Add tests for:
- action=block and action=unblock gettoken
- attempting to block or unblock a user with no token passed.
Patchset2: use a provider to have tests run against both 'block' and
'unblock' actions.
Change-Id: I686348ff4e2fe419c556acea2fa59dd203dc9440
That test has probably always been broken and use an invalid token. The
bug tracking brokenness is http://bugzilla.wikimedia.org/35646
Since the broken test is merged in master, that block the integration of
Jenkins and Gerrit by having any submitted patchset to be reported as
failling :-(
Root cause is https://gerrit.wikimedia.org/r/3434 which made the
Block/Unblock API to actually verify the token previously always
considered valid (bug 34212).
Change-Id: Iecf6b083163c214c734360b2f6d9b4bed8af07dc
Printer friendly version of article must encode URL in unicode.
- Patch originally written by Brion Vibber
https://bugzilla.wikimedia.org/attachment.cgi?id=9593
- introduces wfExpandIRI() global function, uses wfExpandIRI_callback.
- phpunit test.
Change-Id: I348b9f1d2ce65cb14f20d4a5751ac9359c8b8316
That will let us tests all the API tests by using PHPUnit group
filtering such as:
php phpunit.php --group API
Also cleaned some whitespaces
Patchset-4: skipped files that had only whitespace changes
Change-Id: I51e03d910521b061f505e3a9b11a08c7b95f1538
* Remove default id/name attributes
* Remove now redundant tests introduced in r110274
* Add tests to make sure label has no 'for' attribute when label isn't null but name/id are unset
* Update tests to not include id="" and name="" when calling with no arguments
* Updating calls to add name/id if needed, and while at it remove useless 'null' params
* Context:
-- Introduced in r109990, r111376, r111315
-- No callers exist that assume these defaults. Forcing an ID that should be unique is annoying and redundant. The name used in the query when submitting a form should be mentioned in the same file where it is used from the submission, never assume what name="" is from unrelated code
-- Same for ID, this is often used in CSS or JavaScript, shouldn't be assumed. (It should be simple to but two simple namespace selectors on a page without getting DOM conflicts)
* protect/protect and protect/modify were still failing because the output on IRC is not matching the i18n messages. The bots look for comment.match(<i18n-message> + (optionally) followed by ": user comment").
* The problem with protect/protect and protect/modify is that the protection settings are also inserted in somewhere in the message. Before MediaWiki 1.19 these were included in the IRC message as part of $1 (title). That may be ugly, but that's how it has always been. Since the point is to reproduce the old output we need to fix this one as well. This will make the output like:
Krinkle * protected "[[Sandbox [edit=autoconfirmed)]]"
which is exactly how it has been before and that's the the bots rely on (which was presumably done originally like that since it was the only way to make the output match the i18n message)
* Fixes more bug 34508
* assert used message-keys/variables instead of hardcoded values
* Pass $comment to the assertion helper instead of using it by default so that actions that don't use a comment (such as patrol) don't get forced a comment, to better replicate the IRC behavior
* Adding placeholder for Auto-Edit-Summary things that should also be checked in here, not sure how to do those yet
* Adding tests for block/* and protect/*
Exif test are leaking files heavily on our test server. This quick patch make
it sure we delete temporary files.
Ideally, we should have something like a temporary filesystem backend that
would self destruct :-D
requires r112326: wfRecursiveRemoveDir()
- Update the tests to test extra characters and patterns like like \\ and $1
- Also update the tests to make sure that matches that don't have enough data to work fail
- Replace the str_replace and preg_match based code with code based on preg_replace_callback.
ApiUploadTest gives inconsistent results which trigger false alamrs to all
the ops team. We are disabling those tests for now.
See https://bugzilla.wikimedia.org/26169
This patch fix ApiQueryTest::testTitlesGetNormalized() when $wgMetaNamespace
would contains space / underscore.
Example:
-8<-------------8<------------8<------------------8<----------8<------
1) ApiQueryTest::testTitlesGetNormalized
Failed asserting that two arrays are equal.
--- Expected
+++ Actual
@@ @@
Array (
'from' => 'Project:articleA'
- 'to' => 'My_Project:ArticleA'
+ 'to' => 'My Project:ArticleA'
)
-8<-------------8<------------8<------------------8<----------8<------
This patch use Title class logic to forge a normalized title instead of
manually creating it by appending two strings.
Final fix to bug 33663 (comment 21):
https://bugzilla.wikimedia.org/show_bug.cgi?id=33663#c21
* Moved some public static functions from FileBackend to FileBackendBase as the later defines the public API.
* Made splitStoragePath() return null if the backend or container name is empty.
* Made normalizeContainerPath() kill leading directory separators.
* Added more unit tests and made some documentation tweaks.
In FSFileBackend:
* Added resolveContainerName() to disallow '.' a container name, since this would cause a traversal.
Since we're here: nothing uses $namespaceNames, $mNamespaceIds or $namespaceAliases
outside of this class (core or extensions) so lets make it protected.
* r109993 broke the test by overwriting namespaces it tried to preserve
* tearDown should always have the opposite order of the overrides in setUp
* Adding wgLanguageCode while at it, no reason not to, just in case.
-- Follows-up r109993
* Previously it was passing $selectAttribs['name'] to Xml::label, which uses its value for the <label for=""> attribute. This works as long as $selectAttribs['id'] and $selectAttribs['name'] match, but when they don't it fails. <label for=""> always corresponds with <{input,text area,select} id=""> in browsers, never with "name".
* Make name/id match eachother by default to avoid backwards compatibility breakages (they used to match in the Xml class method as well)
* Add HtmlTest.php entries similar to the ones in XmlTest
* Fix E_NOTICE about $params['selected'], default to ''
-- Follows-up r109974, r109698, r109990
-- Bug originally introduced in r41425
-- XmlTest.php still runs successfully
-- HtmlTest.php runs successfully
* Use 'b' param in some fopen() calls as needed for Windows and newline handling.
* Removed some useless padding code in FileBackend::getContainerShard(). Initialized $m to make IDE happy.
* Updated some code comments.
In SwiftFileBackend:
* Manually set the ETag when using php-cloudfiles for creating files to avoid https://github.com/rackspace/php-cloudfiles/issues/59.
* Manually set the content type based on how StreamFile::getType(). This makes it safe to read files directly out of the proxy to end-users. The streamFile() backend functions already uses a similar content-type check.
* This is in preparation for deprecating this in favor of an Html:: method soon, making sure here that tests still match afterwards
* Follows-up r109698
new FileRepo() requires an array of parameters having at least the
'name' and 'backend' key setup.
TODO: 'backend' keyword should probably default to FileBackend.
* Made FileBackend::getFileProps() final.
* Added exception needed in SwiftFileBackend::getConnection().
* Various FileBackendTests fixes and additions. Optimized it a bit by keeping the backend instance in memory.
* In StoreBatchTest: sed proper clean() function to remove temp dirs (follows up r109641). Also removed some commented out code.
* Fixed temp dir leakage in FileBackendTest.
Some other test could have added their own logs, for example
deprecated notice. In that case, the first test will fail because
the log array already has some content :-b
The first ones were failing because the test suite ran on gallium without
MWDebug being initialized. This is now the case since r109033.
The second part was failling because PHPUnit on gallium does not have
the assertCount() method. Use a assertEquals() / count() workaround instead.
Tests added by 108873
Reverts r108877
The MySQL backend, when using temporary tables, does not support referencing
a table which was already open.
http://bugs.mysql.com/bug.php?id=10327
This patch skip BrokenRedirectsPage on MySQL as reported on r102411.
* Added SHA-1 to FileOp::attemptBatch $predicates (since concatenate was removed from FileOp). This lets us get rid of temp file backups, as the remaining failure case is just the backend medium going down (the ops would break in that case anyway). Doing so reduces RTTs and backup file I/O overhead. This also simplifies expiry object support by avoiding having to stash the expiry values of temp backup objects somewhere.
* Improved precheck() and attempt() status logic in FileOp::attemptBatch() a bit. Use separate $subStatus var to check if each op is OK.
* A few minor code cleanups and comment tweaks.
* Fixed MoveFileOp bug found in unit tests when the source/dest are the same and an overwriteDest/overwriteSame param is given (the file would just get deleted). Improved CopyFileOp in this case too.
Other:
* Added more unit tests.
* Made secure() call doPrepare() for caller dummy proofing (especially those that don't check the status).
In FSFileBackend:
* Removed redundant wfMkdirParents() calls in FSFileBackend, we already have prepare() for this purpose. This also keeps it's behavior more consistent with the other backends.
* Made use of 'backend-fail-store' message in doStoreInternal().
Other:
* Updated unit tests and renamed $src => $source in some functions for consistency.
* Added some documentation comments and @since tags.
r106780 to fix bug 33270 introduced a new bug that prevented selected
namespaces for search to be applied. This patch fix the issue.
Credits to Brad Jorsch
This is reverting the work done by MaxSem to support stored procedures
and stored function in MySQL. The reasons are:
- it is not needed yet
- tests are not functionals
- alter the stable include/db/Database.php and drop support for ';;'
So please create a branch to work on it and merge it back in trunk
once we have branched 1.19 :-)
I have opened bug 33654 to track this enhancement request.
Reverts r107376, r107994.
trunk is frozen pending stabilisation so we can release MediaWiki 1.19.
Those changes introduces API changes and new SQL tables, so that sounds like
new feature we do not have time to review right now.
Please reapply changes in branches/concurrency and have code review handled
there. Once the branch has been reviewed, please hold. Once trunk is stable
enough and 1.19 got branched, you are welcome to merge the branch in trunk.
Note: we can have a Jenkins jobs setup to run the branch tests if you need.
Reverts:
r108595 r108591 r108585 r108584 108572 r108564 108560 r108559
These fail only on sqlite, and only when run with the entire suite (they're fine when run individually).
Skipping ApiBlockTest allows ApiConcurrencyTest to run, oddly enough.
Clearly, this will need to be fixed ASAP. I'll take care of it on Thursday when I'm back online, or possibly
before.
They are now in their own files. Rewrite wfTimestamp tests to use
data providers like it should.
Originally commited without history by ^demon as r108420
