_ipv6_multicast): change bound
|
|
interface address because multicast address is not allowed on Linux
|
|
or Windows.
|
|
[ruby-core:53692] [Bug #8159]
|
|
|
|
Thu Apr 25 23:45:02 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* lib/rinda/ring.rb (Rinda::RingServer#initialize): add a socket
|
|
to @sockets in make_socket() to close sockets on shutdown even if
|
|
make_socket() is called after initialize.
|
|
|
|
* lib/rinda/ring.rb (Rinda::RingServer#make_socket): ditto.
|
|
|
|
Thu Apr 25 23:39:42 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* test/rinda/test_rinda.rb (TupleSpaceProxyTest#test_take_bug_8215):
|
|
use KILL on Windows since TERM doen't work and ruby process remains
|
|
after test-all on Windows.
|
|
|
|
Thu Apr 25 23:16:28 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/curses/extconf.rb: Implement
|
|
--with-curses-version={function,variable} configure option for
|
|
cross-compiling.
|
|
|
|
Thu Apr 25 18:15:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Don't use WIDE getaddrinfo by default.
|
|
|
|
Thu Apr 25 17:56:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Remove obsolete options: ---with-ipv6-lib and
|
|
--with-ipv6-libdir.
|
|
|
|
Thu Apr 25 17:43:49 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Implement
|
|
--{enable,disable}-close-fds-by-recvmsg-with-peek configure option
|
|
for cross-compiling.
|
|
Make --{enable,disable}-wide-getaddrinfo configure option
|
|
cross-compiling friendly.
|
|
|
|
Thu Apr 25 16:11:06 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* io.c (rb_io_ext_int_to_encs, parse_mode_enc): bom-prefixed name is
|
|
not a real encoding name, just a fallback. so the proper conversion
|
|
should take place even if if the internal encoding is equal to the
|
|
bom-prefixed name, unless actual encoding is equal to the internal
|
|
encoding. [ruby-core:54563] [Bug #8323]
|
|
|
|
* io.c (io_set_encoding_by_bom): reset extenal encoding if no BOM
|
|
found. [ruby-core:54569]
|
|
|
|
Thu Apr 25 14:35:01 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/openssl/ossl_bn.c (ossl_bn_initialize): allow Fixnum and Bignum.
|
|
[ruby-core:53986] [Feature #8217]
|
|
|
|
Thu Apr 25 14:26:32 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* lib/uri/common.rb (URI.decode_www_form): follow current URL Standard.
|
|
It gets encoding argument to specify the character encoding.
|
|
It now allows loose percent encoded strings, but denies ;-separator.
|
|
[ruby-core:53475] [Bug #8103]
|
|
|
|
* lib/uri/common.rb (URI.decode_www_form): follow current URL Standard.
|
|
It gets encoding argument to convert before percent encode.
|
|
Now UTF-16 strings aren't converted to UTF-8 before percent encode
|
|
by default.
|
|
|
|
Wed Apr 25 14:26:00 2013 Charlie Somerville <charlie@charliesomerville.com>
|
|
|
|
* benchmark/bm_hash_shift.rb: add benchmark for Hash#shift
|
|
|
|
* hash.c (rb_hash_shift): use st_shift if hash is not being iterated to
|
|
delete element without iterating the whole hash.
|
|
|
|
* hash.c (shift_i): remove function
|
|
|
|
* include/ruby/st.h (st_shift): add st_shift function
|
|
|
|
* st.c (st_shift): ditto
|
|
|
|
[Bug #8312] [ruby-core:54524] Patch by funny-falcon
|
|
|
|
Thu Apr 25 12:03:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Extract C programs as toplevel constants.
|
|
|
|
Thu Apr 25 02:23:28 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (RUBY_RM_RECURSIVE): this hack is needed by only
|
|
autoconf 2.69 or earlier on darwin.
|
|
|
|
Thu Apr 25 01:22:41 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/tracer.rb (get_line): simply read by File.readlines.
|
|
|
|
* lib/debug.rb (script_lines): get source lines from SCRIPT_LINES__ or
|
|
read from the file.
|
|
|
|
* lib/debug.rb (display_list): use script_lines instead of recursion.
|
|
[Bug #8318]
|
|
|
|
* lib/debug.rb (line_at): use script_lines same as display_list.
|
|
|
|
* lib/debug.rb (display_list): Fix debug listing when called from the
|
|
same file it has been required. patch by Dario Bertini <berdario AT
|
|
gmail.com> [Bug #8318] [fix GH-280]
|
|
|
|
Wed Apr 24 21:51:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: Check mblen().
|
|
mblen() is optional in uClibc.
|
|
|
|
* eval_intern.h (CharNext): Don't use mblen() is not available.
|
|
|
|
Wed Apr 24 15:55:06 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* io.c (rb_fd_fix_cloexec): use rb_update_max_fd().
|
|
|
|
Wed Apr 24 14:08:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* numeric.c: Fix wiki link on Float imprecision in overview, patched
|
|
by Makoto Kishimoto [Bug #8304] [ruby-dev:47280]
|
|
|
|
Wed Apr 24 14:03:59 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (parser_yylex): disallow $- without following identifier
|
|
character. [ruby-talk:406969]
|
|
|
|
* parse.y (is_special_global_name): mere $- is not a valid global
|
|
variable name.
|
|
|
|
Wed Apr 24 13:54:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* string.c: Document String#setbyte return value by @gjmurakami-10gen
|
|
[Fixes GH-294]
|
|
|
|
Wed Apr 24 13:45:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* class.c: Example of Object#methods by @windwiny [Fixes GH-293]
|
|
* ruby.c: Document return values of Kernel #sub, #gsub, and #chop
|
|
|
|
Wed Apr 24 12:54:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/socket/lib/socket.rb: Doc typos by @vipulnsward [Fixes GH-292]
|
|
|
|
|
|
Wed Apr 24 12:54:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/socket/lib/socket.rb: Doc typos by @vipulnsward [Fixes GH-292]
|
|
|
|
Wed Apr 24 12:27:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* array.c: Fix documentation for Array#index and #replace aliases
|
|
Based on a patch by @phiggins [Fixes GH-282]
|
|
|
|
Tue Apr 23 21:14:38 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* string.c (rb_str_inspect): refix r40413, on Ruby 1.9 usual character
|
|
escape uses hex/Unicode escapes, so fix to use Unicode escape on
|
|
Unicode strings and hex on others. [ruby-core:54458] [Bug #8290]
|
|
|
|
Tue Apr 23 20:10:02 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* missing/isnan.c (isnan): Don't define if isnan() macro is defined.
|
|
This fixes a compilation failure on uClibc based Gentoo system.
|
|
|
|
Tue Apr 23 17:40:40 2013 Martin Duerst <duerst@it.aoyama.ac.jp>
|
|
|
|
* lib/rexml/document.rb, lib/rexml/element.rb,
|
|
lib/rexml/formatters/pretty.rb: remove opinionated
|
|
language in documentation. [Bug #8309],
|
|
reported by Charles Beckmann
|
|
|
|
Tue Apr 23 14:04:44 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* lib/net/imap.rb (getacl_response): parse the mailbox of an ACL
|
|
response correctly. [ruby-core:54365] [Bug #8281]
|
|
|
|
Tue Apr 23 11:58:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_str_scrub): fix for UTF-32. strlen() on strings
|
|
contain NUL returns wrong result, use sizeof operator instead.
|
|
[ruby-dev:45975] [Feature #6752]
|
|
|
|
Tue Apr 23 10:26:50 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* test/ruby/test_module.rb
|
|
(TestModule#test_const_get_invalid_name)
|
|
(test_const_defined_invalid_name): Fix expected values.
|
|
|
|
Tue Apr 23 09:51:26 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* string.c (rb_str_inspect): NUL should not be represented as "\0"
|
|
when octal digits may follow. [ruby-core:54458] [Bug #8290]
|
|
|
|
Mon Apr 22 22:54:00 2013 Charlie Somerville <charlie@charliesomerville.com>
|
|
|
|
* insns.def (opt_mod): Use % operator if both operands are positive for
|
|
a significant performance improvement. Thanks to @samsaffron.
|
|
|
|
Mon Apr 22 17:09:37 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* marshal.c (r_object0): copy all instance variables not only generic
|
|
ivars, before calling post proc. [ruby-core:51163] [Bug #7627]
|
|
|
|
Mon Apr 22 10:25:21 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* util.c (ruby_hdtoa): revert r29729.
|
|
If you want ruby to behave as before on x86, specify to use SSE like
|
|
-msse2 -mfpmath=sse for gcc.
|
|
|
|
Sun Apr 21 23:19:00 2013 Charlie Somerville <charlie@charliesomerville.com>
|
|
|
|
* configure.in: Revert using sigsetjmp by default due to performance
|
|
problems on some systems (eg. older Linux)
|
|
|
|
Sun Apr 21 21:35:00 2013 Charlie Somerville <charlie@charliesomerville.com>
|
|
|
|
* configure.in: Use sigsetjmp by default so jumping out of signal
|
|
handlers properly restores the signal mask and SS_ONSTACK flag.
|
|
[ruby-core:54175] [Bug #8254]
|
|
|
|
* configure.in: Manually check for presence of sigsetjmp. It is not a
|
|
function on some systems, so AC_CHECK_FUNCS cannot be used.
|
|
|
|
Sun Apr 21 08:00:55 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/csv/test_features.rb, test/logger/test_logger.rb
|
|
test/mkmf/test_have_macro.rb, test/net/http/test_http.rb,
|
|
test/openssl/test_config.rb, test/psych/test_encoding.rb,
|
|
test/psych/test_exception.rb, test/psych/test_psych.rb,
|
|
test/psych/test_tainted.rb, test/readline/test_readline.rb,
|
|
test/rexml/test_contrib.rb, test/ruby/test_autoload.rb,
|
|
test/ruby/test_beginendblock.rb, test/ruby/test_exception.rb,
|
|
test/ruby/test_file.rb, test/ruby/test_io.rb,
|
|
test/ruby/test_marshal.rb, test/ruby/test_process.rb,
|
|
test/ruby/test_require.rb, test/ruby/test_rubyoptions.rb,
|
|
test/syslog/test_syslog_logger.rb, test/webrick/test_httpauth.rb,
|
|
test/zlib/test_zlib.rb: Use Tempfile.create.
|
|
|
|
Sun Apr 21 00:15:36 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/tempfile.rb (Tempfile.create): Close when the block exits.
|
|
|
|
Sat Apr 20 23:38:14 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/webrick/httpauth/htpasswd.rb: Use Tempfile.create to avoid
|
|
unintentional unlink() by the finalizer.
|
|
lib/webrick/httpauth/htdigest.rb: Ditto.
|
|
|
|
Sat Apr 20 22:47:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/tempfile.rb (Tempfile.create): New method.
|
|
The method name is proposed by Shugo Maeda. [ruby-dev:47220]
|
|
[ruby-core:41478] [Feature #5707]
|
|
|
|
Sat Apr 20 14:22:10 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* marshal.c (w_object): dump no ivars to the original by marshal_dump.
|
|
[ruby-core:54334] [Bug #8276]
|
|
|
|
* marshal.c (r_object0): copy all ivars of marshal_dump data to the
|
|
result object instead. [ruby-core:51163] [Bug #7627]
|
|
|
|
Sat Apr 20 02:33:27 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* string.c (str_scrub): add ruby method String#scrub which verify and
|
|
fix invalid byte sequence. [ruby-dev:45975] [Feature #6752]
|
|
|
|
* string.c (str_compat_and_valid): check given string is compatible
|
|
and valid with given encoding.
|
|
|
|
* transcode.c (str_transcode0): If invalid: :replace is specified for
|
|
String#encode, replace invalid byte sequence even if the destination
|
|
encoding equals to the source encoding.
|
|
|
|
Fri Apr 19 21:55:40 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* README.EXT.ja (Data_Wrap_Struct): Remove a description about
|
|
orphan argument. Oh, I renamed the argument name without
|
|
changing description at r36180... Sorry....
|
|
Patch by Makoto Kishimoto. Thanks!!! [ruby-dev:47269] [Bug #8292]
|
|
* README.EXT.ja (Data_Make_Struct): Add a sample code that describes
|
|
how it works.
|
|
Patch by Makoto Kishimoto. Thanks!!! [ruby-dev:47269] [Bug #8292]
|
|
|
|
Fri Apr 19 17:54:57 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* lib/net/imap.rb (body_type_msg): should accept
|
|
message/delivery-status with extra data.
|
|
[ruby-core:53741] [Bug #8167]
|
|
|
|
* test/net/imap/test_imap_response_parser.rb: related test.
|
|
|
|
Fri Apr 19 13:03:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* marshal.c (w_object): do not dump encoding which is dumped with
|
|
marshal_dump data. [ruby-core:54334] [Bug #8276]
|
|
|
|
Fri Apr 19 11:36:53 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (stack_protector): control use of -fstack-protector.
|
|
|
|
* configure.in (debugflags): let -fstack-protector precede and disable
|
|
debugflags, because they can't work together on SmartOS. [Bug #8268]
|
|
|
|
Fri Apr 19 07:43:52 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* test/openssl/test_cipher.rb: Correct a typo
|
|
by jgls <joerg@joergleis.com>
|
|
https://github.com/ruby/ruby/pull/291 fix GH-291
|
|
|
|
Thu Apr 18 16:58:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_method.c (rb_mod_public_method): fix visibility on anonymous
|
|
module. set visibility of singleton method, not method in base
|
|
class. [ruby-core:54404] [Bug #8284]
|
|
|
|
Thu Apr 18 16:20:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* dir.c (glob_helper): should skip dot directories only for recursion,
|
|
but should not if matching to the given pattern. [ruby-core:54387]
|
|
[Bug #8283]
|
|
|
|
Thu Apr 18 16:20:21 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* pack.c (pack_unpack): increase buffer size to fix buffer overflow,
|
|
and fix garbage just after unpacking without missing paddings.
|
|
[Bug #8286]
|
|
|
|
Thu Apr 18 13:35:54 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* pack.c (pack_unpack): output characters even if the input doesn't
|
|
have paddings. [Bug #8286]
|
|
|
|
Thu Apr 18 08:20:48 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* common.mk (clean-ext): remove timestamps.
|
|
|
|
Wed Apr 17 22:07:50 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/rubysocket.h (SOCKLEN_MAX): Expression simplified.
|
|
|
|
Wed Apr 17 20:09:19 2013 Aman Gupta <ruby@tmm1.net>
|
|
|
|
* compile.c (iseq_add_mark_object): Use new rb_iseq_add_mark_object().
|
|
|
|
* insns.def (setinlinecache): Ditto.
|
|
|
|
* iseq.c (rb_iseq_add_mark_object): New function to allocate
|
|
iseq->mark_ary on demand. [Bug #8142]
|
|
|
|
* iseq.h (rb_iseq_add_mark_object): Ditto.
|
|
|
|
* iseq.c (prepare_iseq_build): Avoid allocating mark_ary until needed.
|
|
|
|
* iseq.c (rb_iseq_build_for_ruby2cext): Ditto.
|
|
|
|
Wed Apr 17 20:00:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/rubysocket.h (SOCKLEN_MAX): Defined.
|
|
|
|
* ext/socket/raddrinfo.c (ext/socket/raddrinfo.c): Reject too long
|
|
Linux abstract socket name.
|
|
|
|
Wed Apr 17 19:45:27 2013 Aman Gupta <tmm1@ruby-lang.org>
|
|
|
|
* iseq.c (iseq_location_setup): re-use existing string when iseq has
|
|
the same path and absolute_path. [Bug #8149]
|
|
|
|
Wed Apr 17 11:38:37 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
|
|
UNASSIGNED is not a valid message.
|
|
|
|
Wed Apr 17 10:58:18 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* thread.c (sleep_timeval): get rid of overflow on Windows where
|
|
timeval.tv_sec is not time_t but mere long.
|
|
|
|
Tue Apr 16 23:07:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/unixsocket.c (unix_send_io): Suppress a warning by clang.
|
|
(unix_recv_io): Ditto.
|
|
|
|
Tue Apr 16 12:27:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/sdbm/init.c: Fix comment indentation, by windwiny [Fixes GH-277]
|
|
|
|
Tue Apr 16 12:25:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/socket/option.c: Document synonymous methods, by windwiny [GH-277]
|
|
* ext/stringio/stringio.c: ditto
|
|
* ext/io/wait/wait.c: ditto
|
|
* ext/gdbm/gdbm.c: ditto
|
|
* ext/dl/cfunc.c: ditto
|
|
* ext/zlib/zlib.c: ditto
|
|
* ext/win32ole/win32ole.c: ditto
|
|
* ext/dbm/dbm.c: ditto
|
|
* ext/json/generator/generator.c: ditto
|
|
* ext/date/date_core.c: ditto
|
|
|
|
Tue Apr 16 11:23:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/openssl/*: Document synonymous methods, by windwiny [GH-277]
|
|
|
|
Mon Apr 15 22:21:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/fiddle/depend: New file.
|
|
|
|
Mon Apr 15 22:01:02 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el (ruby-electric-insert): Check
|
|
ruby-electric-is-last-command-char-expandable-punct-p here.
|
|
|
|
* misc/ruby-electric.el (ruby-electric-closing-char): New
|
|
interactive function bound to closing characters. Typing one of
|
|
those closing characters right after the matching counterpart
|
|
cancels the effect of automatic closing. For example, typing
|
|
"{" followed by "}" simply makes "{}" instead of "{ } }".
|
|
|
|
Mon Apr 15 12:54:42 2013 Martin Bosslet <Martin.Bosslet@gmail.com>
|
|
|
|
* ext/openssl/ossl_ssl.c: Correct shutdown behavior w.r.t GC.
|
|
|
|
* test/openssl/test_ssl.rb: Add tests to verify correct behavior.
|
|
|
|
[Bug #8240] Patch provided by Shugo Maeda. Thanks!
|
|
|
|
Mon Apr 15 10:23:39 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/coverage/depend: fix id.h place as r40283.
|
|
|
|
* ext/coverage/extconf.rb: add topdir and topsrcdir to VPATH.
|
|
|
|
Sun Apr 14 19:46:14 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/-test-/debug/depend: New file.
|
|
|
|
* ext/-test-/exception/depend: Ditto.
|
|
|
|
* ext/-test-/printf/depend: Ditto.
|
|
|
|
* ext/-test-/string/depend: Ditto.
|
|
|
|
* ext/coverage/depend: Ditto.
|
|
|
|
* ext/io/console/depend: Ditto.
|
|
|
|
* ext/io/nonblock/depend: Ditto.
|
|
|
|
* ext/io/wait/depend: Ditto.
|
|
|
|
* ext/openssl/depend: Ditto.
|
|
|
|
* ext/pathname/depend: Ditto.
|
|
|
|
* ext/psych/depend: Ditto.
|
|
|
|
* ext/zlib/depend: Ditto.
|
|
|
|
Sun Apr 14 02:46:50 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (MakeMakefile#create_makefile): remove {$(VPATH)} other
|
|
than nmake.
|
|
|
|
* ext/ripper/depend: use VPATH expecting removed by above.
|
|
|
|
Sat Apr 13 23:06:20 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (timestamp_file): gather timestamp files in one
|
|
directory from each extension directories.
|
|
|
|
Sat Apr 13 21:09:02 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (MakeMakefile#create_makefile): output new macro
|
|
disthdrdir to specify the path of id.h, parse.h and etc.
|
|
|
|
* ext/ripper/depend: use above macro.
|
|
|
|
Sat Apr 13 20:28:08 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* Merge Onigmo 5.13.4 f22cf2e566712cace60d17f84d63119d7c5764ee.
|
|
[bug] fix problem with optimization of \z (Issue #16) [Bug #8210]
|
|
|
|
Sat Apr 13 18:56:15 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ext/ripper/depend: parse.h and id.h may be created on topdir.
|
|
|
|
Sat Apr 13 12:08:16 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
|
|
|
|
* lib/matrix.rb: Add Vector#cross_product, patch by Luis Ezcurdia
|
|
[fix GH-276] [rubyspec:81eec89a124]
|
|
|
|
Sat Apr 13 10:20:37 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* struct.c (rb_struct_define_without_accessor, rb_struct_define),
|
|
(rb_struct_s_def): hide member names array.
|
|
|
|
* struct.c (anonymous_struct, new_struct, setup_struct): split
|
|
make_struct() for each purpose.
|
|
|
|
Sat Apr 13 09:34:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/mkmf.rb: Add ruby/ruby.h, ruby/missing.h, ruby/intern.h,
|
|
ruby/st.h and ruby/subst.h for ruby_headers in generated Makefile.
|
|
|
|
* ext/-test-/old_thread_select/depend: Update dependencies.
|
|
|
|
* ext/-test-/wait_for_single_fd/depend: Ditto.
|
|
|
|
* ext/bigdecimal/depend: Ditto.
|
|
|
|
* ext/curses/depend: Ditto.
|
|
|
|
* ext/digest/bubblebabble/depend: Ditto.
|
|
|
|
* ext/digest/depend: Ditto.
|
|
|
|
* ext/digest/md5/depend: Ditto.
|
|
|
|
* ext/digest/rmd160/depend: Ditto.
|
|
|
|
* ext/digest/sha1/depend: Ditto.
|
|
|
|
* ext/digest/sha2/depend: Ditto.
|
|
|
|
* ext/dl/callback/depend: Ditto.
|
|
|
|
* ext/dl/depend: Ditto.
|
|
|
|
* ext/etc/depend: Ditto.
|
|
|
|
* ext/nkf/depend: Ditto.
|
|
|
|
* ext/objspace/depend: Ditto.
|
|
|
|
* ext/pty/depend: Ditto.
|
|
|
|
* ext/readline/depend: Ditto.
|
|
|
|
* ext/ripper/depend: Ditto.
|
|
|
|
* ext/sdbm/depend: Ditto.
|
|
|
|
* ext/socket/depend: Ditto.
|
|
|
|
* ext/stringio/depend: Ditto.
|
|
|
|
* ext/strscan/depend: Ditto.
|
|
|
|
* ext/syslog/depend: Ditto.
|
|
|
|
* ext/-test-/num2int/depend: Removed.
|
|
|
|
* ext/dbm/depend: Ditto.
|
|
|
|
* ext/fcntl/depend: Ditto.
|
|
|
|
* ext/gdbm/depend: Ditto.
|
|
|
|
* ext/racc/cparse/depend: Ditto.
|
|
|
|
Sat Apr 13 00:15:54 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/etc/etc.c (Init_etc): move Passwd and Group under Etc namespace
|
|
as primary names.
|
|
|
|
Fri Apr 12 21:06:55 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* common.mk: pack.o depends on internal.h.
|
|
|
|
Fri Apr 12 20:59:24 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (ones): Use __builtin_popcountl if available.
|
|
|
|
* internal.h (GCC_VERSION_SINCE): Macro moved from pack.c.
|
|
|
|
* pack.c: Include internal.h for GCC_VERSION_SINCE.
|
|
|
|
Fri Apr 12 18:29:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* common.mk: version.o depends on $(srcdir)/include/ruby/version.h
|
|
instead of {$(VPATH)}version.h to avoid confusion by VPATH between
|
|
top level version.h and include/ruby/version.h for build in-place.
|
|
[ruby-dev:47249] [Bug #8256]
|
|
|
|
Fri Apr 12 15:21:24 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_insnhelper.c (vm_callee_setup_keyword_arg): non-symbol key is not
|
|
a keyword argument, keep it as a positional argument.
|
|
|
|
Fri Apr 12 11:58:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* array.c: Document synonymous methods, by windwiny [GH-277]
|
|
* bignum.c: ditto
|
|
* complex.c: ditto
|
|
* dir.c: ditto
|
|
* encoding.c: ditto
|
|
* enumerator.c: ditto
|
|
* numeric.c: ditto
|
|
* proc.c: ditto
|
|
* re.c: ditto
|
|
* string.c: ditto
|
|
|
|
Thu Apr 11 23:41:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* common.mk: Add dependencies for include/ruby.h
|
|
|
|
* tool/update-deps: Use "make -p all miniruby ruby golf" to extract
|
|
dependencies in makefiles.
|
|
|
|
Thu Apr 11 23:21:17 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* tool/update-deps: Use "make -p all golf" to extract dependencies in
|
|
makefiles.
|
|
|
|
Thu Apr 11 21:02:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* common.mk: Dependency updated.
|
|
|
|
* tool/update-deps: Rewritten.
|
|
|
|
Thu Apr 11 19:59:48 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* common.mk: partially revert r40183, which breaks building on
|
|
other than source directory. (its commit log also says the same
|
|
thing, but such failure is not reproducible on my environment
|
|
and the commit breaks build on my environment)
|
|
|
|
Thu Apr 11 16:10:01 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/fiddle/closure.c (USE_FFI_CLOSURE_ALLOC): define 0 on
|
|
Mac OS X and Linux [Bug #3371]
|
|
|
|
Thu Apr 11 13:19:22 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/drb/drbtest.rb (Drb{Core,Ary}#teardown): retry Process.kill
|
|
if it fails with Errno::EPERM on Windows (workaround).
|
|
[ruby-dev:47245] [Bug #8251]
|
|
|
|
Thu Apr 11 11:11:38 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* dir.c: Fix a typo.
|
|
|
|
Thu Apr 11 10:39:34 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/fiddle/closure.c (USE_FFI_CLOSURE_ALLOC): add missing case:
|
|
RUBY_LIBFFI_MODVERSION is not defined (usually on Windows).
|
|
|
|
Thu Apr 11 09:27:04 2013 Konstantin Haase <me@rkh.im>
|
|
|
|
* dir.c (file_s_fnmatch): Document File::FNM_EXTGLOB flag.
|
|
|
|
Thu Apr 11 09:17:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* README: Fix typo by Benjamin Winkler [Fixes GH-281]
|
|
|
|
Thu Apr 11 06:15:51 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* regint.h: fix typo: _M_AMD86 -> _M_AMD64.
|
|
|
|
* siphash.c: ditto.
|
|
|
|
* st.c: ditto.
|
|
|
|
Thu Apr 11 06:09:57 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/fiddle/extconf.rb: define RUBY_LIBFFI_MODVERSION macro.
|
|
|
|
* ext/fiddle/closure.c (USE_FFI_CLOSURE_ALLOC): define 0 or 1
|
|
with platform and libffi's version. [Bug #3371]
|
|
|
|
Thu Apr 11 05:30:43 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (pkg_config): Add optional argument "option".
|
|
If it is given, it returns the result of
|
|
`pkg-config --<option> <pkgname>`.
|
|
|
|
Thu Apr 11 03:33:05 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/fiddle/closure.c (initialize): check mprotect's return value.
|
|
If mprotect is failed because of PaX or something, its function call
|
|
will cause SEGV.
|
|
http://c5664.rubyci.org/~chkbuild/ruby-trunk/log/20130401T210301Z.diff.html.gz
|
|
|
|
Wed Apr 10 17:39:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/bigdecimal/bigdecimal.c (VpCtoV): Initialize a local variable
|
|
even when overflow.
|
|
|
|
Wed Apr 10 12:32:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* bignum.c (rb_ll2big): Don't overflow on signed integer negation.
|
|
|
|
* ext/bigdecimal/bigdecimal.c (MUL_OVERFLOW_SIGNED_VALUE_P): New
|
|
macro.
|
|
(AddExponent): Don't overflow on signed integer multiplication.
|
|
(VpCtoV): Don't overflow on signed integer arithmetic.
|
|
(VpCtoV): Don't overflow on signed integer arithmetic.
|
|
|
|
Wed Apr 10 06:32:12 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (MUL_OVERFLOW_INT_P): New macro.
|
|
|
|
* sprintf.c (GETNUM): Don't overflow on signed integer multiplication.
|
|
|
|
Tue Apr 9 20:38:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (MUL_OVERFLOW_SIGNED_INTEGER_P): New macro.
|
|
(MUL_OVERFLOW_FIXNUM_P): Ditto.
|
|
(MUL_OVERFLOW_LONG_P): Ditto.
|
|
|
|
* array.c (rb_ary_product): Don't overflow on signed integer
|
|
multiplication.
|
|
|
|
* numeric.c (fix_mul): Ditto.
|
|
(int_pow): Ditto.
|
|
|
|
* rational.c (f_imul): Ditto.
|
|
|
|
* insns.def (opt_mult): Ditto.
|
|
|
|
* thread.c (sleep_timeval): Don't overflow on signed integer addition.
|
|
|
|
* bignum.c (rb_int2big): Don't overflow on signed integer negation.
|
|
(rb_big2ulong): Ditto.
|
|
(rb_big2long): Ditto.
|
|
(rb_big2ull): Ditto.
|
|
(rb_big2ll): Ditto.
|
|
|
|
Tue Apr 9 19:45:44 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/open-uri.rb: Support multiple fields with same field
|
|
name (like Set-Cookie).
|
|
(OpenURI::Meta#metas): New accessor to obtain fields as a Hash from
|
|
field name (string) to field values (array of strings).
|
|
[ruby-core:37734] [Bug #4964] reported by ren li.
|
|
|
|
Tue Apr 9 15:26:12 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* compile.c (iseq_compile_each): append keyword hash to argument array
|
|
to splat if needed. [ruby-core:54094] [Bug #8236]
|
|
|
|
Tue Apr 9 10:02:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (timestamp_file): gather timestamp files in one
|
|
directory from each extension directories, with considering
|
|
target_prefix.
|
|
|
|
Tue Apr 9 04:57:59 JST 2013 Charles Oliver Nutter <headius@headius.com>
|
|
|
|
* error.c: Capture EAGAIN, EWOULDBLOCK, EINPROGRESS exceptions and
|
|
export them for use in WaitReadable/Writable exceptions.
|
|
* io.c: Create versions of EAGAIN, EWOULDBLOCK, EINPROGRESS that
|
|
include WaitReadable and WaitWritable. Add rb_readwrite_sys_fail
|
|
for nonblocking failures using those exceptions. Use that
|
|
function in io_getpartial and io_write_nonblock instead of
|
|
rb_mod_sys_fail
|
|
* ext/openssl/ossl_ssl.c: Add new SSLError subclasses that include
|
|
WaitReadable and WaitWritable. Use those classes for
|
|
write_would_block and read_would_block instead of rb_mod_sys_fail.
|
|
* ext/socket/ancdata.c: Use rb_readwrite_sys_fail instead of
|
|
rb_mod_sys_fail in bsock_sendmsg_internal and
|
|
bsock_recvmsg_internal.
|
|
* ext/socket/init.c: Use rb_readwrite_sys_fail instead of
|
|
rb_mod_sys_fail in rsock_s_recvfrom_nonblock and
|
|
rsock_s_connect_nonblock.
|
|
* ext/socket/socket.c: Use rb_readwrite_sys_fail instead of
|
|
rb_mod_sys_fail in sock_connect_nonblock.
|
|
* include/ruby/ruby.h: Export rb_readwrite_sys_fail for use instead
|
|
of rb_mod_sys_fail. Introduce new constants RB_IO_WAIT_READABLE and
|
|
RB_IO_WAIT_WRITABLE for first arg to rb_readwrite_sys_fail.
|
|
|
|
Tue Apr 9 02:44:32 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/socket/extconf.rb: $defs needs -D or -U. nothing is added
|
|
otherwize.
|
|
|
|
* ext/socket/extconf.rb: check struct in_addr6, which is defined in
|
|
VC6 instead of in6_addr.
|
|
|
|
* ext/socket/option.c (optname_to_sym): fix macro name.
|
|
|
|
* ext/socket/constants.c (rsock_cmsg_type_arg): fix macro name.
|
|
|
|
Mon Apr 8 23:57:21 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* object.c (id_for_setter): extract common code from const, class
|
|
variable, instance variable setters.
|
|
|
|
Mon Apr 8 23:55:53 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/depend (ENCOBJS, TRANSOBJS): use explicit path to ruby.h for
|
|
nmake.
|
|
|
|
* ext/depend (ENCOBJS, TRANSOBJS): fix header dependency, VPATH has
|
|
$(srcdir)/include/ruby but not $(srcdir)/include, so cannot find out
|
|
ruby/ruby.h. use ruby.h instead and ../ruby for include/ruby.h.
|
|
|
|
Mon Apr 8 20:30:37 2013 Yuki Yugui Sonoda <yugui@google.com>
|
|
|
|
* ext/depend (ENCOBJS, TRANSOBJS): Add missing dependencies.
|
|
|
|
Mon Apr 8 17:19:28 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/win32ole/win32ole.c (fole_missing): should check actual argument
|
|
count before accessing.
|
|
|
|
Mon Apr 8 16:03:55 2013 Yuki Yugui Sonoda <yugui@google.com>
|
|
|
|
Fixes a build failure of ext/ripper/ripper.c on building out of place.
|
|
* common.mk (id.h, id.c): Always generated in $(srcdir).
|
|
(ext/ripper/ripper.c): Passes $(PATH_SEPARATOR) too to the sub make.
|
|
|
|
Mon Apr 8 12:05:02 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* object.c (rb_obj_ivar_set): call to_str for string only once.
|
|
to_str was called from rb_is_const_name and rb_to_id before.
|
|
|
|
* object.c (rb_mod_const_set): ditto.
|
|
|
|
* object.c (rb_mod_cvar_set): ditto.
|
|
|
|
Sun Apr 7 13:56:16 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* test/ruby/test_require.rb (TestRequire#test_require_nonascii_path):
|
|
RUBY_PLATFORM should escape as Regexp,
|
|
because RUBY_PLATFORM may contain '.'.
|
|
|
|
Sun Apr 7 10:44:01 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/defines.h: Simplify the logic to include sys/select.h.
|
|
This fixes a compilation error on Haiku (gcc2 and gcc4).
|
|
|
|
* configure.in: Use shared linker as $(CC) for Haiku.
|
|
This fixes a build error on Haiku (gcc2).
|
|
|
|
Sun Apr 7 10:41:30 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/resolv.rb (MDNSOneShot#sender): Delete an unused variable.
|
|
|
|
Sun Apr 7 03:24:36 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* addr2line.c: use more generic type:
|
|
* u_char -> unsigned char
|
|
* u_short -> unsigned short
|
|
* u_int -> unsigned int
|
|
* u_long -> unsigned long
|
|
* quad_t -> int64_t
|
|
* u_quad_t -> uint64_t
|
|
|
|
* addr2line.c (imax): inline is defined by configure.
|
|
|
|
Sun Apr 7 01:40:39 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el (ruby-electric-hash): New electric
|
|
function that expands a hash sign inside a string or regexp to
|
|
"#{}".
|
|
|
|
* misc/ruby-electric.el (ruby-electric-curlies): Do not insert
|
|
spaces inside when the curly brace is a delimiter of %r, %w,
|
|
etc.
|
|
|
|
* misc/ruby-electric.el (ruby-electric-curlies): Insert another
|
|
space before a closing curly brace when
|
|
ruby-electric-newline-before-closing-bracket is nil.
|
|
|
|
Sun Apr 7 01:01:26 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* strftime.c (rb_strftime_with_timespec): Test yday range.
|
|
[ruby-core:44088] [Bug #6247] reported by Ruby Submit.
|
|
|
|
Sat Apr 6 23:46:54 2013 Naohisa Goto <ngotogenome@gmail.com>
|
|
|
|
* configure.in (AC_CHECK_HEADERS): atomic.h for Solaris atomic_ops.
|
|
|
|
* ruby_atomic.h: Skip using Solaris10 atomic_ops on Solaris 9 or
|
|
earlier if atomic.h is not available. [ruby-dev:47229] [Bug #8228]
|
|
|
|
Sat Apr 6 23:40:40 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/resolv.rb: Support LOC resources.
|
|
[ruby-core:23361] [Feature #1436] by JB Smith.
|
|
|
|
Sat Apr 6 23:38:09 2013 Naohisa Goto <ngotogenome@gmail.com>
|
|
|
|
* addr2line.c: quad_t and u_quad_t is not available on Solaris.
|
|
__inline is not available with old compilers on Solaris.
|
|
[ruby-dev:47229] [Bug #8227]
|
|
|
|
Sat Apr 6 23:31:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/resolv.rb: Add one-shot multicast DNS support.
|
|
[ruby-core:53387] [Feature #8089] by Eric Hodel.
|
|
|
|
Sat Apr 6 22:12:01 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/resolv.rb (Resolv::DNS.fetch_resource): New method to obtain
|
|
full result.
|
|
[ruby-dev:43587] [Feature #4788] proposed by Makoto Kishimoto.
|
|
|
|
Sat Apr 6 20:17:51 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/socket.c (rsock_sys_fail_raddrinfo): Renamed from
|
|
rsock_sys_fail_addrinfo.
|
|
(rsock_sys_fail_raddrinfo_or_sockaddr): Renamed from
|
|
rsock_sys_fail_addrinfo_or_sockaddr.
|
|
|
|
* ext/socket/rubysocket.h: Follow the above change.
|
|
|
|
Sat Apr 6 19:24:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/socket.c (rsock_sys_fail_sockaddr): Takes struct sockaddr
|
|
and socklen_t instead of String object.
|
|
(rsock_sys_fail_addrinfo_or_sockaddr): Follow the above change.
|
|
|
|
* ext/socket/rubysocket.h (rsock_sys_fail_sockaddr): Follow the above
|
|
change.
|
|
|
|
Sat Apr 6 14:28:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/rubysocket.h (SockAddrStringValueWithAddrinfo): New macro.
|
|
(rsock_sockaddr_string_value_with_addrinfo): New declaration.
|
|
(rsock_addrinfo_inspect_sockaddr): Ditto.
|
|
(rsock_sys_fail_addrinfo): Ditto.
|
|
(rsock_sys_fail_sockaddr_or_addrinfo): Ditto.
|
|
|
|
* ext/socket/raddrinfo.c (rsock_addrinfo_inspect_sockaddr): Renamed
|
|
from addrinfo_inspect_sockaddr and exported.
|
|
(rsock_sockaddr_string_value_with_addrinfo): New function to obtain
|
|
string and possibly addrinfo object.
|
|
|
|
* ext/socket/socket.c (rsock_sys_fail_sockaddr): Don't use
|
|
rsock_sys_fail_host_port which is IP dependent. Invoke
|
|
rsock_sys_fail_addrinfo.
|
|
(rsock_sys_fail_addrinfo): New function using
|
|
rsock_addrinfo_inspect_sockaddr.
|
|
(rsock_sys_fail_addrinfo_or_sockaddr): New function.
|
|
(sock_connect): Use SockAddrStringValueWithAddrinfo and
|
|
rsock_sys_fail_addrinfo_or_sockaddr.
|
|
(sock_connect_nonblock): Ditto.
|
|
(sock_bind): Ditto.
|
|
|
|
Sat Apr 6 13:34:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/socket.c (rsock_sys_fail_sockaddr): Delete 2nd argument.
|
|
|
|
* ext/socket/rubysocket.h (rsock_sys_fail_sockaddr): Follow above
|
|
change.
|
|
|
|
Sat Apr 6 13:13:39 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/socket.c (rsock_sys_fail_path): Use rb_str_inspect only
|
|
for String to avoid SEGV.
|
|
|
|
Sat Apr 6 12:40:16 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/rubysocket.h (rsock_sys_fail_host_port): Wrap by NORETURN.
|
|
(rsock_sys_fail_path): Ditto.
|
|
(rsock_sys_fail_sockaddr): Ditto.
|
|
|
|
Sat Apr 6 11:49:35 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/socket.c (rsock_sys_fail_path): Use rb_str_inspect if the
|
|
path contains a NUL.
|
|
|
|
Sat Apr 6 11:39:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket: Improve socket exception message to show socket address.
|
|
[ruby-core:45617] [Feature #6583] proposed Eric Hodel.
|
|
|
|
* ext/socket/rubysocket.h (rsock_sys_fail_host_port): Declared.
|
|
(rsock_sys_fail_path): Ditto.
|
|
(rsock_sys_fail_sockaddr): Ditto.
|
|
|
|
* ext/socket/udpsocket.c (udp_connect): Use rsock_sys_fail_host_port.
|
|
(udp_bind): Ditto.
|
|
(udp_send): Ditto.
|
|
|
|
* ext/socket/init.c (rsock_init_sock): Specify a string for rb_sys_fail
|
|
argument.
|
|
(make_fd_nonblock): Ditto.
|
|
(rsock_s_accept): Ditto.
|
|
|
|
* ext/socket/ipsocket.c (init_inetsock_internal): Use
|
|
rsock_sys_fail_host_port.
|
|
|
|
* ext/socket/socket.c (rsock_sys_fail_host_port): Defined.
|
|
(rsock_sys_fail_path): Ditto.
|
|
(rsock_sys_fail_sockaddr): Ditto.
|
|
(setup_domain_and_type): Use rsock_sys_fail_sockaddr.
|
|
(sock_connect_nonblock): Ditto.
|
|
(sock_bind): Ditto.
|
|
(sock_gethostname): Specify a string for rb_sys_fail argument.
|
|
(socket_s_ip_address_list): Ditto.
|
|
|
|
* ext/socket/basicsocket.c (bsock_shutdown): Specify a string for
|
|
rb_sys_fail argument.
|
|
(bsock_setsockopt): Use rsock_sys_fail_path.
|
|
(bsock_getsockopt): Ditto.
|
|
(bsock_getpeereid): Refine the argument for rb_sys_fail.
|
|
|
|
* ext/socket/unixsocket.c (rsock_init_unixsock): Use
|
|
rsock_sys_fail_path.
|
|
(unix_path): Ditto.
|
|
(unix_send_io): Ditto.
|
|
(unix_recv_io): Ditto.
|
|
(unix_addr): Ditto.
|
|
(unix_peeraddr): Ditto.
|
|
|
|
Sat Apr 6 11:23:18 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* test/ruby/test_require.rb (TestRequire#test_require_nonascii_path):
|
|
fix load path for encoding to run the test as stand-alone.
|
|
|
|
Sat Apr 6 09:54:20 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* pack.c (NATINT_LEN): fix definition order, must be after
|
|
NATINT_PACK.
|
|
|
|
Sat Apr 6 03:11:07 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych/visitors/yaml_tree.rb: fix symbol keys in coder
|
|
emission. Thanks @tjwallace
|
|
* test/psych/test_coder.rb: test for change
|
|
|
|
Sat Apr 6 02:54:08 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych/exception.rb: there should be only one exception
|
|
base class. Fixes tenderlove/psych #125
|
|
* ext/psych/lib/psych.rb: require the correct exception class
|
|
* ext/psych/lib/psych/syntax_error.rb: ditto
|
|
* ext/psych/lib/psych/visitors/to_ruby.rb: ditto
|
|
|
|
Sat Apr 6 02:30:28 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (new_defined): remove all extra parentheses, and return
|
|
"nil" for defined? with empty expression.
|
|
[ruby-core:54024] [Bug #8224]
|
|
|
|
Sat Apr 6 02:06:04 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* ext/psych/lib/psych/visitors/to_ruby.rb: correctly register
|
|
self-referential strings. Fixes tenderlove/psych #135
|
|
|
|
* test/psych/test_string.rb: appropriate test.
|
|
|
|
Sat Apr 6 01:21:56 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/init.c (cloexec_accept): Fix a compile error on
|
|
Debian GNU/kFreeBSD. Consider HAVE_ACCEPT4 is defined
|
|
but SOCK_CLOEXEC is not defined.
|
|
|
|
Sat Apr 6 00:19:30 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* load.c (features_index_add): use rb_str_subseq() to specify C string
|
|
position properly to fix require non ascii path.
|
|
[ruby-core:53733] [Bug #8165]
|
|
|
|
* test/ruby/test_require.rb (TestRequire#test_require_nonascii_path):
|
|
a test for the above.
|
|
|
|
Fri Apr 5 20:41:49 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/defines.h (HAVE_TRUE_LONG_LONG): Defined to distinguish
|
|
availability of long long and availability of 64bit integer type.
|
|
|
|
* pack.c: Use HAVE_TRUE_LONG_LONG to distinguish q! and Q! support.
|
|
|
|
Fri Apr 5 20:19:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* addr2line.c: Include ruby/missing.h to fix compile error on Debian.
|
|
|
|
Fri Apr 5 19:39:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* compile.c (iseq_compile_each): fix of defined? with empty
|
|
expression. [ruby-core:53999] [Bug #8220]
|
|
|
|
Fri Apr 5 13:22:59 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/curses/curses.c (Init_curses): fix implementation function,
|
|
crmode should be same as cbreak. [ruby-core:54013] [Bug #8222]
|
|
|
|
Fri Apr 5 12:06:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/curses/hello.rb: Typo in Curses example by Drew Blas
|
|
[Fixes GH-273]
|
|
|
|
Thu Apr 4 23:45:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/resolv.rb (bind_random_port): Rescue EACCES for SunOS.
|
|
bind() on SunOS for port 2049 (nfs) and 4045 (lockd) causes
|
|
EACCES with unprivileged process. cf. PRIV_SYS_NFS in privileges(5)
|
|
[ruby-core:48064] [Bug #7183] reported by Frank Meier.
|
|
|
|
Thu Apr 4 23:24:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Remove condition for bcc.
|
|
|
|
Thu Apr 4 22:53:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/ruby.h (FIX2LONG): Parenthesize the macro body.
|
|
|
|
Thu Apr 4 22:32:32 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* time.c (time_strftime): Describe %L and %N truncates digits under
|
|
the specified length.
|
|
[ruby-core:52130] [Bug #7829]
|
|
|
|
Thu Apr 4 22:08:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* object.c (rb_mod_cvar_set): Reverted "avoid inadvertent
|
|
symbol creation" to avoid SEGV by
|
|
Class.new.class_variable_set(1, 2).
|
|
|
|
Thu Apr 4 20:07:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/pathname/pathname.c (path_write): New method.
|
|
(path_binwrite): Ditto.
|
|
[ruby-core:49468] [Feature #7378]
|
|
|
|
Thu Apr 4 16:51:29 2013 Yuki Yugui Sonoda <yugui@google.com>
|
|
|
|
* thread_pthread.c: Fixes wrong scopes of #if USE_SLEEPY_TIMER_THREAD
|
|
.. #endif sections. This fixes a build error on NativeClient.
|
|
|
|
Wed Apr 3 17:25:31 2013 Yuki Yugui Sonoda <yugui@google.com>
|
|
|
|
* thread_pthread.c (ruby_init_stack): Avoid using uninitialized value.
|
|
stackaddr and size are not set if get_stack() fails.
|
|
|
|
Thu Apr 4 16:55:08 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* struct.c (make_struct): avoid inadvertent symbol creation.
|
|
(rb_struct_aref): ditto.
|
|
(rb_struct_aset): ditto.
|
|
|
|
Thu Apr 4 16:54:40 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* object.c (rb_mod_const_set): avoid inadvertent symbol creation.
|
|
(rb_obj_ivar_set): ditto.
|
|
(rb_mod_cvar_set): ditto.
|
|
|
|
Thu Apr 4 15:46:48 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enum.c (enum_inject): avoid inadvertent symbol creation.
|
|
|
|
Thu Apr 4 14:37:07 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* thread.c (rb_thread_aref): avoid inadvertent symbol creation.
|
|
(rb_thread_variable_get): ditto.
|
|
(rb_thread_key_p): ditto.
|
|
(rb_thread_variable_p): ditto.
|
|
|
|
Thu Apr 4 11:33:57 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/openssl/ossl_bn.c (ossl_bn_to_i): Use bn2hex to speed up.
|
|
In general, binary to/from decimal needs extra cost.
|
|
|
|
Thu Apr 4 07:24:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Specify arguments to test functions.
|
|
|
|
Thu Apr 4 03:25:09 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/openssl/ossl_bn.c (ossl_bn_initialize): fix can't create from bn.
|
|
|
|
Wed Apr 3 22:09:25 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: Test functions and libraries after headers.
|
|
|
|
Wed Apr 3 21:23:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* io.c (rb_io_seek_m): Accept :CUR, :END, :SET as "whence" argument.
|
|
(interpret_seek_whence): New function.
|
|
[ruby-dev:45818] [Feature #6643]
|
|
|
|
Wed Apr 3 20:52:49 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* process.c: Describe the behavior which Ruby invokes a commandline
|
|
directly without shell if the commandline is simple enough.
|
|
[ruby-core:50459] [Bug #7489]
|
|
|
|
Wed Apr 3 20:27:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/extmk.rb (extmake): Invoke Logging::log_close in a ensure
|
|
clause.
|
|
|
|
Wed Apr 3 18:53:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/extmk.rb (extmake): Use Logging.open to switch stdout and
|
|
stderr. Delay Logging::log_close until the failure message is
|
|
written. Write the failure message only if log file is opened.
|
|
|
|
* lib/mkmf.rb (Logging.log_opened?): New method.
|
|
|
|
[ruby-dev:47215] [Bug #8209]
|
|
|
|
Wed Apr 3 17:11:15 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/win32.c (constat_apply): pass through unknown sequence which
|
|
starts with ESC but is not followed by a bracket. [ruby-core:53879]
|
|
[Bug #8201]
|
|
|
|
Wed Apr 3 16:35:32 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* bignum.c (rb_big_eq): hide intermediate Bignums not just freeing
|
|
memory. [ruby-core:53893] [Bug #8204]
|
|
|
|
* object.c (rb_obj_hide): hide an object by clearing klass.
|
|
|
|
* bignum.c (rb_big_eq): test as Fixnum if possible and get rid of zero
|
|
length Bignum. [ruby-core:53893] [Bug #8204]
|
|
|
|
Tue Apr 2 23:56:03 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/securerandom.rb (SecureRandom.random_bytes): Use
|
|
OpenSSL::Random.random_add instead of OpenSSL::Random.seed and
|
|
specify 0.0 as the entropy.
|
|
[ruby-core:47308] [Bug #6928]
|
|
|
|
Tue Apr 2 20:24:52 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* pack.c: Support Q! and q! for long long.
|
|
(natstr): Moved to toplevel. Add q and Q if there is long long type.
|
|
(endstr): Moved to toplevel.
|
|
(NATINT_PACK): Consider long long.
|
|
(NATINT_LEN_Q): New macro.
|
|
(pack_pack): Support Q! and q!.
|
|
(pack_unpack): Ditto.
|
|
[ruby-dev:43970] [Feature #3946]
|
|
|
|
Tue Apr 2 19:24:26 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/-test-/num2int/num2int.c: Define utility methods
|
|
as module methods of Num2int.
|
|
|
|
* test/-ext-/num2int/test_num2int.rb: Follow the above change.
|
|
|
|
Tue Apr 2 18:49:01 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/securerandom.rb: Don't use Array#to_s.
|
|
[ruby-core:52058] [Bug #7811] fixed by zzak (Zachary Scott).
|
|
|
|
Tue Apr 2 17:38:20 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* re.c (rb_reg_to_s): suppress duplicated charclass warning.
|
|
Regexp#to_s suppress extra its whole regexp options by calling
|
|
onig_new with its source, but it doesn't call rb_reg_preprocess.
|
|
Therefore its Unicode escapes (\u{XXXX}) are given as is,
|
|
and it may cause duplicated charclass warning for example
|
|
"[\u{33}]" (3 is duplicated) or "[\u{a}\u{b}]" (u is duplicated).
|
|
[ruby-core:53649] [Bug #8151]
|
|
|
|
Tue Apr 2 16:00:06 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* vm_dump.c (rb_print_backtrace): separate to ease showing C backtrace.
|
|
|
|
* internal.h (rb_print_backtrace): ditto.
|
|
|
|
Tue Apr 2 15:22:09 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* test/ruby/envutil.rb (assert_separately): stop_auto_run of
|
|
Test::Unit::Runner to prevent auto runner use ARGV.
|
|
|
|
* test/ruby/envutil.rb (assert_separately): add $: to separate process.
|
|
|
|
* test/ruby/envutil.rb (assert_separately): fail if stderr is not
|
|
empty and ignore_stderr is false.
|
|
|
|
Tue Apr 2 06:46:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/-test-/num2int/num2int.c: Rename utility methods
|
|
to global functions to ease manual experiments.
|
|
|
|
* test/-ext-/num2int/test_num2int.rb: Follow the above change.
|
|
|
|
Mon Apr 1 22:26:17 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/zlib/zlib.c (rb_gzfile_set_mtime): Use NUM2UINT.
|
|
The old logic doesn't work well on LP64 platforms as:
|
|
.. -2**63-1 => error,
|
|
-2**63 .. -2**62-1 => success,
|
|
-2**62 .. -2**31-1 => error,
|
|
-2**31 .. 2**31-1 => success,
|
|
2**31 .. 2**62-1 => error,
|
|
2**62 .. 2**64-1 => success,
|
|
2**64 .. => error.
|
|
|
|
Mon Apr 1 22:08:02 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* ext/zlib/zlib.c (Zlib::Inflate.new):
|
|
Fix documentation syntax and naming errors.
|
|
Based on patch by Robin Dupret. Fix GH-271.
|
|
|
|
Mon Apr 1 21:22:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/-ext-/num2int/test_num2int.rb: Test small bignums.
|
|
|
|
Mon Apr 1 21:10:56 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* numeric.c (rb_num2ulong_internal): Don't cast a negative double value
|
|
into unsigned long, which is undefined behavior.
|
|
(rb_num2ull): Don't cast a value bigger than LLONG_MAX into
|
|
long long, which is undefined behavior.
|
|
|
|
Mon Apr 1 20:57:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/-test-/num2int/num2int.c: Return string for result, instead of
|
|
printing.
|
|
|
|
* test/-ext-/num2int/test_num2int.rb: updated to follow above change.
|
|
|
|
Mon Apr 1 20:08:07 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* numeric.c (rb_num2long): Don't use SIGNED_VALUE uselessly.
|
|
(check_int): Ditto.
|
|
(check_short): Ditto.
|
|
(rb_num2fix): Ditto.
|
|
(rb_num2ulong_internal): Add a cast.
|
|
|
|
Mon Apr 1 18:41:35 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in: skip autoconf 2.64 and 2.66, 2.67 seems short-lived
|
|
but stick on it for Debian Squeeze.
|
|
|
|
Mon Apr 1 14:22:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in: check clang version by predefined macro values.
|
|
[Bug #8192]
|
|
|
|
Mon Apr 1 12:05:15 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* numeric.c (check_uint): Take the 1st argument as unsigned long,
|
|
instead of VALUE. Refine the validity test conditions.
|
|
(check_ushort): Ditto.
|
|
|
|
Mon Apr 1 07:15:03 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* configure.in: use quadrigraph to put '[' or ']'. [Bug #8192]
|
|
|
|
Mon Apr 1 04:16:41 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* configure.in: kick old clang. [ruby-dev:47204] [Bug #8192]
|
|
|
|
Mon Apr 1 01:12:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/ruby.h (FIX2ULONG): Make it consistent with NUM2ULONG.
|
|
|
|
* ext/-test-/num2int/num2int.c: Add utility methods for FIX2XXX tests.
|
|
|
|
* test/-ext-/num2int/test_num2int.rb: Add tests for FIX2XXX.
|
|
|
|
Sun Mar 31 17:17:56 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* proc.c (rb_mod_define_method): consider visibility in define_method.
|
|
patch by mashiro <mail AT mashiro.org>. fix GH-268.
|
|
|
|
Sun Mar 31 15:40:30 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/configure.bat: try to fix option arguments split by commas and
|
|
equals here. this batch file no longer run with old command.com.
|
|
|
|
* tool/mkconfig.rb: no hacks for cmd.exe.
|
|
|
|
Sun Mar 31 13:47:04 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* numeric.c (rb_num2ulong_internal): New function similar to
|
|
rb_num2ulong but integer wrap around flag is also returned.
|
|
(rb_num2ulong): Use rb_num2ulong_internal.
|
|
(rb_num2uint): Use rb_num2ulong_internal and the wrap around flag is
|
|
used instead of negative_int_p(val).
|
|
(rb_num2ushort): ditto.
|
|
|
|
Sun Mar 31 06:27:17 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* class.c (HAVE_METACLASS_P): should check FL_SINGLETON flag before get
|
|
instance variable to get rid of wrong warning about __attached__.
|
|
[ruby-core:53839] [Bug #8188]
|
|
|
|
Sat Mar 30 14:11:28 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* bcc32: removed. agreed at
|
|
http://bugs.ruby-lang.org/projects/ruby/wiki/DevelopersMeeting20130223Japan
|
|
|
|
Sat Mar 30 03:58:00 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/file.c (code_page): use cp1252 instead of cp20127 as US-ASCII.
|
|
fix [ruby-core:53079] [Bug #7996]
|
|
reported and patched by mmeltner (Michael Meltner).
|
|
|
|
Sat Mar 30 03:49:21 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* win32/win32.c (wrename): use MoveFileExW instead of MoveFileW,
|
|
because the latter fails on cross device file move of some
|
|
environments.
|
|
fix [ruby-core:53492] [Bug #8109]
|
|
reported by mitchellh (Mitchell Hashimoto).
|
|
|
|
Fri Mar 29 22:09:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* thread.c (rb_mutex_synchronize_m): yield no block params. patch by
|
|
splattael (Peter Suschlik) in [ruby-core:53773] [Bug #8097].
|
|
fix GH-266.
|
|
|
|
Fri Mar 29 16:51:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* io.c (argf_next_argv): set init flag if succeeded to forward, after
|
|
skipping.
|
|
|
|
* io.c (argf_block_call_i, argf_block_call): no more forwarding if
|
|
forwarded after skipping. [ruby-list:49185]
|
|
|
|
* io.c (argf_close): deal with init flag.
|
|
|
|
* io.c (argf_block_call_i, argf_block_call): forward next file if
|
|
skipped while iteration, to get rid of IOError. [ruby-list:49185]
|
|
|
|
Fri Mar 29 11:09:48 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (configuration): not include all CFLAGS in CXXFLAGS, to
|
|
use different set than C for C++. [ruby-core:45273] [Bug #6504]
|
|
|
|
Fri Mar 29 10:24:10 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/io.h: undef POSIX compliant names on AIX, which are no
|
|
longer needed. patch suggested by edelsohn (David Edelsohn) in
|
|
[ruby-core:53815]. [Bug #8174]
|
|
|
|
Fri Mar 29 06:39:42 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* numeric.c (rb_num2ull): Cast double to unsigned LONG_LONG via
|
|
LONG_LONG instead of double to unsigned LONG_LONG directly.
|
|
This is a challenge to fix a test_num2ull(TestNum2int)
|
|
failure (NUM2ULL(-1.0) should be "18446744073709551615" but was "0")
|
|
on Mac OS X with 32bit clang.
|
|
http://a.mrkn.jp/~mrkn/chkbuild/mountain_lion/ruby-trunk-m32-o0/log/20130328T191100Z.diff.html.gz
|
|
|
|
Fri Mar 29 00:54:54 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (MAIN_DOES_NOTHING): ensure symbols for tests to be
|
|
preserved. [ruby-core:53745] [Bug #8169]
|
|
|
|
Thu Mar 28 23:11:25 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/resolv.rb: Test Windows platform by detecting LoadError when
|
|
require 'win32/resolv' suggested by Nobuyoshi Nakada [ruby-core:53389].
|
|
[ruby-core:53388] [Feature #8090] Reported by Charles Nutter.
|
|
|
|
Thu Mar 28 23:10:10 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/io.h: rename SVR3,4 member names as POSIX compliant,
|
|
to get rid of conflict on AIX. [ruby-core:53765] [Bug #8174]
|
|
|
|
Thu Mar 28 18:22:21 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/-ext-/num2int/test_num2int.rb: extract
|
|
assert_num2i_success_internal and assert_num2i_error_internal and
|
|
provide assertion messages as "NUM2XXX(NNN)".
|
|
|
|
Thu Mar 28 07:05:25 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/intern.h: Delete redundant inclusions caused by
|
|
AC_INCLUDES_DEFAULT in defines.h.
|
|
|
|
* include/ruby/defines.h: Ditto.
|
|
|
|
* include/ruby/ruby.h: Ditto.
|
|
|
|
* include/ruby/st.h: Ditto.
|
|
|
|
Thu Mar 28 06:51:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* include/ruby/defines.h: Fix a compilation error on NetBSD,
|
|
"type of formal parameter 1 is incomplete" for the rb_thread_wait_for
|
|
invocation in rb_file_flock, by including header files as
|
|
AC_INCLUDES_DEFAULT of autoconf.
|
|
|
|
Wed Mar 27 22:09:14 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* numeric.c (LONG_MIN_MINUS_ONE_IS_LESS_THAN): New macro.
|
|
(LLONG_MIN_MINUS_ONE_IS_LESS_THAN): Ditto.
|
|
(rb_num2long): Use LONG_MIN_MINUS_ONE_IS_LESS_THAN.
|
|
(rb_num2ulong): Ditto.
|
|
(rb_num2ll): Use LLONG_MIN_MINUS_ONE_IS_LESS_THAN.
|
|
(rb_num2ull): Ditto.
|
|
|
|
* test/-ext-/num2int/test_num2int.rb (assert_num2i_success): Test the
|
|
value converted into a Float if Float can represent the value
|
|
exactly.
|
|
(assert_num2i_error): Ditto.
|
|
|
|
Wed Mar 27 20:59:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/-ext-/num2int/test_num2int.rb (assert_num2i_success): New
|
|
utility method.
|
|
(assert_num2i_error): Ditto.
|
|
|
|
Wed Mar 27 20:37:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* time.c (num_exact): Use to_r method only if to_int method is
|
|
available.
|
|
[ruby-core:53764] [Bug #8173] Reported by Hiro Asari.
|
|
|
|
Wed Mar 27 12:07:40 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/-ext-/num2int/test_num2int.rb (test_num2ll): test LLONG_MIN,
|
|
not LONG_MIN.
|
|
|
|
Wed Mar 27 12:02:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (TIMET_MAX_PLUS_ONE): definition simplified.
|
|
|
|
Wed Mar 27 06:39:41 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (MAIN_DOES_NOTHING): force to refer symbols for tests
|
|
to be preserved. [ruby-core:53745] [Bug #8169]
|
|
|
|
Wed Mar 27 05:15:37 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (RUBY_REPLACE_TYPE): define SIGNEDNESS_OF_type same as
|
|
check_signedness of mkmf.rb.
|
|
|
|
* internal.h (TIMET_MAX, TIMET_MIN, TIMET_MAX_PLUS_ONE): use
|
|
SIGNEDNESS_OF_TIME_T.
|
|
|
|
Wed Mar 27 00:28:45 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h (TIMET_MAX_PLUS_ONE): Defined.
|
|
|
|
* thread.c (double2timeval): Saturate out-of-range values.
|
|
|
|
Tue Mar 26 23:41:18 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* internal.h: Define TIMET_MAX and TIMET_MIN here.
|
|
|
|
* time.c: Remove TIMET_MAX and TIMET_MIN definitions.
|
|
|
|
* thread.c: Ditto.
|
|
|
|
* thread_pthread.c: Remove TIMET_MAX definition.
|
|
|
|
* thread_win32.c: Ditto.
|
|
|
|
Tue Mar 26 22:31:10 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/socket.c (sockaddr_len): return the shortest length for
|
|
unknown socket address.
|
|
|
|
Tue Mar 26 22:14:46 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* thread.c (double2timeval): convert the infinity to TIME_MAX to avoid
|
|
SEGV by Thread.new {}.join(Float::INFINITY) on
|
|
Debian GNU/Linux (amd64).
|
|
|
|
Mon Mar 25 07:09:20 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rinda/tuplespace.rb: Only return tuple entry once on move,
|
|
either through port or regular return, not both. This results in a
|
|
120% speedup when combined with #8125. Patch by Joel VanderWerf.
|
|
[ruby-trunk - Feature #8119]
|
|
|
|
Mon Mar 25 06:59:01 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* test/rinda/test_rinda.rb: Skip IPv6 tests if no IPv6 addresses
|
|
exist. Skip fork-dependent test if fork is not available.
|
|
[ruby-trunk - Bug #8159]
|
|
|
|
Sun Mar 24 10:38:24 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* addr2line.c (putce): suppress unused return value warning.
|
|
|
|
Mon Mar 25 02:01:03 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* proc.c (bm_free): need to clean up the mark flag of a free and
|
|
unlinked method entry. [Bug #8100] [ruby-core:53439]
|
|
|
|
Sun Mar 24 22:13:51 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* string.c (rb_str_rpartition): revert r39903, and convert byte offset
|
|
to char offset; the return value of rb_reg_search is byte offset,
|
|
but other than it of rb_str_rpartition expects char offset.
|
|
[Bug #8138] [ruby-dev:47183]
|
|
|
|
Sun Mar 24 18:29:46 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* string.c (rb_str_rpartition): Fix String#rpartition(/re/)
|
|
against a multibyte string. [Bug #8138] [ruby-dev:47183]
|
|
|
|
Sun Mar 24 13:42:24 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* gc.c (GC_ENABLE_LAZY_SWEEP): new macro to switch lazy sweeping
|
|
for debugging. [Feature #8024] [ruby-dev:47135]
|
|
|
|
Sun Mar 24 12:55:47 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* gc.c: We have no chance to expand the heap when lazy sweeping is
|
|
restricted. So collecting is often invoked if there is not
|
|
enough free space in the heap. Try to expand heap when this is
|
|
the case.
|
|
|
|
Sun Mar 24 11:03:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/ruby/test_require.rb: Remove temporally files in the tests.
|
|
|
|
* test/ruby/test_rubyoptions.rb: Ditto.
|
|
|
|
* test/logger/test_logger.rb: Ditto.
|
|
|
|
* test/psych/test_psych.rb: Ditto.
|
|
|
|
* test/readline/test_readline.rb: Ditto.
|
|
|
|
* test/syslog/test_syslog_logger.rb: Ditto.
|
|
|
|
* test/webrick/test_httpauth.rb: Ditto.
|
|
|
|
* test/zlib/test_zlib.rb: Ditto.
|
|
|
|
Sun Mar 24 05:36:29 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rinda/ring.rb: Added documentation for multicast support.
|
|
|
|
* NEWS: Point to above documentation.
|
|
|
|
Sun Mar 24 05:32:39 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* test/rinda/test_rinda.rb: Restore tests commented out while fixing
|
|
test slowdown bug before r39895.
|
|
|
|
Sun Mar 24 05:03:36 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rinda/ring.rb: Add multicast support to Rinda::RingFinger and
|
|
Rinda::RingServer. [ruby-trunk - Bug #8073]
|
|
* test/rinda/test_rinda.rb: Test for the above.
|
|
|
|
* NEWS: Update with Rinda multicast support
|
|
|
|
Sun Mar 24 04:13:27 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* test/rinda/test_rinda.rb: Fixed test failures in r39890 and r39891
|
|
due to stopping DRb service.
|
|
|
|
Sun Mar 24 03:34:02 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rinda/rinda.rb: Fixed loss of tuple when remote is alive but the
|
|
call stack was unwound. Patch by Joel VanderWerf.
|
|
[ruby-trunk - Bug #8125]
|
|
* test/rinda/test_rinda.rb: Test for the above.
|
|
|
|
Sun Mar 24 02:14:53 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/mkmf/test_have_macro.rb: remove temporally files in the tests.
|
|
|
|
Sat Mar 23 23:50:04 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* addr2line.c (kprintf): added from FreeBSD libstand's printf.
|
|
this is consided as async signal safe function.
|
|
|
|
* addr2line.c (rb_dump_backtrace_with_lines): use kfprintf.
|
|
[Bug #8144] [ruby-core:53632]
|
|
|
|
Sat Mar 23 23:28:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* ext/bigdecimal/bigdecimal.c (BigDecimal_divide): Use Qnil and NIL_P
|
|
instead of (VALUE)0 as a return value.
|
|
|
|
* ext/bigdecimal/bigdecimal.c (BigDecimal_div): ditto.
|
|
|
|
* ext/bigdecimal/bigdecimal.c (BigDecimal_divremain): ditto.
|
|
|
|
* ext/bigdecimal/bigdecimal.c (BigDecimal_remainder): ditto.
|
|
|
|
Sat Mar 23 17:39:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_eval.c (check_funcall_respond_to): preserve passed_block, which
|
|
is modified in vm_call0_body() via vm_call0(), and caused a bug of
|
|
rb_check_funcall() by false negative result of rb_block_given_p().
|
|
re-fix [ruby-core:53650] [Bug #8153].
|
|
[ruby-core:53653] [Bug #8154]
|
|
|
|
Fri Mar 22 17:48:34 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/forwardable.rb (Forwardable::FILE_REGEXP): create regexp object
|
|
outside sources for eval, to reduce allocations in def_delegators
|
|
wrappers. //o option does not make each regexps shared. patch by
|
|
tmm1 (Aman Gupta) in [ruby-core:53620] [Bug #8143].
|
|
|
|
Fri Mar 22 17:38:42 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* load.c (rb_feature_p), vm_core.h (rb_vm_struct): turn
|
|
loaded_features_index into st_table. patches by tmm1 (Aman Gupta)
|
|
in [ruby-core:53251] and [ruby-core:53274] [Bug #8048]
|
|
|
|
Fri Mar 22 10:29:00 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/bigdecimal/bigdecimal.c: Fix style.
|
|
|
|
Fri Mar 22 05:30:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (ambiguous_operator): refine warning message, since this
|
|
warning is shown after literal too.
|
|
|
|
Fri Mar 22 04:51:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_insnhelper.c (vm_callee_setup_keyword_arg): should check required
|
|
keyword arguments even if rest hash is defined. [ruby-core:53608]
|
|
[Bug #8139]
|
|
|
|
Fri Mar 22 01:00:17 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* process.c (rb_execarg_addopt, run_exec_pgroup): use rb_pid_t
|
|
instead of pid_t.
|
|
|
|
* ext/pty/pty.c (raise_from_check, pty_check): ditto.
|
|
|
|
Fri Mar 22 00:04:15 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* addr2line.c (rb_dump_backtrace_with_lines): output line at once.
|
|
|
|
Thu Mar 21 23:17:08 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* thread.c (ruby_kill): get rid of deadlock on signal 0.
|
|
[ruby-dev:47182] [Bug #8137]
|
|
|
|
Thu Mar 21 22:39:46 2013 Naohisa Goto <ngotogenome@gmail.com>
|
|
|
|
* marshal.c (marshal_dump, marshal_load): workaround for segv on
|
|
Intel Solaris compiled with Oracle SolarisStudio 12.3.
|
|
Partly revert r38174. [ruby-core:52042] [Bug #7805]
|
|
|
|
Thu Mar 21 16:48:06 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (simple_re_meta): escape all closing characters, not only
|
|
round parenthesis. [ruby-core:53578] [Bug #8133]
|
|
|
|
Thu Mar 21 13:50:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_core.h (UNINITIALIZED_VAR): suppress warnings by clang 4.2.
|
|
[ruby-core:51742] [Bug #7756]
|
|
|
|
Thu Mar 21 07:34:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/date/date_core.c: Typo in Date::MONTHNAMES by Matt Gauger
|
|
[GH fixes #261]
|
|
|
|
Wed Mar 20 22:53:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (find_library): fix to format message.
|
|
[ruby-core:53568] [Bug #8130]
|
|
|
|
Wed Mar 20 22:52:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (install_dirs, with_destdir): prefix with DESTDIR
|
|
directories to install only unless bundled extension libraries.
|
|
[ruby-core:53502] [Bug #8115]
|
|
|
|
Wed Mar 20 17:47:53 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/win32ole/test_err_in_callback.rb (TestErrInCallBack#setup):
|
|
allow using different root for source and build directories.
|
|
this may fixes a minor problem of r39834.
|
|
|
|
Wed Mar 20 16:40:48 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* test/ruby/test_signal.rb (test_hup_me): skip if HUP isn't supported.
|
|
On Windows this test causes ArgumentError.
|
|
|
|
Wed Mar 20 16:24:12 2013 Hiroshi Shirosaki <h.shirosaki@gmail.com>
|
|
|
|
* test/rubygems/test_gem_installer.rb (test_install_extension_flat):
|
|
use ruby in build directory in case ruby is not installed.
|
|
[ruby-core:53265] [Bug #8058]
|
|
|
|
Wed Mar 20 15:22:07 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/win32ole/test_err_in_callback.rb (TestErrInCallBack#setup): use
|
|
relative path to get rid of "too long commandline" error.
|
|
|
|
Wed Mar 20 04:27:42 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* test/rinda/test_rinda.rb: remove unused variables.
|
|
patched by Vipul A M <vipulnsward@gmail.com>
|
|
|
|
Wed Mar 20 04:15:32 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* ext/bigdecimal/bigdecimal.c: fixed typo.
|
|
patched by Vipul A M <vipulnsward@gmail.com>
|
|
|
|
Sat Mar 16 03:40:49 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* test/ruby/test_signal.rb (test_hup_me): added a few comments.
|
|
|
|
Sat Mar 16 03:39:38 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread.c (ruby_kill): added a few comments.
|
|
|
|
Sat Mar 16 03:36:56 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread.c (ruby_kill): release GVL while waiting signal delivered.
|
|
|
|
Tue Mar 19 19:50:48 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ruby_kill (internal.h, thread.c): use rb_pid_t instead of pid_t.
|
|
this fixes the build failure of mswin introduced at r39819.
|
|
|
|
Tue Mar 19 17:09:30 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* string.c (rb_str_conv_enc_opts): convert with one converter, instead
|
|
of re-creating converters for each buffer expansion.
|
|
|
|
Tue Mar 19 17:06:50 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* dir.c (glob_helper): compose HFS file names from UTF8-MAC.
|
|
[ruby-core:48745] [Bug #7267]
|
|
|
|
Sat Mar 16 01:44:29 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* internal.h: added a declaration of ruby_kill().
|
|
* thread.c (ruby_kill): helper function of kill().
|
|
|
|
* signal.c (rb_f_kill): use ruby_kill() instead of kill().
|
|
* signal.c (rb_f_kill): call rb_thread_execute_interrupts()
|
|
to ensure that make SignalException if sent a signal
|
|
to myself. [Bug #7951] [ruby-core:52864]
|
|
|
|
* vm_core.h (typedef struct rb_thread_struct): added
|
|
th->interrupt_cond.
|
|
* thread.c (rb_threadptr_interrupt_common): added to
|
|
initialization of th->interrupt_cond.
|
|
* thread.c (thread_create_core): ditto.
|
|
|
|
* test/ruby/test_signal.rb (TestSignal#test_hup_me): test for
|
|
the above.
|
|
|
|
Sat Mar 16 00:42:39 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* io.c (linux_iocparm_len): enable only exist _IOC_SIZE().
|
|
Because musl libc doesn't have it. [Bug #8051] [ruby-core:53229]
|
|
|
|
Tue Mar 19 10:05:04 2013 Shota Fukumori <her@sorah.jp>
|
|
|
|
* ext/objspace/objspace.c: Fix typo in doc. Patch by Sho Hashimoto.
|
|
[Bug #8116] [ruby-dev:47177]
|
|
|
|
Tue Mar 19 02:13:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* configure.in: set ac_cv_prog_cxx if CXX is supplied.
|
|
|
|
Tue Mar 19 01:18:00 2013 Kenta Murata <mrkn@mrkn.jp>
|
|
|
|
* configure.in: Fix c++ compiler auto-selection not only for
|
|
Darwin 11.x, but also the other versions of Darwin.
|
|
|
|
Tue Mar 19 00:26:22 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* gc.c: Improve accuracy of objspace_live_num() and
|
|
allocated/freed counters. patched by tmm1(Aman Gupta).
|
|
[Bug #8092] [ruby-core:53392]
|
|
|
|
Mon Mar 18 21:42:48 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* gc.c: Avoid unnecessary heap growth. patched by tmm1(Aman Gupta).
|
|
[Bug #8093] [ruby-core:53393]
|
|
|
|
Mon Mar 18 17:58:36 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* gc.c: Fix unlimited memory growth with large values of
|
|
RUBY_FREE_MIN. patched by tmm1(Aman Gupta).
|
|
[Bug #8095] [ruby-core:53405]
|
|
|
|
Mon Mar 18 14:46:19 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/win32ole/test_err_in_callback.rb
|
|
(TestErrInCallBack#test_err_in_callback): shouldn't create a file in
|
|
the top of build directory.
|
|
|
|
Mon Mar 18 13:29:52 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* vm_dump.c (backtrace): on darwin use custom backtrace() to trace
|
|
beyond _sigtramp. darwin's backtrace can't trace beyond signal
|
|
trampoline with sigaltstack.
|
|
|
|
* configure.in: check execinfo.h on darwin.
|
|
|
|
Mon Mar 18 11:03:23 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* vm_exec.h (END_INSN): revert r39517 because the segv seems fixed by
|
|
r39806.
|
|
|
|
Mon Mar 18 10:41:06 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* vm_exec.c: Correct predefined macro name. This typo is introduced by
|
|
r36534 and should be backported to ruby_2_0_0.
|
|
|
|
Mon Mar 18 03:18:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* array.c: Typo in Array#delete by Timo Sand [GH fixes #258]
|
|
|
|
Mon Mar 18 01:14:56 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* io.c (io_fillbuf): show fd number on failure to debug.
|
|
http://c5632.rubyci.org/~chkbuild/ruby-trunk/log/20130316T050302Z.diff.html.gz
|
|
|
|
Sun Mar 17 02:38:21 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* ext/date/date_core.c: include sys/time.h for avoiding implicit
|
|
declaration of gettimeofday().
|
|
|
|
Sun Mar 17 00:55:31 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* include/ruby/missing.h: removed __linux__. it's unnecessary.
|
|
|
|
Fri Mar 15 14:57:16 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread.c: disabled _FORTIFY_SOURCE for avoid to hit glibc bug.
|
|
[Bug #8080] [ruby-core:53349]
|
|
* test/ruby/test_io.rb (TestIO#test_io_select_with_many_files):
|
|
test for the above.
|
|
|
|
Wed Mar 13 15:16:35 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* include/ruby/missing.h (__syscall): moved to...
|
|
* io.c: here. because __syscall() is only used from io.c.
|
|
|
|
* include/ruby/missing.h: move "#include <sys/type.h>" to ....
|
|
* include/ruby/intern.h: here. because it was introduced for
|
|
fixing NFDBITS issue. [ruby-core:05179].
|
|
|
|
Wed Mar 13 14:38:53 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* include/ruby/missing.h (struct timespec): include <sys/time.h>
|
|
|
|
Wed Mar 13 13:54:45 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in: check struct timeval exist or not.
|
|
* include/ruby/missing.h (struct timeval): check HAVE_STRUCT_TIMEVAL
|
|
properly. and don't include sys/time.h if struct timeval exist.
|
|
|
|
* file.c: include sys/time.h explicitly.
|
|
* random.c: ditto.
|
|
* thread_pthread.c: ditto.
|
|
* time.c: ditto.
|
|
* ext/date/date_strftime.c: ditto.
|
|
|
|
Fri Mar 15 14:45:02 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* configure.in (_FORTIFY_SOURCE): added a few comments.
|
|
|
|
Fri Mar 15 14:17:55 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread_pthread.c (numberof): renamed from ARRAY_SIZE() because
|
|
other all files use numberof().
|
|
|
|
Say Mar 15 01:33:00 2013 Charles Oliver Nutter <headius@headius.com>
|
|
|
|
* test/ruby/test_lazy_enumerator.rb (TestLazyEnumerator#test_drop_while):
|
|
Modify while condition to show dropping remains off after first false
|
|
value. This change was made in 39711.
|
|
|
|
Fri Mar 15 23:06:18 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* time.c (GetTimeval): check if already initialized instance.
|
|
|
|
* time.c (GetNewTimeval): check if newly created instance.
|
|
|
|
* time.c (time_init_0, time_init_1, time_init_copy, time_mload): must
|
|
be newly created instance. [ruby-core:53436] [Bug #8099]
|
|
|
|
Fri Mar 15 14:51:33 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* file.c (rb_sys_fail_path_with_func): share same function, and path
|
|
may be nil.
|
|
|
|
Fri Mar 15 08:24:51 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* io.c (rb_sys_fail_path): define & use rb_sys_fail_path0 like r39752
|
|
|
|
Fri Mar 15 04:08:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* proc.c: Typo in Proc.arity found by Jack Nagel [Bug #8094]
|
|
|
|
Thu Mar 14 16:59:09 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (rb_cv_function_name_string): macro for function name
|
|
string predefined identifier, __func__ in C99, or __FUNCTION__ in
|
|
gcc.
|
|
|
|
* file.c (rb_sys_fail_path): use RUBY_FUNCTION_NAME_STRING.
|
|
|
|
Thu Mar 14 14:12:34 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* file.c (rb_sys_fail_path): use rb_sys_fail_path0 only on GCC.
|
|
__func__ is C99 feature.
|
|
|
|
Thu Mar 14 12:59:59 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* file.c (rb_sys_fail_path0): add to append the name of called function
|
|
to ease debugging for example blow umask_spec failure.
|
|
http://fbsd.rubyci.org/~chkbuild/ruby-trunk/log/20130309T010202Z.diff.html.gz
|
|
|
|
* file.c (rb_sys_fail_path): use rb_sys_fail_path0.
|
|
|
|
Thu Mar 14 12:53:15 2013 Luis Lavena <luislavena@gmail.com>
|
|
|
|
* win32/file.c (get_user_from_path): add internal function that retrieves
|
|
username from supplied path (refactored).
|
|
* win32/file.c (rb_file_expand_path_internal): refactor expansion of user
|
|
home to use get_user_from_path and cover dir_string corner cases.
|
|
[ruby-core:53168] [Bug #8034]
|
|
|
|
Thu Mar 14 11:53:01 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* NEWS: describe RUBY_HEAP_SLOTS_GROWTH_FACTOR.
|
|
|
|
Thu Mar 14 10:01:12 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* doc/globals.rdoc: $? is thread-local
|
|
|
|
Wed Mar 13 23:25:59 2013 Narihiro Nakamura <authornari@gmail.com>
|
|
|
|
* gc.c: allow to tune growth of heap by environment variable
|
|
RUBY_HEAP_SLOTS_GROWTH_FACTOR. patched by tmm1(Aman Gupta).
|
|
[Feature #8015] [ruby-core:53131]
|
|
|
|
Wed Mar 13 19:43:46 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* doc/irb/irb.rd.ja: fix typo
|
|
|
|
* ext/tk/MANUAL_tcltklib.eng: fix typos
|
|
|
|
* ext/tk/sample/tktextframe.rb (Tk#component_delegates): fix typo
|
|
|
|
Wed Mar 13 15:13:04 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* class.c (rb_obj_singleton_methods): collect methods from the origin
|
|
class. [ruby-core:53207] [Bug #8044]
|
|
|
|
Wed Mar 13 14:51:26 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_method.c (rb_export_method): directly override the flag of method
|
|
defined in prepending class too, not adding zsuper entry.
|
|
[ruby-core:53106] [Bug #8005]
|
|
|
|
Wed Mar 13 13:06:26 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (rm, shvar_to_cpp, unexpand_shvar): local is not
|
|
available on old shells.
|
|
|
|
* configure.in (shvar_to_cpp): escape quotes for old shells.
|
|
[Bug #7959] [Bug #8071]
|
|
|
|
Wed Mar 13 11:11:07 2013 Shugo Maeda <shugo@ruby-lang.org>
|
|
|
|
* object.c (Init_Object): remove Module#used, which has been
|
|
introduced in Ruby 2.0 by mistake. [Bug #7916] [ruby-core:52719]
|
|
|
|
Wed Mar 13 05:49:29 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/irb.rb: Fix typo
|
|
|
|
Tue Mar 12 22:20:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* compile.c (iseq_set_arguments, iseq_compile_each): support required
|
|
keyword arguments. [ruby-core:51454] [Feature #7701]
|
|
|
|
* iseq.c (rb_iseq_parameters): ditto.
|
|
|
|
* parse.y (f_kw, f_block_kw): ditto. this syntax is still
|
|
experimental, the notation may change.
|
|
|
|
* vm_core.h (rb_iseq_struct): ditto.
|
|
|
|
* vm_insnhelper.c (vm_callee_setup_keyword_arg): ditto.
|
|
|
|
Tue Mar 12 17:02:53 2013 TAKANO Mitsuhiro <tak@no32.tk>
|
|
|
|
* date_core.c: clearly specify operator precedence.
|
|
|
|
Tue Mar 12 17:00:45 2013 TAKANO Mitsuhiro <tak@no32.tk>
|
|
|
|
* insns.def: fix condition.
|
|
|
|
Tue Mar 12 16:48:19 2013 TAKANO Mitsuhiro <tak@no32.tk>
|
|
|
|
* rational.c: fix dangling if, else-if and else.
|
|
|
|
Tue Mar 12 06:27:59 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/commands/setup_command.rb: Don't delete non-rubygems
|
|
files when installing RubyGems.
|
|
* test/rubygems/test_gem_commands_setup_command.rb: Test for the
|
|
above.
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb: Use full path to siteconf.rb
|
|
in case the extconf.rb changes directories (like memcached does).
|
|
|
|
* lib/rubygems/package.rb: Remove double slash from path.
|
|
* test/rubygems/test_gem_package.rb: Test for the above.
|
|
* test/rubygems/test_gem_package_old.rb: ditto.
|
|
|
|
* lib/rubygems/source.rb: Revert automatic HTTPS upgrade
|
|
* lib/rubygems/spec_fetcher.rb: ditto.
|
|
* test/rubygems/test_gem_remote_fetcher.rb: ditto.
|
|
* test/rubygems/test_gem_source.rb: ditto.
|
|
* test/rubygems/test_gem_spec_fetcher.rb: ditto.
|
|
|
|
Tue Mar 12 02:25:19 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/net/smtp.rb: Added Net::SMTP#rset method to implement the SMTP
|
|
RSET command. [ruby-trunk - Feature #5373]
|
|
* NEWS: ditto.
|
|
* test/net/smtp/test_smtp.rb: Test for the above.
|
|
|
|
Mon Mar 11 22:44:57 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/resolv-replace.rb (TCPSocket#initialize): resolve the 3rd
|
|
argument only if non-nil value is given.
|
|
[ruby-dev:47150] [ruby-trunk - Bug #8054] reported and analyzed by
|
|
mrkn.
|
|
|
|
Mon Mar 11 19:22:54 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* test/mkmf/base.rb: class name conflict.
|
|
|
|
Mon Mar 11 18:45:09 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* enumerator.c (enumerator_with_index): try to convert given offset to
|
|
integer. fix bug introduced in r39594.
|
|
|
|
Mon Mar 11 17:27:57 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* test/ruby/envutil.rb (EnvUtil.with_default_external): add for
|
|
changing Encoding.default_external without warnings.
|
|
|
|
* test/ruby/envutil.rb (EnvUtil.with_default_internal): ditto.
|
|
|
|
* test/ruby/test_io_m17n.rb: use above with_default_external.
|
|
|
|
Mon Mar 11 16:57:00 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* io.c (extract_binmode): raise error even if binmode and textmode
|
|
don't conflict. [Bug #5918] [ruby-core:42199]
|
|
|
|
Mon Mar 11 12:25:12 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* Merge Onigmo d4bad41e16e3eccd97ccce6f1f96712e557c4518.
|
|
fix lookbehind assertion fails with /m mode enabled. [Bug #8023]
|
|
fix \Z matches where it shouldn't. [Bug #8001]
|
|
|
|
Mon Mar 11 11:53:35 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (MakeMakefile#dir_config, MakeMakefile#_libdir_basename):
|
|
defer use of instance variable until needed. [Bug #8074]
|
|
|
|
Thu Mar 7 10:42:28 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* lib/thread.rb (Queue#clear): return self.
|
|
Patch by Cubing Cube. Thank you! [Bug #7947] [ruby-dev:47098]
|
|
* lib/thread.rb (Queue#push): ditto.
|
|
* lib/thread.rb (SizedQueue#push): ditto.
|
|
* test/thread/test_queue.rb: add tests for the above.
|
|
|
|
Thu Mar 7 10:40:49 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* tool/change_maker.rb (#diff2index): check Encoding::BINARY.
|
|
BASERUBY may still be 1.8.x.
|
|
|
|
Thu Mar 7 08:47:42 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* NEWS (Mutex#owned?): no longer experimental.
|
|
|
|
Sun Mar 10 23:38:15 2013 Luis Lavena <luislavena@gmail.com>
|
|
|
|
* win32/file.c (rb_file_expand_path_internal): Expand home directory when
|
|
used as second parameter (dir_string). [ruby-core:53168] [Bug #8034]
|
|
* test/ruby/test_file_exhaustive.rb: add test to verify.
|
|
|
|
Sun Mar 10 23:27:05 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
|
|
it is impossible to predict which file will be installed to where,
|
|
by the arguments, so use intermediate destination directory always.
|
|
[Bug #7698]
|
|
|
|
Sun Mar 10 17:00:22 2013 Tadayoshi Funaba <tadf@dotrb.org>
|
|
|
|
* complex.c: edited rdoc.
|
|
* rational.c: ditto.
|
|
|
|
Sun Mar 10 15:02:39 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* process.c (setup_communication_pipe): remove unused function.
|
|
it was unintentionally added r39683.
|
|
|
|
Wed Mar 6 00:30:40 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* tool/gen_ruby_tapset.rb: add tapset generator.
|
|
|
|
Wed Mar 6 03:27:43 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* probes.d (symbol-create): change argument name `string' to
|
|
`str'. `string' is a keyword for systemtap.
|
|
|
|
Tue Mar 5 22:23:01 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* probes.d: added argument name
|
|
|
|
Thu Mar 7 01:17:00 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* test/thread/test_queue.rb (TestQueue#test_thr_kill): reduce
|
|
iterations from 2000 to 250. When running on uniprocessor
|
|
systems, every th.kill needs TIME_QUANTUM_USEC time (i.e.
|
|
100msec on posix systems). Because, "r.read 1" is 3 steps
|
|
operations that 1) release GVL 2) read 3) acquire gvl and
|
|
(1) invoke context switch to main thread. and then, main
|
|
thread's th.kill resume (1), but not (2). Thus read interrupt
|
|
need TIME_QUANTUM_USEC. Then maximum iteration is 30sec/100msec
|
|
= 300.
|
|
|
|
Thu Mar 7 00:14:51 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* io.c (rb_update_max_fd): use ATOMIC_CAS because this function
|
|
is used from timer thread too.
|
|
|
|
Wed Mar 6 23:30:21 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread_pthread.c (ARRAY_SIZE): new.
|
|
* thread_pthread.c (gvl_acquire_common): use low priority
|
|
notification for avoiding timer thread interval confusion.
|
|
If we use timer_thread_pipe[1], every gvl_yield() request
|
|
one more gvl_yield(). It lead to thread starvation.
|
|
[Bug #7999] [ruby-core:53095]
|
|
* thread_pthread.c (rb_reserved_fd_p): adds timer_thread_pipe_low
|
|
to reserved fds.
|
|
|
|
Wed Mar 6 22:36:19 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread_pthread.c (rb_thread_wakeup_timer_thread_fd): add fd
|
|
argument and remove hardcoded dependency of timer_thread_pipe[1].
|
|
* thread_pthread.c (consume_communication_pipe): add fd argument.
|
|
* thread_pthread.c (close_communication_pipe): ditto.
|
|
|
|
* thread_pthread.c (timer_thread_sleep): adjust the above changes.
|
|
|
|
* thread_pthread.c (setup_communication_pipe_internal): factor
|
|
out pipe initialize logic.
|
|
|
|
Wed Mar 6 22:56:14 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread_pthread.c (ubf_select): add to small comments why we
|
|
need to call rb_thread_wakeup_timer_thread().
|
|
|
|
Wed Mar 6 21:42:24 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread_pthread.c (rb_thread_create_timer_thread): factor out
|
|
creating communication pipe logic into separate function.
|
|
* thread_pthread.c (setup_communication_pipe): new helper function.
|
|
* thread_pthread.c (set_nonblock): moves a definition before
|
|
setup_communication_pipe.
|
|
|
|
Sun Mar 3 02:42:29 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread_pthread.c (consume_communication_pipe): retry when
|
|
read returned CCP_READ_BUFF_SIZE.
|
|
|
|
Wed Mar 6 21:31:35 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread_pthread.c (timer_thread_sleep): use poll() instead of
|
|
select(). select doesn't work if timer_thread_pipe[0] is
|
|
greater than FD_SETSIZE.
|
|
* thread_pthread.c (USE_SLEEPY_TIMER_THREAD): add a dependency
|
|
against poll.
|
|
|
|
Wed Mar 6 21:00:23 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread_pthread.c (USE_SLEEPY_TIMER_THREAD): use more accurate
|
|
ifdef conditions.
|
|
|
|
Sun Mar 3 02:30:36 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* thread_pthread.c (set_nonblock): new helper function for set
|
|
O_NONBLOCK.
|
|
* thread_pthread.c (rb_thread_create_timer_thread): set O_NONBLOCK
|
|
to timer_thread_pipe[0] too.
|
|
|
|
Sun Mar 10 09:12:51 2013 Tadayoshi Funaba <tadf@dotrb.org>
|
|
|
|
* complex.c: described syntax of string form.
|
|
* rational.c: ditto.
|
|
|
|
Sat Mar 9 11:58:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* marshal.c (w_extended): check for prepended object.
|
|
[ruby-core:53206] [Bug #8043]
|
|
|
|
Sat Mar 9 08:36:58 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* load.c (features_index_add_single, rb_feature_p): store single index
|
|
as Fixnum to reduce the number of arrays for the indexes. based on
|
|
the patch by tmm1 (Aman Gupta) in [ruby-core:53216] [Bug #8048].
|
|
|
|
Sat Mar 9 00:25:57 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* marshal.c (r_object0): load prepended objects. treat the class of
|
|
extended object in the included modules as prepended singleton
|
|
class. [ruby-core:53202] [Bug #8041]
|
|
|
|
Fri Mar 8 19:44:00 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* man/rake.1, man/ruby.1: Use the Pa macro to make URLs stand out.
|
|
|
|
Fri Mar 8 13:20:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/pathname/pathname.c (path_f_pathname): rdoc for Pathname()
|
|
|
|
Fri Mar 8 12:00:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* man/rake.1: Document ENVIRONMENT variables on RAKE(1) manpage
|
|
|
|
Fri Mar 8 10:44:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/webrick/httpproxy.rb: Fix typos in HTTPProxyServer [Bug #8013]
|
|
Patch by Nobuhiro IMAI [ruby-core:53127]
|
|
|
|
Fri Mar 8 03:16:15 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
|
|
|
|
* class.c (rb_mod_ancestors): Include singleton_class in ancestors
|
|
list [Feature #8035]
|
|
|
|
* test/ruby/test_module.rb (class): test for above
|
|
|
|
* test/ruby/marshaltestlib.rb (module): adapt test
|
|
|
|
* NEWS: list change
|
|
|
|
Thu Mar 7 14:21:37 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* compile.c (iseq_compile_each): pass keyword arguments to zsuper,
|
|
with current values. [ruby-core:53114] [Bug #8008]
|
|
|
|
Thu Mar 7 12:53:47 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/commands/setup_command.rb: Install .pem files.
|
|
* test/rubygems/test_gem_commands_setup_command.rb: Test for the
|
|
above.
|
|
|
|
* lib/rubygems/spec_fetcher.rb: Test HTTPS upgrade with URI::HTTPS,
|
|
not URI::HTTP. Fixes bug in automatic HTTPS upgrade.
|
|
* test/rubygems/test_gem_spec_fetcher.rb: Test for the above.
|
|
|
|
* lib/rubygems.rb: Version 2.0.2
|
|
|
|
* lib/rubygems/test_utilities.rb: Ensure scheme and uri class match.
|
|
|
|
Thu Mar 7 10:39:04 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* tool/rbinstall.rb (gem): Gem.ensure_gem_subdirectories now has mode
|
|
option since r39607. refix of r38870.
|
|
|
|
Wed Mar 6 13:14:28 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* test/rubygems/test_gem_spec_fetcher.rb: Removed unused variable.
|
|
|
|
Wed Mar 6 08:10:15 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* test/rubygems/test_require.rb: Fix tests when 'a.rb' exists.
|
|
[ruby-trunk - Bug #7749]
|
|
|
|
Wed Mar 6 08:00:59 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems.rb: Allow specification of directory permissions.
|
|
[ruby-trunk - Bug #7713]
|
|
* test/rubygems/test_gem.rb: Test for the above.
|
|
|
|
Wed Mar 6 07:40:21 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/commands/query_command.rb: Only fetch remote specs when
|
|
showing details. [ruby-trunk - Bug #8019] RubyGems bug #487
|
|
* lib/rubygems/remote_fetcher.rb: ditto.
|
|
* lib/rubygems/security/policy.rb: ditto.
|
|
* test/rubygems/test_gem_commands_query_command.rb: Test for the
|
|
above.
|
|
|
|
* lib/rubygems/security.rb: Make OpenSSL optional for RubyGems.
|
|
* lib/rubygems/commands/cert_command.rb: ditto.
|
|
|
|
* lib/rubygems/config_file.rb: Display file with YAML error, not
|
|
~/.gemrc
|
|
|
|
* lib/rubygems/remote_fetcher.rb: Only create gem subdirectories when
|
|
installing gems.
|
|
* lib/rubygems/dependency_resolver.rb: ditto.
|
|
* lib/rubygems/test_utilities.rb: ditto.
|
|
* test/rubygems/test_gem_commands_fetch_command.rb: Test for the
|
|
above.
|
|
|
|
* lib/rubygems/spec_fetcher.rb: Only try to upgrade
|
|
http://rubygems.org to HTTPS
|
|
* test/rubygems/test_gem_spec_fetcher.rb: Test for the above.
|
|
|
|
* lib/rubygems.rb: Update win_platform? check for JRuby compatibility.
|
|
|
|
* test/rubygems/test_gem_installer.rb: Update for Ruby 1.9.2
|
|
compatibility
|
|
|
|
Wed Mar 6 01:19:28 2013 Kazuhiro NISHIYAMA <zn@mbf.nifty.com>
|
|
|
|
* enumerator.c (enumerator_with_index, lazy_take): use INT2FIX(0)
|
|
instead of INT2NUM(0).
|
|
|
|
* ext/bigdecimal/bigdecimal.c (BigMath_s_exp): ditto.
|
|
|
|
* ext/fiddle/function.c (function_call): ditto.
|
|
|
|
* ext/openssl/ossl_x509store.c (ossl_x509store_initialize): ditto.
|
|
|
|
* process.c (proc_getsid): ditto.
|
|
|
|
* transcode.c (econv_finish): ditto.
|
|
|
|
Tue Mar 5 21:36:43 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* class.c (rb_prepend_module): check redefinition of built-in optimized
|
|
methods. [ruby-dev:47124] [Bug #7983]
|
|
|
|
* vm.c (rb_vm_check_redefinition_by_prepend): ditto.
|
|
|
|
Tue Mar 5 20:29:25 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* proc.c (mnew): revert r39224. [ruby-core:53038] [Bug #7988]
|
|
|
|
Tue Mar 5 20:23:54 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/intern.h (rb_check_arity): make a static inline
|
|
function so it can be used as an expression and argc would be
|
|
evaluated only once.
|
|
|
|
Tue Mar 5 12:30:55 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems.rb: Bump version to 2.0.1 for upcoming bugfix release
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb: Restore ruby 1.8 compatibility
|
|
for [Bug #7698]
|
|
* test/rubygems/test_gem_installer.rb: Ditto.
|
|
|
|
* lib/rubygems/package.rb: Restore ruby 1.8 compatibility.
|
|
|
|
* test/rubygems/test_gem_dependency_installer.rb: Fix warnings
|
|
|
|
Tue Mar 5 12:24:23 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* enumerator.c (enumerator_with_index): Restore handling of a nil memo
|
|
from r39594.
|
|
|
|
Tue Mar 5 10:40:22 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/objspace/objspace.c (count_nodes): count also newly added nodes,
|
|
and fix key for unknown node. patch by tmm1 (Aman Gupta) in
|
|
[ruby-core:53130] [Bug #8014]
|
|
|
|
Tue Mar 5 10:20:16 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enumerator.c (enumerator_with_index_i): allow Bignum as offset, to
|
|
get rid of conversion exception and integer overflow.
|
|
[ruby-dev:47131] [Bug #8010]
|
|
|
|
* numeric.c (rb_int_succ, rb_int_pred): shortcut optimization for
|
|
Bignum.
|
|
|
|
Tue Mar 5 10:02:48 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
|
|
clear DESTDIR so RUBYARCHDIR and RUBYLIBDIR are not be overridden.
|
|
[Bug #7698]
|
|
|
|
Mon Mar 4 15:33:40 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
|
|
fix for unusual cases again. install to a temporary directory once
|
|
and move installed files to the destination directory, if it is same
|
|
as the current directory. [Bug #7698]
|
|
|
|
Mon Mar 4 14:13:36 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* Makefile.in (miniruby, ruby): move MAINLIBC because linker arguments
|
|
must appear after object files with newer versions of gcc. patch by
|
|
tmm1 (Aman Gupta) in [ruby-core:53121] [Bug #8009]
|
|
|
|
Mon Mar 4 10:23:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* encoding.c: Typo in Encoding overview by Tom Wardrop [GH fixes #255]
|
|
|
|
Sun Mar 3 12:35:08 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (MakeMakefile#libpath_env): set runtime library path for
|
|
the case rpath is disabled.
|
|
|
|
Sun Mar 3 12:17:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb
|
|
(Gem::Ext::ExtConfBuilder.hack_for_obsolete_style_gems): remove
|
|
circular dependencies in install-so too. [ruby-core:52882]
|
|
[Bug #7698]
|
|
|
|
Sun Mar 3 07:33:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/socket/tcpserver.c: Grammar for TCPServer.new from r39554
|
|
|
|
Sun Mar 3 01:17:20 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb
|
|
(Gem::Ext::ExtConfBuilder.hack_for_obsolete_style_gems): remove
|
|
circular dependencies for old style gems which locate extconf.rb on
|
|
the toplevel. [ruby-core:53059] [ruby-trunk - Bug #7698]
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
|
|
use RUBYOPT instead of -r option, and revert some tests. [Bug #7698]
|
|
|
|
* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
|
|
revert use of temporary directory for build, to work some buggy
|
|
extconf.rb which cannot build outside the source directory.
|
|
[ruby-core:53056] [Bug #7698]
|
|
|
|
Sun Mar 3 00:04:20 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enc/depend (CPPFLAGS), lib/mkmf.rb (MakeMakefile#create_makefile):
|
|
define RUBY_EXPORT for static-linked-ext mswin. [Bug #7960]
|
|
|
|
Sat Mar 2 22:49:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* win32/Makefile.sub (ENCOBJS, EXTOBJS, config.h): definitions for
|
|
static-linked-ext. [Bug #7960]
|
|
|
|
Sat Mar 2 17:34:19 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* lib/webrick/utils.rb: use Socket.tcp_server_sockets to create server
|
|
sockets.
|
|
fix [Bug #7100] https://bugs.ruby-lang.org/issues/7100
|
|
reported by sho-h (Sho Hashimoto).
|
|
|
|
Sat Mar 2 02:45:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* array.c: typo in comment patch by Nami-Doc [Github fixes #253]
|
|
|
|
Sat Mar 2 01:33:17 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* Merge Onigmo 0fe387da2fee089254f6b04990541c731a26757f
|
|
v5.13.3 [Bug#7972] [Bug#7974]
|
|
|
|
Fri Mar 1 11:09:06 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/fileutils.rb: Revert r34669 which altered the way
|
|
metaprogramming in FileUtils occurred. [ruby-trunk - Bug #7958]
|
|
|
|
* test/fileutils/visibility_tests.rb: Refactored tests of FileUtils
|
|
options modules to expose bug found in #7958
|
|
* test/fileutils/test_dryrun.rb: ditto.
|
|
* test/fileutils/test_nowrite.rb: ditto.
|
|
* test/fileutils/test_verbose.rb: ditto.
|
|
|
|
Fri Mar 1 09:18:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/psych.rb: specify in rdoc what object is returned in parser
|
|
By Adam Stankiewicz [Github tenderlove/psych#133]
|
|
|
|
Fri Mar 1 07:21:41 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/ext/builder.rb: Fix incompatibilities when installing
|
|
extensions. Patch by Nobu.
|
|
[ruby-trunk - Bug #7698] [ruby-trunk - Bug #7971]
|
|
* lib/rubygems/ext/ext_conf_builder.rb: ditto.
|
|
* lib/rubygems/installer.rb: ditto.
|
|
* test/rubygems/test_gem_ext_ext_conf_builder.rb: Test for the above.
|
|
* test/rubygems/test_gem_installer.rb: ditto.
|
|
|
|
* lib/rubygems/commands/sources_command.rb: Prefer HTTPS over HTTP.
|
|
* lib/rubygems/defaults.rb: ditto
|
|
* lib/rubygems/dependency_resolver.rb: Ditto.
|
|
* lib/rubygems/source.rb: ditto.
|
|
* lib/rubygems/spec_fetcher.rb: ditto.
|
|
* lib/rubygems/specification.rb: ditto.
|
|
* lib/rubygems/test_utilities.rb: ditto.
|
|
* test/rubygems/test_gem.rb: Test for the above.
|
|
* test/rubygems/test_gem_commands_sources_command.rb: ditto.
|
|
* test/rubygems/test_gem_dependency_resolver_api_set.rb: ditto.
|
|
* test/rubygems/test_gem_remote_fetcher.rb: ditto.
|
|
* test/rubygems/test_gem_source.rb: ditto.
|
|
* test/rubygems/test_gem_spec_fetcher.rb: ditto.
|
|
|
|
Fri Mar 1 03:25:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/psych/lib/psych.rb: rdoc for Psych overview by Adam Stankiewicz
|
|
[Github tenderlove/psych#134]
|
|
|
|
Thu Feb 28 22:57:48 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* compile.c (iseq_compile_each): remove redundant trace(line)
|
|
instruction. for example, at the following script
|
|
def m()
|
|
p:xyzzy
|
|
1
|
|
2
|
|
end
|
|
compiler ignores `1' because there is no effect. However,
|
|
`trace(line)' instruction remains in bytecode.
|
|
This modification removes such redundant trace(line) instruction.
|
|
|
|
* test/ruby/test_iseq.rb: add a test.
|
|
|
|
Thu Feb 28 22:23:27 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/raddrinfo.c (inspect_sockaddr): don't show that Unix
|
|
domain socket filename is bigger than sizeof(sun_path).
|
|
This limit is not rigid on some platforms such as Darwin and SunOS.
|
|
|
|
Thu Feb 28 21:33:01 2013 WATANABE Hirofumi <eban@ruby-lang.org>
|
|
|
|
* configure.in(AC_DISABLE_OPTION_CHECKING): avoid warning "WARNING:
|
|
Unrecognized options: --with-PACKAGE".
|
|
|
|
Thu Feb 28 20:22:04 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* iseq.c (iseq_data_to_ary): fix condition.
|
|
r34303 introduces a bug to avoid all line information from
|
|
a result of ISeq#to_a. This is a regression problem from 2.0.0p0.
|
|
|
|
* test/ruby/test_iseq.rb: add a test of lines after ISeq#to_a.
|
|
|
|
Thu Feb 28 08:20:33 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/available_set.rb: Undent for style
|
|
|
|
* lib/rubygems/dependency_installer.rb: Pick latest prerelease gem to
|
|
install. Fixes RubyGems bug #468.
|
|
* test/rubygems/test_gem_dependency_installer.rb: Test for the above.
|
|
|
|
* lib/rubygems/dependency_installer.rb: Don't display "Done installing
|
|
documentation" if documentation will not be installed.
|
|
* lib/rubygems/rdoc.rb: ditto
|
|
|
|
* lib/rubygems/dependency_list.rb: Use Array#concat for Ruby 1.x
|
|
performance.
|
|
|
|
* lib/rubygems/installer.rb: Use formatted program name when comparing
|
|
executables. RubyGems pull request #471
|
|
* test/rubygems/test_gem_installer.rb: Test for the above.
|
|
|
|
* lib/rubygems/package.rb: Use more explicit feature check to work
|
|
around JRuby bug #552
|
|
|
|
* lib/rubygems/ssl_certs/GeoTrust_Global_CA.pem: Added GeoTrust root
|
|
certificate.
|
|
|
|
* test/rubygems/test_gem_source_list.rb: Use "example" instead of real
|
|
hostname
|
|
|
|
Thu Feb 28 05:57:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* thread.c: rdoc formatting for Thread, ThreadGroup, and ThreadError
|
|
|
|
Thu Feb 28 02:42:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* vm.c: Typo in overview for example of Thread#status returning false
|
|
Reported by Lee Jarvis
|
|
|
|
Wed Feb 27 22:54:27 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/rubysocket.h (union_sockaddr): make it longer for SunOS
|
|
and Darwin.
|
|
|
|
Wed Feb 27 21:14:34 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/security.rb (REXML::Security): create.
|
|
* lib/rexml/rexml.rb: move entity_expansion_limit and
|
|
entity_expansion_text_limit accessors to ...
|
|
* lib/rexml/security.rb: ... here.
|
|
* lib/rexml/document.rb: use REXML::Security.
|
|
* lib/rexml/text.rb: use REXML::Security.
|
|
* test/rexml/test_document.rb: use REXML::Security.
|
|
|
|
Wed Feb 27 19:53:32 2013 Benoit Daloze <eregontp@gmail.com>
|
|
|
|
* vm.c (Thread): fix typos in overview
|
|
|
|
Wed Feb 27 13:21:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* vm.c (Thread): Typo in overview, swap setting and getting
|
|
|
|
Wed Feb 27 13:02:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* vm.c (Thread): Documentation overview of Thread class
|
|
|
|
Wed Feb 27 12:57:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* thread.c (rb_thread_wakeup): rdoc formatting
|
|
|
|
Wed Feb 27 12:53:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* thread.c (rb_thread_group): rdoc formatting
|
|
|
|
Wed Feb 27 12:33:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/ostruct.rb: Typo in OpenStruct overview [Github Fixes #251]
|
|
Patch by Chun-wei Kuo
|
|
|
|
Wed Feb 27 12:13:32 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* vm_exec.h (END_INSN): llvm-gcc may optimize out reg_cfp and cause
|
|
Stack/cfp consistency error when the instruction doesn't use reg_cfp.
|
|
Usually instructions use PUSH() but for example trace doesn't.
|
|
This hack cause speed down but you shouldn't use llvm-gcc, use clang.
|
|
[Bug #7938]
|
|
|
|
Wed Feb 27 10:23:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* thread.c (thread_raise_m): rdoc formatting
|
|
|
|
Tue Feb 26 23:32:44 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/document.rb: move entity_expansion_limit accessor to ...
|
|
* lib/rexml/rexml.rb: ... here for consistency.
|
|
* lib/rexml/document.rb (REXML::Document.entity_expansion_limit):
|
|
deprecated.
|
|
* lib/rexml/document.rb (REXML::Document.entity_expansion_limit=):
|
|
deprecated.
|
|
|
|
Tue Feb 26 23:26:13 2013 Kouhei Sutou <kou@cozmixng.org>
|
|
|
|
* lib/rexml/document.rb: move entity_expansion_text_limit accessor to ...
|
|
* lib/rexml/rexml.rb: ... here to make rexml/text independent from
|
|
REXML::Document. It causes circular require.
|
|
* lib/rexml/document.rb (REXML::Document.entity_expansion_text_limit):
|
|
deprecated.
|
|
* lib/rexml/document.rb (REXML::Document.entity_expansion_text_limit=):
|
|
deprecated.
|
|
* lib/rexml/text.rb: add missing require "rexml/rexml" for
|
|
REXML.entity_expansion_text_limit.
|
|
Reported by Robert Ulejczyk. Thanks!!! [ruby-core:52895] [Bug #7961]
|
|
|
|
Tue Feb 26 15:12:11 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* tool/mkconfig.rb: reconstruct comma separated list values. a
|
|
command line to Windows batch file is split not only by spaces
|
|
and equal signs but also by commas and semicolons.
|
|
|
|
Tue Feb 26 15:04:19 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (unexpand_shvar): get rid of non-portable shell
|
|
behavior on OpenBSD, so no extra quotes. [Bug #7959]
|
|
|
|
Tue Feb 26 10:24:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* parse.y (IS_LABEL_POSSIBLE): allow labels for keyword arguments just
|
|
after method definition without a parenthesis. [ruby-core:52820]
|
|
[Bug #7942]
|
|
|
|
Tue Feb 26 04:50:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* error.c: clarify reason for sleep in SignalException example
|
|
|
|
Tue Feb 26 03:47:00 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* error.c: clarify a document of SignalException. Process.kill()
|
|
doesn't have any guarantee when signal will be delivered.
|
|
[Bug #7951] [ruby-core:52864]
|
|
|
|
Mon Feb 25 23:51:04 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/version.h: bump RUBY_API_VERSION same as RUBY_VERSION.
|
|
|
|
Mon Feb 25 21:03:34 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* string.c (str_byte_substr): don't set coderange if it's not known.
|
|
[Bug #7954] [ruby-dev:47108]
|
|
|
|
Mon Feb 25 16:47:02 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* common.mk (realclean-local): miniprelude.c is made by srcs, so it
|
|
should not removed by distclean but by realclean. [Bug #6807]
|
|
|
|
Mon Feb 25 16:30:30 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/config_file.rb: Lazily load .gem/credentials to only
|
|
check permissions when necessary. RubyGems bug #465
|
|
* test/rubygems/test_gem_config_file.rb: Test for the above.
|
|
|
|
* test/rubygems/test_gem_commands_push_command.rb: Remove duplicated
|
|
test.
|
|
|
|
Mon Feb 25 15:47:18 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* enc/depend (ARFLAGS): VisualC++ linker does not allow spaces between
|
|
output option and the output file name. [Bug #7950]
|
|
|
|
* enc/depend (RANLIB): set default command to do nothing, or make the
|
|
entire line a label on Windows.
|
|
|
|
Mon Feb 25 14:41:07 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (MakeMakefile#init_mkmf): default libdirname to libdir.
|
|
|
|
* tool/rbinstall.rb: ditto.
|
|
|
|
Mon Feb 25 13:12:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (setup): find Setup file from target_os 1. by
|
|
suffix (e.g. Setup.nacl, Setup.atheos), 2. by "platform"
|
|
option (e.g. Setup.nt, Setup.emx), and 3. default Setup. And
|
|
Setup.dj had been removed.
|
|
|
|
Mon Feb 25 12:48:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* thread.c: Document Thread::new, clean up ::fork and mention calling
|
|
super if subclassing Thread
|
|
|
|
Mon Feb 25 12:38:50 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: don't test ss_family and ss_len member of
|
|
struct sockaddr_storage. They are not used now except SunOS
|
|
specific code.
|
|
|
|
Mon Feb 25 11:03:38 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* configure.in (unexpand_shvar): Use the numeric comparison
|
|
operator instead of '==' which is a ksh extension. [Bug #7941]
|
|
|
|
Mon Feb 25 02:37:56 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket: define and use union_sockaddr instead of struct
|
|
sockaddr_storage for less casts.
|
|
|
|
* ext/socket/rubysocket.h (union_sockaddr): defined.
|
|
|
|
* ext/socket/socket.c (sock_accept): use union_sockaddr.
|
|
(sock_accept_nonblock): ditto.
|
|
(sock_sysaccept): ditto.
|
|
(sock_s_getnameinfo): ditto.
|
|
|
|
* ext/socket/basicsocket.c (bsock_getsockname): ditto.
|
|
(bsock_getpeername): ditto.
|
|
(bsock_local_address): ditto.
|
|
(bsock_remote_address): ditto.
|
|
|
|
* ext/socket/ancdata.c (bsock_recvmsg_internal): ditto.
|
|
|
|
* ext/socket/init.c (recvfrom_arg): ditto.
|
|
(recvfrom_blocking): ditto.
|
|
(rsock_s_recvfrom): ditto.
|
|
(rsock_s_recvfrom_nonblock): ditto.
|
|
(rsock_getfamily): ditto.
|
|
|
|
* ext/socket/raddrinfo.c (rb_addrinfo_t): ditto.
|
|
(ai_get_afamily): ditto.
|
|
(inspect_sockaddr): ditto.
|
|
(addrinfo_mdump): ditto.
|
|
(addrinfo_mload): ditto.
|
|
(addrinfo_getnameinfo): ditto.
|
|
(addrinfo_ip_port): ditto.
|
|
(extract_in_addr): ditto.
|
|
(addrinfo_ipv6_to_ipv4): ditto.
|
|
(addrinfo_unix_path): ditto.
|
|
|
|
* ext/socket/tcpserver.c (tcp_accept): ditto.
|
|
(tcp_accept_nonblock): ditto.
|
|
(tcp_sysaccept): ditto.
|
|
|
|
* ext/socket/ipsocket.c (ip_addr): ditto.
|
|
(ip_peeraddr): ditto.
|
|
(ip_s_getaddress): ditto.
|
|
|
|
Sun Feb 24 21:15:05 2013 Tadayoshi Funaba <tadf@dotrb.org>
|
|
|
|
* ext/date/date_core.c: [ruby-core:52303]
|
|
|
|
Sun Feb 24 15:33:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* random.c (rb_random_ulong_limited): limit is inclusive, but generic
|
|
rand method should return a number less than it, so increase for the
|
|
difference. [ruby-core:52779] [Bug #7935]
|
|
|
|
Sun Feb 24 15:32:36 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* random.c (rb_random_ulong_limited): limit is inclusive, but generic
|
|
rand method should return a number less than it, so increase for the
|
|
difference. [ruby-core:52779] [Bug #7935]
|
|
|
|
Sun Feb 24 15:14:43 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/net/http.rb: Removed duplicate Accept-Encoding in Net::HTTP#get.
|
|
[ruby-trunk - Bug #7924]
|
|
* test/net/http/test_http.rb: Test for the above.
|
|
|
|
Wed Feb 20 14:28:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* thread.c: Document ThreadGroup::Default
|
|
|
|
Wed Feb 20 14:23:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* thread.c: Grammar for #backtrace_locations and ::handle_interrupt
|
|
|
|
Sun Feb 24 13:35:57 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_insnhelper.c (vm_call_method): block level control frame does not
|
|
have method entry, so obtain the method entry from method top-level
|
|
control frame to be compared with refined method entry.
|
|
[ruby-core:52750] [Bug #7925]
|
|
|
|
Wed Feb 20 13:23:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* object.c: Document methods receiving string and convert to symbol
|
|
Patch by Stefan Rusterholz
|
|
* vm_eval.c: ditto
|
|
* vm_method.c: ditto
|
|
|
|
Wed Feb 20 07:20:56 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* signal.c (sigsegv): suppress unused result warning. Because
|
|
write(2) is marked __warn_unused_result__ on Linux glibc.
|
|
|
|
Sun Feb 24 07:50:53 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* compile.c (iseq_set_arguments): no keyword check if any keyword rest
|
|
argument exists, even unnamed. [ruby-core:52744] [Bug #7922]
|
|
|
|
Sat Feb 23 16:51:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* thread.c: Documentation for Thread#backtrace_locations
|
|
|
|
Sat Feb 23 16:05:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* vm.c: Typo in ObjectSpace::WeakMap overview
|
|
|
|
Sat Feb 23 16:00:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* thread.c: Improved rdoc for ::handle_interrupt, ::pending_interrupt?
|
|
and #pending_interrupt?
|
|
|
|
Sat Feb 23 12:26:43 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* misc/ruby-electric.el (ruby-electric-curlies)
|
|
(ruby-electric-matching-char, ruby-electric-bar): Avoid electric
|
|
insertion when there is a prefix argument.
|
|
|
|
* misc/ruby-electric.el (ruby-electric-insert)
|
|
(ruby-electric-cua-replace-region-p)
|
|
(ruby-electric-cua-replace-region): Avoid electric insertion and
|
|
fall back when cua-mode is enabled and a region is active.
|
|
|
|
Sat Feb 23 12:35:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* array.c: Document #<=> return values and formatting
|
|
* bignum.c: ditto
|
|
* file.c: ditto
|
|
* object.c: ditto
|
|
* numeric.c: ditto
|
|
* rational.c: ditto
|
|
* string.c: ditto
|
|
* time.c: ditto
|
|
|
|
Sat Feb 23 10:50:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* array.c (rb_ary_diff, rb_ary_and, rb_ary_or): Document return order
|
|
[RubySpec #7803]
|
|
|
|
Sat Feb 23 10:17:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* object.c (rb_obj_comp): Documenting Object#<=> return values
|
|
Patch by Stefan Rusterholz
|
|
|
|
Sat Feb 23 09:48:41 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* dir.c (file_s_fnmatch, fnmatch_brace): encoding-incompatible pattern
|
|
and string do not match, instead of exception. [ruby-dev:47069]
|
|
[Bug #7911]
|
|
|
|
Sat Feb 23 08:57:46 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
|
|
|
|
* doc/NEWS-*: Update NEWS from their respective branches
|
|
|
|
Sat Feb 23 08:14:43 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
|
|
|
|
* NEWS: many additions for Ruby 2.0.0
|
|
|
|
* object.c: Add doc for Module.prepended
|
|
|
|
Sat Feb 23 07:52:53 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* template/ruby.pc.in: reorder library flags which may refer library
|
|
names. [Bug #7913]
|
|
|
|
Fri Feb 22 23:46:20 2013 CHIKANAGA Tomoyuki <nagachika@ruby-lang.org>
|
|
|
|
* lib/rexml/document.rb (REXML::Document.entity_expansion_text_limit):
|
|
fix a typo in comment in r39384.
|
|
|
|
Fri Feb 22 18:31:46 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* lib/rexml/document.rb (REXML::Document.entity_expansion_text_limit):
|
|
new attribute to read/write entity expansion text limit. the default
|
|
limit is 10Kb.
|
|
|
|
* lib/rexml/text.rb (REXML::Text.unnormalize): check above attribute.
|
|
|
|
Fri Feb 22 17:36:23 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* test/test_rbconfig.rb (TestRbConfig): fix r39372.
|
|
It must see RbConfig::CONFIG instead of CONFIG.
|
|
|
|
Fri Feb 22 14:55:41 2013 Naohisa Goto <ngotogenome@gmail.com>
|
|
|
|
* signal.c (ruby_abort): fix typo in r39354 [Bug #5014]
|
|
|
|
Fri Feb 22 12:46:41 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* random.c (rb_random_ulong_limited): fix error message for negative
|
|
value. [ruby-dev:47061] [Bug #7903]
|
|
|
|
Fri Feb 22 11:36:45 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* test/test_rbconfig.rb (TestRbConfig): skip user defined values by
|
|
configuration options. [Bug #7902]
|
|
|
|
Fri Feb 22 11:33:42 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (MakeMakefile#init_mkmf): adjust default library path
|
|
for multiarch. [Bug #7874]
|
|
|
|
Fri Feb 22 11:10:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* enum.c (Enumerable#chunk: Improved examples, grammar, and formatting
|
|
Patch by Dan Bernier and Rich Bruchal of newhaven.rb
|
|
[Github documenting-ruby/ruby#8]
|
|
|
|
Fri Feb 22 11:00:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* numeric.c: Examples and formatting for Numeric and Float
|
|
Based on a patch by Zach Morek and Oren K of newhaven.rb
|
|
[Github documenting-ruby/ruby#5]
|
|
|
|
Fri Feb 22 07:04:41 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/installer.rb (build_extensions): Create extension
|
|
install destination before building extension. Patch by Kenta Murata.
|
|
[ruby-trunk - Bug #7897]
|
|
* test/rubygems/test_gem_installer.rb: Test for the above.
|
|
|
|
Fri Feb 22 06:30:57 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* doc/globals.rdoc: Document what setting $DEBUG does.
|
|
|
|
* doc/globals.rdoc: Added pointer to $-d for full documentation.
|
|
|
|
Fri Feb 22 06:27:07 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* doc/globals.rdoc: Document what setting $VERBOSE does. [Bug #7899]
|
|
|
|
* doc/globals.rdoc: Added pointer to $-w and $-v for full
|
|
documentation.
|
|
|
|
Fri Feb 22 02:33:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/abbrev.rb: Add words parameter to Abbrev::abbrev
|
|
Patch by Devin Weaver [Github documenting-ruby/ruby#7]
|
|
|
|
Thu Feb 21 17:28:14 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* tool/merger.rb: add interaction when only ChangeLog is modified.
|
|
|
|
Thu Feb 21 16:34:46 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* signal.c (check_stack_overflow): extract duplicated code and get rid
|
|
of declaration-after-statement. [Bug #5014]
|
|
|
|
Thu Feb 21 14:14:13 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* signal.c (sigsegv): avoid to use async signal unsafe functions
|
|
when nested sigsegv is happen.
|
|
[Bug #5014] [ruby-dev:44082]
|
|
|
|
Thu Feb 21 13:47:59 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* file.c (rb_group_member): added an error check. SUS says,
|
|
getgroups(small_value) may return EINVAL.
|
|
|
|
Thu Feb 21 13:37:07 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* process.c (RB_MAX_GROUPS): moved to
|
|
* internal.h (RB_MAX_GROUPS): here.
|
|
|
|
* file.c (rb_group_member): use RB_MAX_GROUPS instead of
|
|
RUBY_GROUP_MAX. They are the same.
|
|
|
|
Thu Feb 21 13:15:40 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* file.c (access_internal): removed.
|
|
* file.c (rb_file_readable_real): use access() instead of
|
|
access_internal().
|
|
* file.c (rb_file_writable_real): ditto.
|
|
* file.c (rb_file_executable_real): ditto.
|
|
|
|
Thu Feb 21 13:04:59 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* file.c (eaccess): use access() when not using setuid nor setgid.
|
|
This is minor optimization.
|
|
|
|
Thu Feb 21 12:56:19 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* file.c (rb_group_member): get rid of NGROUPS dependency.
|
|
[Bug #7886] [ruby-core:52537]
|
|
|
|
Thu Feb 21 12:45:03 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ruby.c (ruby_init_loadpath_safe): try two levels upper for stripping
|
|
libdir name. [Bug #7874]
|
|
|
|
* configure.in (libdir_basename): expand with multiarch in configure,
|
|
not to defer the expansion till ruby.pc.in and mkmf.rb. [Bug #7874]
|
|
|
|
* configure.in (libdir_basename): also -rpath and -install_name flags
|
|
are affected when libruby directory changes. [Bug #7874]
|
|
|
|
Wed Feb 20 19:27:02 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* include/ruby/ruby.h (HAVE_RB_SCAN_ARGS_OPTIONAL_HASH): for
|
|
rb_scan_args() optional hash feature. [Bug #7861]
|
|
|
|
Wed Feb 20 18:02:26 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (target_os): do not strip -gnu suffix on Linux if
|
|
--target is given explicitly. [Bug #7874]
|
|
|
|
* configure.in (libdirname): adjust library path name which libruby
|
|
files will be installed. [Bug #7874]
|
|
|
|
* tool/rbinstall.rb (libdir): ditto.
|
|
|
|
Wed Feb 20 13:37:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/pty/pty.c: Documentation for the PTY module
|
|
|
|
Wed Feb 20 12:18:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* object.c: Document Data class [Bug #7890] [ruby-core:52549]
|
|
Patch by Matthew Mongeau
|
|
|
|
Wed Feb 20 11:50:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/mutex_m.rb: Add rdoc for Mutex_m module
|
|
|
|
Wed Feb 20 09:34:43 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/commands/update_command.rb: Create the installer after
|
|
options are processed. [ruby-trunk - Bug #7779]
|
|
* test/rubygems/test_gem_commands_update_command.rb: Test for the
|
|
above.
|
|
|
|
Wed Feb 20 07:51:19 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/installer.rb: Use gsub instead of gsub! to avoid
|
|
altering @bin_dir. Fixes tests on windows. [ruby-trunk - Bug #7885]
|
|
|
|
Tue Feb 19 20:50:00 2013 Kenta MURATA <mrkn@mrkn.jp>
|
|
|
|
* ext/bigdecimal/bigdecimal.gemspec: bump to 1.2.0.
|
|
[ruby-core:51777] [Bug #7761]
|
|
|
|
Tue Feb 19 13:07:25 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* ext/syslog/syslog.c (Init_syslog): Define inspect as a singleton
|
|
method and remove it as an instance method. [Bug #6502]
|
|
|
|
Tue Feb 19 12:30:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* object.c: rdoc formatting for Kernel#Array()
|
|
* array.c: Add rdoc for Array() method to Creating Arrays section
|
|
|
|
Tue Feb 19 10:35:52 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* ext/openssl/ossl.c (class OpenSSL): Use only inner parenthesis in
|
|
create_extension examples.
|
|
|
|
Tue Feb 19 10:27:12 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* ext/openssl/ossl.c (class OpenSSL): Fixed ExtensionFactory example.
|
|
Patch by Richard Bradley. [ruby-trunk - Bug #7551]
|
|
|
|
Tue Feb 19 08:32:11 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm_eval.c (vm_call0_body): check interrupts after method dispatch
|
|
from C methods. [Bug #7878]
|
|
|
|
Tue Feb 19 08:14:40 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/installer.rb: Fixed placement of executables with
|
|
--user-install. [ruby-trunk - Bug #7779]
|
|
* test/rubygems/test_gem_installer.rb: Test for above.
|
|
|
|
Tue Feb 19 06:04:06 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* vm_dump: FreeBSD ports' libexecinfo's backtrace(3) can't trace
|
|
beyond signal trampoline, and as described in r38342 it can't
|
|
trace on -O because it see stack frame pointers.
|
|
libunwind unw_backtrace see dwarf information in the binary
|
|
and it works with -O (without frame pointers).
|
|
|
|
* configure.in: remove r38342's hack and check libunwind.
|
|
|
|
Tue Feb 19 04:26:29 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* configure.in: check whether backtrace(3) works well or not.
|
|
|
|
* vm_dump.c: set HAVE_BACKTRACE 0 if BROKEN_BACKTRACE.
|
|
|
|
Mon Feb 18 16:30:18 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/ipaddr.rb (IPAddr#in6_addr): Fix a typo with the closing
|
|
parenthesis.
|
|
|
|
Mon Feb 18 12:32:24 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* lib/ipaddr.rb (IPAddr#in6_addr): Fix the parser so that it can
|
|
recognize IPv6 addresses with only one edge 16-bit piece
|
|
compressed, like [::2:3:4:5:6:7:8] or [1:2:3:4:5:6:7::].
|
|
[Bug #7477]
|
|
|
|
Mon Feb 18 10:09:54 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (unexpand_shvar): regularize a shell variable by
|
|
unexpanding shell variables in it.
|
|
|
|
Sun Feb 17 20:55:44 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* compar.c (rb_invcmp): compare by inversed comparison, with preventing
|
|
from infinite recursion. [ruby-core:52305] [Bug #7870]
|
|
|
|
* string.c (rb_str_cmp_m), time.c (time_cmp): get rid of infinite
|
|
recursion.
|
|
|
|
Sun Feb 17 17:23:22 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb: remove extra topdir in VPATH, which was in
|
|
win32/Makefile.sub for some reason and moved from there.
|
|
[ruby-dev:46998] [Bug #7864]
|
|
|
|
Sun Feb 17 01:19:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* ext/psych/lib/psych/y.rb: Document Kernel#y by Adam Stankiewicz
|
|
[Github tenderlove/psych#127]
|
|
|
|
Sun Feb 17 00:52:14 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* tool/mkconfig.rb: remove prefix from rubyarchdir.
|
|
r39267 expands variables, it changes expansion timing,
|
|
breaks RbConfig::CONFIG["includedir"] and building
|
|
extension libraries with installed ruby.
|
|
|
|
Sat Feb 16 20:51:17 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* vm.c (ENV_IN_HEAP_P): fix off-by-one error.
|
|
|
|
Sat Feb 16 20:47:16 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* configure.in (LIBRUBY_DLDFLAGS): Fix a bug where --with-opt-dir
|
|
options given were not reflected to LIBRUBY_DLDFLAGS on many
|
|
platforms including Linux and other GNU-based systems, NetBSD,
|
|
AIX and BeOS.
|
|
|
|
Sat Feb 16 20:43:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/ancdata.c (rsock_recvmsg): ignore truncated part of
|
|
socket address returned from recvmsg().
|
|
|
|
* ext/socket/init.c (recvfrom_blocking): ignore truncated part of
|
|
socket address returned from recvfrom().
|
|
(rsock_s_recvfrom_nonblock): ditto.
|
|
|
|
Sat Feb 16 20:05:26 2013 Ayumu AIZAWA <ayumu.aizawa@gmail.com>
|
|
|
|
* test/ruby/test_thread.rb: fixed typo
|
|
patched by Hiroki Matsue via https://github.com/ruby/ruby/pull/248
|
|
|
|
Sat Feb 16 16:08:35 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* vm.c (rb_thread_mark): mark a working Proc of bmethod
|
|
(a method defined by define_method) even if the method was removed.
|
|
We could not trace working Proc object which represents the body
|
|
of bmethod if the method was removed (alias/undef/overridden).
|
|
Simply, it was mark miss.
|
|
This patch by Kazuki Tsujimoto. [Bug #7825]
|
|
|
|
NOTE: We can brush up this marking because we do not need to mark
|
|
`me' on each living control frame. We need to mark `me's
|
|
only if `me' was free'ed. This is future work after Ruby 2.0.0.
|
|
|
|
* test/ruby/test_method.rb: add a test.
|
|
|
|
Sat Feb 16 15:45:56 2013 Koichi Sasada <ko1@atdot.net>
|
|
|
|
* proc.c (rb_binding_new_with_cfp): create binding object even if
|
|
the frame is IFUNC. But return a ruby-level binding to keep
|
|
compatibility.
|
|
This patch fix degradation introduced from r39067.
|
|
[Bug #7774] [ruby-dev:46960]
|
|
|
|
* test/ruby/test_settracefunc.rb: add a test.
|
|
|
|
Sat Feb 16 13:40:13 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (shvar_to_cpp): do not substitute exec_prefix itself
|
|
with RUBY_EXEC_PREFIX, which cause recursive definition.
|
|
[ruby-core:52296] [Bug #7860]
|
|
|
|
Sat Feb 16 13:13:04 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* ext/io/console/io-console.gemspec: bump to 0.4.2. now explicitly
|
|
requires ruby 1.9.3 or later. [Bug #7847]
|
|
|
|
* ext/io/console/console.c (console_dev): compatibility with ruby 1.8.
|
|
|
|
* ext/io/console/console.c (rawmode_opt, console_dev): compatibility
|
|
with ruby 1.9. [ruby-core:52220] [Bug #7847]
|
|
|
|
Sat Feb 16 12:45:50 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in: unexpand arch sitearch and exec_prefix values, so
|
|
directly specified bindir, libdir, rubyprefix, etc can be properly
|
|
substituted. [ruby-core:52296] [Bug #7860]
|
|
|
|
Sat Feb 16 12:15:20 2013 Aaron Patterson <aaron@tenderlovemaking.com>
|
|
|
|
* parse.y: add dtrace probe for symbol create.
|
|
|
|
* probes.d: ditto
|
|
|
|
Sat Feb 16 09:27:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: don't test sys/feature_tests.h which is not
|
|
used now.
|
|
It was included in r7901 as "bug of gcc 3.0 on Solaris 8 ?".
|
|
|
|
Sat Feb 16 09:24:37 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: reorder header tests to consider inclusion
|
|
order in rubysocket.h.
|
|
|
|
Sat Feb 16 08:42:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in, ext/socket/extconf.rb: test netinet/in_systm.h in
|
|
ext/socket/extconf.rb instead of configure.in.
|
|
|
|
Originally, netinet/in_systm.h is included for NextStep, OpenStep,
|
|
and Rhapsody. [ruby-core:1596]
|
|
|
|
Sat Feb 16 07:55:40 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: don't test xti.h here.
|
|
|
|
* ext/socket/extconf.rb: test xti.h here.
|
|
|
|
Originally, xti.h is included for IRIX [ruby-core:14447].
|
|
|
|
Sat Feb 16 07:16:49 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: test struct sockaddr_un and its member,
|
|
sun_len.
|
|
|
|
* ext/socket/sockport.h (INIT_SOCKADDR_UN): new macro defined.
|
|
|
|
* ext/socket/socket.c (sock_s_pack_sockaddr_un): use INIT_SOCKADDR_UN.
|
|
|
|
* ext/socket/unixsocket.c (rsock_init_unixsock): ditto.
|
|
|
|
* ext/socket/raddrinfo.c (init_unix_addrinfo): ditto.
|
|
(addrinfo_mload): ditto.
|
|
|
|
Sat Feb 16 07:05:59 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/sockport.h (INIT_SOCKADDR_IN): don't need family
|
|
argument. it is always AF_INET.
|
|
|
|
* ext/socket/raddrinfo.c (make_inetaddr): follow INIT_SOCKADDR_IN
|
|
change.
|
|
(addrinfo_ipv6_to_ipv4): ditto.
|
|
|
|
Sat Feb 16 04:21:07 2013 NAKAMURA Usaku <usa@ruby-lang.org>
|
|
|
|
* ext/socket/extconf.rb: workaround for mswin/mingw build problem.
|
|
sendmsg emulation in win32/win32.c is not enough.
|
|
|
|
Sat Feb 16 00:19:20 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: use all all tested available headers for
|
|
have_func.
|
|
|
|
Fri Feb 15 22:21:37 2013 Akinori MUSHA <knu@iDaemons.org>
|
|
|
|
* configure.in: Fix a bug introduced in r38342 that the cflagspat
|
|
substitution is messed up by the way CFLAGS and optflags are
|
|
modified, which affected FreeBSD and NetBSD/amd64 when
|
|
configured to use libexecinfo. This bug resulted in CFLAGS and
|
|
CXXFLAGS in RbConfig::CONFIG having warnflags expanded in them,
|
|
forcing third-party C/C++ extensions to follow what warnflags
|
|
demands, like ANSI/ISO-C90 conformance. ref [Bug #7101]
|
|
|
|
Fri Feb 15 20:29:11 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/sockport.h (SET_SIN_LEN): defined for strict-aliasing
|
|
rule.
|
|
(INIT_SOCKADDR_IN): ditto.
|
|
|
|
* ext/socket/raddrinfo.c (make_inetaddr): use INIT_SOCKADDR_IN.
|
|
(addrinfo_ipv6_to_ipv4): ditto.
|
|
|
|
Fri Feb 15 18:24:48 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* lib/mkmf.rb (MakeMakefile#try_run): bail out explicitly if cross
|
|
compiling, because it cannot work of course.
|
|
|
|
Fri Feb 15 12:34:58 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: test struct sockaddr_storage directly.
|
|
|
|
* ext/socket/rubysocket.h: use HAVE_TYPE_STRUCT_SOCKADDR_STORAGE.
|
|
|
|
Fri Feb 15 12:26:13 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/getaddrinfo.c (GET_AI): don't cast 1st argument for
|
|
INIT_SOCKADDR.
|
|
|
|
Fri Feb 15 08:12:11 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/sockport.h (SET_SS_LEN): removed.
|
|
(SET_SIN_LEN): removed.
|
|
(INIT_SOCKADDR): new macro.
|
|
|
|
* ext/socket/ancdata.c (extract_ipv6_pktinfo): use INIT_SOCKADDR.
|
|
|
|
* ext/socket/raddrinfo.c (make_inetaddr): use INIT_SOCKADDR.
|
|
(addrinfo_ipv6_to_ipv4): ditto.
|
|
|
|
* ext/socket/getaddrinfo.c (GET_AI): use INIT_SOCKADDR.
|
|
|
|
Fri Feb 15 07:49:27 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rdoc.rb: Update to release version of 4.0.0
|
|
|
|
* lib/rubygems.rb: Update to release version of 2.0.0
|
|
|
|
Fri Feb 15 07:07:27 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/sockport.h (SA_LEN): removed because unused now.
|
|
(SS_LEN): ditto.
|
|
(SIN_LEN): ditto.
|
|
|
|
Thu Feb 14 10:45:31 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* test/ruby/test_process.rb (test_setsid): Added a workaround for
|
|
MacOS X. Patch by nagachika. [Bug #7826] [ruby-core:52126]
|
|
|
|
Fri Feb 15 00:15:31 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/sockport.h (VALIDATE_SOCKLEN): new macro to validate
|
|
sa_len member of 4.4BSD socket address.
|
|
|
|
* ext/socket/getnameinfo.c (getnameinfo): use VALIDATE_SOCKLEN,
|
|
instead of SA_LEN.
|
|
|
|
* ext/socket/socket.c (sock_s_getnameinfo): use VALIDATE_SOCKLEN
|
|
instead of SS_LEN.
|
|
|
|
Thu Feb 14 22:25:54 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/socket.c (sockaddr_len): extracted from sockaddr_obj.
|
|
(sockaddr_obj): add an argument to length of socket address.
|
|
(socket_s_ip_address_list): call sockaddr_obj with actual socket
|
|
address length if given, use sockaddr_len otherwise.
|
|
|
|
Thu Feb 14 20:11:23 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket: always operate length of socket address companion with
|
|
socket address.
|
|
|
|
* ext/socket/rubysocket.h (rsock_make_ipaddr): add an argument for
|
|
socket address length.
|
|
(rsock_ipaddr): ditto.
|
|
|
|
* ext/socket/ipsocket.c (ip_addr): pass length to rsock_ipaddr.
|
|
(ip_peeraddr): ditto.
|
|
(ip_s_getaddress): pass length to rsock_make_ipaddr.
|
|
|
|
* ext/socket/socket.c (make_addrinfo): pass length to rsock_ipaddr.
|
|
(sock_s_getnameinfo): pass actual address length to rb_getnameinfo.
|
|
(sock_s_unpack_sockaddr_in): pass length to rsock_make_ipaddr.
|
|
|
|
* ext/socket/init.c (rsock_s_recvfrom): pass length to rsock_ipaddr.
|
|
(rsock_s_recvfrom_nonblock): ditto.
|
|
|
|
* ext/socket/tcpsocket.c (tcp_sockaddr): pass length to
|
|
rsock_make_ipaddr.
|
|
|
|
* ext/socket/raddrinfo.c (make_ipaddr0): add an argument for socket
|
|
address length. pass the length to rb_getnameinfo.
|
|
(rsock_ipaddr): ditto.
|
|
(rsock_make_ipaddr): add an argument for socket address length.
|
|
pass the length to make_ipaddr0.
|
|
(make_inetaddr): pass length to make_ipaddr0.
|
|
a local variable renamed.
|
|
(host_str): a local variable renamed.
|
|
(port_str): ditto.
|
|
|
|
Thu Feb 14 14:31:43 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/net/http.rb: Removed OpenSSL dependency from Net::HTTP.
|
|
|
|
* test/net/http/test_http.rb: Remove Zlib dependency from tests.
|
|
* test/net/http/test_http_request.rb: ditto.
|
|
|
|
Thu Feb 14 11:08:15 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* class.c (include_modules_at): detect cyclic prepend with original
|
|
method table. [ruby-core:52205] [Bug #7841]
|
|
|
|
Thu Feb 14 10:30:41 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* vm_method.c: call method_removed hook on called class, not on
|
|
prepending iclass. [ruby-core:52207] [Bug #7843]
|
|
|
|
Thu Feb 14 10:05:57 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/net/http: Do not handle Content-Encoding when the user sets
|
|
Accept-Encoding. This allows users to handle Content-Encoding for
|
|
themselves. This restores backwards-compatibility with Ruby 1.x.
|
|
[ruby-trunk - Bug #7831]
|
|
* lib/net/http/generic_request.rb: ditto.
|
|
* lib/net/http/response.rb: ditto
|
|
* test/net/http/test_http.rb: Test for the above.
|
|
* test/net/http/test_http_request.rb: ditto.
|
|
* test/net/http/test_httpresponse.rb: ditto.
|
|
|
|
Thu Feb 14 08:18:47 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: don't define HAVE_SA_LEN and HAVE_SA_LEN.
|
|
use HAVE_STRUCT_SOCKADDR_SA_LEN and HAVE_STRUCT_SOCKADDR_IN_SIN_LEN
|
|
instead.
|
|
|
|
Wed Feb 13 20:59:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: don't define socklen_t here, just test.
|
|
|
|
* ext/socket/rubysocket.h: define socklen_t if not available.
|
|
|
|
Wed Feb 13 18:37:50 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* proc.c (mnew): skip prepending modules and return the method bound
|
|
on the given class. [ruby-core:52160] [Bug #7836]
|
|
|
|
Wed Feb 13 18:11:59 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* proc.c (method_original_name): new methods Method#original_name and
|
|
UnboundMethod#original_name. [ruby-core:52048] [Bug #7806]
|
|
[EXPERIMENTAL]
|
|
|
|
* proc.c (method_inspect): show the given name primarily, and
|
|
original_id if aliased. [ruby-core:52048] [Bug #7806]
|
|
|
|
Wed Feb 13 17:56:39 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (warnflags): disable -Werror by default unless
|
|
development. [ruby-core:52131] [Bug #7830]
|
|
|
|
Wed Feb 13 06:05:52 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems.rb: Return BINARY strings from Gem.gzip and Gem.gunzip.
|
|
Fixes intermittent test failures. RubyGems issue #450 by Jeremey
|
|
Kemper.
|
|
* test/rubygems/test_gem.rb: Test for the above.
|
|
|
|
Wed Feb 13 05:49:21 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: test functions just after struct members.
|
|
|
|
Tue Feb 12 12:02:35 2013 NARUSE, Yui <naruse@ruby-lang.org>
|
|
|
|
* ext/json: merge JSON 1.7.7.
|
|
This includes security fix. [CVE-2013-0269]
|
|
https://github.com/flori/json/commit/d0a62f3ced7560daba2ad546d83f0479a5ae2cf2
|
|
https://groups.google.com/d/topic/rubyonrails-security/4_YvCpLzL58/discussion
|
|
|
|
Mon Feb 11 23:08:48 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: enable rb_cv_page_size_log test for MirOS BSD.
|
|
|
|
Mon Feb 11 20:06:38 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: use -pthread on mirbsd*.
|
|
|
|
Mon Feb 11 16:07:09 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: add SOLIBS and LIBRUBY_SO definition for mirbsd*.
|
|
|
|
Mon Feb 11 13:17:20 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in (rubysitearchprefix): sitearchdir and vendorarchdir
|
|
should use sitearch, not arch. [ruby-dev:46964] [Bug #7823]
|
|
|
|
* win32/Makefile.sub (config.status): site and vendor directories
|
|
should use sitearch, not arch. [ruby-dev:46964] [Bug #7823]
|
|
|
|
Mon Feb 11 12:31:25 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: move OS specific header/function knowledge before
|
|
automatic header tests.
|
|
|
|
Mon Feb 11 11:04:29 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: move the test for -march=i486 just after
|
|
RUBY_UNIVERSAL_ARCH/RUBY_DEFAULT_ARCH.
|
|
|
|
Sun Feb 10 23:42:26 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: test structure members just after types test.
|
|
|
|
Sun Feb 10 20:58:17 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: test types just after headers test.
|
|
|
|
Sun Feb 10 16:00:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/rake/doc/MIT-LICENSE: Add license file from upstream
|
|
* lib/rake/doc/README.rdoc: Link to license file from Rake README
|
|
* lib/rake/version.rb: Include README rdoc for Rake module overview
|
|
|
|
Sun Feb 10 15:26:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/rake/doc/*: Sync Rake rdoc files from upstream
|
|
|
|
Sun Feb 10 15:50:02 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* vm_exec.h (DISPATCH_ARCH_DEPEND_WAY): use __asm__ __volatile__
|
|
instead of asm volatile.
|
|
|
|
Sun Feb 10 15:50:02 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
|
|
|
|
* gc.h (SET_MACHINE_STACK_END): use __volatile__ instead of volatile.
|
|
|
|
Sun Feb 10 14:25:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* doc/rake/, lib/rake/doc/: Move Rake rdoc files to lib/rake
|
|
|
|
Sun Feb 10 12:10:25 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* ext/socket/extconf.rb: test headers at first.
|
|
|
|
Sun Feb 10 12:00:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* doc/rake/*: Removed stale Rake static files
|
|
|
|
Sun Feb 10 09:10:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* lib/pp.rb, lib/prettyprint.rb: Documentation for PP and PrettyPrint
|
|
Based on a patch by Vincent Batts [ruby-core:51253] [Bug #7656]
|
|
|
|
Sat Feb 9 21:11:21 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: move header files check to the beginning of
|
|
"header and library section".
|
|
test rlim_t with sys/types.h and sys/time.h for MirOS BSD.
|
|
sys/types.h and sys/time.h is guarded by #ifdef and the above
|
|
move is required for this change.
|
|
|
|
Sat Feb 9 17:45:58 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in, version.c: prevent duplicated load paths by empty
|
|
version string, it does not work right now.
|
|
|
|
Sat Feb 9 17:38:41 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
|
|
|
|
* configure.in: fix arch parameters in help message. [Bug #7804]
|
|
|
|
Sat Feb 9 13:13:00 2013 Zachary Scott <zachary@zacharyscott.net>
|
|
|
|
* vm_trace.c: Note about TracePoint events set, and comment on
|
|
Kernel#set_trace_func to prefer new TracePoint API
|
|
|
|
Sat Feb 9 10:07:47 2013 Kazuki Tsujimoto <kazuki@callcc.net>
|
|
|
|
* BSDL: update copyright notice for 2013.
|
|
|
|
Sat Feb 9 09:24:38 2013 Eric Hodel <drbrain@segment7.net>
|
|
|
|
* lib/rubygems/package/old.rb: Fix behavior only on ruby 1.8.
|
|
|
|
* lib/rubygems/package.rb: Include checksums.yaml.gz signatures for
|
|
verification.
|
|
* test/rubygems/test_gem_package.rb: Test for the above.
|
|
|
|
Sat Feb 9 01:23:24 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* test/fiddle/helper.rb: specify libc and libm locations for MirOS BSD.
|
|
|
|
* test/dl/test_base.rb: ditto.
|
|
|
|
Fri Feb 8 23:25:33 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: change CFLAGS temporally to test
|
|
ARCH_FLAG="-march=i486".
|
|
|
|
Fri Feb 8 21:19:41 2013 Tanaka Akira <akr@fsij.org>
|
|
|
|
* configure.in: don't define ARCH_FLAG="-march=i486" if it causes
|
|
compilation problem.
|
|
|
|
For the changes before 2.0.0, see doc/ChangeLog-2.0.0
|
|
For the changes before 1.9.3, see doc/ChangeLog-1.9.3
|
|
For the changes before 1.8.0, see doc/ChangeLog-1.8.0
|
|
|
|
Local variables:
|
|
coding: us-ascii
|
|
add-log-time-format: (lambda ()
|
|
(let* ((time (current-time))
|
|
(system-time-locale "C")
|
|
(diff (+ (cadr time) 32400))
|
|
(lo (% diff 65536))
|
|
(hi (+ (car time) (/ diff 65536))))
|
|
(format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t)))
|
|
indent-tabs-mode: t
|
|
tab-width: 8
|
|
change-log-indent-text: 2
|
|
end:
|
|
vim: tabstop=8 shiftwidth=2
|