Duplicate code is one of the most pungent code smells. A rule that is often used is to re-structure code once it is duplicated in three or more places.
Common duplication problems, and corresponding solutions are:
Complex classes like DjVuImage often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use DjVuImage, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
36 | class DjVuImage { |
||
37 | /** |
||
38 | * @const DJVUTXT_MEMORY_LIMIT Memory limit for the DjVu description software |
||
39 | */ |
||
40 | const DJVUTXT_MEMORY_LIMIT = 300000; |
||
41 | |||
42 | /** |
||
43 | * Constructor |
||
44 | * |
||
45 | * @param string $filename The DjVu file name. |
||
46 | */ |
||
47 | function __construct( $filename ) { |
||
50 | |||
51 | /** |
||
52 | * Check if the given file is indeed a valid DjVu image file |
||
53 | * @return bool |
||
54 | */ |
||
55 | public function isValid() { |
||
60 | |||
61 | /** |
||
62 | * Return data in the style of getimagesize() |
||
63 | * @return array|bool Array or false on failure |
||
64 | */ |
||
65 | public function getImageSize() { |
||
78 | |||
79 | // --------- |
||
80 | |||
81 | /** |
||
82 | * For debugging; dump the IFF chunk structure |
||
83 | */ |
||
84 | function dump() { |
||
85 | $file = fopen( $this->mFilename, 'rb' ); |
||
86 | $header = fread( $file, 12 ); |
||
87 | $arr = unpack( 'a4magic/a4chunk/NchunkLength', $header ); |
||
88 | $chunk = $arr['chunk']; |
||
89 | $chunkLength = $arr['chunkLength']; |
||
90 | echo "$chunk $chunkLength\n"; |
||
91 | $this->dumpForm( $file, $chunkLength, 1 ); |
||
92 | fclose( $file ); |
||
93 | } |
||
94 | |||
95 | private function dumpForm( $file, $length, $indent ) { |
||
96 | $start = ftell( $file ); |
||
97 | $secondary = fread( $file, 4 ); |
||
98 | echo str_repeat( ' ', $indent * 4 ) . "($secondary)\n"; |
||
99 | while ( ftell( $file ) - $start < $length ) { |
||
100 | $chunkHeader = fread( $file, 8 ); |
||
101 | if ( $chunkHeader == '' ) { |
||
102 | break; |
||
103 | } |
||
104 | $arr = unpack( 'a4chunk/NchunkLength', $chunkHeader ); |
||
105 | $chunk = $arr['chunk']; |
||
106 | $chunkLength = $arr['chunkLength']; |
||
107 | echo str_repeat( ' ', $indent * 4 ) . "$chunk $chunkLength\n"; |
||
108 | |||
109 | if ( $chunk == 'FORM' ) { |
||
110 | $this->dumpForm( $file, $chunkLength, $indent + 1 ); |
||
111 | View Code Duplication | } else { |
|
112 | fseek( $file, $chunkLength, SEEK_CUR ); |
||
113 | if ( $chunkLength & 1 == 1 ) { |
||
114 | // Padding byte between chunks |
||
115 | fseek( $file, 1, SEEK_CUR ); |
||
116 | } |
||
117 | } |
||
118 | } |
||
119 | } |
||
120 | |||
121 | function getInfo() { |
||
122 | MediaWiki\suppressWarnings(); |
||
123 | $file = fopen( $this->mFilename, 'rb' ); |
||
124 | MediaWiki\restoreWarnings(); |
||
125 | if ( $file === false ) { |
||
126 | wfDebug( __METHOD__ . ": missing or failed file read\n" ); |
||
127 | |||
128 | return false; |
||
129 | } |
||
130 | |||
131 | $header = fread( $file, 16 ); |
||
132 | $info = false; |
||
133 | |||
134 | if ( strlen( $header ) < 16 ) { |
||
135 | wfDebug( __METHOD__ . ": too short file header\n" ); |
||
136 | } else { |
||
137 | $arr = unpack( 'a4magic/a4form/NformLength/a4subtype', $header ); |
||
138 | |||
139 | $subtype = $arr['subtype']; |
||
140 | if ( $arr['magic'] != 'AT&T' ) { |
||
141 | wfDebug( __METHOD__ . ": not a DjVu file\n" ); |
||
142 | } elseif ( $subtype == 'DJVU' ) { |
||
143 | // Single-page document |
||
144 | $info = $this->getPageInfo( $file ); |
||
145 | } elseif ( $subtype == 'DJVM' ) { |
||
146 | // Multi-page document |
||
147 | $info = $this->getMultiPageInfo( $file, $arr['formLength'] ); |
||
148 | } else { |
||
149 | wfDebug( __METHOD__ . ": unrecognized DJVU file type '{$arr['subtype']}'\n" ); |
||
150 | } |
||
151 | } |
||
152 | fclose( $file ); |
||
153 | |||
154 | return $info; |
||
155 | } |
||
156 | |||
157 | private function readChunk( $file ) { |
||
158 | $header = fread( $file, 8 ); |
||
159 | if ( strlen( $header ) < 8 ) { |
||
160 | return [ false, 0 ]; |
||
161 | } else { |
||
162 | $arr = unpack( 'a4chunk/Nlength', $header ); |
||
163 | |||
164 | return [ $arr['chunk'], $arr['length'] ]; |
||
165 | } |
||
166 | } |
||
167 | |||
168 | private function skipChunk( $file, $chunkLength ) { |
||
169 | fseek( $file, $chunkLength, SEEK_CUR ); |
||
170 | |||
171 | View Code Duplication | if ( $chunkLength & 0x01 == 1 && !feof( $file ) ) { |
|
172 | // padding byte |
||
173 | fseek( $file, 1, SEEK_CUR ); |
||
174 | } |
||
175 | } |
||
176 | |||
177 | private function getMultiPageInfo( $file, $formLength ) { |
||
178 | // For now, we'll just look for the first page in the file |
||
179 | // and report its information, hoping others are the same size. |
||
180 | $start = ftell( $file ); |
||
181 | do { |
||
182 | list( $chunk, $length ) = $this->readChunk( $file ); |
||
183 | if ( !$chunk ) { |
||
184 | break; |
||
185 | } |
||
186 | |||
187 | if ( $chunk == 'FORM' ) { |
||
188 | $subtype = fread( $file, 4 ); |
||
189 | if ( $subtype == 'DJVU' ) { |
||
190 | wfDebug( __METHOD__ . ": found first subpage\n" ); |
||
191 | |||
192 | return $this->getPageInfo( $file ); |
||
193 | } |
||
194 | $this->skipChunk( $file, $length - 4 ); |
||
195 | } else { |
||
196 | wfDebug( __METHOD__ . ": skipping '$chunk' chunk\n" ); |
||
197 | $this->skipChunk( $file, $length ); |
||
198 | } |
||
199 | } while ( $length != 0 && !feof( $file ) && ftell( $file ) - $start < $formLength ); |
||
200 | |||
201 | wfDebug( __METHOD__ . ": multi-page DJVU file contained no pages\n" ); |
||
202 | |||
203 | return false; |
||
204 | } |
||
205 | |||
206 | private function getPageInfo( $file ) { |
||
207 | list( $chunk, $length ) = $this->readChunk( $file ); |
||
208 | if ( $chunk != 'INFO' ) { |
||
209 | wfDebug( __METHOD__ . ": expected INFO chunk, got '$chunk'\n" ); |
||
210 | |||
211 | return false; |
||
212 | } |
||
213 | |||
214 | if ( $length < 9 ) { |
||
215 | wfDebug( __METHOD__ . ": INFO should be 9 or 10 bytes, found $length\n" ); |
||
216 | |||
217 | return false; |
||
218 | } |
||
219 | $data = fread( $file, $length ); |
||
220 | if ( strlen( $data ) < $length ) { |
||
221 | wfDebug( __METHOD__ . ": INFO chunk cut off\n" ); |
||
222 | |||
223 | return false; |
||
224 | } |
||
225 | |||
226 | $arr = unpack( |
||
227 | 'nwidth/' . |
||
228 | 'nheight/' . |
||
229 | 'Cminor/' . |
||
230 | 'Cmajor/' . |
||
231 | 'vresolution/' . |
||
232 | 'Cgamma', $data ); |
||
233 | |||
234 | # Newer files have rotation info in byte 10, but we don't use it yet. |
||
235 | |||
236 | return [ |
||
237 | 'width' => $arr['width'], |
||
238 | 'height' => $arr['height'], |
||
239 | 'version' => "{$arr['major']}.{$arr['minor']}", |
||
240 | 'resolution' => $arr['resolution'], |
||
241 | 'gamma' => $arr['gamma'] / 10.0 ]; |
||
242 | } |
||
243 | |||
244 | /** |
||
245 | * Return an XML string describing the DjVu image |
||
246 | * @return string|bool |
||
247 | */ |
||
248 | function retrieveMetaData() { |
||
249 | global $wgDjvuToXML, $wgDjvuDump, $wgDjvuTxt; |
||
250 | |||
251 | if ( !$this->isValid() ) { |
||
252 | return false; |
||
253 | } |
||
254 | |||
255 | if ( isset( $wgDjvuDump ) ) { |
||
256 | # djvudump is faster as of version 3.5 |
||
257 | # https://sourceforge.net/p/djvu/bugs/71/ |
||
258 | $cmd = wfEscapeShellArg( $wgDjvuDump ) . ' ' . wfEscapeShellArg( $this->mFilename ); |
||
259 | $dump = wfShellExec( $cmd ); |
||
260 | $xml = $this->convertDumpToXML( $dump ); |
||
261 | } elseif ( isset( $wgDjvuToXML ) ) { |
||
262 | $cmd = wfEscapeShellArg( $wgDjvuToXML ) . ' --without-anno --without-text ' . |
||
263 | wfEscapeShellArg( $this->mFilename ); |
||
264 | $xml = wfShellExec( $cmd ); |
||
265 | } else { |
||
266 | $xml = null; |
||
267 | } |
||
268 | # Text layer |
||
269 | if ( isset( $wgDjvuTxt ) ) { |
||
270 | $cmd = wfEscapeShellArg( $wgDjvuTxt ) . ' --detail=page ' . wfEscapeShellArg( $this->mFilename ); |
||
271 | wfDebug( __METHOD__ . ": $cmd\n" ); |
||
272 | $retval = ''; |
||
273 | $txt = wfShellExec( $cmd, $retval, [], [ 'memory' => self::DJVUTXT_MEMORY_LIMIT ] ); |
||
274 | if ( $retval == 0 ) { |
||
275 | # Strip some control characters |
||
276 | $txt = preg_replace( "/[\013\035\037]/", "", $txt ); |
||
277 | $reg = <<<EOR |
||
278 | /\(page\s[\d-]*\s[\d-]*\s[\d-]*\s[\d-]*\s*" |
||
279 | ((?> # Text to match is composed of atoms of either: |
||
280 | \\\\. # - any escaped character |
||
281 | | # - any character different from " and \ |
||
282 | [^"\\\\]+ |
||
283 | )*?) |
||
284 | "\s*\) |
||
285 | | # Or page can be empty ; in this case, djvutxt dumps () |
||
286 | \(\s*()\)/sx |
||
287 | EOR; |
||
288 | $txt = preg_replace_callback( $reg, [ $this, 'pageTextCallback' ], $txt ); |
||
289 | $txt = "<DjVuTxt>\n<HEAD></HEAD>\n<BODY>\n" . $txt . "</BODY>\n</DjVuTxt>\n"; |
||
290 | $xml = preg_replace( "/<DjVuXML>/", "<mw-djvu><DjVuXML>", $xml, 1 ); |
||
291 | $xml = $xml . $txt . '</mw-djvu>'; |
||
292 | } |
||
293 | } |
||
294 | |||
295 | return $xml; |
||
296 | } |
||
297 | |||
298 | function pageTextCallback( $matches ) { |
||
304 | |||
305 | /** |
||
306 | * Hack to temporarily work around djvutoxml bug |
||
307 | * @param string $dump |
||
308 | * @return string |
||
309 | */ |
||
310 | function convertDumpToXML( $dump ) { |
||
370 | |||
371 | function parseFormDjvu( $line, &$xml ) { |
||
410 | } |
||
411 |
In PHP it is possible to write to properties without declaring them. For example, the following is perfectly valid PHP code:
Generally, it is a good practice to explictly declare properties to avoid accidental typos and provide IDE auto-completion: