All of the behavior :environment was giving (that db:schema:load needed)
was provided as well with :load_config.
This will address an issue introduced in
https://github.com/rails/rails/pull/15394. The fact that db:schema:load
now drops and creates the database causes the Octopus gem to have [an
issue](https://github.com/tchandy/octopus/issues/273) during the drop
step for the test database (which wasn't happening in db:schema:load
before). The error looks like:
ActiveRecord::StatementInvalid: PG::ObjectInUse: ERROR: cannot drop the currently open database
: DROP DATABASE IF EXISTS "app_test"
Because of the timing, this issue is present in master, 4-2-*, and
4.1.8.
A note to forlorn developers who might see this: "Additionally" in a
commit message means you should have a separate commit, with a separate
justification for changes. Small commits with big messages are your
friends.
This `# :nodoc:` had the effect of hiding every method that follows.
This meant that the API page for `ActiveRecord::Core` only contained
`configurations` and none of the following methods.
Furthermore this `# :nodoc:` had no effect on `maintain_test_schema`.
Those `mattr_accessor` inside the `included` block are not picked up
by rdoc.
/cc @zzak
When running the following migration:
change_table(:table_name) { |t| t/timestamps }
The following error was produced:
wrong number of arguments (2 for 1) .... /connection_adapters/abstract/schema_statements.rb:851:in `remove_timestamps'
This is due to `arguments` containing an empty hash as its second
argument.
[ci skip]
This is due to the fact that `.delete` is directly translated to SQL.
It tries to follow the same rules as `.delete_all` which is not able
to verify that records are `#readonly?`.
This ensures that we're handling all forms of nested tables the same way.
We're aware that the `convert_dot_notation_to_hash` method will cause a
performance hit, and we intend to come back to it once we've refactored some of
the surrounding code.
[Melissa Xie & Melanie Gilman]
The detection of in-place changes caused a weird unexpected issue with
numericality validations. That validator (out of necessity) works on the
`_before_type_cast` version of the attribute, since on an `:integer`
type column, a non-numeric string would type cast to 0.
However, strings are mutable, and we changed strings to ensure that the
post type cast version of the attribute was a different instance than
the before type cast version (so the mutation detection can work
properly).
Even though strings are the only mutable type for which a numericality
validation makes sense, special casing strings would feel like a strange
change to make here. Instead, we can make the assumption that for all
mutable types, we should work on the post-type-cast version of the
attribute, since all cases which would return 0 for non-numeric strings
are immutable.
Fixes#17852
We never actually make use of it on the table, since we're constructing
the select manager manually. It looks like if we ever actually were
grabbing it from the table, we're grossly misusing it since it's meant
to vary by AR class.
Its existence on `Arel::Table` appears to be purely for convenience
methods that are never used outside of tests. However, in production
code it just complicates construction of the tables on the rails side,
and the plan is to remove it from `Arel::Table` entirely. I'm not
convinced it needs to live on `SelectManager`, etc either.
This fixes a regression where preloading association throws an
exception if one of the associations in the preloading hash doesn't
exist for one record.
Fixes#16070
Since 3e30c5d, it started ignoring the given error message. This commit
changes the behavior of AR::RecordNotSaved#initialize so that it no
longer loses the given error message.