| Commit message (Collapse) | Author | Age | Files | Lines |
|\
| |
| |
| | |
Make ActiveSupport frozen-string-literal friendly.
|
| |
| |
| |
| |
| |
| |
| | |
The ActiveSupport test suite only passes currently if it uses the latest unreleased commits for dalli, and a patch for Builder:
https://github.com/tenderlove/builder/pull/6
Beyond that, all external dependencies (at least, to the extent they’re used by ActiveSupport) are happy, including Nokogiri as of 1.8.0.
|
| | |
|
|\ \
| | |
| | | |
Remove useless class checking for `ActiveSupport::Callbacks`s result_lambda
|
| |/ |
|
| | |
|
|/
|
|
| |
This deprecation warning message will be more useful if it indicates what the string was doing — being eval'd — and what the non-deprecated options for callback conditionals are.
|
|\
| |
| | |
Log subscriber - avoid rescuing certain exceptions
|
| | |
|
| | |
|
| |
| |
| |
| | |
Follow up of #29263
|
|\ \
| | |
| | | |
Default Message Encryptor Cipher to AES-256-GCM From AES-256-CBC
|
| | |
| | |
| | |
| | | |
- Introduce a method to select default cipher, and maintain backward compatibility
|
| | |
| | |
| | |
| | |
| | |
| | | |
No need to pass `#cleanup` options through to `LocalCache#clear`.
Fixes #29081. References #25628.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
See https://github.com/rails/rails/issues/29067#issuecomment-301342084
for rationale.
This reverts commit b76f82d714e590c20370e72fa36fa574c4f17650.
Fixes #29067. Fixes #29081.
|
|\ \ \
| | | |
| | | | |
Remove unreachable code
|
| |/ /
| | |
| | |
| | |
| | |
| | |
| | | |
`Time.find_zone!` raise `ArgumentError` if invalid value is specified.
https://github.com/rails/rails/blob/379a0b42daf0d8e14130db7fd886d05d8d88e3f2/activesupport/lib/active_support/core_ext/time/zones.rb#L97..L99
Therefore, the return value never becomes nil.
|
|/ /
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Rails.cache.write_multi foo: 'bar', baz: 'qux'
Plus faster `fetch_multi` with stores that implement `write_multi_entries`.
Keys that aren't found may be written to the cache store in one shot
instead of separate writes.
The default implementation simply calls `write_entry` for each entry.
Stores may override if they're capable of one-shot bulk writes, like
Redis `MSET`.
|
|\ \
| | |
| | | |
Introduce mattr_accessor default option
|
| | | |
|
| | | |
|
|/ / |
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
* Allow a default value to be declared for class_attribute
* Convert to using class_attribute default rather than explicit setter
* Removed instance_accessor option by mistake
* False is a valid default value
* Documentation
|
|\ \
| | |
| | | |
round before calculating exponent in number_to_human_converter
|
| | |
| | |
| | |
| | | |
fixes #25664
|
| | | |
|
|\ \ \
| | | |
| | | |
| | | |
| | | | |
kamipo/remove_returning_true_in_internal_callbacks
Remove returning true in internal callbacks
|
| | | |
| | | |
| | | |
| | | |
| | | | |
`display_deprecation_warning_for_false_terminator` was removed since
3a25cdc.
|
|\ \ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
EilisHamilton/fix_uncountable_pluralization_locale
Fix pluralization of uncountables when given a locale
|
| | |_|/
| |/| |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Previously apply_inflections would only use the :en uncountables
rather then the ones for the locale that was passed to pluralize or
singularize.
This changes apply_inflections to take a locale which it will use to
find the uncountables.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
If users added an attribute or otherwise changed a CurrentAttributes subclass
they'd see exceptions on the next page load.
Because `ActiveSupport::CurrentAttributes.current_instances` would keep
references to the old instances from the previous request.
We can fix this by clearing out the `current_attributes` before we unload
constants. Then any change to the model can be autoloaded again since its
slot isn't taken by an old instance.
We'll still have to call reset before we clear so external collaborators,
like Time.zone, won't linger with their current value throughout other code.
|
| | | |
| | | |
| | | |
| | | |
| | | | |
Since we're generating a key through the class name we can combine
the two Thread.current calls into a single hash version.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
`find` raises when it can't find a record, so we'll never reach the
else. Switch to `find_by` which returns nil when no record can be
found.
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
singleton (#29180)
* Add ActiveSupport::CurrentAttributes to provide a thread-isolated attributes singleton
* Need to require first
* Move stubs into test namespace.
Thus they won't conflict with other Current and Person stubs.
* End of the line for you, whitespace!
* Support super in attribute methods.
Define instance level accessors in an included module such that
`super` in an overriden accessor works, akin to Active Model.
* Spare users the manual require.
Follow the example of concerns, autoload in the top level Active Support file.
* Add bidelegation support
* Rename #expose to #set. Simpler, clearer
* Automatically reset every instance.
Skips the need for users to actively embed something that resets
their CurrentAttributes instances.
* Fix test name; add tangible name value when blank.
* Try to ensure we run after a request as well.
* Delegate all missing methods to the instance
This allows regular `delegate` to serve, so we don't need bidelegate.
* Properly test resetting after execution cycle.
Also remove the stale puts debugging.
* Update documentation to match new autoreset
|
|\ \ \ \
| | | | |
| | | | | |
Define path with __dir__
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
".. with __dir__ we can restore order in the Universe." - by @fxn
Related to 5b8738c2df003a96f0e490c43559747618d10f5f
|
|/ / / / |
|
|\ \ \ \
| | | | |
| | | | | |
Fix implicit calculations with scalars and durations
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Previously calculations where the scalar is first would be converted
to a duration of seconds but this causes issues with dates being
converted to times, e.g:
Time.zone = "Beijing" # => Asia/Shanghai
date = Date.civil(2017, 5, 20) # => Mon, 20 May 2017
2 * 1.day # => 172800 seconds
date + 2 * 1.day # => Mon, 22 May 2017 00:00:00 CST +08:00
Now the `ActiveSupport::Duration::Scalar` calculation methods will try
to maintain the part structure of the duration where possible, e.g:
Time.zone = "Beijing" # => Asia/Shanghai
date = Date.civil(2017, 5, 20) # => Mon, 20 May 2017
2 * 1.day # => 2 days
date + 2 * 1.day # => Mon, 22 May 2017
Fixes #29160, #28970.
|
|/ / / / |
|
|/ / /
| | |
| | |
| | |
| | |
| | | |
This retains the existing behavior of
ActiveSupport::Cache.expand_cache_key (as used by etaging) where the
cache key includes the version.
|
| | | |
|
|\ \ \
| | | |
| | | |
| | | |
| | | | |
Message encryptor auth tag check
Fixes MessageEncryptor when used in AEAD mode. Specifically, we need to check if the `auth_tag` is nil. This may arise when an AEAD encryptor is used to decrypt a ciphertext generated from a different mode, such as CBC-HMAC. Basically, the number of double dashes will differ and `auth_tag` may be nil in this case.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
When MessageEncryptor tries to +decrypt_and_verify+ ciphertexts
generated in a different mode (such CBC-HMAC), the +auth_tag+ may be
+nil+ and must explicitly check for it.
See the discussion here:
https://github.com/rails/rails/pull/28132#discussion_r116388462
|
| |/ /
|/| |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Some cache backends support the `unless_exist` option, which tells them
not to overwrite an existing entry. The local cache currently always
stores the new value, even though the backend may have rejected it.
Since we can't tell which value will end up in the backend cache, we
should delete the key from the local cache, so that the next read for
that key will go to the backend and pick up the correct value.
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
* Fix rdoc code formatting — `tt`, not backticks
* Fix/simplify sentence grammar — should at least just be “and the like”, not “likes”, but this is just general tightening up.
* Add note that delegated methods must be public. Tested here: https://github.com/rails/rails/blob/7ff5ccae94ce2aff76b5f8a31a28e305a047d642/activesupport/test/core_ext/module_test.rb#L359-L365
* Simplify example code for delegate_missing_to. The example had complexity that wasn’t necessary for demonstrating `delegate_missing_to`. This gets rid of a bunch of cruft so the example is more obvious about what’s going on regarding the feature itself.
[ci skip]
|