aboutsummaryrefslogtreecommitdiffstats
path: root/docs
Commit message (Collapse)AuthorAgeFilesLines
* DomainEvent: rename EventIngressBase to DomainEventIngressAaron Schulz2025-04-031-1/+1
| | | | | Bug: T390735 Change-Id: I0ebec537bb15925e8507ee6934cd4a17973c536a
* Parsoid Fragment Support v3: make mStripExtTags a persistent Parser propertyC. Scott Ananian2025-04-021-1/+2
| | | | | | | | | | | | | | | | | | | Rather than reset `Parser::$mStripExtTags` after post-processing, ensure that it is `false` for any frame expansion done using the legacy parser kept by Parsoid's DataAccess. This was causing issues with Extension:Babel, which was invoking $parser->replaceVariables() recursively when the {{#babel}} parser function was expanded by Parsoid. This new behavior is enabled by setting $wgParsoidFragmentSupport to 'v3', in order to allow us to run round-trip testing with this configuration without disturbing production. The configuration variable is temporary, and will be cleaned up in Ib5365c87ab594a2c21a84ec8bc2a64a71799085f. Bug: T390420 Change-Id: I8f45ea027776c3bb0c9f4468afa00465e41b6dec
* Namespace all remaining files in includes/changetagsJames D. Forrester2025-03-251-12/+12
| | | | | Bug: T353458 Change-Id: I3cf44dfe5425f2efb8409c83571c427447b053af
* Broaden type of DataAccess::preprocessWikitext()C. Scott Ananian2025-03-242-0/+18
| | | | | | | This prepares the way for a revision of this API in the follow-up patch Ie543457d5a2eba2ef1f1f4b7622531582c48c3e4. Change-Id: I29b843daeb614d1f48009e1ade93c16fe2f16736
* Merge "notifications: Introduce Notification Middleware and ↵jenkins-bot2025-03-211-0/+19
|\ | | | | | | NotificationEnvelope"
| * notifications: Introduce Notification Middleware and NotificationEnvelopePiotr Miazga2025-03-211-0/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | To allow ourselves esier processing/modifying Notifications lets introduce an idea of NotificationsEnvelope which represents a Notification being sent and list of recipients. The middleware approach will allow us to modify the Notification behaviour by letting extensions to inject/modify the Notifications. Each Middleware will retrieve a list of Envelopes MediaWiki wants to send. Middlewares should iterate over envelopes and decide if those want to add/remove/replace Notifications and/or Recipients. Bug: T387996 Change-Id: Ib3ee35c75b2f4dcfdc516b9259a852dc73c4a778
* | Namespace all remaining files in includes/jobqueueJames D. Forrester2025-03-201-17/+17
| | | | | | | | | | Bug: T353458 Change-Id: I95690a312e356c45dbeed607d32fb0e4626690cf
* | Namespace all remaining files in includes/recentchangesJames D. Forrester2025-03-201-1/+1
| | | | | | | | | | Bug: T353458 Change-Id: I2ae4577de79832b082adca282ff73cfabc8f9392
* | Merge "extension.json: DomainEventIngresses instead of DomainEventSubscribers"jenkins-bot2025-03-201-1/+8
|\ \
| * | extension.json: DomainEventIngresses instead of DomainEventSubscribersdaniel2025-03-201-1/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Why: - "ingress" is the more meaningful concept when implementing listeners What: - Add support for DomainEventSubscribers in extension.json Bug: T389033 Change-Id: I458bd7cd439a2e3213458d994cf87affa5da966b
* | | Namespace all remaining files in includes/loggingJames D. Forrester2025-03-191-43/+43
|/ / | | | | | | | | Bug: T353458 Change-Id: Ibe1810f1c71316a9124e1dc6ae405097dafd5267
* / objectcache: Introduce dataRedundancy to SqlBagOStuffAmir Sarabadani2025-03-111-0/+8
|/ | | | | | | | | | | | | | | | | | | | | | | | | | | | This forces SqlBagOStuff to instead of sharding keys, write to n out of m servers instead. It also reads from those servers as well and in case of incosistency, picks the value with the highest exptime. This is mostly for mainstash and allows us to provide stronger consistency guarantees while allowing for a section to be depooled and put to maintenance. It basically implements the logic already used by NoSQL database systems such as Cassandra (There are two types to solve conflict, quorum or timestamp, Cassandra is using quorum while we are using timestamp). There will be some edge cases that it might still pick the wrong value: - if TTL is set to INDEF - if the TTL gets shortened for various reasons. - If we go with two clusters, value is set, one gets depooled, a new value is set, the depooled one gets pooled and the other depooled and then read happens. But all of these are extremely rare edge cases and we should be fine. This also means if data redundancy is set, locking means all sections will be locked and removal means all sections must allow the unblock. Otherwise, the lock will be kept. Bug: T383327 Change-Id: I80da12396858ee4fc58ae257f6c154b3050df696
* Deprecate wgEnotifImpersonal and wgEnotifMaxRecipsJames D. Forrester2025-03-101-0/+2
| | | | | Bug: T387856 Change-Id: I1e206dc4ad8bca2fae78ca4fdf919e3a8ee3c4b5
* Merge "Introduce thumbnail steps"jenkins-bot2025-03-032-0/+34
|\
| * Introduce thumbnail stepsAmir Sarabadani2025-03-022-0/+34
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This would make it when the editor requests thumbnail with the width of 200px, and we have defined the steps as [...,150px,250px,...], the thumbnail with the size of 250px to be picked but still the width attribute in HTML will be set to force the requested size. This change will massively reduce the storage of thumbnails that have been causing issues for us recently and improves cache hit ratio in every layer (from client-side to CDN frontend and backend). Tested locally and it worked just fine. Bug: T360589 Change-Id: I9110d4ac9bcd421b07f13deeae5d863ef1ef9c31
* | Replace remaining gendered pronounsTacsipacsi2025-03-021-4/+4
|/ | | | | | | | | | | | | | | | | | | | | | To respect all genders, fix comments to not assume users use binary pronouns (or even that users are “he”s), but rather use singular “they”. Also fix some typos that happened to result in gendered pronouns, and a few incorrect commas and missing articles in comments near the fixed pronouns. I skipped four files: - HISTORY – the release notes were made with the wording they were made with, I’m not sure if rewording them afterwards is okay - tests/phpunit/data/preprocess/All_system_messages.{txt,expected} – these are test cases generated from somewhere, I’d regenerate them rather than updating - languages/i18n/qqq.json – fixed on Translatewiki instead to make their edit histories more useful Bug: T387626 Change-Id: I282406a0e1407be548e917735fe7eb9a6bf8b136
* Merge "Remove $wgAllowAuthenticatedCrossOrigin flag"jenkins-bot2025-02-272-17/+0
|\
| * Remove $wgAllowAuthenticatedCrossOrigin flagLucas Werkmeister2025-02-182-17/+0
| | | | | | | | | | Bug: T322944 Change-Id: I4e142ec5eba2dc05afe947f138bea043e0667151
* | Merge "Enable $wgAllowAuthenticatedCrossOrigin by default"developmentjenkins-bot2025-02-251-1/+1
|\|
| * Enable $wgAllowAuthenticatedCrossOrigin by defaultLucas Werkmeister2025-02-181-1/+1
| | | | | | | | | | Bug: T322944 Change-Id: I6de31143e67e14d14aeaf7df04f1cbe257cf56bb
* | Merge "Declare $wgAllowAuthenticatedCrossOrigin non-experimental"jenkins-bot2025-02-241-3/+0
|\|
| * Declare $wgAllowAuthenticatedCrossOrigin non-experimentalLucas Werkmeister2025-02-181-3/+0
| | | | | | | | | | | | | | | | The feature is enabled and working in Wikimedia production – I think that’s good enough to call it properly supported. Bug: T322944 Change-Id: I8fc310d4ab4fc3e17cbeaadfc8eadb4d2120ebda
* | Merge "Add 'lang=en' to 'Powered by MediaWiki' footer icon"jenkins-bot2025-02-221-1/+1
|\ \
| * | Add 'lang=en' to 'Powered by MediaWiki' footer iconReedy2025-01-111-1/+1
| | | | | | | | | | | | | | | Bug: T383501 Change-Id: I9fb473c0ebbc7b002aff513b0630d18d9cbd68d3
* | | user: Implement batch user registration lookupsMáté Szabó2025-02-191-1/+1
| |/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Why: - The TSP team would like to change the way expired temporary account user links are displayed, which requires an efficient way to fetch their registration timestamps. - On WMF wikis, which use CentralAuth, this requires fetching the first (i.e. global) registration timestamp of the account, rather than the naïve approach of using the registration timestamp from the local user table. - MediaWiki provides the UserRegistrationLookup facade to transparently fetch the earliest registration timestamp for a single user, but offers no batch interface to do the same. - Since user links are often rendered in large pagers, a batch interface is needed. What: - Add IUserRegistrationProvider::fetchRegistrationBatch(), which takes an iterable of UserIdentities and returns a map of their registration timestamps (or null if not available), keyed by user ID. Although this interface is marked as stable to implement, its sole non-core implementor according to codesearch is CentralAuth. - Add UserRegistrationLookup::getFirstRegistrationBatch(), which delegates to fetchRegistrationBatch() on configured registration providers and returns the earliest registration timestamp for each user in the batch. - To avoid potential interface incompatibility in WMF production, this depends on CentralAuth implementing the new IUserRegistrationProvider method first. Bug: T358469 Depends-On: Ibe28163e962161567d486607e36d999a36a1e604 Change-Id: I1f6af2693a8f0c5c854b8a6b04edd1eb21934007
* | Merge "Add a built-in way to send notifications"jenkins-bot2025-02-181-0/+26
|\ \
| * | Add a built-in way to send notificationsBartosz Dziewoński2025-02-121-0/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Create a provider pattern for sending notifications (T383992). There is no base notifications handler in core; they can only be provided by extensions. A handler in the Echo extension, compatible with the existing Echo notifications system, will be implemented in T383993. Co-Authored-By: Piotr Miazga <pmiazga@wikimedia.org> Bug: T383992 Change-Id: I16b309935c3d29c3cde4459b5e36abce063a8534
* | | block: Add $wgAutoblockExemptionsTaavi Väänänen2025-02-182-0/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | Add a new option to exempt users from autoblocks in the configuration, instead of editing a MediaWiki space page on every wiki. The use case for this is WMCS ranges (see T386689). Bug: T240542 Change-Id: I704b34b81214e7a1ac819fefa7ad3c2c87305647
* | | Merge "Set correct merge strategy for $wgAddGroups and $wgRemoveGroups"jenkins-bot2025-02-141-0/+2
|\ \ \
| * | | Set correct merge strategy for $wgAddGroups and $wgRemoveGroupsDaimona Eaytoy2025-02-141-0/+2
| |/ / | | | | | | | | | | | | | | | | | | | | | These are 2D lists, so the default array_merge will overwrite values instead of merging. Bug: T386210 Change-Id: Id001462b17ff43964af4f627ca40f07cb198eab2
* / / logging: Use LinkTarget in LogFormatter::getPreloadTitles overridesUmherirrender2025-02-131-13/+13
|/ / | | | | | | | | | | | | Also avoid null as array item, as that is not allowed according to the return type Change-Id: I4083c55a69d6186448a13f35f18d96bfe9ffd23c
* | RenameUser: Introduce RenameUserJob to run rename jobxtex2025-02-111-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | RenameUserJob is moved to Job/RenameUserTableJob because there are two kinds of jobs now. The newly added RenameUserDerivedJob is used for performing user-renames across a wiki family using virtual domains or shared tables. Most code are moved from SpecialRenameUser and maintenance/renameUser. The new service, RenameUserFactory is added for constructing RenameUser easier. When a global rename happen, the central wiki will enqueue RenameUserDerivedJobs for other wikis in the same family. The derived jobs will check if the central wiki has the same user table as local, and perform updates to local tables. A new user-right 'renameuser-global' is also added because wiki families may want global users to be renamed only by a limit set of users or on a certain global wiki. Bug: T104830 Change-Id: Ic4120cbd9a4850dfe22d009daa171199fe7c5e39
* | Remove temporary $wgParsoidNewTemplateExpansionMode configurationC. Scott Ananian2025-02-032-17/+0
| | | | | | | | | | | | | | | | | | This was used to test an experimental parsoid feature before deployment, but the testing was successful. Bug: T382464 Follows-Up: I194a9550500bf7ece215791c51d6feb78a80b1a8 Change-Id: Ib91a17868352722dc3570b07856423733f1b2368
* | Allow non-anonymous CORS for some session providersLucas Werkmeister2025-01-312-0/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If a session provider is safe against CSRF (e.g. OAuth), we can allow cross-origin requests to be non-anonymous. This makes it possible to have fully client-side web applications that authenticate users via an OAuth 2.0 client (necessarily a non-confidential client) and then make authorized requests against wikis using the Authorization header. To opt into this new mode of CORS requests, we use a new boolean parameter called "crossorigin". (An earlier version of this change reused the existing "origin=*" parameter for this, but the change to its previous “always anonymous” behavior was not welcomed during code review.) The parameter is disabled by default via a config setting, which is currently declared experimental; if this works out in practice, we’ll presumably want to at least change it to non-experimental, though I don’t know if we want to enable the feature by default (or even unconditionally) or keep the setting as it is. Note that the preflight request doesn’t send the real Authorization header (it just includes its name in Access-Control-Request-Headers), so the session provider in the preflight request is still the normal cookie provider (which is why handleCORS() has to bypass the safeAgainstCsrf() check in that case). This shouldn’t be an issue, because executeActionWithErrorHandling() returns quite early if the request is an OPTIONS request (immediately after handleCORS()), but to be sure that the unsafe session isn’t used during the preflight request, I added a "crossorigin" check to lacksSameOriginSecurity(). (That method is called by the constructor before the param validator has been set up, so $this->getParameter() is not available – hence the call to $request->getCheck() instead, just as for the 'callback' parameter.) Bug: T322944 Change-Id: I41200852ee5d22a36429ffadb049ec3076804c78
* | Introduce ParsoidExperimentalParserFunctionOutput config flagSubramanya Sastry2025-01-302-0/+19
| | | | | | | | | | | | | | | | | | * Plumb this value into Parsoid's SiteConfig so that the Parsoid library code can access this. Bug: T373253 Bug: T385129 Change-Id: If119ff94e65325fc446ca068e0b2d2434c070a2e
* | Merge "Add ParserOutputFlags::{HAS_ASYNC_CONTENT,ASYNC_NOT_READY}"jenkins-bot2025-01-292-0/+14
|\ \
| * | Add ParserOutputFlags::{HAS_ASYNC_CONTENT,ASYNC_NOT_READY}C. Scott Ananian2025-01-292-0/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This allows Parsoid to mark parses which contain async content which is "not ready yet". At the moment this output is cached with a reduced TTL, although in the future it might still be treated as uncacheable, cached until evicted, or some other option. The HAS_ASYNC_CONTENT flag along with ParserOutput::hasReducedExpiry() ensures that RefreshLinksJob is opportunistically reinvoked whenever the page is reparsed, since the asynchronous content may change the metadata for the page when it becomes ready. As describe in T373256, ::hasReducedExpiry() is misnamed now, and a follow-up patch will probably rename it to ::hasDynamicContent() or something like that. What it really means is "RefreshLinksJob must be re-run on every parse, because the content may change on each parse". In the past we would *also* reduce the cache time for pages like this. But for asynchronous content, "the content may change on each parse" only *until* the asynchronous content is "ready". Once it is ready the contents will no longer change, and the cache lifetime can be raised again -- but ::hasDynamicContent() still needs to be set, which in the future will mean "you need to check that RefreshLinksJob has last run" not "you must always run RefreshLinksJob". Asynchronous content will always set HAS_ASYNC_CONTENT, even after the content is "ready", but will only set ASYNC_NOT_READY if it needed to use placeholder content in this render. Bug: T373256 Change-Id: I71e10f8a9133c16ebd9120c23c965b9ff20dabd2
* | | Remove Google search fallback for external searchSam Wilson2025-01-291-1/+0
|/ / | | | | | | | | | | | | | | | | Remove the 'googlesearch' and 'search-external' messages, so that there is no fallback search form if $wgDisableTextSearch=true and $wgSearchForwardUrl=false. Bug: T384678 Change-Id: I20a3fe8484424427de5dcc55098a09114fedaf66
* | Change default value of wgParsoidNewTemplateExpansionMode to trueSubramanya Sastry2025-01-271-1/+1
| | | | | | | | | | | | | | * This has to ride the train with Parsoid's changes Bug: T382464 Change-Id: I92a81d41d284a9b272d3f0d6cbdc5b022d051f57
* | Merge "categorylinks: Add support for write both"jenkins-bot2025-01-232-0/+16
|\ \
| * | categorylinks: Add support for write bothAmir Sarabadani2025-01-212-0/+16
| | | | | | | | | | | | | | | Bug: T299951 Change-Id: Ifd9876bcb452e412b7335741e74cfc4c820aa248
* | | Merge "For Parsoid calls, treat preprocessing as starting in SOL state"jenkins-bot2025-01-232-0/+17
|\ \ \
| * | | For Parsoid calls, treat preprocessing as starting in SOL stateSubramanya Sastry2025-01-172-0/+17
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * If Parsoid calls the preprocessor, initialize lineStart to true. Track this through: - parser function calls that return expandable template messages int: parser function is an example in core Extensions seem to define a number of other such parser functions - template-arg substitions So {{templatename|mytemplate}} with text {{{{{1}}}}} which is effectively a call to {{mytemplate}} continues to set sol-state to true across the expansion. See test "Preprocessor precedence 5: tplarg takes precedence over template" in preprocessor.txt which exercises this use case. - However, note that this is a best-faith effort because this flag is set while building the preprocessor DOM tree before templates are expanded. So, this is mostly a source syntax flag and constructs that expand to empty strings can blind the preprocessor to the true value of SOL state in the expanded string. This is true for both the legacy parser and Parsoid, and as such T2529 behavior is a hack with a set of associated edge cases. * Parsoid models templates as independent documents as always starting in start-of-line state (and does some patch up for b/c reasons where this assumption fails). So, there is no reason to add newlines for some set of wikitext characters (per T2529) when Parsoid is involved. * This lets us eliminate some hacks in Parsoid to strip these added newlines when Parsoid was already in SOL state but which then introduces edge cases. See discussion in T382464 where Parsoid currently cannot distinguish between a couple of test cases. * But, with this change, where Parsoid no longer gets a newline added, Parsoid doesn't have to heuristically remove the newline (and incorrectly as in the edge case in the bug report) which eliminates the edge case from the bug report. * This change has to be backed by a change in Parsoid to undo the T2529 newline removal hack in TokenStreamPatcher to ensure Parsoid CI doesn't break with this change. * To let us safely test this in Parsoid's round-trip testing and safely (and conservatively) roll this out to production, this change is backed by a new config flag (ParsoidTemplateExpansionMode) which defaults to false. We unconditionally set this to true in the ParserTestRunner for all parser tests. This flag will be removed once we roll out this change and the Parsoid change to production. Bug: T382464 Change-Id: I194a9550500bf7ece215791c51d6feb78a80b1a8
* / / resourceloader: Minor documentation correctionsBartosz Dziewoński2025-01-202-11/+12
|/ / | | | | | | | | | | | | | | Issues spotted while working on I03a9a6945ab27e9888ea21b03985ed713f0a9b50. Some code style improvements too. Change-Id: I409d0a1805aa7430cc86e53633f4f85ef8a76dcf
* / Tag page creations of previously deleted pageKgraessle2025-01-141-0/+2
|/ | | | | | | | adds a new tag in core titled mw-recreated to note when a new page is a recreation Bug: T56145 Co-Authored-by: Rockingpenny4 <rockingpenny4@gmail.com> Change-Id: Ib8ffe3fba73d0464f3fd353138456b07e7afc7d7
* Merge "extension.json schema: use ObjectSpec definition"jenkins-bot2025-01-072-308/+138
|\
| * extension.json schema: use ObjectSpec definitiondaniel2025-01-062-308/+138
| | | | | | | | | | | | | | Make use of a reusable definition of ObjectSpec structures, for consistency across all parts of extension.json that uses object specs. Change-Id: Ie09b933b9419523cdc62dcba79f86c5bf4242ac3
* | Merge "DnsBlacklistUrls: Remove sorbs.net"jenkins-bot2025-01-061-2/+1
|\ \
| * | DnsBlacklistUrls: Remove sorbs.netReedy2025-01-061-2/+1
| | | | | | | | | | | | | | | | | | | | | Service is no longer running, so it's a default that doesn't do anything Bug: T382987 Change-Id: I3a21c12ba689928d38e410cbe2547ab7e616ac8a
* | | file: Basic support for writing to the new file tablesAmir Sarabadani2025-01-062-0/+16
|/ / | | | | | | | | Bug: T368113 Change-Id: I8d98d187ba4f1342167820b5710f5382b2ac4831
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-25 13:44:52 +0000