Page MenuHomePhabricator

Remove core fallbacks to global $wgUser [1.36]
Closed, ResolvedPublic

Description

T245482: Remove core fallbacks to global $wgUser [1.35] || T262707: Remove core fallbacks to global $wgUser [1.37]

See T242935: Deprecate core fallbacks to $wgUser for deprecations

Following hard deprecations in 1.35, some fallbacks should be removed.

Subtasks are for each fallback that should be removed. All deployed callers have already been updated, but ideally calls from non-deployed extensions will still be updated.

Related Objects

StatusSubtypeAssignedTask
OpenNone
OpenNone
OpenDannyS712
OpenNone
OpenDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
InvalidNone
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedYaron_Koren
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedPRODUCTION ERRORcscott
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712
ResolvedDannyS712

Event Timeline

There are a very large number of changes, so older changes are hidden. Show Older Changes

@Pchelolo @daniel I'd like to do the first batch of removals all together around the start of 1.36. Would that be okay, or should the removals be done separately? I just think it'll be easier to make progress on the remaining uses of $wgUser if the deprecated uses that have no remaining uses are removed first. Thoughts?

I'm not entirely sure what you mean. Do you mean remove all fallbacks in core with a single patch? I don't really care, whatever is more convenient. Given that the fallbacks you'd be removing were already hard-deprecated, it seems safe to do.

I'm not entirely sure what you mean. Do you mean remove all fallbacks in core with a single patch? I don't really care, whatever is more convenient. Given that the fallbacks you'd be removing were already hard-deprecated, it seems safe to do.

Yes, I mean doing as much as possible in one patch

DannyS712 changed the status of subtask T248509: Remove deprecated ways to call WikiPage::doDeleteArticleReal from Stalled to Open.
DannyS712 changed the status of subtask T248424: Remove Revision::newFromPageId from Stalled to Open.
DannyS712 changed the status of subtask T248192: Remove PageArchive::undelete from Stalled to Open.
DannyS712 changed the status of subtask T248423: Remove Revision::getTimestampFromId from Stalled to Open.
DannyS712 changed the status of subtask T248036: Remove Revision::getParentLengths from Stalled to Open.
DannyS712 changed the status of subtask T248022: Remove deprecated api token hooks from Stalled to Open.
DannyS712 changed the status of subtask T247889: Remove LocalFile::delete(Old)? from Stalled to Open.
DannyS712 changed the status of subtask T247143: Remove the Revision class [1.37] from Stalled to Open.
DannyS712 changed the status of subtask T246413: Remove LogEventsList::getExcludeClause fallback to $wgUser from Stalled to Open.
DannyS712 changed the status of subtask T246411: Remove Revision::userCanBitfield from Stalled to Open.
DannyS712 changed the status of subtask T246141: Remove old SpecialPageFactory from Stalled to Open.
DannyS712 changed the status of subtask T246192: Remove RecentChange::markPatrolled, deprecated and unused from Stalled to Open.
DannyS712 changed the status of subtask T246138: Remove deprecated title permissions functions from Stalled to Open.
DannyS712 changed the status of subtask T245777: Remove WebRequest::getLimitOffset from Stalled to Open.
DannyS712 changed the status of subtask T245703: Remove Revision::newNullRevision from Stalled to Open.
DannyS712 changed the status of subtask T245355: Remove ApiTestCase::doLogin from Stalled to Open.

Other than beginning to remove the Revision related code, this looks mostly done?
Haven't sent patches for the revisions pending resolution of the -2 on https://gerrit.wikimedia.org/r/c/mediawiki/core/+/614027

@Legoktm left a -2 saying

These hooks were only hard deprecated in 1.35, so please wait until at least 1.37 per "It is RECOMMENDED to emit hard deprecation notices for at least two major MediaWiki versions".

Recent discussions on wikitech-l suggest that waiting 2 versions is not required. @Legoktm can you reconsider your -2?

Nothing more to remove before 1.36 is tagged, new work is at T262707: Remove core fallbacks to global $wgUser [1.37]