Index: trunk/extensions/PagedTiffHandler/PagedTiffHandler.image.php |
— | — | @@ -29,215 +29,230 @@ |
30 | 30 | */ |
31 | 31 | |
32 | 32 | class PagedTiffImage { |
33 | | - protected $_meta = NULL; |
34 | | - protected $mFilename; |
35 | | - |
36 | | - function __construct( $filename ) { |
37 | | - $this->mFilename = $filename; |
38 | | - } |
39 | | - |
40 | | - /** |
41 | | - * Customize the exiv shell command. |
42 | | - */ |
43 | | - static function exivCommand( &$cmd, $filename ) { |
44 | | - return true; |
45 | | - } |
46 | | - |
47 | | - /** |
48 | | - * Called by MimeMagick functions. |
49 | | - */ |
50 | | - public function isValid() { |
51 | | - return count($this->retrieveMetaData()); |
52 | | - } |
53 | | - |
54 | | - /** |
55 | | - * Returns an array that corresponds to the native PHP function getimagesize(). |
56 | | - */ |
57 | | - public function getImageSize() { |
58 | | - $data = $this->retrieveMetaData(); |
59 | | - $size = $this->getPageSize( $data, 1 ); |
60 | | - |
61 | | - if( $size ) { |
62 | | - $width = $size['width']; |
63 | | - $height = $size['height']; |
64 | | - return array( $width, $height, 'Tiff', |
65 | | - "width=\"$width\" height=\"$height\"" ); |
66 | | - } |
67 | | - return false; |
68 | | - } |
69 | | - |
70 | | - /** |
71 | | - * Returns an array with width and height of the tiff page. |
72 | | - */ |
73 | | - public static function getPageSize( $data, $page ) { |
74 | | - if( isset( $data['pages'][$page] ) ) { |
75 | | - return array('width' => $data['pages'][$page]['width'], |
76 | | - 'height' => $data['pages'][$page]['height']); |
77 | | - } |
78 | | - return false; |
79 | | - } |
80 | | - |
81 | | - /** |
82 | | - * Reads metadata of the tiff file via shell command and returns an associative array. |
83 | | - * layout: |
84 | | - * meta['Pages'] = amount of pages |
85 | | - * meta['pages'] = metadata per page |
86 | | - * meta['exif'] = Exif, XMP and IPTC |
87 | | - * meta['errors'] = identify-errors |
88 | | - * meta['warnings'] = identify-warnings |
89 | | - */ |
90 | | - public function retrieveMetaData() { |
91 | | - global $wgImageMagickIdentifyCommand, $wgTiffExivCommand, $wgTiffUseExiv, $wgMemc, $wgTiffErrorCacheTTL; |
92 | | - |
93 | | - $imgKey = wfMemcKey('PagedTiffHandler-ThumbnailGeneration', $this->mFilename); |
94 | | - $isCached = $wgMemc->get($imgKey); |
95 | | - if($isCached) { |
96 | | - return -1; |
97 | | - } |
98 | | - $wgMemc->add($imgKey, 1, $wgTiffErrorCacheTTL); |
99 | | - |
100 | | - if($this->_meta === NULL) { |
101 | | - if ( $wgImageMagickIdentifyCommand ) { |
102 | | - |
103 | | - wfProfileIn( 'PagedTiffImage' ); |
104 | | - /** |
105 | | - * ImageMagick is used in order to get the basic metadata of embedded files. |
106 | | - * This is not reliable in exiv2m since it is not possible to name a set of required fields. |
107 | | - */ |
108 | | - $cmd = wfEscapeShellArg( $wgImageMagickIdentifyCommand ) . |
109 | | - ' -format "[BEGIN]page=%p\nalpha=%A\nheight=%h\nwidth=%w\ndepth=%z[END]" ' . |
110 | | - wfEscapeShellArg( $this->mFilename ) . ' 2>&1'; |
111 | | - |
112 | | - wfProfileIn( 'identify' ); |
113 | | - wfDebug( __METHOD__.": $cmd\n" ); |
114 | | - $dump = wfShellExec( $cmd, $retval ); |
115 | | - wfProfileOut( 'identify' ); |
116 | | - if($retval) { |
117 | | - return false; |
118 | | - } |
119 | | - $this->_meta = $this->convertDumpToArray( $dump ); |
120 | | - $this->_meta['exif'] = array(); |
121 | | - |
122 | | - if($wgTiffUseExiv) { |
123 | | - $cmd = wfEscapeShellArg( $wgTiffExivCommand ) . |
124 | | - ' -u -psix -Pnt ' . // read EXIF, XMP, IPTC as name-tag => interpreted data -ignore unknown fields |
125 | | - // exiv2-doc @link http://www.exiv2.org/sample.html |
126 | | - wfEscapeShellArg( $this->mFilename ); |
127 | | - |
128 | | - wfRunHooks( "PagedTiffHandlerExivCommand", array( &$cmd, $this->mFilename ) ); |
129 | | - |
130 | | - wfProfileIn( 'exiv2' ); |
131 | | - wfDebug( __METHOD__.": $cmd\n" ); |
132 | | - $dump = wfShellExec( $cmd, $retval ); |
133 | | - wfProfileOut( 'exiv2' ); |
134 | | - $result = array(); |
135 | | - preg_match_all('/(\w+)\s+(.+)/', $dump, $result, PREG_SET_ORDER); |
136 | | - |
137 | | - foreach($result as $data) { |
138 | | - $this->_meta['exif'][$data[1]] = $data[2]; |
139 | | - } |
140 | | - } |
141 | | - else { |
142 | | - $cmd = wfEscapeShellArg( $wgImageMagickIdentifyCommand ) . |
143 | | - ' -verbose ' . |
144 | | - wfEscapeShellArg( $this->mFilename )."[0]"; |
145 | | - |
146 | | - wfProfileIn( 'identify -verbose' ); |
147 | | - wfDebug( __METHOD__.": $cmd\n" ); |
148 | | - $dump = wfShellExec( $cmd, $retval ); |
149 | | - wfProfileOut( 'identify -verbose' ); |
150 | | - $this->_meta['exif'] = $this->parseVerbose($dump); |
151 | | - } |
152 | | - wfProfileOut( 'PagedTiffImage' ); |
153 | | - } |
154 | | - } |
155 | | - unset($this->_meta['exif']['Image']); |
156 | | - unset($this->_meta['exif']['filename']); |
157 | | - unset($this->_meta['exif']['Base filename']); |
158 | | - return $this->_meta; |
159 | | - } |
160 | | - |
161 | | - /** |
162 | | - * helper function of retrieveMetaData(). |
163 | | - * parses shell return from identify-command into an array. |
164 | | - */ |
165 | | - protected function convertDumpToArray( $dump ) { |
166 | | - global $wgTiffIdentifyRejectMessages, $wgTiffIdentifyBypassMessages; |
167 | | - if ( strval( $dump ) == '' ) return false; |
168 | | - $infos = NULL; |
169 | | - preg_match_all('/\[BEGIN\](.+?)\[END\]/si', $dump, $infos, PREG_SET_ORDER); |
170 | | - $data = array(); |
171 | | - $data['Pages'] = count($infos); |
172 | | - $data['pages'] = array(); |
173 | | - foreach($infos as $info) { |
174 | | - $entry = array(); |
175 | | - $lines = explode("\n", $info[1]); |
176 | | - foreach($lines as $line) { |
177 | | - if(trim($line) == '') { |
178 | | - continue; |
179 | | - } |
180 | | - $parts = explode('=', $line); |
181 | | - $entry[trim($parts[0])] = trim($parts[1]); |
182 | | - } |
183 | | - $entry['pixels'] = $entry['height'] * $entry['width']; |
184 | | - $data['pages'][$entry['page']] = $entry; |
185 | | - } |
186 | | - |
187 | | - $dump = preg_replace('/\[BEGIN\](.+?)\[END\]/si', '', $dump); |
188 | | - if(strlen($dump)) { |
189 | | - $errors = explode("\n", $dump); |
190 | | - foreach($errors as $error) { |
191 | | - $knownError = false; |
192 | | - foreach($wgTiffIdentifyRejectMessages as $msg) { |
193 | | - if (preg_match($msg, trim($error))) { |
194 | | - $data['errors'][] = $error; |
195 | | - $knownError = true; |
196 | | - break; |
197 | | - } |
198 | | - } |
199 | | - if(!$knownError) { |
200 | | - foreach($wgTiffIdentifyBypassMessages as $msg) { |
201 | | - if (preg_match($msg, trim($error))) { |
202 | | - $data['warnings'][] = $error; |
203 | | - $knownError = true; |
204 | | - break; |
205 | | - } |
206 | | - } |
207 | | - } |
208 | | - if(!$knownError) { |
209 | | - $data['unknownErrors'][] = $error; |
210 | | - } |
211 | | - } |
212 | | - } |
213 | | - return $data; |
214 | | - } |
215 | | - |
216 | | - /** |
217 | | - * helper function of retrieveMetaData(). |
218 | | - * parses shell return from identify-verbose-command into an array. |
219 | | - */ |
220 | | - protected function parseVerbose($dump) { |
221 | | - $data = array(); |
222 | | - $dump = explode("\n", $dump); |
223 | | - $lastwhite = 0; |
224 | | - $lastkey = false; |
225 | | - foreach($dump as $line) { |
226 | | - if(preg_match('/^(\s*?)(\w([\w\s]+?)?):(.*?)$/sim', $line, $res)) { |
227 | | - if($lastwhite == 0 || strlen($res[1]) == $lastwhite) { |
228 | | - if(strlen(trim($res[4]))) { |
229 | | - $data[trim($res[2])] = trim($res[4]); |
230 | | - } |
231 | | - else { |
232 | | - $data[trim($res[2])] = " Data:\n"; |
233 | | - } |
234 | | - $lastkey = trim($res[2]); |
235 | | - $lastwhite = strlen($res[1]); |
236 | | - } |
237 | | - else { |
238 | | - $data[$lastkey] .= $line."\n"; |
239 | | - } |
240 | | - } |
241 | | - } |
242 | | - return $data; |
243 | | - } |
| 33 | + protected $_meta = NULL; |
| 34 | + protected $mFilename; |
| 35 | + |
| 36 | + function __construct( $filename ) { |
| 37 | + $this->mFilename = $filename; |
| 38 | + } |
| 39 | + |
| 40 | + /** |
| 41 | + * Customize the exiv shell command. |
| 42 | + */ |
| 43 | + static function exivCommand( &$cmd, $filename ) { |
| 44 | + return true; |
| 45 | + } |
| 46 | + |
| 47 | + /** |
| 48 | + * Called by MimeMagick functions. |
| 49 | + */ |
| 50 | + public function isValid() { |
| 51 | + return count($this->retrieveMetaData()); |
| 52 | + } |
| 53 | + |
| 54 | + /** |
| 55 | + * Returns an array that corresponds to the native PHP function getimagesize(). |
| 56 | + */ |
| 57 | + public function getImageSize() { |
| 58 | + $data = $this->retrieveMetaData(); |
| 59 | + $size = $this->getPageSize( $data, 1 ); |
| 60 | + |
| 61 | + if( $size ) { |
| 62 | + $width = $size['width']; |
| 63 | + $height = $size['height']; |
| 64 | + return array( $width, $height, 'Tiff', |
| 65 | + "width=\"$width\" height=\"$height\"" ); |
| 66 | + } |
| 67 | + return false; |
| 68 | + } |
| 69 | + |
| 70 | + /** |
| 71 | + * Returns an array with width and height of the tiff page. |
| 72 | + */ |
| 73 | + public static function getPageSize( $data, $page ) { |
| 74 | + if( isset( $data['pages'][$page] ) ) { |
| 75 | + return array('width' => $data['pages'][$page]['width'], |
| 76 | + 'height' => $data['pages'][$page]['height']); |
| 77 | + } |
| 78 | + return false; |
| 79 | + } |
| 80 | + |
| 81 | + /** |
| 82 | + * Reads metadata of the tiff file via shell command and returns an associative array. |
| 83 | + * layout: |
| 84 | + * meta['Pages'] = amount of pages |
| 85 | + * meta['pages'] = metadata per page |
| 86 | + * meta['exif'] = Exif, XMP and IPTC |
| 87 | + * meta['errors'] = identify-errors |
| 88 | + * meta['warnings'] = identify-warnings |
| 89 | + */ |
| 90 | + public function retrieveMetaData() { |
| 91 | + global $wgImageMagickIdentifyCommand, $wgTiffExivCommand, $wgTiffUseExiv, $wgMemc, $wgTiffErrorCacheTTL; |
| 92 | + |
| 93 | + $imgKey = wfMemcKey('PagedTiffHandler-ThumbnailGeneration', $this->mFilename); |
| 94 | + $isCached = $wgMemc->get($imgKey); |
| 95 | + if($isCached) { |
| 96 | + return -1; |
| 97 | + } |
| 98 | + $wgMemc->add($imgKey, 1, $wgTiffErrorCacheTTL); |
| 99 | + |
| 100 | + if($this->_meta === NULL) { |
| 101 | + if ( $wgImageMagickIdentifyCommand ) { |
| 102 | + |
| 103 | + wfProfileIn( 'PagedTiffImage' ); |
| 104 | + /** |
| 105 | + * ImageMagick is used in order to get the basic metadata of embedded files. |
| 106 | + * This is not reliable in exiv2m since it is not possible to name a set of required fields. |
| 107 | + */ |
| 108 | + $cmd = wfEscapeShellArg( $wgImageMagickIdentifyCommand ) . |
| 109 | + ' -format "[BEGIN]page=%p\nalpha=%A\nalpha2=%r\nheight=%h\nwidth=%w\ndepth=%z[END]" ' . |
| 110 | + wfEscapeShellArg( $this->mFilename ) . ' 2>&1'; |
| 111 | + |
| 112 | + wfProfileIn( 'identify' ); |
| 113 | + wfDebug( __METHOD__.": $cmd\n" ); |
| 114 | + $dump = wfShellExec( $cmd, $retval ); |
| 115 | + wfProfileOut( 'identify' ); |
| 116 | + if($retval) { |
| 117 | + return false; |
| 118 | + } |
| 119 | + $this->_meta = $this->convertDumpToArray( $dump ); |
| 120 | + $this->_meta['exif'] = array(); |
| 121 | + |
| 122 | + if($wgTiffUseExiv) { |
| 123 | + $cmd = wfEscapeShellArg( $wgTiffExivCommand ) . |
| 124 | + ' -u -psix -Pnt ' . // read EXIF, XMP, IPTC as name-tag => interpreted data -ignore unknown fields |
| 125 | + // exiv2-doc @link http://www.exiv2.org/sample.html |
| 126 | + wfEscapeShellArg( $this->mFilename ); |
| 127 | + |
| 128 | + wfRunHooks( "PagedTiffHandlerExivCommand", array( &$cmd, $this->mFilename ) ); |
| 129 | + |
| 130 | + wfProfileIn( 'exiv2' ); |
| 131 | + wfDebug( __METHOD__.": $cmd\n" ); |
| 132 | + $dump = wfShellExec( $cmd, $retval ); |
| 133 | + wfProfileOut( 'exiv2' ); |
| 134 | + $result = array(); |
| 135 | + preg_match_all('/(\w+)\s+(.+)/', $dump, $result, PREG_SET_ORDER); |
| 136 | + |
| 137 | + foreach($result as $data) { |
| 138 | + $this->_meta['exif'][$data[1]] = $data[2]; |
| 139 | + } |
| 140 | + } |
| 141 | + else { |
| 142 | + $cmd = wfEscapeShellArg( $wgImageMagickIdentifyCommand ) . |
| 143 | + ' -verbose ' . |
| 144 | + wfEscapeShellArg( $this->mFilename )."[0]"; |
| 145 | + |
| 146 | + wfProfileIn( 'identify -verbose' ); |
| 147 | + wfDebug( __METHOD__.": $cmd\n" ); |
| 148 | + $dump = wfShellExec( $cmd, $retval ); |
| 149 | + wfProfileOut( 'identify -verbose' ); |
| 150 | + $this->_meta['exif'] = $this->parseVerbose($dump); |
| 151 | + } |
| 152 | + wfProfileOut( 'PagedTiffImage' ); |
| 153 | + } |
| 154 | + } |
| 155 | + unset($this->_meta['exif']['Image']); |
| 156 | + unset($this->_meta['exif']['filename']); |
| 157 | + unset($this->_meta['exif']['Base filename']); |
| 158 | + return $this->_meta; |
| 159 | + } |
| 160 | + |
| 161 | + /** |
| 162 | + * helper function of retrieveMetaData(). |
| 163 | + * parses shell return from identify-command into an array. |
| 164 | + */ |
| 165 | + protected function convertDumpToArray( $dump ) { |
| 166 | + global $wgTiffIdentifyRejectMessages, $wgTiffIdentifyBypassMessages; |
| 167 | + if ( strval( $dump ) == '' ) return false; |
| 168 | + $infos = NULL; |
| 169 | + preg_match_all('/\[BEGIN\](.+?)\[END\]/si', $dump, $infos, PREG_SET_ORDER); |
| 170 | + $data = array(); |
| 171 | + $data['Pages'] = count($infos); |
| 172 | + $data['pages'] = array(); |
| 173 | + foreach($infos as $info) { |
| 174 | + $entry = array(); |
| 175 | + $lines = explode("\n", $info[1]); |
| 176 | + foreach($lines as $line) { |
| 177 | + if(trim($line) == '') { |
| 178 | + continue; |
| 179 | + } |
| 180 | + $parts = explode('=', $line); |
| 181 | + if(trim($parts[0]) == 'alpha' && trim($parts[1]) == '%A') { |
| 182 | + continue; |
| 183 | + } |
| 184 | + if(trim($parts[0]) == 'alpha2' && !isset($entry['alpha'])) { |
| 185 | + switch(trim($parts[1])) { |
| 186 | + case 'DirectClassRGBMatte': |
| 187 | + case 'DirectClassRGBA': |
| 188 | + $entry['alpha'] = 'true'; |
| 189 | + break; |
| 190 | + default: |
| 191 | + $entry['alpha'] = 'false'; |
| 192 | + break; |
| 193 | + } |
| 194 | + continue; |
| 195 | + } |
| 196 | + $entry[trim($parts[0])] = trim($parts[1]); |
| 197 | + } |
| 198 | + $entry['pixels'] = $entry['height'] * $entry['width']; |
| 199 | + $data['pages'][$entry['page']] = $entry; |
| 200 | + } |
| 201 | + |
| 202 | + $dump = preg_replace('/\[BEGIN\](.+?)\[END\]/si', '', $dump); |
| 203 | + if(strlen($dump)) { |
| 204 | + $errors = explode("\n", $dump); |
| 205 | + foreach($errors as $error) { |
| 206 | + $knownError = false; |
| 207 | + foreach($wgTiffIdentifyRejectMessages as $msg) { |
| 208 | + if (preg_match($msg, trim($error))) { |
| 209 | + $data['errors'][] = $error; |
| 210 | + $knownError = true; |
| 211 | + break; |
| 212 | + } |
| 213 | + } |
| 214 | + if(!$knownError) { |
| 215 | + foreach($wgTiffIdentifyBypassMessages as $msg) { |
| 216 | + if (preg_match($msg, trim($error))) { |
| 217 | + $data['warnings'][] = $error; |
| 218 | + $knownError = true; |
| 219 | + break; |
| 220 | + } |
| 221 | + } |
| 222 | + } |
| 223 | + if(!$knownError) { |
| 224 | + $data['unknownErrors'][] = $error; |
| 225 | + } |
| 226 | + } |
| 227 | + } |
| 228 | + return $data; |
| 229 | + } |
| 230 | + |
| 231 | + /** |
| 232 | + * helper function of retrieveMetaData(). |
| 233 | + * parses shell return from identify-verbose-command into an array. |
| 234 | + */ |
| 235 | + protected function parseVerbose($dump) { |
| 236 | + $data = array(); |
| 237 | + $dump = explode("\n", $dump); |
| 238 | + $lastwhite = 0; |
| 239 | + $lastkey = false; |
| 240 | + foreach($dump as $line) { |
| 241 | + if(preg_match('/^(\s*?)(\w([\w\s]+?)?):(.*?)$/sim', $line, $res)) { |
| 242 | + if($lastwhite == 0 || strlen($res[1]) == $lastwhite) { |
| 243 | + if(strlen(trim($res[4]))) { |
| 244 | + $data[trim($res[2])] = trim($res[4]); |
| 245 | + } |
| 246 | + else { |
| 247 | + $data[trim($res[2])] = " Data:\n"; |
| 248 | + } |
| 249 | + $lastkey = trim($res[2]); |
| 250 | + $lastwhite = strlen($res[1]); |
| 251 | + } |
| 252 | + else { |
| 253 | + $data[$lastkey] .= $line."\n"; |
| 254 | + } |
| 255 | + } |
| 256 | + } |
| 257 | + return $data; |
| 258 | + } |
244 | 259 | } |