Add missing type to @param documentation
[lhc/web/wiklou.git] / includes / content / TextContent.php
1 <?php
2 /**
3 * Content object implementation for representing flat text.
4 *
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2 of the License, or
8 * (at your option) any later version.
9 *
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
14 *
15 * You should have received a copy of the GNU General Public License along
16 * with this program; if not, write to the Free Software Foundation, Inc.,
17 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 * http://www.gnu.org/copyleft/gpl.html
19 *
20 * @since 1.21
21 *
22 * @file
23 * @ingroup Content
24 *
25 * @author Daniel Kinzler
26 */
27
28 /**
29 * Content object implementation for representing flat text.
30 *
31 * TextContent instances are immutable
32 *
33 * @ingroup Content
34 */
35 class TextContent extends AbstractContent {
36
37 /**
38 * @param string $text
39 * @param string $model_id
40 * @throws MWException
41 */
42 public function __construct( $text, $model_id = CONTENT_MODEL_TEXT ) {
43 parent::__construct( $model_id );
44
45 if ( $text === null || $text === false ) {
46 wfWarn( "TextContent constructed with \$text = " . var_export( $text, true ) . "! "
47 . "This may indicate an error in the caller's scope.", 2 );
48
49 $text = '';
50 }
51
52 if ( !is_string( $text ) ) {
53 throw new MWException( "TextContent expects a string in the constructor." );
54 }
55
56 $this->mText = $text;
57 }
58
59 /**
60 * @note Mutable subclasses MUST override this to return a copy!
61 *
62 * @return Content $this
63 */
64 public function copy() {
65 return $this; # NOTE: this is ok since TextContent are immutable.
66 }
67
68 public function getTextForSummary( $maxlength = 250 ) {
69 global $wgContLang;
70
71 $text = $this->getNativeData();
72
73 $truncatedtext = $wgContLang->truncate(
74 preg_replace( "/[\n\r]/", ' ', $text ),
75 max( 0, $maxlength ) );
76
77 return $truncatedtext;
78 }
79
80 /**
81 * Returns the text's size in bytes.
82 *
83 * @return int
84 */
85 public function getSize() {
86 $text = $this->getNativeData();
87
88 return strlen( $text );
89 }
90
91 /**
92 * Returns true if this content is not a redirect, and $wgArticleCountMethod
93 * is "any".
94 *
95 * @param bool|null $hasLinks If it is known whether this content contains links,
96 * provide this information here, to avoid redundant parsing to find out.
97 *
98 * @return bool
99 */
100 public function isCountable( $hasLinks = null ) {
101 global $wgArticleCountMethod;
102
103 if ( $this->isRedirect() ) {
104 return false;
105 }
106
107 if ( $wgArticleCountMethod === 'any' ) {
108 return true;
109 }
110
111 return false;
112 }
113
114 /**
115 * Returns the text represented by this Content object, as a string.
116 *
117 * @return string The raw text.
118 */
119 public function getNativeData() {
120 return $this->mText;
121 }
122
123 /**
124 * Returns the text represented by this Content object, as a string.
125 *
126 * @return string The raw text.
127 */
128 public function getTextForSearchIndex() {
129 return $this->getNativeData();
130 }
131
132 /**
133 * Returns attempts to convert this content object to wikitext,
134 * and then returns the text string. The conversion may be lossy.
135 *
136 * @note this allows any text-based content to be transcluded as if it was wikitext.
137 *
138 * @return string|bool The raw text, or false if the conversion failed.
139 */
140 public function getWikitextForTransclusion() {
141 $wikitext = $this->convert( CONTENT_MODEL_WIKITEXT, 'lossy' );
142
143 if ( $wikitext ) {
144 return $wikitext->getNativeData();
145 } else {
146 return false;
147 }
148 }
149
150 /**
151 * Do a "\r\n" -> "\n" and "\r" -> "\n" transformation
152 * as well as trim trailing whitespace
153 *
154 * This was formerly part of Parser::preSaveTransform, but
155 * for non-wikitext content models they probably still want
156 * to normalize line endings without all of the other PST
157 * changes.
158 *
159 * @since 1.28
160 * @param string $text
161 * @return string
162 */
163 public static function normalizeLineEndings( $text ) {
164 return str_replace( [ "\r\n", "\r" ], "\n", rtrim( $text ) );
165 }
166
167 /**
168 * Returns a Content object with pre-save transformations applied.
169 *
170 * At a minimum, subclasses should make sure to call TextContent::normalizeLineEndings()
171 * either directly or part of Parser::preSaveTransform().
172 *
173 * @param Title $title
174 * @param User $user
175 * @param ParserOptions $popts
176 *
177 * @return Content
178 */
179 public function preSaveTransform( Title $title, User $user, ParserOptions $popts ) {
180 $text = $this->getNativeData();
181 $pst = self::normalizeLineEndings( $text );
182
183 return ( $text === $pst ) ? $this : new static( $pst, $this->getModel() );
184 }
185
186 /**
187 * Diff this content object with another content object.
188 *
189 * @since 1.21
190 *
191 * @param Content $that The other content object to compare this content object to.
192 * @param Language $lang The language object to use for text segmentation.
193 * If not given, $wgContentLang is used.
194 *
195 * @return Diff A diff representing the changes that would have to be
196 * made to this content object to make it equal to $that.
197 */
198 public function diff( Content $that, Language $lang = null ) {
199 global $wgContLang;
200
201 $this->checkModelID( $that->getModel() );
202
203 // @todo could implement this in DifferenceEngine and just delegate here?
204
205 if ( !$lang ) {
206 $lang = $wgContLang;
207 }
208
209 $otext = $this->getNativeData();
210 $ntext = $that->getNativeData();
211
212 # Note: Use native PHP diff, external engines don't give us abstract output
213 $ota = explode( "\n", $lang->segmentForDiff( $otext ) );
214 $nta = explode( "\n", $lang->segmentForDiff( $ntext ) );
215
216 $diff = new Diff( $ota, $nta );
217
218 return $diff;
219 }
220
221 /**
222 * Fills the provided ParserOutput object with information derived from the content.
223 * Unless $generateHtml was false, this includes an HTML representation of the content
224 * provided by getHtml().
225 *
226 * For content models listed in $wgTextModelsToParse, this method will call the MediaWiki
227 * wikitext parser on the text to extract any (wikitext) links, magic words, etc.
228 *
229 * Subclasses may override this to provide custom content processing.
230 * For custom HTML generation alone, it is sufficient to override getHtml().
231 *
232 * @param Title $title Context title for parsing
233 * @param int $revId Revision ID (for {{REVISIONID}})
234 * @param ParserOptions $options Parser options
235 * @param bool $generateHtml Whether or not to generate HTML
236 * @param ParserOutput &$output The output object to fill (reference).
237 */
238 protected function fillParserOutput( Title $title, $revId,
239 ParserOptions $options, $generateHtml, ParserOutput &$output
240 ) {
241 global $wgParser, $wgTextModelsToParse;
242
243 if ( in_array( $this->getModel(), $wgTextModelsToParse ) ) {
244 // parse just to get links etc into the database, HTML is replaced below.
245 $output = $wgParser->parse( $this->getNativeData(), $title, $options, true, true, $revId );
246 }
247
248 if ( $generateHtml ) {
249 $html = $this->getHtml();
250 } else {
251 $html = '';
252 }
253
254 $output->setText( $html );
255 }
256
257 /**
258 * Generates an HTML version of the content, for display. Used by
259 * fillParserOutput() to provide HTML for the ParserOutput object.
260 *
261 * Subclasses may override this to provide a custom HTML rendering.
262 * If further information is to be derived from the content (such as
263 * categories), the fillParserOutput() method can be overridden instead.
264 *
265 * For backwards-compatibility, this default implementation just calls
266 * getHighlightHtml().
267 *
268 * @return string An HTML representation of the content
269 */
270 protected function getHtml() {
271 return $this->getHighlightHtml();
272 }
273
274 /**
275 * Generates an HTML version of the content, for display.
276 *
277 * This default implementation returns an HTML-escaped version
278 * of the raw text content.
279 *
280 * @note The functionality of this method should really be implemented
281 * in getHtml(), and subclasses should override getHtml() if needed.
282 * getHighlightHtml() is kept around for backward compatibility with
283 * extensions that already override it.
284 *
285 * @deprecated since 1.24. Use getHtml() instead. In particular, subclasses overriding
286 * getHighlightHtml() should override getHtml() instead.
287 *
288 * @return string An HTML representation of the content
289 */
290 protected function getHighlightHtml() {
291 return htmlspecialchars( $this->getNativeData() );
292 }
293
294 /**
295 * This implementation provides lossless conversion between content models based
296 * on TextContent.
297 *
298 * @param string $toModel The desired content model, use the CONTENT_MODEL_XXX flags.
299 * @param string $lossy Flag, set to "lossy" to allow lossy conversion. If lossy conversion is not
300 * allowed, full round-trip conversion is expected to work without losing information.
301 *
302 * @return Content|bool A content object with the content model $toModel, or false if that
303 * conversion is not supported.
304 *
305 * @see Content::convert()
306 */
307 public function convert( $toModel, $lossy = '' ) {
308 $converted = parent::convert( $toModel, $lossy );
309
310 if ( $converted !== false ) {
311 return $converted;
312 }
313
314 $toHandler = ContentHandler::getForModelID( $toModel );
315
316 if ( $toHandler instanceof TextContentHandler ) {
317 // NOTE: ignore content serialization format - it's just text anyway.
318 $text = $this->getNativeData();
319 $converted = $toHandler->unserializeContent( $text );
320 }
321
322 return $converted;
323 }
324
325 }