Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
0.00% |
0 / 141 |
|
0.00% |
0 / 13 |
CRAP | |
0.00% |
0 / 1 |
| CompareLanguageConverterOutput | |
0.00% |
0 / 138 |
|
0.00% |
0 / 13 |
462 | |
0.00% |
0 / 1 |
| __construct | |
0.00% |
0 / 12 |
|
0.00% |
0 / 1 |
2 | |||
| execute | |
0.00% |
0 / 19 |
|
0.00% |
0 / 1 |
20 | |||
| newHtmlOutputRendererHelper | |
0.00% |
0 / 10 |
|
0.00% |
0 / 1 |
2 | |||
| getParserOptions | |
0.00% |
0 / 5 |
|
0.00% |
0 / 1 |
2 | |||
| getParserOutput | |
0.00% |
0 / 13 |
|
0.00% |
0 / 1 |
6 | |||
| getParsoidOutput | |
0.00% |
0 / 7 |
|
0.00% |
0 / 1 |
2 | |||
| getWords | |
0.00% |
0 / 3 |
|
0.00% |
0 / 1 |
2 | |||
| getBody | |
0.00% |
0 / 6 |
|
0.00% |
0 / 1 |
6 | |||
| compareOutput | |
0.00% |
0 / 10 |
|
0.00% |
0 / 1 |
2 | |||
| getConverterUsed | |
0.00% |
0 / 7 |
|
0.00% |
0 / 1 |
6 | |||
| mb_sprintf | |
0.00% |
0 / 12 |
|
0.00% |
0 / 1 |
2 | |||
| outputSimilarity | |
0.00% |
0 / 12 |
|
0.00% |
0 / 1 |
2 | |||
| outputDiff | |
0.00% |
0 / 22 |
|
0.00% |
0 / 1 |
12 | |||
| 1 | <?php |
| 2 | /** |
| 3 | * This program is free software; you can redistribute it and/or modify |
| 4 | * it under the terms of the GNU General Public License as published by |
| 5 | * the Free Software Foundation; either version 2 of the License, or |
| 6 | * (at your option) any later version. |
| 7 | * |
| 8 | * This program is distributed in the hope that it will be useful, |
| 9 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 10 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 11 | * GNU General Public License for more details. |
| 12 | * |
| 13 | * You should have received a copy of the GNU General Public License along |
| 14 | * with this program; if not, write to the Free Software Foundation, Inc., |
| 15 | * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. |
| 16 | * http://www.gnu.org/copyleft/gpl.html |
| 17 | * |
| 18 | * @file |
| 19 | * @ingroup Maintenance |
| 20 | */ |
| 21 | |
| 22 | use MediaWiki\Parser\ParserOutput; |
| 23 | use MediaWiki\Rest\Handler\Helper\HtmlOutputRendererHelper; |
| 24 | use MediaWiki\Revision\SlotRecord; |
| 25 | use MediaWiki\Title\Title; |
| 26 | use MediaWiki\User\User; |
| 27 | use Wikimedia\Bcp47Code\Bcp47Code; |
| 28 | use Wikimedia\Diff\ArrayDiffFormatter; |
| 29 | use Wikimedia\Diff\ComplexityException; |
| 30 | use Wikimedia\Diff\Diff; |
| 31 | |
| 32 | require_once __DIR__ . '/Maintenance.php'; |
| 33 | |
| 34 | /** |
| 35 | * Maintenance script that compares variant conversion output between Parser and |
| 36 | * HtmlOutputRendererHelper. |
| 37 | * |
| 38 | * @ingroup Maintenance |
| 39 | */ |
| 40 | class CompareLanguageConverterOutput extends Maintenance { |
| 41 | public function __construct() { |
| 42 | parent::__construct(); |
| 43 | $this->addDescription( 'Compares variant conversion output between Parser and HtmlOutputRendererHelper' ); |
| 44 | $this->addArg( |
| 45 | 'page-title', |
| 46 | 'Name of the page to be parsed and compared', |
| 47 | true |
| 48 | ); |
| 49 | $this->addArg( |
| 50 | 'target-variant', |
| 51 | 'Target variant language code to transform the content to', |
| 52 | true |
| 53 | ); |
| 54 | } |
| 55 | |
| 56 | public function execute() { |
| 57 | $mwInstance = $this->getServiceContainer(); |
| 58 | |
| 59 | $pageName = $this->getArg( 'page-title' ); |
| 60 | $pageTitle = Title::newFromText( $pageName ); |
| 61 | |
| 62 | if ( !$pageTitle || !$pageTitle->exists() ) { |
| 63 | $this->fatalError( "Title with name $pageName not found" ); |
| 64 | } |
| 65 | |
| 66 | $targetVariantCode = $this->getArg( 'target-variant' ); |
| 67 | $languageNameUtils = $mwInstance->getLanguageNameUtils(); |
| 68 | if ( !$languageNameUtils->isValidBuiltInCode( $targetVariantCode ) ) { |
| 69 | $this->fatalError( "$targetVariantCode is not a supported variant" ); |
| 70 | } |
| 71 | $targetVariant = $mwInstance->getLanguageFactory()->getLanguage( |
| 72 | $targetVariantCode |
| 73 | ); |
| 74 | |
| 75 | $user = User::newSystemUser( User::MAINTENANCE_SCRIPT_USER, [ 'steal' => true ] ); |
| 76 | $baseLanguage = $pageTitle->getPageLanguage(); |
| 77 | |
| 78 | $parserOutput = $this->getParserOutput( $pageTitle, $baseLanguage, $targetVariant ); |
| 79 | $parsoidOutput = $this->getParsoidOutput( $pageTitle, $targetVariant, $user ); |
| 80 | $converterUsed = $this->getConverterUsed( $parsoidOutput ); |
| 81 | |
| 82 | $this->compareOutput( $parserOutput, $parsoidOutput, $converterUsed ); |
| 83 | |
| 84 | return true; |
| 85 | } |
| 86 | |
| 87 | private function newHtmlOutputRendererHelper(): HtmlOutputRendererHelper { |
| 88 | $services = $this->getServiceContainer(); |
| 89 | |
| 90 | $helper = new HtmlOutputRendererHelper( |
| 91 | $services->getParsoidOutputStash(), |
| 92 | new NullStatsdDataFactory(), |
| 93 | $services->getParsoidOutputAccess(), |
| 94 | $services->getHtmlTransformFactory(), |
| 95 | $services->getContentHandlerFactory(), |
| 96 | $services->getLanguageFactory() |
| 97 | ); |
| 98 | return $helper; |
| 99 | } |
| 100 | |
| 101 | private function getParserOptions( Language $language ): ParserOptions { |
| 102 | $parserOpts = ParserOptions::newFromAnon(); |
| 103 | $parserOpts->setTargetLanguage( $language ); |
| 104 | $parserOpts->disableContentConversion( false ); |
| 105 | $parserOpts->disableTitleConversion( false ); |
| 106 | |
| 107 | return $parserOpts; |
| 108 | } |
| 109 | |
| 110 | private function getParserOutput( |
| 111 | Title $pageTitle, |
| 112 | Language $baseLanguage, |
| 113 | Language $targetVariant |
| 114 | ): ParserOutput { |
| 115 | // We update the default language variant because we want Parser to |
| 116 | // perform variant conversion to it. |
| 117 | global $wgDefaultLanguageVariant; |
| 118 | $wgDefaultLanguageVariant = $targetVariant->getCode(); |
| 119 | |
| 120 | $mwInstance = $this->getServiceContainer(); |
| 121 | |
| 122 | $languageFactory = $mwInstance->getLanguageFactory(); |
| 123 | $parser = $mwInstance->getParser(); |
| 124 | $parserOptions = $this->getParserOptions( |
| 125 | $languageFactory->getParentLanguage( $baseLanguage ) |
| 126 | ); |
| 127 | |
| 128 | $content = $mwInstance->getRevisionLookup() |
| 129 | ->getRevisionByTitle( $pageTitle ) |
| 130 | ->getContent( SlotRecord::MAIN ); |
| 131 | $wikiContent = ( $content instanceof TextContent ) ? $content->getText() : ''; |
| 132 | |
| 133 | return $parser->parse( $wikiContent, $pageTitle, $parserOptions ); |
| 134 | } |
| 135 | |
| 136 | private function getParsoidOutput( |
| 137 | Title $pageTitle, |
| 138 | Bcp47Code $targetVariant, |
| 139 | User $user |
| 140 | ): ParserOutput { |
| 141 | $htmlOutputRendererHelper = $this->newHtmlOutputRendererHelper(); |
| 142 | $htmlOutputRendererHelper->init( $pageTitle, [ |
| 143 | 'stash' => false, |
| 144 | 'flavor' => 'view', |
| 145 | ], $user ); |
| 146 | $htmlOutputRendererHelper->setVariantConversionLanguage( $targetVariant ); |
| 147 | |
| 148 | return $htmlOutputRendererHelper->getHtml(); |
| 149 | } |
| 150 | |
| 151 | private function getWords( string $output ): array { |
| 152 | $tagsRemoved = strip_tags( $output ); |
| 153 | $words = preg_split( '/\s+/', trim( $tagsRemoved ), -1, PREG_SPLIT_NO_EMPTY ); |
| 154 | return $words; |
| 155 | } |
| 156 | |
| 157 | private function getBody( string $output ): string { |
| 158 | $dom = new DOMDocument(); |
| 159 | // phpcs:disable Generic.PHP.NoSilencedErrors.Discouraged |
| 160 | @$dom->loadHTML( $output ); |
| 161 | $body = $dom->getElementsByTagName( 'body' )->item( 0 ); |
| 162 | if ( $body === null ) { |
| 163 | // Body element not present |
| 164 | return $output; |
| 165 | } |
| 166 | |
| 167 | return $body->textContent; |
| 168 | } |
| 169 | |
| 170 | private function compareOutput( |
| 171 | ParserOutput $parserOutput, |
| 172 | ParserOutput $parsoidOutput, |
| 173 | string $converterUsed |
| 174 | ): void { |
| 175 | $parsoidText = $parsoidOutput->getText( [ 'deduplicateStyles' => false ] ); |
| 176 | $parserText = $parserOutput->getText( [ 'deduplicateStyles' => false ] ); |
| 177 | |
| 178 | $parsoidWords = $this->getWords( $this->getBody( $parsoidText ) ); |
| 179 | $parserWords = $this->getWords( $parserText ); |
| 180 | |
| 181 | $parserWordCount = count( $parserWords ); |
| 182 | $parsoidWordCount = count( $parsoidWords ); |
| 183 | $this->output( "Word count: Parsoid: $parsoidWordCount; Parser: $parserWordCount\n" ); |
| 184 | |
| 185 | $this->outputSimilarity( $parsoidWords, $parserWords ); |
| 186 | $this->output( "\n" ); |
| 187 | $this->outputDiff( $parsoidWords, $parserWords, $converterUsed ); |
| 188 | } |
| 189 | |
| 190 | private function getConverterUsed( ParserOutput $parsoidOutput ): string { |
| 191 | $isCoreConverterUsed = strpos( |
| 192 | $parsoidOutput->getRawText(), |
| 193 | 'Variant conversion performed using the core LanguageConverter' |
| 194 | ); |
| 195 | |
| 196 | if ( $isCoreConverterUsed ) { |
| 197 | return 'Core LanguageConverter'; |
| 198 | } else { |
| 199 | return 'Parsoid LanguageConverter'; |
| 200 | } |
| 201 | } |
| 202 | |
| 203 | // Inspired from: https://stackoverflow.com/a/55927237/903324 |
| 204 | private function mb_sprintf( string $format, ...$args ): string { |
| 205 | $params = $args; |
| 206 | |
| 207 | return sprintf( |
| 208 | preg_replace_callback( |
| 209 | '/(?<=%|%-)\d+(?=s)/', |
| 210 | static function ( array $matches ) use ( &$params ) { |
| 211 | $value = array_shift( $params ); |
| 212 | |
| 213 | return (string)( strlen( $value ) - mb_strlen( $value ) + $matches[0] ); |
| 214 | }, |
| 215 | $format |
| 216 | ), |
| 217 | ...$args |
| 218 | ); |
| 219 | } |
| 220 | |
| 221 | private function outputSimilarity( array $parsoidWords, array $parserWords ): void { |
| 222 | $parsoidOutput = implode( ' ', $parsoidWords ); |
| 223 | $parserOutput = implode( ' ', $parserWords ); |
| 224 | $this->output( |
| 225 | 'Total characters: Parsoid: ' . strlen( $parsoidOutput ) . |
| 226 | '; Parser: ' . strlen( $parserOutput ) . "\n" |
| 227 | ); |
| 228 | |
| 229 | $similarityPercent = 0; |
| 230 | $similarCharacters = similar_text( $parsoidOutput, $parserOutput, $similarityPercent ); |
| 231 | $similarityPercent = round( $similarityPercent, 2 ); |
| 232 | |
| 233 | $this->output( |
| 234 | "Similarity via similar_text(): $similarityPercent%; Similar characters: $similarCharacters" |
| 235 | ); |
| 236 | } |
| 237 | |
| 238 | private function outputDiff( array $parsoidWords, array $parserWords, string $converterUsed ): void { |
| 239 | $out = str_repeat( '-', 96 ) . "\n"; |
| 240 | $out .= sprintf( "| %5s | %-35s | %-35s | %-8s |\n", 'Line', 'Parsoid', 'Parser', 'Diff' ); |
| 241 | $out .= sprintf( "| %5s | %-35s | %-35s | %-8s |\n", '', "($converterUsed)", '', '' ); |
| 242 | $out .= str_repeat( '-', 96 ) . "\n"; |
| 243 | |
| 244 | try { |
| 245 | $diff = new Diff( $parsoidWords, $parserWords ); |
| 246 | } catch ( ComplexityException $e ) { |
| 247 | $this->output( $e->getMessage() ); |
| 248 | $this->error( 'Encountered ComplexityException while computing diff' ); |
| 249 | } |
| 250 | |
| 251 | // Print the difference between the words |
| 252 | $wordDiffFormat = ( new ArrayDiffFormatter() )->format( $diff ); |
| 253 | foreach ( $wordDiffFormat as $index => $wordDiff ) { |
| 254 | $action = $wordDiff['action']; |
| 255 | $old = $wordDiff['old'] ?? null; |
| 256 | $new = $wordDiff['new'] ?? null; |
| 257 | |
| 258 | $out .= $this->mb_sprintf( |
| 259 | "| %5s | %-35s | %-35s | %-8s |\n", |
| 260 | str_pad( (string)( $index + 1 ), 5, ' ', STR_PAD_LEFT ), |
| 261 | mb_strimwidth( $old ?? '- N/A -', 0, 35, '…' ), |
| 262 | mb_strimwidth( $new ?? '- N/A -', 0, 35, '…' ), |
| 263 | $action |
| 264 | ); |
| 265 | } |
| 266 | |
| 267 | // Print the footer. |
| 268 | $out .= str_repeat( '-', 96 ) . "\n"; |
| 269 | $this->output( "\n" . $out ); |
| 270 | } |
| 271 | } |
| 272 | |
| 273 | $maintClass = CompareLanguageConverterOutput::class; |
| 274 | require_once RUN_MAINTENANCE_IF_MAIN; |