aboutsummaryrefslogtreecommitdiffstats
path: root/activesupport/test
Commit message (Collapse)AuthorAgeFilesLines
* Partially revert 14b20ceGodfrey Chan2015-11-271-0/+5
| | | | | The failure have returned, so disabling the tests again. (The upgrade is fine to keep.)
* Upgrade to listen 3.0.5, re-enable testsGodfrey Chan2015-11-261-5/+0
| | | | | | | | | | In listen 3.0.4 and below, the `#stop` method on the notification backends are a no-op, meaning that we are leaking them per test. This ended up triggering another bug in Ruby that causes our builds to fail randomly on CI. listen bug: https://github.com/guard/listen/issues/353 ruby bug: https://bugs.ruby-lang.org/issues/11744
* Only run listen tests in isolation, for nowGodfrey Chan2015-11-261-0/+5
| | | | | | There is a known issue with the listen tests that casuses files to be incorrectly GC'ed even when they are still in-use. The current is to only run them in isolation to avoid randomly failing our test suite.
* Remove unused codeWill Jessop2015-11-241-4/+1
| | | | Looking at the history of this file it seems this code was never actually used.
* add deprecations for a smooth transition after #22215Michael Grosser2015-11-191-0/+27
|
* Merge pull request #22215 from grosser/grosser/normalize_keyRafael França2015-11-161-5/+5
|\ | | | | send normalized keys to the cache backends so they do not need to man…
| * send normalized keys to the cache backends so they do not need to manage ↵Michael Grosser2015-11-101-5/+5
| | | | | | | | this themselves
| * Add days_in_year methodJon Pascoe2015-11-101-0/+19
| |
* | Fix instance variable not defined warning from Active Support test suiteJuanitoFatas2015-11-121-1/+1
| | | | | | | | | | | | | | | | | | | | Before ./Users/Juan/dev/rails/activesupport/test/file_update_checker_shared_tests.rb:20: warning: instance variable @tmpdir not initialized After No warnings
* | revises the implementation of Pathname#ascendant_of? (in refinement)Xavier Noria2015-11-111-0/+29
| |
* | Add days_in_year methodJon Pascoe2015-11-111-0/+19
| |
* | modernizes the test definitions in the file monitors suiteXavier Noria2015-11-111-16/+17
| |
* | applies code style guidelinesXavier Noria2015-11-113-7/+7
| |
* | renames the module with shared tests for file monitorsXavier Noria2015-11-103-5/+5
| |
* | the evented monitor filters out descendantsXavier Noria2015-11-101-0/+30
| |
* | better tmp file management in the monitors suiteXavier Noria2015-11-101-36/+88
| |
* | adds another test case in the monitors suiteXavier Noria2015-11-101-1/+17
| |
* | editorial pass over test method namesXavier Noria2015-11-091-5/+5
| |
* | improves waiting in the file monitors suiteXavier Noria2015-11-093-23/+21
| |
* | adds more tests for the file monitorsXavier Noria2015-11-091-0/+60
| |
* | s/watcher/checker/gXavier Noria2015-11-081-3/+3
| | | | | | | | "checker" is the name being used everywhere.
* | more ad-hoc sleepsXavier Noria2015-11-081-2/+4
| | | | | | | | This sucks, but otherwise I get occasional Fs on Mac OS X.
* | create the tmpdir under testXavier Noria2015-11-081-1/+1
| | | | | | | | | | Mac OS X tries by all means to hide that /var is /private/var, and that is what FSEvents reports back.
* | encapsulate sleep margin when touching filesXavier Noria2015-11-081-5/+6
| |
* | renames the monitor creation helperXavier Noria2015-11-083-11/+11
| |
* | refactors the file monitors test suiteXavier Noria2015-11-083-57/+47
| | | | | | | | | | In particular files are no longer created in the current working directory, but in a temporary folder.
* | let listen stop all listeners on teardownXavier Noria2015-11-081-0/+5
| |
* | stop ascending at the longest common subpathXavier Noria2015-11-081-0/+60
| | | | | | | | This commit also bases everything on Pathname internally.
* | remove unused constants in the file monitor suitesXavier Noria2015-11-082-4/+0
| |
* | initial edit pass over the evented file checker patchXavier Noria2015-11-083-41/+67
| |
* | implements an evented file update checker [Puneet Agarwal]Xavier Noria2015-11-083-103/+135
|/ | | | | | | | | | | | This is the implementation of the file update checker written by Puneet Agarwal for GSoC 2015 (except for the tiny version of the listen gem, which was 3.0.2 in the original patch). Puneet's branch became too out of sync with upstream. This is the final work in one single clean commit. Credit goes in the first line using a convention understood by the contrib app.
* Merge pull request #21897 from swaathi/masterKasper Timm Hansen2015-11-073-4/+90
|\ | | | | Parameterize with options to preserve the case of string
| * Parameterize with options to preserve case of stringSwaathi K2015-11-073-4/+90
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Added test cases Using kwargs instead of three seperate functions Updated parameterize in transliterate.rb Updated parameterize in transliterate.rb Added deprecation warnings and updating RDoc+Guide Misspelled separtor. Fixed. Deprecated test cases and added support to parameterize with keyword parameters Squashing commits. Fixed test cases and added deprecated test cases Small changes to Gemfile.lock and CHANGELOG Update Gemfile.lock
* | Merge pull request #22197 from grosser/grosser/fetchKasper Timm Hansen2015-11-071-2/+9
|\ \ | | | | | | do not override fetch on local cache
| * | do not override fetch on local cacheMichael Grosser2015-11-071-2/+9
| |/ | | | | | | | | | | fetch is supposed to behave differently, this was a mistake merged in https://github.com/rails/rails/pull/22194
* | Merge pull request #21806 from kachick/fix-transformer-sizeEileen M. Uchitelle2015-11-072-4/+8
|\ \ | |/ |/| Return a sized Enumerator from Hash#transform_values{!}
| * Return a sized Enumerator from Hash#{transform_values{!}|transform_keys{!}}Kenichi Kamiya2015-09-292-4/+8
| |
* | Merge pull request #22194 from grosser/grosser/read-nilSean Griffin2015-11-051-0/+8
|\ \ | | | | | | cache nil replies from backend cache so misses are fast too
| * | cache nil replies from backend cache so misses are fast tooMichael Grosser2015-11-051-0/+8
| | |
* | | Require only necessary concurrent-ruby classes.Jerry D'Antonio2015-11-041-1/+1
| | |
* | | :scissors:Rafael Mendonça França2015-10-301-1/+0
| | |
* | | Merge pull request #11872 from AvnerCohen/log_namespaceSean Griffin2015-10-301-0/+13
|\ \ \ | | | | | | | | When testing cache issues, it is useful to log the actual key, including namespace
| * | | When logging Cache key, in debug mode, also log namespace, to create the ↵Avner Cohen2015-06-231-0/+13
| | | | | | | | | | | | | | | | full key that actually used by the underline cache implementation
* | | | Deprecate `HWIDA.new_from_hash_copying_default`Sean Griffin2015-10-291-1/+1
| | | | | | | | | | | | | | | | This method was already niche, and is now redundant with `.new`
* | | | Merge pull request #16357 from gchan/hwia-respects-to-hash-defaultSean Griffin2015-10-291-0/+18
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | `HashWithIndifferentAccess.new` respects the default value or proc on objects that respond to `#to_hash`
| * | | | `HashWithIndifferentAccess.new` respects the default value or proc on ↵Gordon Chan2014-07-311-0/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | objects that respond to `#to_hash`. Builds on the work of #12550 where `.new` will convert the object (that respond to `#to_hash`) to a hash and add that hash's keys and values to itself. This change will also make `.new` respect the default value or proc of objects that respond to `#to_hash`. In other words, this `.new` behaves exactly like `.new_from_hash_copying_default`. `.new_from_hash_copying_default` now simply invokes `.new` and any references to `.new_from_hash_copying_default` are replaced with `.new`. Added tests confirm behavior.
* | | | | Make `Module#redefine_method` to keep method visibilityyui-knk2015-10-261-1/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Before this commit `Module#redefine_method` always changes visibility of redefined method to `public`. This commit changes behavior of Module#redefine_method` to keep method visibility.
* | | | | Tweaked wording used in some tests.Sebastian McKenzie2015-10-251-6/+6
| | | | |
* | | | | Add Module#remove_possible_singleton_methodAndrew White2015-10-211-3/+15
| | | | | | | | | | | | | | | | | | | | | | | | | This is primarily to fix method redefinition warnings in class_attribute but may be of use in other places where we define singleton methods.
* | | | | Merge pull request #19889 from cedrics/long-uri-encoded-keysSean Griffin2015-10-201-0/+5
|\ \ \ \ \ | | | | | | | | | | | | FileStore: Long cache keys may result in too long paths due to encoding