MediaWiki  1.28.0
DjVu.php
Go to the documentation of this file.
1 <?php
29 class DjVuHandler extends ImageHandler {
30  const EXPENSIVE_SIZE_LIMIT = 10485760; // 10MiB
31 
35  function isEnabled() {
37  if ( !$wgDjvuRenderer || ( !$wgDjvuDump && !$wgDjvuToXML ) ) {
38  wfDebug( "DjVu is disabled, please set \$wgDjvuRenderer and \$wgDjvuDump\n" );
39 
40  return false;
41  } else {
42  return true;
43  }
44  }
45 
50  public function mustRender( $file ) {
51  return true;
52  }
53 
59  public function isExpensiveToThumbnail( $file ) {
60  return $file->getSize() > static::EXPENSIVE_SIZE_LIMIT;
61  }
62 
67  public function isMultiPage( $file ) {
68  return true;
69  }
70 
74  public function getParamMap() {
75  return [
76  'img_width' => 'width',
77  'img_page' => 'page',
78  ];
79  }
80 
86  public function validateParam( $name, $value ) {
87  if ( $name === 'page' && trim( $value ) !== (string)intval( $value ) ) {
88  // Extra junk on the end of page, probably actually a caption
89  // e.g. [[File:Foo.djvu|thumb|Page 3 of the document shows foo]]
90  return false;
91  }
92  if ( in_array( $name, [ 'width', 'height', 'page' ] ) ) {
93  if ( $value <= 0 ) {
94  return false;
95  } else {
96  return true;
97  }
98  } else {
99  return false;
100  }
101  }
102 
107  public function makeParamString( $params ) {
108  $page = isset( $params['page'] ) ? $params['page'] : 1;
109  if ( !isset( $params['width'] ) ) {
110  return false;
111  }
112 
113  return "page{$page}-{$params['width']}px";
114  }
115 
120  public function parseParamString( $str ) {
121  $m = false;
122  if ( preg_match( '/^page(\d+)-(\d+)px$/', $str, $m ) ) {
123  return [ 'width' => $m[2], 'page' => $m[1] ];
124  } else {
125  return false;
126  }
127  }
128 
133  function getScriptParams( $params ) {
134  return [
135  'width' => $params['width'],
136  'page' => $params['page'],
137  ];
138  }
139 
148  function doTransform( $image, $dstPath, $dstUrl, $params, $flags = 0 ) {
150 
151  if ( !$this->normaliseParams( $image, $params ) ) {
152  return new TransformParameterError( $params );
153  }
154  $width = $params['width'];
155  $height = $params['height'];
156  $page = $params['page'];
157 
158  if ( $flags & self::TRANSFORM_LATER ) {
159  $params = [
160  'width' => $width,
161  'height' => $height,
162  'page' => $page
163  ];
164 
165  return new ThumbnailImage( $image, $dstUrl, $dstPath, $params );
166  }
167 
168  if ( !wfMkdirParents( dirname( $dstPath ), null, __METHOD__ ) ) {
169  return new MediaTransformError(
170  'thumbnail_error',
171  $width,
172  $height,
173  wfMessage( 'thumbnail_dest_directory' )->text()
174  );
175  }
176 
177  // Get local copy source for shell scripts
178  // Thumbnail extraction is very inefficient for large files.
179  // Provide a way to pool count limit the number of downloaders.
180  if ( $image->getSize() >= 1e7 ) { // 10MB
181  $work = new PoolCounterWorkViaCallback( 'GetLocalFileCopy', sha1( $image->getName() ),
182  [
183  'doWork' => function () use ( $image ) {
184  return $image->getLocalRefPath();
185  }
186  ]
187  );
188  $srcPath = $work->execute();
189  } else {
190  $srcPath = $image->getLocalRefPath();
191  }
192 
193  if ( $srcPath === false ) { // Failed to get local copy
194  wfDebugLog( 'thumbnail',
195  sprintf( 'Thumbnail failed on %s: could not get local copy of "%s"',
196  wfHostname(), $image->getName() ) );
197 
198  return new MediaTransformError( 'thumbnail_error',
199  $params['width'], $params['height'],
200  wfMessage( 'filemissing' )->text()
201  );
202  }
203 
204  # Use a subshell (brackets) to aggregate stderr from both pipeline commands
205  # before redirecting it to the overall stdout. This works in both Linux and Windows XP.
206  $cmd = '(' . wfEscapeShellArg(
207  $wgDjvuRenderer,
208  "-format=ppm",
209  "-page={$page}",
210  "-size={$params['physicalWidth']}x{$params['physicalHeight']}",
211  $srcPath );
212  if ( $wgDjvuPostProcessor ) {
213  $cmd .= " | {$wgDjvuPostProcessor}";
214  }
215  $cmd .= ' > ' . wfEscapeShellArg( $dstPath ) . ') 2>&1';
216  wfDebug( __METHOD__ . ": $cmd\n" );
217  $retval = '';
218  $err = wfShellExec( $cmd, $retval );
219 
220  $removed = $this->removeBadFile( $dstPath, $retval );
221  if ( $retval != 0 || $removed ) {
222  $this->logErrorForExternalProcess( $retval, $err, $cmd );
223  return new MediaTransformError( 'thumbnail_error', $width, $height, $err );
224  } else {
225  $params = [
226  'width' => $width,
227  'height' => $height,
228  'page' => $page
229  ];
230 
231  return new ThumbnailImage( $image, $dstUrl, $dstPath, $params );
232  }
233  }
234 
242  function getDjVuImage( $image, $path ) {
243  if ( !$image ) {
244  $deja = new DjVuImage( $path );
245  } elseif ( !isset( $image->dejaImage ) ) {
246  $deja = $image->dejaImage = new DjVuImage( $path );
247  } else {
248  $deja = $image->dejaImage;
249  }
250 
251  return $deja;
252  }
253 
261  private function getUnserializedMetadata( File $file ) {
262  $metadata = $file->getMetadata();
263  if ( substr( $metadata, 0, 3 ) === '<?xml' ) {
264  // Old style. Not serialized but instead just a raw string of XML.
265  return $metadata;
266  }
267 
268  MediaWiki\suppressWarnings();
269  $unser = unserialize( $metadata );
270  MediaWiki\restoreWarnings();
271  if ( is_array( $unser ) ) {
272  if ( isset( $unser['error'] ) ) {
273  return false;
274  } elseif ( isset( $unser['xml'] ) ) {
275  return $unser['xml'];
276  } else {
277  // Should never ever reach here.
278  throw new MWException( "Error unserializing DjVu metadata." );
279  }
280  }
281 
282  // unserialize failed. Guess it wasn't really serialized after all,
283  return $metadata;
284  }
285 
292  public function getMetaTree( $image, $gettext = false ) {
293  if ( $gettext && isset( $image->djvuTextTree ) ) {
294  return $image->djvuTextTree;
295  }
296  if ( !$gettext && isset( $image->dejaMetaTree ) ) {
297  return $image->dejaMetaTree;
298  }
299 
300  $metadata = $this->getUnserializedMetadata( $image );
301  if ( !$this->isMetadataValid( $image, $metadata ) ) {
302  wfDebug( "DjVu XML metadata is invalid or missing, should have been fixed in upgradeRow\n" );
303 
304  return false;
305  }
306 
307  MediaWiki\suppressWarnings();
308  try {
309  // Set to false rather than null to avoid further attempts
310  $image->dejaMetaTree = false;
311  $image->djvuTextTree = false;
312  $tree = new SimpleXMLElement( $metadata, LIBXML_PARSEHUGE );
313  if ( $tree->getName() == 'mw-djvu' ) {
315  foreach ( $tree->children() as $b ) {
316  if ( $b->getName() == 'DjVuTxt' ) {
317  // @todo File::djvuTextTree and File::dejaMetaTree are declared
318  // dynamically. Add a public File::$data to facilitate this?
319  $image->djvuTextTree = $b;
320  } elseif ( $b->getName() == 'DjVuXML' ) {
321  $image->dejaMetaTree = $b;
322  }
323  }
324  } else {
325  $image->dejaMetaTree = $tree;
326  }
327  } catch ( Exception $e ) {
328  wfDebug( "Bogus multipage XML metadata on '{$image->getName()}'\n" );
329  }
330  MediaWiki\restoreWarnings();
331  if ( $gettext ) {
332  return $image->djvuTextTree;
333  } else {
334  return $image->dejaMetaTree;
335  }
336  }
337 
338  function getImageSize( $image, $path ) {
339  return $this->getDjVuImage( $image, $path )->getImageSize();
340  }
341 
342  function getThumbType( $ext, $mime, $params = null ) {
344  static $mime;
345  if ( !isset( $mime ) ) {
346  $magic = MimeMagic::singleton();
347  $mime = $magic->guessTypesForExtension( $wgDjvuOutputExtension );
348  }
349 
350  return [ $wgDjvuOutputExtension, $mime ];
351  }
352 
353  function getMetadata( $image, $path ) {
354  wfDebug( "Getting DjVu metadata for $path\n" );
355 
356  $xml = $this->getDjVuImage( $image, $path )->retrieveMetaData();
357  if ( $xml === false ) {
358  // Special value so that we don't repetitively try and decode a broken file.
359  return serialize( [ 'error' => 'Error extracting metadata' ] );
360  } else {
361  return serialize( [ 'xml' => $xml ] );
362  }
363  }
364 
365  function getMetadataType( $image ) {
366  return 'djvuxml';
367  }
368 
369  function isMetadataValid( $image, $metadata ) {
370  return !empty( $metadata ) && $metadata != serialize( [] );
371  }
372 
373  function pageCount( File $image ) {
374  $info = $this->getDimensionInfo( $image );
375 
376  return $info ? $info['pageCount'] : false;
377  }
378 
380  $index = $page - 1; // MW starts pages at 1
381 
382  $info = $this->getDimensionInfo( $image );
383  if ( $info && isset( $info['dimensionsByPage'][$index] ) ) {
384  return $info['dimensionsByPage'][$index];
385  }
386 
387  return false;
388  }
389 
390  protected function getDimensionInfo( File $file ) {
392  return $cache->getWithSetCallback(
393  $cache->makeKey( 'file-djvu', 'dimensions', $file->getSha1() ),
394  $cache::TTL_INDEFINITE,
395  function () use ( $file ) {
396  $tree = $this->getMetaTree( $file );
397  if ( !$tree ) {
398  return false;
399  }
400 
401  $dimsByPage = [];
402  $count = count( $tree->xpath( '//OBJECT' ) );
403  for ( $i = 0; $i < $count; $i++ ) {
404  $o = $tree->BODY[0]->OBJECT[$i];
405  if ( $o ) {
406  $dimsByPage[$i] = [
407  'width' => (int)$o['width'],
408  'height' => (int)$o['height'],
409  ];
410  } else {
411  $dimsByPage[$i] = false;
412  }
413  }
414 
415  return [ 'pageCount' => $count, 'dimensionsByPage' => $dimsByPage ];
416  },
417  [ 'pcTTL' => $cache::TTL_INDEFINITE ]
418  );
419  }
420 
426  function getPageText( File $image, $page ) {
427  $tree = $this->getMetaTree( $image, true );
428  if ( !$tree ) {
429  return false;
430  }
431 
432  $o = $tree->BODY[0]->PAGE[$page - 1];
433  if ( $o ) {
434  $txt = $o['value'];
435 
436  return $txt;
437  } else {
438  return false;
439  }
440  }
441 }
isMultiPage($file)
Definition: DjVu.php:67
static getMainWANInstance()
Get the main WAN cache object.
parseParamString($str)
Definition: DjVu.php:120
getPageText(File $image, $page)
Definition: DjVu.php:426
getDimensionInfo(File $file)
Definition: DjVu.php:390
removeBadFile($dstPath, $retval=0)
Check for zero-sized thumbnails.
getUnserializedMetadata(File $file)
Get metadata, unserializing it if neccessary.
Definition: DjVu.php:261
Apache License January AND DISTRIBUTION Definitions License shall mean the terms and conditions for use
wfMkdirParents($dir, $mode=null, $caller=null)
Make directory, and make all parent directories if they don't exist.
getParamMap()
Definition: DjVu.php:74
static singleton()
Get an instance of this class.
Definition: MimeMagic.php:29
div flags Integer display flags(NO_ACTION_LINK, NO_EXTRA_USER_LINKS) 'LogException'returning false will NOT prevent logging $e
Definition: hooks.txt:2102
execute($skipcache=false)
Get the result of the work (whatever it is), or the result of the error() function.
wfHostname()
Fetch server name for use in error reporting etc.
$value
getMetadata($image, $path)
Definition: DjVu.php:353
pageCount(File $image)
Definition: DjVu.php:373
Handler for DjVu images.
Definition: DjVu.php:29
if($ext== 'php'||$ext== 'php5') $mime
Definition: router.php:65
it s the revision text itself In either if gzip is the revision text is gzipped $flags
Definition: hooks.txt:2703
getThumbType($ext, $mime, $params=null)
Definition: DjVu.php:342
wfShellExec($cmd, &$retval=null, $environ=[], $limits=[], $options=[])
Execute a shell command, with time and memory limits mirrored from the PHP configuration if supported...
when a variable name is used in a it is silently declared as a new local masking the global
Definition: design.txt:93
getPageDimensions(File $image, $page)
Definition: DjVu.php:379
wfDebug($text, $dest= 'all', array $context=[])
Sends a line to the debug log if enabled or, optionally, to a comment in output.
logErrorForExternalProcess($retval, $err, $cmd)
Log an error that occurred in an external process.
isEnabled()
Definition: DjVu.php:35
Shortcut class for parameter validation errors.
getImageSize($image, $path)
Definition: DjVu.php:338
Support for detecting/validating DjVu image files and getting some basic file metadata (resolution et...
Definition: DjVuImage.php:36
wfDebugLog($logGroup, $text, $dest= 'all', array $context=[])
Send a line to a supplementary debug log file, if configured, or main debug log if not...
Convenience class for dealing with PoolCounters using callbacks.
getDjVuImage($image, $path)
Cache an instance of DjVuImage in an Image object, return that instance.
Definition: DjVu.php:242
unserialize($serialized)
Definition: ApiMessage.php:102
either a unescaped string or a HtmlArmor object after in associative array form externallinks including delete and has completed for all link tables whether this was an auto creation default is conds Array Extra conditions for the No matching items in log is displayed if loglist is empty msgKey Array If you want a nice box with a set this to the key of the message First element is the message additional optional elements are parameters for the key that are processed with wfMessage() -> params() ->parseAsBlock()-offset Set to overwrite offset parameter in $wgRequest set to ''to unsetoffset-wrap String Wrap the message in html(usually something like"&lt
const EXPENSIVE_SIZE_LIMIT
Definition: DjVu.php:30
getMetaTree($image, $gettext=false)
Cache a document tree for the DjVu XML metadata.
Definition: DjVu.php:292
getMetadata()
Get handler-specific metadata Overridden by LocalFile, UnregisteredLocalFile STUB.
Definition: File.php:638
Media transform output for images.
isMetadataValid($image, $metadata)
Definition: DjVu.php:369
$wgDjvuRenderer
Path of the ddjvu DJVU renderer Enable this and $wgDjvuDump to enable djvu rendering example: $wgDjvu...
$cache
Definition: mcc.php:33
$params
$wgDjvuOutputExtension
File extension for the DJVU post processor output.
getSha1()
Get the SHA-1 base 36 hash of the file.
Definition: File.php:2116
validateParam($name, $value)
Definition: DjVu.php:86
design txt This is a brief overview of the new design More thorough and up to date information is available on the documentation wiki at etc Handles the details of getting and saving to the user table of the and dealing with sessions and cookies OutputPage Encapsulates the entire HTML page that will be sent in response to any server request It is used by calling its functions to add text
Definition: design.txt:12
isExpensiveToThumbnail($file)
True if creating thumbnails from the file is large or otherwise resource-intensive.
Definition: DjVu.php:59
This document is intended to provide useful advice for parties seeking to redistribute MediaWiki to end users It s targeted particularly at maintainers for Linux since it s been observed that distribution packages of MediaWiki often break We ve consistently had to recommend that users seeking support use official tarballs instead of their distribution s and this often solves whatever problem the user is having It would be nice if this could such as
Definition: distributors.txt:9
getMetadataType($image)
Definition: DjVu.php:365
makeParamString($params)
Definition: DjVu.php:107
$wgDjvuDump
Path of the djvudump executable Enable this and $wgDjvuRenderer to enable djvu rendering example: $wg...
injection txt This is an overview of how MediaWiki makes use of dependency injection The design described here grew from the discussion of RFC T384 The term dependency this means that anything an object needs to operate should be injected from the the object itself should only know narrow no concrete implementation of the logic it relies on The requirement to inject everything typically results in an architecture that based on two main types of and essentially stateless service objects that use other service objects to operate on the value objects As of the beginning MediaWiki is only starting to use the DI approach Much of the code still relies on global state or direct resulting in a highly cyclical dependency which acts as the top level factory for services in MediaWiki which can be used to gain access to default instances of various services MediaWikiServices however also allows new services to be defined and default services to be redefined Services are defined or redefined by providing a callback the instantiator that will return a new instance of the service When it will create an instance of MediaWikiServices and populate it with the services defined in the files listed by thereby bootstrapping the DI framework Per $wgServiceWiringFiles lists includes ServiceWiring php
Definition: injection.txt:35
Media handler abstract base class for images.
mustRender($file)
Definition: DjVu.php:50
$wgDjvuToXML
Path of the djvutoxml executable This works like djvudump except much, much slower as of version 3...
this hook is for auditing only or null if authentication failed before getting that far or null if we can t even determine that probably a stub it is not rendered in wiki pages or galleries in category pages allow injecting custom HTML after the section Any uses of the hook need to handle escaping see BaseTemplate::getToolbox and BaseTemplate::makeListItem for details on the format of individual items inside of this array or by returning and letting standard HTTP rendering take place modifiable or by returning false and taking over the output modifiable modifiable after all normalizations have been except for the $wgMaxImageArea check $image
Definition: hooks.txt:802
$count
wfEscapeShellArg()
Version of escapeshellarg() that works better on Windows.
doTransform($image, $dstPath, $dstUrl, $params, $flags=0)
Definition: DjVu.php:148
normaliseParams($image, &$params)
serialize()
Definition: ApiMessage.php:94
Implements some public methods and some protected utility functions which are required by multiple ch...
Definition: File.php:50
getScriptParams($params)
Definition: DjVu.php:133
please add to it if you re going to add events to the MediaWiki code where normally authentication against an external auth plugin would be creating a local account incomplete not yet checked for validity & $retval
Definition: hooks.txt:242
Basic media transform error class.
$wgDjvuPostProcessor
Shell command for the DJVU post processor Default: pnmtojpeg, since ddjvu generates ppm output Set th...
do that in ParserLimitReportFormat instead use this to modify the parameters of the image and a DIV can begin in one section and end in another Make sure your code can handle that case gracefully See the EditSectionClearerLink extension for an example zero but section is usually empty its values are the globals values before the output is cached $page
Definition: hooks.txt:2491
Allows to change the fields on the form that will be generated $name
Definition: hooks.txt:300