merged from master
[lhc/web/wiklou.git] / includes / api / ApiParse.php
1 <?php
2 /**
3 * Created on Dec 01, 2007
4 *
5 * Copyright © 2007 Yuri Astrakhan <Firstname><Lastname>@gmail.com
6 *
7 * This program is free software; you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation; either version 2 of the License, or
10 * (at your option) any later version.
11 *
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU General Public License for more details.
16 *
17 * You should have received a copy of the GNU General Public License along
18 * with this program; if not, write to the Free Software Foundation, Inc.,
19 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
20 * http://www.gnu.org/copyleft/gpl.html
21 *
22 * @file
23 */
24
25 /**
26 * @ingroup API
27 */
28 class ApiParse extends ApiBase {
29 private $section, $text, $pstText = null;
30
31 public function __construct( $main, $action ) {
32 parent::__construct( $main, $action );
33 }
34
35 public function execute() {
36 // The data is hot but user-dependent, like page views, so we set vary cookies
37 $this->getMain()->setCacheMode( 'anon-public-user-private' );
38
39 // Get parameters
40 $params = $this->extractRequestParams();
41 $text = $params['text'];
42 $title = $params['title'];
43 $page = $params['page'];
44 $pageid = $params['pageid'];
45 $oldid = $params['oldid'];
46
47 if ( !is_null( $page ) && ( !is_null( $text ) || $title != 'API' ) ) {
48 $this->dieUsage( 'The page parameter cannot be used together with the text and title parameters', 'params' );
49 }
50
51 $prop = array_flip( $params['prop'] );
52
53 if ( isset( $params['section'] ) ) {
54 $this->section = $params['section'];
55 } else {
56 $this->section = false;
57 }
58
59 // The parser needs $wgTitle to be set, apparently the
60 // $title parameter in Parser::parse isn't enough *sigh*
61 // TODO: Does this still need $wgTitle?
62 global $wgParser, $wgTitle, $wgLang;
63
64 // Currently unnecessary, code to act as a safeguard against any change in current behaviour of uselang breaks
65 $oldLang = null;
66 if ( isset( $params['uselang'] ) && $params['uselang'] != $wgLang->getCode() ) {
67 $oldLang = $wgLang; // Backup wgLang
68 $wgLang = Language::factory( $params['uselang'] );
69 }
70
71 $popts = ParserOptions::newFromContext( $this->getContext() );
72 $popts->setTidy( true );
73 $popts->enableLimitReport( !$params['disablepp'] );
74
75 $redirValues = null;
76
77 // Return result
78 $result = $this->getResult();
79
80 if ( !is_null( $oldid ) || !is_null( $pageid ) || !is_null( $page ) ) {
81 if ( !is_null( $oldid ) ) {
82 // Don't use the parser cache
83 $rev = Revision::newFromID( $oldid );
84 if ( !$rev ) {
85 $this->dieUsage( "There is no revision ID $oldid", 'missingrev' );
86 }
87 if ( !$rev->userCan( Revision::DELETED_TEXT, $this->getUser() ) ) {
88 $this->dieUsage( "You don't have permission to view deleted revisions", 'permissiondenied' );
89 }
90
91 $titleObj = $rev->getTitle();
92
93 $wgTitle = $titleObj;
94
95 // If for some reason the "oldid" is actually the current revision, it may be cached
96 if ( $titleObj->getLatestRevID() === intval( $oldid ) ) {
97 // May get from/save to parser cache
98 $p_result = $this->getParsedSectionOrText( $titleObj, $popts, $pageid,
99 isset( $prop['wikitext'] ) ) ;
100 } else { // This is an old revision, so get the text differently
101 $this->text = $rev->getText( Revision::FOR_THIS_USER, $this->getUser() );
102
103 if ( $this->section !== false ) {
104 $this->text = $this->getSectionText( $this->text, 'r' . $rev->getId() );
105 }
106
107 // Should we save old revision parses to the parser cache?
108 $p_result = $wgParser->parse( $this->text, $titleObj, $popts );
109 }
110 } else { // Not $oldid, but $pageid or $page
111 if ( $params['redirects'] ) {
112 $reqParams = array(
113 'action' => 'query',
114 'redirects' => '',
115 );
116 if ( !is_null ( $pageid ) ) {
117 $reqParams['pageids'] = $pageid;
118 } else { // $page
119 $reqParams['titles'] = $page;
120 }
121 $req = new FauxRequest( $reqParams );
122 $main = new ApiMain( $req );
123 $main->execute();
124 $data = $main->getResultData();
125 $redirValues = isset( $data['query']['redirects'] )
126 ? $data['query']['redirects']
127 : array();
128 $to = $page;
129 foreach ( (array)$redirValues as $r ) {
130 $to = $r['to'];
131 }
132 $titleObj = Title::newFromText( $to );
133 } else {
134 if ( !is_null ( $pageid ) ) {
135 $reqParams['pageids'] = $pageid;
136 $titleObj = Title::newFromID( $pageid );
137 } else { // $page
138 $to = $page;
139 $titleObj = Title::newFromText( $to );
140 }
141 }
142 if ( !is_null ( $pageid ) ) {
143 if ( !$titleObj ) {
144 // Still throw nosuchpageid error if pageid was provided
145 $this->dieUsageMsg( array( 'nosuchpageid', $pageid ) );
146 }
147 } elseif ( !$titleObj || !$titleObj->exists() ) {
148 $this->dieUsage( "The page you specified doesn't exist", 'missingtitle' );
149 }
150 $wgTitle = $titleObj;
151
152 if ( isset( $prop['revid'] ) ) {
153 $oldid = $titleObj->getLatestRevID();
154 }
155
156 // Potentially cached
157 $p_result = $this->getParsedSectionOrText( $titleObj, $popts, $pageid,
158 isset( $prop['wikitext'] ) ) ;
159 }
160 } else { // Not $oldid, $pageid, $page. Hence based on $text
161
162 if ( is_null( $text ) ) {
163 $this->dieUsage( 'The text parameter should be passed with the title parameter. Should you be using the "page" parameter instead?', 'params' );
164 }
165 $this->text = $text;
166 $titleObj = Title::newFromText( $title );
167 if ( !$titleObj ) {
168 $this->dieUsageMsg( array( 'invalidtitle', $title ) );
169 }
170 $wgTitle = $titleObj;
171
172 if ( $this->section !== false ) {
173 $this->text = $this->getSectionText( $this->text, $titleObj->getText() );
174 }
175
176 if ( $params['pst'] || $params['onlypst'] ) {
177 $this->pstText = $wgParser->preSaveTransform( $this->text, $titleObj, $this->getUser(), $popts );
178 }
179 if ( $params['onlypst'] ) {
180 // Build a result and bail out
181 $result_array = array();
182 $result_array['text'] = array();
183 $result->setContent( $result_array['text'], $this->pstText );
184 if ( isset( $prop['wikitext'] ) ) {
185 $result_array['wikitext'] = array();
186 $result->setContent( $result_array['wikitext'], $this->text );
187 }
188 $result->addValue( null, $this->getModuleName(), $result_array );
189 return;
190 }
191 // Not cached (save or load)
192 $p_result = $wgParser->parse( $params['pst'] ? $this->pstText : $this->text, $titleObj, $popts ); #FIXME: use Content object¡
193 }
194
195 $result_array = array();
196
197 $result_array['title'] = $titleObj->getPrefixedText();
198
199 if ( !is_null( $oldid ) ) {
200 $result_array['revid'] = intval( $oldid );
201 }
202
203 if ( $params['redirects'] && !is_null( $redirValues ) ) {
204 $result_array['redirects'] = $redirValues;
205 }
206
207 if ( isset( $prop['text'] ) ) {
208 $result_array['text'] = array();
209 $result->setContent( $result_array['text'], $p_result->getText() );
210 }
211
212 if ( !is_null( $params['summary'] ) ) {
213 $result_array['parsedsummary'] = array();
214 $result->setContent( $result_array['parsedsummary'], Linker::formatComment( $params['summary'], $titleObj ) );
215 }
216
217 if ( isset( $prop['langlinks'] ) ) {
218 $result_array['langlinks'] = $this->formatLangLinks( $p_result->getLanguageLinks() );
219 }
220 if ( isset( $prop['languageshtml'] ) ) {
221 $languagesHtml = $this->languagesHtml( $p_result->getLanguageLinks() );
222 $result_array['languageshtml'] = array();
223 $result->setContent( $result_array['languageshtml'], $languagesHtml );
224 }
225 if ( isset( $prop['categories'] ) ) {
226 $result_array['categories'] = $this->formatCategoryLinks( $p_result->getCategories() );
227 }
228 if ( isset( $prop['categorieshtml'] ) ) {
229 $categoriesHtml = $this->categoriesHtml( $p_result->getCategories() );
230 $result_array['categorieshtml'] = array();
231 $result->setContent( $result_array['categorieshtml'], $categoriesHtml );
232 }
233 if ( isset( $prop['links'] ) ) {
234 $result_array['links'] = $this->formatLinks( $p_result->getLinks() );
235 }
236 if ( isset( $prop['templates'] ) ) {
237 $result_array['templates'] = $this->formatLinks( $p_result->getTemplates() );
238 }
239 if ( isset( $prop['images'] ) ) {
240 $result_array['images'] = array_keys( $p_result->getImages() );
241 }
242 if ( isset( $prop['externallinks'] ) ) {
243 $result_array['externallinks'] = array_keys( $p_result->getExternalLinks() );
244 }
245 if ( isset( $prop['sections'] ) ) {
246 $result_array['sections'] = $p_result->getSections();
247 }
248
249 if ( isset( $prop['displaytitle'] ) ) {
250 $result_array['displaytitle'] = $p_result->getDisplayTitle() ?
251 $p_result->getDisplayTitle() :
252 $titleObj->getPrefixedText();
253 }
254
255 if ( isset( $prop['headitems'] ) || isset( $prop['headhtml'] ) ) {
256 $context = $this->getContext();
257 $context->setTitle( $titleObj );
258 $context->getOutput()->addParserOutputNoText( $p_result );
259
260 if ( isset( $prop['headitems'] ) ) {
261 $headItems = $this->formatHeadItems( $p_result->getHeadItems() );
262
263 $css = $this->formatCss( $context->getOutput()->buildCssLinksArray() );
264
265 $scripts = array( $context->getOutput()->getHeadScripts() );
266
267 $result_array['headitems'] = array_merge( $headItems, $css, $scripts );
268 }
269
270 if ( isset( $prop['headhtml'] ) ) {
271 $result_array['headhtml'] = array();
272 $result->setContent( $result_array['headhtml'], $context->getOutput()->headElement( $context->getSkin() ) );
273 }
274 }
275
276 if ( isset( $prop['iwlinks'] ) ) {
277 $result_array['iwlinks'] = $this->formatIWLinks( $p_result->getInterwikiLinks() );
278 }
279
280 if ( isset( $prop['wikitext'] ) ) {
281 $result_array['wikitext'] = array();
282 $result->setContent( $result_array['wikitext'], $this->text );
283 if ( !is_null( $this->pstText ) ) {
284 $result_array['psttext'] = array();
285 $result->setContent( $result_array['psttext'], $this->pstText );
286 }
287 }
288
289 $result_mapping = array(
290 'redirects' => 'r',
291 'langlinks' => 'll',
292 'categories' => 'cl',
293 'links' => 'pl',
294 'templates' => 'tl',
295 'images' => 'img',
296 'externallinks' => 'el',
297 'iwlinks' => 'iw',
298 'sections' => 's',
299 'headitems' => 'hi',
300 );
301 $this->setIndexedTagNames( $result_array, $result_mapping );
302 $result->addValue( null, $this->getModuleName(), $result_array );
303
304 if ( !is_null( $oldLang ) ) {
305 $wgLang = $oldLang; // Reset $wgLang to $oldLang
306 }
307 }
308
309 /**
310 * @param $titleObj Title
311 * @param $popts ParserOptions
312 * @param $pageId Int
313 * @param $getWikitext Bool
314 * @return ParserOutput
315 */
316 private function getParsedSectionOrText( $titleObj, $popts, $pageId = null, $getWikitext = false ) {
317 global $wgParser;
318
319 $page = WikiPage::factory( $titleObj );
320
321 if ( $this->section !== false ) { #FIXME: get section Content, get parser output, ...
322 $this->text = $this->getSectionText( $page->getRawText(), !is_null( $pageId )
323 ? 'page id ' . $pageId : $titleObj->getText() ); #FIXME: get section...
324
325 // Not cached (save or load)
326 return $wgParser->parse( $this->text, $titleObj, $popts );
327 } else {
328 // Try the parser cache first
329 // getParserOutput will save to Parser cache if able
330 $pout = $page->getParserOutput( $popts );
331 if ( !$pout ) {
332 $this->dieUsage( "There is no revision ID {$page->getLatest()}", 'missingrev' );
333 }
334 if ( $getWikitext ) {
335 $this->content = $page->getContent( Revision::RAW ); #FIXME: use $this->content everywhere
336 $this->text = ContentHandler::getContentText( $this->content ); #FIXME: serialize, get format from params; or use object structure in result?
337 }
338 return $pout;
339 }
340 }
341
342 private function getSectionText( $text, $what ) { #FIXME: replace with Content::getSection
343 global $wgParser;
344 // Not cached (save or load)
345 $text = $wgParser->getSection( $text, $this->section, false );
346 if ( $text === false ) {
347 $this->dieUsage( "There is no section {$this->section} in " . $what, 'nosuchsection' );
348 }
349 return $text;
350 }
351
352 private function formatLangLinks( $links ) {
353 $result = array();
354 foreach ( $links as $link ) {
355 $entry = array();
356 $bits = explode( ':', $link, 2 );
357 $title = Title::newFromText( $link );
358
359 $entry['lang'] = $bits[0];
360 if ( $title ) {
361 $entry['url'] = wfExpandUrl( $title->getFullURL(), PROTO_CURRENT );
362 }
363 $this->getResult()->setContent( $entry, $bits[1] );
364 $result[] = $entry;
365 }
366 return $result;
367 }
368
369 private function formatCategoryLinks( $links ) {
370 $result = array();
371 foreach ( $links as $link => $sortkey ) {
372 $entry = array();
373 $entry['sortkey'] = $sortkey;
374 $this->getResult()->setContent( $entry, $link );
375 $result[] = $entry;
376 }
377 return $result;
378 }
379
380 private function categoriesHtml( $categories ) {
381 $context = $this->getContext();
382 $context->getOutput()->addCategoryLinks( $categories );
383 return $context->getSkin()->getCategories();
384 }
385
386 /**
387 * @deprecated since 1.18 No modern skin generates language links this way, please use language links
388 * data to generate your own HTML.
389 * @param $languages array
390 * @return string
391 */
392 private function languagesHtml( $languages ) {
393 wfDeprecated( __METHOD__, '1.18' );
394
395 global $wgContLang, $wgHideInterlanguageLinks;
396
397 if ( $wgHideInterlanguageLinks || count( $languages ) == 0 ) {
398 return '';
399 }
400
401 $s = htmlspecialchars( wfMsg( 'otherlanguages' ) . wfMsg( 'colon-separator' ) );
402
403 $langs = array();
404 foreach ( $languages as $l ) {
405 $nt = Title::newFromText( $l );
406 $text = Language::fetchLanguageName( $nt->getInterwiki() );
407
408 $langs[] = Html::element( 'a',
409 array( 'href' => $nt->getFullURL(), 'title' => $nt->getText(), 'class' => "external" ),
410 $text == '' ? $l : $text );
411 }
412
413 $s .= implode( htmlspecialchars( wfMsgExt( 'pipe-separator', 'escapenoentities' ) ), $langs );
414
415 if ( $wgContLang->isRTL() ) {
416 $s = Html::rawElement( 'span', array( 'dir' => "LTR" ), $s );
417 }
418
419 return $s;
420 }
421
422 private function formatLinks( $links ) {
423 $result = array();
424 foreach ( $links as $ns => $nslinks ) {
425 foreach ( $nslinks as $title => $id ) {
426 $entry = array();
427 $entry['ns'] = $ns;
428 $this->getResult()->setContent( $entry, Title::makeTitle( $ns, $title )->getFullText() );
429 if ( $id != 0 ) {
430 $entry['exists'] = '';
431 }
432 $result[] = $entry;
433 }
434 }
435 return $result;
436 }
437
438 private function formatIWLinks( $iw ) {
439 $result = array();
440 foreach ( $iw as $prefix => $titles ) {
441 foreach ( array_keys( $titles ) as $title ) {
442 $entry = array();
443 $entry['prefix'] = $prefix;
444
445 $title = Title::newFromText( "{$prefix}:{$title}" );
446 if ( $title ) {
447 $entry['url'] = wfExpandUrl( $title->getFullURL(), PROTO_CURRENT );
448 }
449
450 $this->getResult()->setContent( $entry, $title->getFullText() );
451 $result[] = $entry;
452 }
453 }
454 return $result;
455 }
456
457 private function formatHeadItems( $headItems ) {
458 $result = array();
459 foreach ( $headItems as $tag => $content ) {
460 $entry = array();
461 $entry['tag'] = $tag;
462 $this->getResult()->setContent( $entry, $content );
463 $result[] = $entry;
464 }
465 return $result;
466 }
467
468 private function formatCss( $css ) {
469 $result = array();
470 foreach ( $css as $file => $link ) {
471 $entry = array();
472 $entry['file'] = $file;
473 $this->getResult()->setContent( $entry, $link );
474 $result[] = $entry;
475 }
476 return $result;
477 }
478
479 private function setIndexedTagNames( &$array, $mapping ) {
480 foreach ( $mapping as $key => $name ) {
481 if ( isset( $array[$key] ) ) {
482 $this->getResult()->setIndexedTagName( $array[$key], $name );
483 }
484 }
485 }
486
487 public function getAllowedParams() {
488 return array(
489 'title' => array(
490 ApiBase::PARAM_DFLT => 'API',
491 ),
492 'text' => null,
493 'summary' => null,
494 'page' => null,
495 'pageid' => array(
496 ApiBase::PARAM_TYPE => 'integer',
497 ),
498 'redirects' => false,
499 'oldid' => array(
500 ApiBase::PARAM_TYPE => 'integer',
501 ),
502 'prop' => array(
503 ApiBase::PARAM_DFLT => 'text|langlinks|categories|links|templates|images|externallinks|sections|revid|displaytitle',
504 ApiBase::PARAM_ISMULTI => true,
505 ApiBase::PARAM_TYPE => array(
506 'text',
507 'langlinks',
508 'languageshtml',
509 'categories',
510 'categorieshtml',
511 'links',
512 'templates',
513 'images',
514 'externallinks',
515 'sections',
516 'revid',
517 'displaytitle',
518 'headitems',
519 'headhtml',
520 'iwlinks',
521 'wikitext',
522 )
523 ),
524 'pst' => false,
525 'onlypst' => false,
526 'uselang' => null,
527 'section' => null,
528 'disablepp' => false,
529 );
530 }
531
532 public function getParamDescription() {
533 $p = $this->getModulePrefix();
534 return array(
535 'text' => 'Wikitext to parse',
536 'summary' => 'Summary to parse',
537 'redirects' => "If the {$p}page or the {$p}pageid parameter is set to a redirect, resolve it",
538 'title' => 'Title of page the text belongs to',
539 'page' => "Parse the content of this page. Cannot be used together with {$p}text and {$p}title",
540 'pageid' => "Parse the content of this page. Overrides {$p}page",
541 'oldid' => "Parse the content of this revision. Overrides {$p}page and {$p}pageid",
542 'prop' => array(
543 'Which pieces of information to get',
544 ' text - Gives the parsed text of the wikitext',
545 ' langlinks - Gives the language links in the parsed wikitext',
546 ' categories - Gives the categories in the parsed wikitext',
547 ' categorieshtml - Gives the HTML version of the categories',
548 ' languageshtml - Gives the HTML version of the language links',
549 ' links - Gives the internal links in the parsed wikitext',
550 ' templates - Gives the templates in the parsed wikitext',
551 ' images - Gives the images in the parsed wikitext',
552 ' externallinks - Gives the external links in the parsed wikitext',
553 ' sections - Gives the sections in the parsed wikitext',
554 ' revid - Adds the revision ID of the parsed page',
555 ' displaytitle - Adds the title of the parsed wikitext',
556 ' headitems - Gives items to put in the <head> of the page',
557 ' headhtml - Gives parsed <head> of the page',
558 ' iwlinks - Gives interwiki links in the parsed wikitext',
559 ' wikitext - Gives the original wikitext that was parsed',
560 ),
561 'pst' => array(
562 'Do a pre-save transform on the input before parsing it',
563 'Ignored if page, pageid or oldid is used'
564 ),
565 'onlypst' => array(
566 'Do a pre-save transform (PST) on the input, but don\'t parse it',
567 'Returns the same wikitext, after a PST has been applied. Ignored if page, pageid or oldid is used'
568 ),
569 'uselang' => 'Which language to parse the request in',
570 'section' => 'Only retrieve the content of this section number',
571 'disablepp' => 'Disable the PP Report from the parser output',
572 );
573 }
574
575 public function getDescription() {
576 return 'Parses wikitext and returns parser output';
577 }
578
579 public function getPossibleErrors() {
580 return array_merge( parent::getPossibleErrors(), array(
581 array( 'code' => 'params', 'info' => 'The page parameter cannot be used together with the text and title parameters' ),
582 array( 'code' => 'params', 'info' => 'The text parameter should be passed with the title parameter. Should you be using the "page" parameter instead?' ),
583 array( 'code' => 'missingrev', 'info' => 'There is no revision ID oldid' ),
584 array( 'code' => 'permissiondenied', 'info' => 'You don\'t have permission to view deleted revisions' ),
585 array( 'code' => 'missingtitle', 'info' => 'The page you specified doesn\'t exist' ),
586 array( 'code' => 'nosuchsection', 'info' => 'There is no section sectionnumber in page' ),
587 array( 'nosuchpageid' ),
588 array( 'invalidtitle', 'title' ),
589 ) );
590 }
591
592 public function getExamples() {
593 return array(
594 'api.php?action=parse&text={{Project:Sandbox}}'
595 );
596 }
597
598 public function getHelpUrls() {
599 return 'https://www.mediawiki.org/wiki/API:Parsing_wikitext#parse';
600 }
601
602 public function getVersion() {
603 return __CLASS__ . ': $Id$';
604 }
605 }