wikimedia/mediawiki-extensions-TextExtracts
1
0
Fork 0
mirror of https://gerrit.wikimedia.org/r/mediawiki/extensions/TextExtracts synced 2024-11-23 15:56:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge "ApiQueryExtracts: Use raw text from ParserOutput"

Browse source
This commit is contained in:
jenkins-bot 2024-11-15 15:36:35 +00:00 committed by Gerrit Code Review
parent 56b3c275d6 207242f93d
commit bc13671078
5 changed files with 8 additions and 41 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
extension.json
View file

@ -56,7 +56,9 @@
"input",
"style",
"ul.gallery",
".mw-editsection",
"mw\\:editsection",
"editsection",
"meta",
"sup.reference",
"ol.references",
".error",

9
includes/ApiQueryExtracts.php
View file

@ -27,7 +27,7 @@ class ApiQueryExtracts extends ApiQueryBase {
/**
* Bump when memcache needs clearing
*/
private const CACHE_VERSION = 2;
private const CACHE_VERSION = 3;
private const PREFIX = 'ex';
@ -221,10 +221,9 @@ class ApiQueryExtracts extends ApiQueryBase {
*/
private function getFirstSection( $text, $plainText ) {
if ( $plainText ) {
$regexp = '/^.*?(?=' . ExtractFormatter::SECTION_MARKER_START .
'(?!.' . ExtractFormatter::SECTION_MARKER_END . '<h2 id="mw-toc-heading"))/s';
$regexp = '/^(.*?)(?=' . ExtractFormatter::SECTION_MARKER_START . ')/s';
} else {
$regexp = '/^.*?(?=<h[1-6]\b(?! id="mw-toc-heading"))/s';
$regexp = '/^(.*?)(?=<h[1-6]\b)/s';
}
if ( preg_match( $regexp, $text, $matches ) ) {
$text = $matches[0];
@ -246,7 +245,7 @@ class ApiQueryExtracts extends ApiQueryBase {
);
if ( $status->isOK() ) {
$pout = $status->getValue();
$text = $pout->getText( [ 'unwrap' => true ] );
$text = $pout->getRawText();
if ( $this->params['intro'] ) {
$text = $this->getFirstSection( $text, false );
}

15
includes/ExtractFormatter.php
View file

@ -4,7 +4,6 @@ namespace MediaWiki\Extension\TextExtracts;
use DOMElement;
use HtmlFormatter\HtmlFormatter;
use Wikimedia\Parsoid\Utils\DOMCompat;
/**
* Provides text-only or limited-HTML extracts of page HTML
@ -81,21 +80,9 @@ class ExtractFormatter extends HtmlFormatter {
* @return array Array of removed DOMElements
*/
public function filterContent(): array {
$doc = $this->getDoc();
// Headings in a DIV wrapper may get removed by $wgExtractsRemoveClasses,
// move it outside the header to rescue it (T363445)
// https://www.mediawiki.org/wiki/Heading_HTML_changes
$headings = DOMCompat::querySelectorAll( $doc->documentElement, 'h1, h2, h3, h4, h5, h6' );
foreach ( $headings as $heading ) {
// @phan-suppress-next-line PhanTypeMismatchArgumentSuperType
if ( DOMCompat::getClassList( $heading->parentNode )->contains( 'mw-heading' ) ) {
$heading->parentNode->parentNode->insertBefore( $heading, $heading->parentNode );
}
}
$removed = parent::filterContent();
$doc = $this->getDoc();
$spans = $doc->getElementsByTagName( 'span' );
/** @var DOMElement $span */

10
tests/phpunit/ApiQueryExtractsTest.php
View file

@ -160,16 +160,6 @@ class ApiQueryExtractsTest extends \MediaWikiIntegrationTestCase {
false,
'Example <h11>...',
],
'__TOC__ before intro (HTML)' => [
'<h2 id="mw-toc-heading">Contents</h2>Intro<h2>Actual heading</h2>...',
false,
'<h2 id="mw-toc-heading">Contents</h2>Intro',
],
'__TOC__ before intro (plaintext)' => [
"\1\2_\2\1<h2 id=\"mw-toc-heading\">Contents</h2>Intro\1\2_\2\1<h2>Actual heading</h2>...",
true,
"\1\2_\2\1<h2 id=\"mw-toc-heading\">Contents</h2>Intro",
],
];
}

11
tests/phpunit/ExtractFormatterTest.php
View file

@ -67,17 +67,6 @@ class ExtractFormatterTest extends MediaWikiIntegrationTestCase {
'<span class="foo"><span style="bar: baz;" lang="qux">quux</span></span>',
false,
],
[
// New heading markup from T13555 (T363445)
"<p>hello</p><h2 id=\"heading\">heading</h2><p>bye</p>",
'<p>hello</p>' .
'<div class="mw-heading mw-heading2">' .
'<h2 id="heading">heading</h2>' .
'<span class="mw-editsection">xxx</span>' .
'</div>' .
'<p>bye</p>',
false,
],
[
// Verify that TOC is properly removed (HTML mode)
"Lead\n<h1>Section</h1>\n<p>Section text</p>",