aboutsummaryrefslogtreecommitdiffstats
path: root/includes
Commit message (Collapse)AuthorAgeFilesLines
* Merge "Removed redundant doc comments."jenkins-bot2013-06-181-107/+0
|\
| * Removed redundant doc comments.Aaron Schulz2013-06-151-107/+0
| | | | | | | | Change-Id: Iced16c6d32ef0e4d58376ed3c092a500ec917446
* | Merge "Make rc_cur_id have proper value for upload log entries."jenkins-bot2013-06-181-5/+14
|\ \
| * | Make rc_cur_id have proper value for upload log entries.Brian Wolff2013-06-171-5/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Primary motivation here is to make Special:Recentchangelinked work correctly for uploads where the description page doesn't exist yet. (Previously it was mostly only showing reupload log entries) Mild downside to this approach is adding the recentchanges row is now in a different commit... Change-Id: I74c68b972c47c03066d3065e3748769a27a654fb
* | | Pass along tagFilter to ContribsPager, it already expects itMatthias Mullie2013-06-181-0/+1
|/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ContribsPager already has this below line: $this->tagFilter = isset( $options['tagfilter'] ) ? $options['tagfilter'] : false; The reason the code already works fine is because, in ChangeTags.php, there's this fallback: if ( $filter_tag === false ) { $filter_tag = $wgRequest->getVal( 'tagfilter' ); } Which essentially boils down to the same result as passing along tagfilter, which in SpecialContributions is read from $wgRequest, passed to ContribsPager, and (well, currently not) passed to ChangeTags Change-Id: I12c0336ae31c5524e41fa0087afe2ba104dc7d37
* | Merge "Do not output numberofviews, if $wgDisableCounters = true"Parent54462013-06-171-1/+2
|\ \
| * | Do not output numberofviews, if $wgDisableCounters = trueumherirrender2013-06-171-1/+2
| | | | | | | | | | | | | | | | | | | | | Follow up I9975dceedeffafadf560ed71238c51b94eb8c3ad But now for the parser function Change-Id: I18261b884b7f3be57a42c0b10594384fab6f9c9a
* | | Merge "Remove Selenium tests from core"jenkins-bot2013-06-173-234/+0
|\ \ \ | |/ / |/| |
| * | Remove Selenium tests from coreChad Horohoe2013-06-143-234/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | This never really worked properly, doesn't conform to coding conventions, and isn't actually used by anyone for testing. Our selenium stuff these days are in the qa/browsertests repo Change-Id: I8f1efaa118fe41821fb3f4a6099d75a33681f17b
* | | Fixed mysql warning about expecting a resource.Aaron Schulz2013-06-161-1/+1
| | | | | | | | | | | | Change-Id: I799d063fec7a8e96f49db34ce669453bd26c3fb3
* | | Merge "Localisation updates from http://translatewiki.net."Translation updater bot2013-06-161-0/+22
|\ \ \
| * | | Localisation updates from http://translatewiki.net.Translation updater bot2013-06-161-0/+22
| | | | | | | | | | | | | | | | Change-Id: I352ed34fa912240395a72e84e60f19c42fe10fb3
* | | | Merge "Don't reinvent the wheel in SpecialRecentchangeslinked::getExtraOptions"TheDJ2013-06-162-13/+16
|\ \ \ \ | |/ / / |/| | |
| * | | Don't reinvent the wheel in SpecialRecentchangeslinked::getExtraOptionsMatmaRex2013-05-242-13/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Just call parent::getExtraOptions() first and then simply add the one additional field. The 'SpecialRecentChangesPanel' hook wasn't called from SpecialRecentchangeslinked previously, I've wrapped the call in an if to keep the behavior the same. Changes this causes: * If $wgAllowCategorizedRecentChanges is true, Recentchangeslinked will include the category filter form as well. * The target filter on Recentchangeslinked will be displayed at the end of the form, below the tag filter. Change-Id: I4436a63356adb1e0b0daa1aa52c179974a036fa1
* | | | ResourceLoader: Serve stylesheets with "Access-Control-Allow-Origin: *"Derk-Jan Hartman2013-06-161-0/+1
| |_|/ |/| | | | | | | | | | | | | | | | | | | | | | | This allows Javascript to access our styling information (cssRules and related cross domain protected dom information) even if it is hosted on a different domain. Bug: 25886 Change-Id: I8e05c13ae1a1589fd120d5c439b1a7128ce2b659
* | | Localisation updates from http://translatewiki.net.Translation updater bot2013-06-151-1/+4
| | | | | | | | | | | | Change-Id: Ice334feb284bd21c83c26ea92705a1345d10b225
* | | Merge "Cleaned up variable/type order in docs."jenkins-bot2013-06-1410-104/+114
|\ \ \
| * | | Cleaned up variable/type order in docs.Aaron Schulz2013-06-1310-104/+114
| | | | | | | | | | | | | | | | | | | | | | | | * A few other minor doc cleanups and obvious type hints. Change-Id: I454fad0c1942068b4f3632ebd1b605bfbb9dfe7c
* | | | Fix typo ('associed' => 'associated').Ori Livneh2013-06-141-1/+1
| |_|/ |/| | | | | | | | Change-Id: I5ea2eec92ebe397f2d269562f1650e59cb969a19
* | | (bug 47449) {{#special:}} should always return a usable titleMormegil2013-06-141-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | {{#special:}} should always return a usable title in NS_SPECIAL, even when given an unknown special page title. Currently, it returns "No such special page" text for such unknown names, which is surprising. Bug: 47449 Change-Id: I7f184956ec9d56c849f201f1230736fcedd0fadf
* | | Merge "(Bug 48952) Show debug output in toolbar"jenkins-bot2013-06-141-1/+1
|\ \ \
| * | | (Bug 48952) Show debug output in toolbarphysikerwelt2013-06-131-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The debug output written with the function wfDebugLog should be printed to the debug toolbar in the same way as wfDebug does. Change-Id: Ib979df080332e16c8e0e6a89eab8741986ab39c3
* | | | API: Fix action=parse without any page or title or textBrad Jorsch2013-06-131-3/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | action=parse was raising an error about missing 'text' with 'title' even when 'title' wasn't actually passed, due to a fix for bug 33865. But this broke using action=parse to parse an edit summary without also giving it wikitext to parse. Instead, let's give the client a warning (unless it seems clear they know what they are doing) and assume the empty string for 'text'. Also, while we're editing the file, add some more examples as also requested in bug 48319. Bug: 48319 Change-Id: I03c1fbcb0bd31dea8bd84e164104f7ced0ace449
* | | | Fix patrolling where NP patrol isn't explicitly enabledMarius Hoch2013-06-131-2/+2
| |/ / |/| | | | | | | | | | | | | | Hot fix! Follow-up fc5bcf1147f9 Change-Id: Iba7e70e1a1ddd9826f6e5f4022b88ccafc5f571d
* | | API: Fix IPv6 handling in list=blocksBrad Jorsch2013-06-131-13/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current handling of the bkip parameter assumes IPv4, and breaks for IPv6 CIDR ranges (it also isn't working right for IPv6 non-CIDR, but not in an obvious way). This rewrite handles IPv6 correctly. It also necessarily adds validation for the bkip parameter, which would formerly return (not very sensible) results when passed invalid values. Bug: 48129 Change-Id: I02471bb32c3a217004d07a79d9f98b62133b31ef
* | | Merge "Add section preview mode to API action=parse"jenkins-bot2013-06-131-2/+4
|\ \ \ | |/ / |/| |
| * | Add section preview mode to API action=parseMax Semenik2013-06-131-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | In https://gerrit.wikimedia.org/r/#/c/67266/, it was enabled automatically when a section was specified in preview mode. However, this only works for previews of a saved page's content, failing for user input previews which might contain several sections of various degrees of nestedness. Change-Id: I987144e5e315ed9e0a67f599083461a78879c0df
* | | Prepare DatabaseMysql for mysqliJakub Vrana2013-06-123-928/+1108
|/ / | | | | | | | | | | | | | | | | | | | | PHP extensions mysql and mysqli have lots in common. In order to support mysqli, this diff separates common MySQL behavior to abstract class DatabaseMysqlBase and implements the mysql-extension specific in DatabaseMysql. Outside behavior remains the same. I've tried to minimize future code duplication by introducing mysql*() methods even if the parent method just unwraps the result. Bug: 45288 Change-Id: I905d4a4550377bc849a860f0962dad710d9dc71f
* | Merge "Remove non UTF-8 chars from debug output"jenkins-bot2013-06-121-1/+1
|\ \
| * | Remove non UTF-8 chars from debug outputphysikerwelt2013-06-061-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The debug output might contain non UTF-8 chars. Theses bad characters are not compatible with the debug toolbar. As a result the debug toolbar disappears. Non UTF-8 chars are being produced by the database log that prints out binary fields for example. Bug: 48951 Change-Id: I42f7a5c913b378c05b68970646c75894ca068ed9
* | | Removed redundant documentation for some methods.Aaron Schulz2013-06-121-80/+0
| | | | | | | | | | | | Change-Id: I72f132c0b6e9a300efff6c15957f163d84e7d9c2
* | | Merge "Localisation updates from http://translatewiki.net."Translation updater bot2013-06-121-0/+2
|\ \ \
| * | | Localisation updates from http://translatewiki.net.Translation updater bot2013-06-121-0/+2
| | | | | | | | | | | | | | | | Change-Id: I599770c4c8b53d0d46c74f1426168a5c3d963a81
* | | | Removed redundant documentation for some methods.Aaron Schulz2013-06-123-233/+0
| | | | | | | | | | | | | | | | Change-Id: Iad422716f6796cc4b69405182f4a74daa0c93663
* | | | support setting Memcached::OPT_SERVER_FAILURE_LIMIT, ↵asher2013-06-121-0/+10
|/ / / | | | | | | | | | | | | | | | Memcached::OPT_RETRY_TIMEOUT Change-Id: Ib1fcae92b65e8fd23baee7af7ed64abd908b600e
* | | Fix return type of MessageCache::getMsgFromNamespace for existingTyler Anthony Romeo2013-06-121-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Functions expect the message cache to return a string if a message exists, even empty, and false if it does not exist. This adds casting to the substr() function, which would return false for existing messages that were just blank. Bug: 14176 Change-Id: Id91914a3701fe53f1e2e894824512489392c628b
* | | Localisation updates from http://translatewiki.net.Translation updater bot2013-06-111-33/+56
| | | | | | | | | | | | Change-Id: I7458f257824c9b1a4f14d6cc7a9567ee454749c7
* | | Merge "Only show notoken as possible API error once"jenkins-bot2013-06-111-1/+6
|\ \ \
| * | | Only show notoken as possible API error onceMarius Hoch2013-06-111-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently notoken is being listed as possible API error twice if it's explicitly set as required parameter and needsToken() returns true. See: https://www.wikidata.org/w/api.php?action=paraminfo&modules=edit Change-Id: Ia17c5cfa634919b43affa146df0d1dc0ff06b758
* | | | Merge "Remove unnecessary exception from Title::getLatestRevID()"jenkins-bot2013-06-111-15/+13
|\ \ \ \
| * | | | Remove unnecessary exception from Title::getLatestRevID()Tim Starling2013-06-111-15/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Bug 37209: An exception was thrown where the data cached in the Title object was inconsistent with the data in LinkCache. It shouldn't be surprising that this happens, since there is no guarantee that the data was derived from the same transaction or even the same DB server. But I don't think it is a problem worth troubling the user over, since with $flags=0, the slave DB server will be used, and no special guarantee of consistency should be expected by callers. If callers do need consistency, then they should make their own arrangements to get it, such as clearing the LinkCache. Since we have to pick a winner, and the choice is mostly arbitrary since it's not possible to tell which is fresher, I think LinkCache is a better choice since the lifetime of its cache entries can be controlled. Change-Id: I0add48463341e56fe8c155b1007487278ad2705d
* | | | | Merge changes I4950496e,I849c4bd1jenkins-bot2013-06-112-16/+2
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * changes: Remove all uses of obsolete .editsection class Remove Wikimedia-specific hack in ParserOutput
| * | | | | Remove all uses of obsolete .editsection classMatmaRex2013-06-071-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Cleanup after I6a6c12a9. To be merged after appropriate caches are purged. Change-Id: I4950496e584735a3ceeb0e1c4123dfe6597f2ae1
| * | | | | Remove Wikimedia-specific hack in ParserOutputMatmaRex2013-06-071-15/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Cleanup after I6a6c12a9. To be merged after appropriate caches are purged. Change-Id: I849c4bd1fe59e582f9f16789e8b511719e2a5d06
* | | | | | Allow patrolling pages by revision idMarius Hoch2013-06-102-16/+57
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This became necessary as it's quite hard to get the rc id of a change from the html after I1e24733c. Bug: 49259 Change-Id: Ia7d3960cf11bf8ae0fc06ae1a0f7fcfb3c080f21
* | | | | | [Database] Tweaked LoadMonitor::getLagTimes() to further reduce stampedesAaron Schulz2013-06-111-8/+16
| | | | | | | | | | | | | | | | | | | | | | | | Change-Id: I79e3232e3e87bb298ae1a3fe214fcd2506818884
* | | | | | Drop __autoload.Daniel Friesen2013-06-111-9/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | spl_autoload_register exists in >=5.1.2, we require >=5.3.2 We no longer support hphpc and it looks like hhvm supports spl_autoload_register We've got no reason to keep using __autoload. We'll also need to exclusively use spl_autoload_register when we start using composer.json stuff. Change-Id: I694b7b96825e5d136fa76461511efc9a002149fa
* | | | | | Add new hook AbortTalkPageEmailNotificationbsitu2013-06-101-3/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This hook will allow extension to disable the regular talk page email notification, Echo is doing this with a hackish solution of setting global variable wgEnotifUserTalk to false on the fly, this approach is problematic and will be replaced with this hook Change-Id: Ie043d9a1a771d64e4b05b7ddeca75c7542a6c575
* | | | | | Merge "Add user rights 'editmyuserjs' and 'editmyusercss'"jenkins-bot2013-06-103-36/+17
|\ \ \ \ \ \
| * | | | | | Add user rights 'editmyuserjs' and 'editmyusercss'Brad Jorsch2013-06-103-36/+17
| | |/ / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | These are needed for OAuth grants. Change-Id: I52f8e4a5cb48573cb2dbc26fc508e61a95d748c3
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-24 15:43:13 +0000