mediawiki-extensions-Discus.../includes/Notifications/EventDispatcher.php
C. Scott Ananian 25272e7a4a Don't refer directly to PHP dom extension classes; avoid nonstandard behavior
These changes ensure that DiscussionTools is independent of DOM
library choice, and will not break if/when Parsoid switches to an
alternate (more standards-compliant) DOM library.

We run `phan` against the Dodo standards-compliant DOM library,
so this ends up flagging uses of non-standard PHP extensions to
the DOM.  These will be suppressed for now with a "Nonstandard DOM"
comment that can be grepped for, since they will eventually
will need to be rewritten or worked around.

Most frequent issues:

* Node::nodeValue and Node::textContent and Element::getAttribute()
can return null in a spec-compliant implementation.  Add `?? ''` to
make spec-compliant results consistent w/ what PHP returns.

* DOMXPath doesn't accept anything except DOMDocument.  These uses
should be replaced with DOMCompat::querySelectorAll() or similar
(which end up using DOMXPath under the covers for DOMDocument any way,
but are implemented more efficiently in a spec-compliant
implementation).

* A couple of times we have code like:
  `while ($node->firstChild!==null) { $node = $node->firstChild; }`
and phan's analysis isn't strong enough to determine that $node is still
non-null after the while.  This same issue should appear with DOMDocument
but phan doesn't complain for some reason.

One apparently legit issue:

* Node::insertBefore() is once called in a funny way which leans on
the fact that the second option is optional in PHP.  This seems to be
a workaround for an ancient PHP bug, and can probably be safely
removed.

Bug: T287611
Bug: T217867
Change-Id: I3c4f41c3819770f85d68157c9f690d650b7266a3
2021-07-30 18:15:40 -04:00

209 lines
6.3 KiB
PHP

<?php
/**
* DiscussionTools event dispatcher
*
* @file
* @ingroup Extensions
* @license MIT
*/
namespace MediaWiki\Extension\DiscussionTools\Notifications;
use EchoEvent;
use Error;
use IDBAccessObject;
use Iterator;
use MediaWiki\Extension\DiscussionTools\CommentParser;
use MediaWiki\Extension\DiscussionTools\HeadingItem;
use MediaWiki\Extension\DiscussionTools\Hooks\HookUtils;
use MediaWiki\Extension\DiscussionTools\SubscriptionItem;
use MediaWiki\MediaWikiServices;
use MediaWiki\Page\PageIdentity;
use MediaWiki\Revision\RevisionRecord;
use MediaWiki\User\UserIdentity;
use ParserOptions;
use Title;
use Wikimedia\Assert\Assert;
use Wikimedia\Parsoid\DOM\Element;
use Wikimedia\Parsoid\Utils\DOMUtils;
class EventDispatcher {
/**
* @param RevisionRecord $revRecord
* @return CommentParser
*/
private static function getParsedRevision( RevisionRecord $revRecord ): CommentParser {
$services = MediaWikiServices::getInstance();
$pageRecord = $services->getPageStore()->getPageByReference( $revRecord->getPage() );
Assert::postcondition( $pageRecord !== null, 'Revision had no page' );
// If the $revRecord was fetched from the primary database, this will also fetch the content
// from the primary database (using the same query flags)
$status = $services->getParserOutputAccess()->getParserOutput(
$pageRecord,
ParserOptions::newCanonical( 'canonical' ),
$revRecord
);
if ( !$status->isOK() ) {
throw new Error( 'Could not load revision for notifications' );
}
$parserOutput = $status->getValue();
$html = $parserOutput->getText();
$doc = DOMUtils::parseHTML( $html );
$container = $doc->getElementsByTagName( 'body' )->item( 0 );
if ( !( $container instanceof Element ) ) {
throw new Error( 'Could not load revision for notifications' );
}
return CommentParser::newFromGlobalState( $container );
}
/**
* @param array &$events
* @param RevisionRecord $newRevRecord
*/
public static function generateEventsForRevision( array &$events, RevisionRecord $newRevRecord ) {
$services = MediaWikiServices::getInstance();
$dtConfig = $services->getConfigFactory()->makeConfig( 'discussiontools' );
if ( !$dtConfig->get( 'DiscussionToolsEnableTopicSubscriptionBackend' ) ) {
// Feature disabled for all users
return;
}
$revisionStore = $services->getRevisionStore();
$userFactory = $services->getUserFactory();
$oldRevRecord = $revisionStore->getPreviousRevision( $newRevRecord, IDBAccessObject::READ_LATEST );
if ( $oldRevRecord === null ) {
// TODO: Handle page creation (oldRevRecord = null?)
return;
}
$title = Title::newFromLinkTarget(
$newRevRecord->getPageAsLinkTarget()
);
if ( !HookUtils::isAvailableForTitle( $title ) ) {
// Not a talk page
return;
}
$user = $newRevRecord->getUser();
if ( !$user ) {
// User can be null if the user is deleted, but this is unlikely
// to be the case if the user just made an edit
return;
}
$oldParser = self::getParsedRevision( $oldRevRecord );
$newParser = self::getParsedRevision( $newRevRecord );
self::generateEventsFromParsers( $events, $oldParser, $newParser, $newRevRecord, $title, $user );
}
/**
* Helper for generateEventsForRevision(), separated out for easier testing.
*
* @param array &$events
* @param CommentParser $oldParser
* @param CommentParser $newParser
* @param RevisionRecord $newRevRecord
* @param PageIdentity $title
* @param UserIdentity $user
*/
protected static function generateEventsFromParsers(
array &$events,
CommentParser $oldParser,
CommentParser $newParser,
RevisionRecord $newRevRecord,
PageIdentity $title,
UserIdentity $user
) {
$newComments = [];
foreach ( $newParser->getCommentItems() as $newComment ) {
if (
$newComment->getAuthor() === $user->getName() &&
// Compare comments by name, as ID could be changed by a parent comment
// being moved/deleted. The downside is that multiple replies within the
// same minute will only fire one notification.
count( $oldParser->findCommentsByName( $newComment->getName() ) ) === 0
) {
$newComments[] = $newComment;
}
}
$mentionedUsers = [];
foreach ( $events as $event ) {
if ( $event['type'] === 'mention' || $event['type'] === 'mention-summary' ) {
// Array is keyed by user id so we can do a simple array merge
$mentionedUsers += $event['extra']['mentioned-users'];
}
}
foreach ( $newComments as $newComment ) {
$heading = $newComment->getHeading();
// Find a level 2 heading, because the interface doesn't allow subscribing to other headings.
// (T286736)
while ( $heading instanceof HeadingItem && $heading->getHeadingLevel() !== 2 ) {
$heading = $heading->getParent();
}
if ( !( $heading instanceof HeadingItem ) ) {
continue;
}
// Check if the name corresponds to a section that contain no comments (only sub-sections).
// The interface doesn't allow subscribing to them either, because they can't be distinguished
// from each other. (T285796)
if ( $heading->getName() === 'h-' ) {
continue;
}
$events[] = [
'type' => 'dt-subscribed-new-comment',
'title' => $title,
'extra' => [
'subscribed-comment-name' => $heading->getName(),
'comment-id' => $newComment->getId(),
'comment-name' => $newComment->getName(),
'content' => $newComment->getBodyText( true ),
'section-title' => $heading->getText(),
'revid' => $newRevRecord->getId(),
'mentioned-users' => $mentionedUsers,
],
'agent' => $user,
];
}
}
/**
* Return all users subscribed to a comment
*
* @param EchoEvent $event
* @param int $batchSize
* @return UserIdentity[]|Iterator<UserIdentity>
*/
public static function locateSubscribedUsers( EchoEvent $event, $batchSize = 500 ) {
$commentName = $event->getExtraParam( 'subscribed-comment-name' );
$subscriptionStore = MediaWikiServices::getInstance()->getService( 'DiscussionTools.SubscriptionStore' );
$subscriptionItems = $subscriptionStore->getSubscriptionItemsForTopic(
$commentName,
1
);
// Update notified timestamps
$subscriptionStore->updateSubscriptionNotifiedTimestamp(
null,
$commentName
);
// TODD: Have this return an Iterator instead?
$users = array_map( static function ( SubscriptionItem $item ) {
return $item->getUserIdentity();
}, $subscriptionItems );
return $users;
}
}