Reverting for now in the sake of productivity and restoration of
the passing unit test status of extensions' master branches.
Would recommend re-applying when the CI gate extensions are
passing and ideally some notification ahead of time to wikitech-l.
This reverts commit 0d3712dd60.
Change-Id: I9142e6161257459cd11db0ca4cdd57793cfee8fa
It's not perfect since module i18n can vary on configuration, but it
should catch many cases of forgotten i18n.
Change-Id: Ida7e5720fe304c5168fe752b1b5485fe4cb2cb45
Depends-On: I4a0a6c47afdd62e1c9d0b29f066d2a6d1791b52d
Depends-On: I378d112a36c0981ac8d5366e3e1805ead71a85c8
Depends-On: If9e1ae5f1e360385e70bd68ec9847230bd274d9d
Depends-On: If2694de72a62e46cfdc29401c4d2d3756897c998
* Remove ResourceLoaderFileModule::getLessCompiler(). There is no reason for a
module to need to get a compiler in a different manner than
ResourceLoader::getLessCompiler().
* Add ResourceLoaderModule::getLessVars(). This method provides a means for
subclasses to easily inject custom LESS variables. The default implementation
simply returns an empty array.
* Make the $context parameter for ResourceLoaderFileModule::readStyleFiles()
non-optional (via graceful deprecation). The only callers I found were either
already calling it with a ResourceLoader context, or had a perfectly usable
ResourceLoaderContext in local scope.
* Make ResourceLoaderFileModule::{readStyleFile,getLessCompiler} require a
context. These methods are protected, so I didn't bother with a deprecation.
* Call ksort() on the LESS variables array in the only place it matters -- when
hashing its serialized representation to construct a cache lookup key. This
relieves getLessVars() subclasses from having to remember to re-sort the
variables array if they modify it.
* These changes make it possible to substantially simplify
ResourceLoaderEditToolbarModule, because the only thing it needs to do now is
implement its own getLessVars() method.
* This also allows it to be versioned like any other ResourceLoaderFileModule,
rather than having to use enableModuleContentVersion().
Change-Id: Ic3eab71691e502bfe19bdf4eb6f82cc679a7782f
Change I5a3cdeb7d53485f161ccf8133e76850cdf5b4579 introduced tests
for existence of all local files referenced in CSS. It's currently
failing changes even when the referenced files have been commented
out. This change ensures that commented-out files are not included
in the testing list.
Change-Id: I2ad74c13a9cf36e96a7d7451bf229856bcd18422
The way this is implemented is really dirty... but it found us
a few pre-existing bugs already (T111518, T111519, T111771).
I think it might be worth it.
* CSSMin: Add new method getAllLocalFileReferences() which skips the
file_exists() check.
* ResourceLoaderModule: Make use of it to track missing files too.
* ResourcesTest: Verify that the missing files are missing.
Change-Id: I5a3cdeb7d53485f161ccf8133e76850cdf5b4579
Send the ResourceLoaderContext param in calls to getDependencies() in
PHPUnit tests for ResourceLoader. (This so an RL module in CentralNotice
can emit a warning when that parameter is not provided, and still pass
tests. Sending the parameter here will also be necessary when the parameter
is made non-optional in the future.)
Change-Id: I6033e03e2d8a46b4bd7f7a04ae13fd9f387ebd10
- Removed space after casts
- Removed spaces in array index
- Added spaces around string concat
- Added space after words: switch, foreach
- else if -> elseif
- Removed parentheses around require_once, because it is not a function
- Added newline at end of file
- Removed double spaces
- Added spaces around operations
- Removed repeated newlines
Bug: T102609
Change-Id: Ib860222b24f8ad8e9062cd4dc42ec88dc63fb49e
wfSuppressWarnings() and wfRestoreWarnings() were split out into a
separate library. All usages in core were replaced with the new
functions, and the wf* global functions are marked as deprecated.
Additionally, some uses of @ were replaced due to composer's autoloader
being loaded even earlier.
Ie1234f8c12693408de9b94bf6f84480a90bd4f8e adds the library to
mediawiki/vendor.
Bug: T100923
Change-Id: I5c35079a0a656180852be0ae6b1262d40f6534c4
Modules now track their version via getVersionHash() instead of getModifiedTime().
== Background ==
While some resources have observeable timestamps (e.g. files stored on disk),
many other resources do not. E.g. config variables, and module definitions.
For static file modules, one can e.g. revert one of more files in a module to a
previous version and not affect the max timestamp.
Wiki modules include pages only if they exist. The user module supports common.js
and skin.js. By default neither exists. If a user has both, and then the
less-recently modified one is deleted, the max-timestamp remains unchanged.
For client-side caching, batch requests use "Math.max" on the relevant timestamps.
Again, if a module changes but another module is more recent (e.g. out-of-order
deployment, or out-of-order discovery), the change would not result in a cache miss.
More scenarios can be found in the associated Phabricator tasks.
== Version hash ==
Previously we virtually mapped these variables to a timestamp by storing the current
time alongside a hash of the value in ObjectCache. Considering the number of
possible request contexts (wikis * modules * users * skins * languages) this doesn't
work well. It results in needless cache invalidation when the first time observation
is purged due to LRU algorithms. It also has other minor bugs leading to fewer
cache hits.
All modules automatically get the benefits of version hashing with this change.
The old getDefinitionMtime() and getHashMtime() have been replaced with dummies
that return 1. These functions are often called from getModifiedTime() in subclasses.
For backward-compatibility, their respective values (definition summary and hash)
are now included in getVersionHash directly.
As examples, the following modules have been updated to use getVersionHash directly.
Other modules still work fine and can be updated later.
* ResourceLoaderFileModule
* ResourceLoaderEditToolbarModule
* ResourceLoaderStartUpModule
* ResourceLoaderWikiModule
The presence of hashes in place of timestamps increases the startup module size on
a default MediaWiki install from 4.4k to 5.8k (after gzip and minification).
== ETag ==
Since timestamps are no longer tracked, we need a different way to implement caching
for cache proxies (e.g. Varnish) and web browsers. Previously we used the
Last-Modified header (in combination with Cache-Control and Expires).
Instead of Last-Modified (and If-Modified-Since), we use ETag (and If-None-Match).
Entity tags (new in HTTP/1.1) are much stricter than Last-Modified by default.
They instruct browsers to allow usage of partial Range requests. Since our responses
are dynamically generated, we need to use the Weak version of ETag.
While this sounds bad, it's no different than Last-Modified. As reassured by
RFC 2616 <http://www.w3.org/Protocols/rfc2616/rfc2616-sec13.html#sec13.3.3> the
specified behaviour behind Last-Modified follows the same "Weak" caching logic as
Entity tags. It's just that entity tags are capable of a stricter mode (whereas
Last-Modified is inherently weak).
== File cache ==
If $wgUseFileCache is enabled, ResourceLoader uses ResourceFileCache to cache
load.php responses. While the blind TTL handling (during the allowed expiry period)
is still maxage/timestamp based, tryRespondNotModified() now requires the caller to
know the expected ETag.
For this to work, the FileCache handling had to be moved from the top of
ResoureLoader::respond() to after the expected ETag is computed.
This also allows us to remove the duplicate tryRespondNotModified() handling since
that's is already handled by ResourceLoader::respond() meanwhile.
== Misc ==
* Remove redundant modifiedTime cache in ResourceLoaderFileModule.
* Change bugzilla references to Phabricator.
* Centralised inclusion of wgCacheEpoch using getDefinitionSummary. Previously this
logic was duplicated in each place the modified timestamp was used.
* It's easy to forget calling the parent class in getDefinitionSummary().
Previously this method only tracked 'class' by default. As such, various
extensions hardcoded that one value instead of calling the parent and extending
the array. To better prevent this in the future, getVersionHash() now asserts
that the '_cacheEpoch' property made it through.
* tests: Don't use getDefinitionSummary() as an API.
Fix ResourceLoaderWikiModuleTest to call getPages properly.
* In tests, the default timestamp used to be 1388534400000 (which is the unix time
of 20140101000000; the unit tests' CacheEpoch). The new version hash of these
modules is "XyCC+PSK", which is the base64 encoded prefix of the SHA1 digest of:
'{"_class":"ResourceLoaderTestModule","_cacheEpoch":"20140101000000"}'
* Add sha1.js library for client-side hash generation.
Compared various different implementations for code size (after minfication/gzip),
and speed (when used for short hexidecimal strings).
https://jsperf.com/sha1-implementations
- CryptoJS <https://code.google.com/p/crypto-js/#SHA-1> (min+gzip: 2.5k)
http://crypto-js.googlecode.com/svn/tags/3.1.2/build/rollups/sha1.js
Chrome: 45k, Firefox: 89k, Safari: 92k
- jsSHA <https://github.com/Caligatio/jsSHA>
https://github.com/Caligatio/jsSHA/blob/3c1d4f2e/src/sha1.js (min+gzip: 1.8k)
Chrome: 65k, Firefox: 53k, Safari: 69k
- phpjs-sha1 <https://github.com/kvz/phpjs> (RL min+gzip: 0.8k)
https://github.com/kvz/phpjs/blob/1eaab15d/functions/strings/sha1.js
Chrome: 200k, Firefox: 280k, Safari: 78k
Modern browsers implement the HTML5 Crypto API. However, this API is asynchronous,
only enabled when on HTTPS in Chromium, and is quite low-level. It requires boilerplate
code to actually use with TextEncoder, ArrayBuffer and Uint32Array. Due this being
needed in the module loader, we'd have to load the fallback regardless. Considering
this is not used in a critical path for performance, it's not worth shipping two
implementations for this optimisation.
May also resolve:
* T44094
* T90411
* T94810
Bug: T94074
Change-Id: Ibb292d2416839327d1807a66c78fd96dac0637d0
* Remove redundant @licence/@license from test suite files.
They already have full licence headers. And @licence raises a
warning in Doxygen.
* Fix weird messes of comments inside comments and other things.
Change-Id: I38da8ca76330f72b8dc22b0ecf1ea69d5ea55ede
Avoids:
1) AutoLoaderTest::testWrongCaseClass
Failed asserting that false is true.
2) AutoLoaderTest::testWrongCaseSerializedClass
unserialize() can load classes case-insensitively.
Failed asserting that true is false.
Change-Id: Ic5c271906fb9b2e27639aa8ecfc04ffe6788d3eb
When a dependency is missing the test
ResourcesTest::testUnsatisfiableDependencies gives a error because of
Undefined index and stops further checks. Skip this situation because
the existing test ResourcesTest::testMissingDependencies already gives a
failure with a better explanation in this situation.
Change-Id: I71a88551df67d52ea96bf7283b5de892692f9336
Previously, it just failed with "No such file or directory" without explaining
what's wrong where why.
Change-Id: I776e8fe61f74fd21295da48a45c86ffed9489e23
This change has not been advertised and cause random extensions
to start failing giving little clue to users.
Please have this change discussed on wikitech-l and announced
ahead of time to minimize disruptions. It brokes MathSearch at least:
https://gerrit.wikimedia.org/r/#/c/189442/
This reverts commit d333cd8f7b.
Change-Id: If9a28b1386fca946d8ba351c16134cdf7da8a148
Because extensions often don't add their rights to $wgAvailableRights
or via the 'UserGetAllRights' hook, thus User::getAllRights is incomplete.
Change-Id: Id9ae9eff71e822ec5c038c1c1f990ac36f05cea8
The newly introduced $wgResourceModuleSkinStyles global enables skins to
provide additional stylesheets to existing ResourceLoader module.
This both makes it easier (or at all possible) to override default
styles and lowers the style footprint by making it possible not to
load styles unused on most pages.
----
Example:
Use the file 'foo-styles.css' for the 'mediawiki.foo' module when using
the MySkin skin:
$wgResourceModuleSkinStyles['myskin'] = array(
'mediawiki.foo' => 'foo-styles.css',
'remoteSkinPath' => 'MySkin',
'localBasePath' => __DIR__,
);
For detailed documentation, see the doc comment in DefaultSettings.php.
For a practical usage example, see Vector.php.
----
Implementation notes:
* The values defined in $wgResourceModuleSkinStyles are embedded into
the modules as late as possible (in ResourceLoader::register()).
* Only plain file modules are supported, setting module skin styles
for other module types has no effect.
* ResourceLoader and ResourceLoaderFileModule now support loading
files from arbitrary paths to make this possible, defined using
ResourceLoaderFilePath objects.
* This required some adjustments in seemingly unrelated places for
code which didn't handle the paths fully correctly before.
* ResourceLoader and ResourceLoaderFileModule are now a bit more
tightly coupled than before :(
* Included a tiny example change for the Vector skin, a lot more of
similar cleanup is possible and planned for the future.
* Many of the non-essential mediawiki.* modules defined in
Resources.php should be using `'skinStyles' => array( 'default' => … )`
instead of `'styles' => …` to allow more customizations, this is
also planned for the future after auditing which ones would actually
benefit from this.
Change-Id: Ica4ff9696b490e35f60288d7ce1295766c427e87
- Removed spaces after not operator (!)
- Removed spaces inside array index
- use tab as indent instead of spaces
- Add newline at end of file
- Removed spaces after casts
Change-Id: I9ba17c4385fcb43d38998d45f89cf42952bc791b
Whenever module A depends on module B, module B must exist.
Whenever module A depends on module B, the possible targets of A must
be a subset of targets of module B.
Bug: 64513
Change-Id: I9de291d20d08df2813eb55be9a1f7e5efd907d6f
* Changed the regex to match class aliases, and added a check to
also ensure that all aliases for a class are in the same file as
the class (see cb77e59f7c for explanation).
* Removed support for the seemingly unmaintained Parsekit extension
(no release since 2009, and known not to compile against PHP 5.4).
The test would currently fail if Parsekit were used anyway because
the class MyLocalSettingsGenerator is commented out.
Change-Id: I7be2352f4659abc7b4952a9f74a14d0cef911614
Seems good in theory, though not quite usable in practice.
As with subclassing, the aliases would not work reliably
for code that uses type hints and/or instanceof.
See <https://bugs.php.net/bug.php?id=61422>, which I was
previously unaware of.
The best alternative is to add class_alias() calls to the
class files themselves, as well as duplicate entries to the
AutoLoader (just like when multiple classes are in the same
file).
There is no good way to show deprecation warnings, which
likely would have to be implemented in PHP itself. For now,
renamed classes should be indicated in RELEASE-NOTES. Maybe
I will file an RfC (at <https://wiki.php.net/rfc>) for this
and related improvements (e.g. making class names case
sensitive).
This reverts commit c61fdb4ef5.
Change-Id: I9771b4239543b543cfa078f357db1cd3918f081e
Sometimes it is desirable to change a class name that is still referenced
in extensions or config files (e.g. for consistency, as in I507ba00a).
PHP's class_alias() function can help in preserving backward compatibility;
however, creating an alias first requires that the class be loaded.
Hence this is implemented in AutoLoader. Lazy loading continues to work,
the list of class names is still maintained in a central location, and
optionally, deprecation warnings can be generated.
Change-Id: I1d3fb04a448647b5be10bed7fec05238b9fc6fc7
- Removed double spaces
- Added space after if/switch/foreach
- Removed space on elseif
- Added space around parentheses
- Added newline at end of file
- Removed space before semicolon at end of line
Change-Id: Id40b87e04786c6111e6686d7f7eea1e588bdf37d
Make protected, add method to allow tests to clear the protected class
property, and remove a now unneeded class from the test.
Change-Id: I269cab8d3f4419c91dfe15d5722e19ff6da828d1
When MediaWiki autoloading fails, we should gracefully return false.
Instead, we have been calling strtolower roughly 1,000 times in the hope
of finding a case-insensitive match.
This patch preserves the legacy case-insensitivity, but improves its
performance by approximately 100x, by storing the case-insensitive class
lookups as a static variable.
There is a new global $wgAutoloadAttemptLowercase which will switch the
behavior if desired. The default is to support case-insensitive loading.
Change-Id: Ifb12e05614a48390b730167e9d4ddcd8545db764
phpunit interprets '@author' documentation statements to generate group.
Appending the year to the author name created duplicates entry:
$ php phpunit.php --list-groups|grep Antoine
Antoine Musso
Antoine Musso, 2012
$
Removing the date fix the duplication. I have also inserted
corresponding '@copyright' statements.
Change-Id: I65b19adb59c4894314ce68a0c815c5d3e4c30df0
The ResourcesTest class under ./resources/ were not registered anywhere.
This patch move it under the `structure` testsuite and phaseout the, now
empty, `resources` directory.
Change-Id: I53410e1dc83263c4c541f4dc278a9e616265d7b1
The phpunit root directory has two test file:
AutoLoaderTest.php
StructureTest.php
The later was registered in phpunit under the `structure` test suite
while the former was not registered and hence never run (bug 47750).
This patch moves both files under the `structure` subdirectory and
change the suite to look in that directory. That will avoid us having
to manually maintain a list of test files.
Updated the __DIR__ in StructureTest.php.
Change-Id: I419c9157f32bdf7e1ff26a42f4bb3f3922b7be37
