1
0
Fork 0
mirror of https://github.com/ruby/ruby.git synced 2022-11-09 12:17:21 -05:00

string.c: Added option parsing/checking for upcase/downcase/

capitalize/swapcase (with Kimihito Matsui

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53503 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This commit is contained in:
duerst 2016-01-12 07:03:31 +00:00
parent 3a7c12c21a
commit 2788cd9849
2 changed files with 98 additions and 36 deletions

View file

@ -1,3 +1,8 @@
Mon Jan 12 16:03:03 2016 Martin Duerst <duerst@it.aoyama.ac.jp>
* string.c: Added option parsing/checking for upcase/downcase/
capitalize/swapcase (with Kimihito Matsui)
Mon Jan 11 21:28:28 2016 Martin Duerst <duerst@it.aoyama.ac.jp>
* include/ruby/oniguruma.h: Added flags needed for upcase/downcase

129
string.c
View file

@ -171,6 +171,9 @@ str_make_independent(VALUE str)
str_make_independent_expand((str), len, 0L, termlen);
}
/* symbols for [up|down|swap]case/capitalize options */
static VALUE sym_ascii, sym_turkic, sym_lithuanian, sym_fold;
static rb_encoding *
get_actual_encoding(const int encidx, VALUE str)
{
@ -5589,6 +5592,46 @@ rb_str_check_dummy_enc(rb_encoding *enc)
}
}
static OnigCaseFoldType
check_case_options(int argc, VALUE *argv, OnigCaseFoldType flag)
{
if (argc==0)
return flag;
if (argc>2)
rb_raise(rb_eArgError, "too many options");
if (argv[0]==sym_turkic) {
flag &= ONIGENC_CASE_FOLD_TURKISH_AZERI;
if (argc==2) {
if (argv[1]==sym_lithuanian)
flag &= ONIGENC_CASE_FOLD_LITHUANIAN;
else
rb_raise(rb_eArgError, "invalid second option");
}
}
else if (argv[0]==sym_lithuanian) {
flag &= ONIGENC_CASE_FOLD_LITHUANIAN;
if (argc==2) {
if (argv[1]==sym_turkic)
flag &= ONIGENC_CASE_FOLD_TURKISH_AZERI;
else
rb_raise(rb_eArgError, "invalid second option");
}
}
else if (argc>1)
rb_raise(rb_eArgError, "too many options");
else if (argv[0]==sym_ascii)
flag &= ONIGENC_CASE_ASCII_ONLY;
else if (argv[0]==sym_fold) {
if ((flag & (ONIGENC_CASE_UPCASE|ONIGENC_CASE_DOWNCASE)) == ONIGENC_CASE_DOWNCASE)
flag &= ONIGENC_CASE_FOLD;
else
rb_raise(rb_eArgError, "option :fold only allowed for downcasing");
}
else
rb_raise(rb_eArgError, "invalid option");
return flag;
}
/*
* call-seq:
* str.upcase! -> str or nil
@ -5599,13 +5642,15 @@ rb_str_check_dummy_enc(rb_encoding *enc)
*/
static VALUE
rb_str_upcase_bang(VALUE str)
rb_str_upcase_bang(int argc, VALUE *argv, VALUE str)
{
rb_encoding *enc;
char *s, *send;
int modify = 0;
int n;
OnigCaseFoldType flag = ONIGENC_CASE_UPCASE;
flag = check_case_options(argc, argv, flag);
str_modify_keep_cr(str);
enc = STR_ENC_GET(str);
rb_str_check_dummy_enc(enc);
@ -5664,10 +5709,10 @@ rb_str_upcase_bang(VALUE str)
*/
static VALUE
rb_str_upcase(VALUE str)
rb_str_upcase(int argc, VALUE *argv, VALUE str)
{
str = rb_str_dup(str);
rb_str_upcase_bang(str);
rb_str_upcase_bang(argc, argv, str);
return str;
}
@ -5682,12 +5727,14 @@ rb_str_upcase(VALUE str)
*/
static VALUE
rb_str_downcase_bang(VALUE str)
rb_str_downcase_bang(int argc, VALUE *argv, VALUE str)
{
rb_encoding *enc;
char *s, *send;
int modify = 0;
OnigCaseFoldType flag = ONIGENC_CASE_DOWNCASE;
flag = check_case_options(argc, argv, flag);
str_modify_keep_cr(str);
enc = STR_ENC_GET(str);
rb_str_check_dummy_enc(enc);
@ -5747,10 +5794,10 @@ rb_str_downcase_bang(VALUE str)
*/
static VALUE
rb_str_downcase(VALUE str)
rb_str_downcase(int argc, VALUE *argv, VALUE str)
{
str = rb_str_dup(str);
rb_str_downcase_bang(str);
rb_str_downcase_bang(argc, argv, str);
return str;
}
@ -5770,14 +5817,17 @@ rb_str_downcase(VALUE str)
*/
static VALUE
rb_str_capitalize_bang(VALUE str)
rb_str_capitalize_bang(int argc, VALUE *argv, VALUE str)
{
rb_encoding *enc;
char *s, *send;
int modify = 0;
unsigned int c;
int n;
OnigCaseFoldType flag = ONIGENC_CASE_UPCASE |
ONIGENC_CASE_TITLECASE | ONIGENC_CASE_ONCEONLY;
flag = check_case_options(argc, argv, flag);
str_modify_keep_cr(str);
enc = STR_ENC_GET(str);
rb_str_check_dummy_enc(enc);
@ -5818,10 +5868,10 @@ rb_str_capitalize_bang(VALUE str)
*/
static VALUE
rb_str_capitalize(VALUE str)
rb_str_capitalize(int argc, VALUE *argv, VALUE str)
{
str = rb_str_dup(str);
rb_str_capitalize_bang(str);
rb_str_capitalize_bang(argc, argv, str);
return str;
}
@ -5836,13 +5886,15 @@ rb_str_capitalize(VALUE str)
*/
static VALUE
rb_str_swapcase_bang(VALUE str)
rb_str_swapcase_bang(int argc, VALUE *argv, VALUE str)
{
rb_encoding *enc;
char *s, *send;
int modify = 0;
int n;
OnigCaseFoldType flag = ONIGENC_CASE_UPCASE | ONIGENC_CASE_DOWNCASE;
flag = check_case_options(argc, argv, flag);
str_modify_keep_cr(str);
enc = STR_ENC_GET(str);
rb_str_check_dummy_enc(enc);
@ -5881,10 +5933,10 @@ rb_str_swapcase_bang(VALUE str)
*/
static VALUE
rb_str_swapcase(VALUE str)
rb_str_swapcase(int argc, VALUE *argv, VALUE str)
{
str = rb_str_dup(str);
rb_str_swapcase_bang(str);
rb_str_swapcase_bang(argc, argv, str);
return str;
}
@ -9257,54 +9309,54 @@ sym_empty(VALUE sym)
/*
* call-seq:
* sym.upcase -> symbol
* sym.upcase [options] -> symbol
*
* Same as <code>sym.to_s.upcase.intern</code>.
*/
static VALUE
sym_upcase(VALUE sym)
sym_upcase(int argc, VALUE *argv, VALUE sym)
{
return rb_str_intern(rb_str_upcase(rb_sym2str(sym)));
return rb_str_intern(rb_str_upcase(argc, argv, rb_sym2str(sym)));
}
/*
* call-seq:
* sym.downcase -> symbol
* sym.downcase [options] -> symbol
*
* Same as <code>sym.to_s.downcase.intern</code>.
*/
static VALUE
sym_downcase(VALUE sym)
sym_downcase(int argc, VALUE *argv, VALUE sym)
{
return rb_str_intern(rb_str_downcase(rb_sym2str(sym)));
return rb_str_intern(rb_str_downcase(argc, argv, rb_sym2str(sym)));
}
/*
* call-seq:
* sym.capitalize -> symbol
* sym.capitalize [options] -> symbol
*
* Same as <code>sym.to_s.capitalize.intern</code>.
*/
static VALUE
sym_capitalize(VALUE sym)
sym_capitalize(int argc, VALUE *argv, VALUE sym)
{
return rb_str_intern(rb_str_capitalize(rb_sym2str(sym)));
return rb_str_intern(rb_str_capitalize(argc, argv, rb_sym2str(sym)));
}
/*
* call-seq:
* sym.swapcase -> symbol
* sym.swapcase [options] -> symbol
*
* Same as <code>sym.to_s.swapcase.intern</code>.
*/
static VALUE
sym_swapcase(VALUE sym)
sym_swapcase(int argc, VALUE *argv, VALUE sym)
{
return rb_str_intern(rb_str_swapcase(rb_sym2str(sym)));
return rb_str_intern(rb_str_swapcase(argc, argv, rb_sym2str(sym)));
}
/*
@ -9423,15 +9475,20 @@ Init_String(void)
rb_define_method(rb_cString, "inspect", rb_str_inspect, 0);
rb_define_method(rb_cString, "dump", rb_str_dump, 0);
rb_define_method(rb_cString, "upcase", rb_str_upcase, 0);
rb_define_method(rb_cString, "downcase", rb_str_downcase, 0);
rb_define_method(rb_cString, "capitalize", rb_str_capitalize, 0);
rb_define_method(rb_cString, "swapcase", rb_str_swapcase, 0);
sym_ascii = ID2SYM(rb_intern("ascii"));
sym_turkic = ID2SYM(rb_intern("turkic"));
sym_lithuanian = ID2SYM(rb_intern("lithuanian"));
sym_fold = ID2SYM(rb_intern("fold"));
rb_define_method(rb_cString, "upcase!", rb_str_upcase_bang, 0);
rb_define_method(rb_cString, "downcase!", rb_str_downcase_bang, 0);
rb_define_method(rb_cString, "capitalize!", rb_str_capitalize_bang, 0);
rb_define_method(rb_cString, "swapcase!", rb_str_swapcase_bang, 0);
rb_define_method(rb_cString, "upcase", rb_str_upcase, -1);
rb_define_method(rb_cString, "downcase", rb_str_downcase, -1);
rb_define_method(rb_cString, "capitalize", rb_str_capitalize, -1);
rb_define_method(rb_cString, "swapcase", rb_str_swapcase, -1);
rb_define_method(rb_cString, "upcase!", rb_str_upcase_bang, -1);
rb_define_method(rb_cString, "downcase!", rb_str_downcase_bang, -1);
rb_define_method(rb_cString, "capitalize!", rb_str_capitalize_bang, -1);
rb_define_method(rb_cString, "swapcase!", rb_str_swapcase_bang, -1);
rb_define_method(rb_cString, "hex", rb_str_hex, 0);
rb_define_method(rb_cString, "oct", rb_str_oct, 0);
@ -9538,10 +9595,10 @@ Init_String(void)
rb_define_method(rb_cSymbol, "empty?", sym_empty, 0);
rb_define_method(rb_cSymbol, "match", sym_match, 1);
rb_define_method(rb_cSymbol, "upcase", sym_upcase, 0);
rb_define_method(rb_cSymbol, "downcase", sym_downcase, 0);
rb_define_method(rb_cSymbol, "capitalize", sym_capitalize, 0);
rb_define_method(rb_cSymbol, "swapcase", sym_swapcase, 0);
rb_define_method(rb_cSymbol, "upcase", sym_upcase, -1);
rb_define_method(rb_cSymbol, "downcase", sym_downcase, -1);
rb_define_method(rb_cSymbol, "capitalize", sym_capitalize, -1);
rb_define_method(rb_cSymbol, "swapcase", sym_swapcase, -1);
rb_define_method(rb_cSymbol, "encoding", sym_encoding, 0);