matecat /
xliff-parser
| Total Complexity | 164 |
| Total Lines | 811 |
| Duplicated Lines | 0 % |
| Changes | 6 | ||
| Bugs | 0 | Features | 0 |
Complex classes like XliffSAXTranslationReplacer 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.
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 XliffSAXTranslationReplacer, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 9 | class XliffSAXTranslationReplacer extends AbstractXliffReplacer { |
||
| 10 | /** |
||
| 11 | * @var int |
||
| 12 | */ |
||
| 13 | private $mdaGroupCounter = 0; |
||
| 14 | |||
| 15 | /** |
||
| 16 | * @var array |
||
| 17 | */ |
||
| 18 | private $nodesToCopy = [ |
||
| 19 | 'source', |
||
| 20 | 'mda:metadata', |
||
| 21 | 'memsource:additionalTagData', |
||
| 22 | 'originalData', |
||
| 23 | 'seg-source', |
||
| 24 | 'value', |
||
| 25 | 'bpt', |
||
| 26 | 'ept', |
||
| 27 | 'ph', |
||
| 28 | 'st', |
||
| 29 | 'note', |
||
| 30 | 'context', |
||
| 31 | ]; |
||
| 32 | |||
| 33 | public function replaceTranslation() { |
||
| 34 | fwrite( $this->outputFP, '<?xml version="1.0" encoding="UTF-8"?>' ); |
||
| 35 | |||
| 36 | //create Sax parser |
||
| 37 | $xmlParser = $this->initSaxParser(); |
||
| 38 | |||
| 39 | while ( $this->currentBuffer = fread( $this->originalFP, 4096 ) ) { |
||
| 40 | /* |
||
| 41 | preprocess file |
||
| 42 | */ |
||
| 43 | // obfuscate entities because sax automatically does html_entity_decode |
||
| 44 | $temporary_check_buffer = preg_replace( "/&(.*?);/", self::$INTERNAL_TAG_PLACEHOLDER . '$1' . self::$INTERNAL_TAG_PLACEHOLDER, $this->currentBuffer ); |
||
| 45 | |||
| 46 | //avoid cutting entities in half: |
||
| 47 | //the last fread could have truncated an entity (say, '<' in '&l'), thus invalidating the escaping |
||
| 48 | //***** and if there is an & that it is not an entity, this is an infinite loop !!!!! |
||
| 49 | |||
| 50 | $escape_AMP = false; |
||
| 51 | |||
| 52 | // 9 is the max length of an entity. So, suppose that the & is at the end of buffer, |
||
| 53 | // add 9 Bytes and substitute the entities, if the & is present, and it is not at the end |
||
| 54 | //it can't be an entity, exit the loop |
||
| 55 | |||
| 56 | while ( true ) { |
||
| 57 | $_ampPos = strpos( $temporary_check_buffer, '&' ); |
||
| 58 | |||
| 59 | //check for real entity or escape it to safely exit from the loop!!! |
||
| 60 | if ( $_ampPos === false || strlen( substr( $temporary_check_buffer, $_ampPos ) ) > 9 ) { |
||
| 61 | $escape_AMP = true; |
||
| 62 | break; |
||
| 63 | } |
||
| 64 | |||
| 65 | //if an entity is still present, fetch some more and repeat the escaping |
||
| 66 | $this->currentBuffer .= fread( $this->originalFP, 9 ); |
||
| 67 | $temporary_check_buffer = preg_replace( "/&(.*?);/", self::$INTERNAL_TAG_PLACEHOLDER . '$1' . self::$INTERNAL_TAG_PLACEHOLDER, $this->currentBuffer ); |
||
| 68 | } |
||
| 69 | |||
| 70 | //free stuff outside the loop |
||
| 71 | unset( $temporary_check_buffer ); |
||
| 72 | |||
| 73 | $this->currentBuffer = preg_replace( "/&(.*?);/", self::$INTERNAL_TAG_PLACEHOLDER . '$1' . self::$INTERNAL_TAG_PLACEHOLDER, $this->currentBuffer ); |
||
| 74 | if ( $escape_AMP ) { |
||
| 75 | $this->currentBuffer = str_replace( "&", self::$INTERNAL_TAG_PLACEHOLDER . 'amp' . self::$INTERNAL_TAG_PLACEHOLDER, $this->currentBuffer ); |
||
| 76 | } |
||
| 77 | |||
| 78 | //get length of chunk |
||
| 79 | $this->len = strlen( $this->currentBuffer ); |
||
| 80 | |||
| 81 | //parse chunk of text |
||
| 82 | if ( !xml_parse( $xmlParser, $this->currentBuffer, feof( $this->originalFP ) ) ) { |
||
| 83 | //if unable, raise an exception |
||
| 84 | throw new RuntimeException( sprintf( |
||
| 85 | "XML error: %s at line %d", |
||
| 86 | xml_error_string( xml_get_error_code( $xmlParser ) ), |
||
| 87 | xml_get_current_line_number( $xmlParser ) |
||
| 88 | ) ); |
||
| 89 | } |
||
| 90 | //get accumulated this->offset in document: as long as SAX pointer advances, we keep track of total bytes it has seen so far; this way, we can translate its global pointer in an address local to the current buffer of text to retrieve last char of tag |
||
| 91 | $this->offset += $this->len; |
||
| 92 | } |
||
| 93 | |||
| 94 | // close Sax parser |
||
| 95 | $this->closeSaxParser( $xmlParser ); |
||
| 96 | |||
| 97 | } |
||
| 98 | |||
| 99 | /** |
||
| 100 | * @inheritDoc |
||
| 101 | */ |
||
| 102 | protected function tagOpen( $parser, $name, $attr ) { |
||
| 103 | // check if we are entering into a <trans-unit> (xliff v1.*) or <unit> (xliff v2.*) |
||
| 104 | if ( $this->tuTagName === $name ) { |
||
| 105 | $this->inTU = true; |
||
| 106 | |||
| 107 | // get id |
||
| 108 | // trim to first 100 characters because this is the limit on Matecat's DB |
||
| 109 | $this->currentTransUnitId = substr( $attr[ 'id' ], 0, 100 ); |
||
| 110 | |||
| 111 | // `translate` attribute can be only yes or no |
||
| 112 | if ( isset( $attr[ 'translate' ] ) && $attr[ 'translate' ] === 'no' ) { |
||
| 113 | $attr[ 'translate' ] = 'no'; |
||
| 114 | } else { |
||
| 115 | $attr[ 'translate' ] = 'yes'; |
||
| 116 | } |
||
| 117 | |||
| 118 | // current 'translate' attribute of the current trans-unit |
||
| 119 | $this->currentTransUnitTranslate = $attr[ 'translate' ]; |
||
| 120 | } |
||
| 121 | |||
| 122 | if ( 'source' === $name ) { |
||
| 123 | $this->sourceAttributes = $attr; |
||
| 124 | } |
||
| 125 | |||
| 126 | if ( 'mda:metadata' === $name ) { |
||
| 127 | $this->unitContainsMda = true; |
||
| 128 | } |
||
| 129 | |||
| 130 | // check if we are entering into a <target> |
||
| 131 | if ( 'target' === $name ) { |
||
| 132 | |||
| 133 | if ( $this->currentTransUnitTranslate === 'no' ) { |
||
| 134 | $this->inTarget = false; |
||
| 135 | } else { |
||
| 136 | $this->inTarget = true; |
||
| 137 | } |
||
| 138 | } |
||
| 139 | |||
| 140 | // check if we are inside a <target>, obviously this happen only if there are targets inside the trans-unit |
||
| 141 | // <target> must be stripped to be replaced, so this check avoids <target> reconstruction |
||
| 142 | if ( !$this->inTarget ) { |
||
| 143 | |||
| 144 | $tag = ''; |
||
| 145 | |||
| 146 | // |
||
| 147 | // ============================================ |
||
| 148 | // only for Xliff 2.* |
||
| 149 | // ============================================ |
||
| 150 | // |
||
| 151 | // In xliff v2 we MUST add <mda:metadata> BEFORE <notes>/<originalData>/<segment>/<ignorable> |
||
| 152 | // |
||
| 153 | // As documentation says, <unit> contains: |
||
| 154 | // |
||
| 155 | // - elements from other namespaces, OPTIONAL |
||
| 156 | // - Zero or one <notes> elements followed by |
||
| 157 | // - Zero or one <originalData> element followed by |
||
| 158 | // - One or more <segment> or <ignorable> elements in any order. |
||
| 159 | // |
||
| 160 | // For more info please refer to: |
||
| 161 | // |
||
| 162 | // http://docs.oasis-open.org/xliff/xliff-core/v2.0/os/xliff-core-v2.0-os.html#unit |
||
| 163 | // |
||
| 164 | if ( $this->xliffVersion === 2 && ( $name === 'notes' || $name === 'originalData' || $name === 'segment' || $name === 'ignorable' ) && $this->unitContainsMda === false ) { |
||
| 165 | if ( isset( $this->transUnits[ $this->currentTransUnitId ] ) && !empty( $this->transUnits[ $this->currentTransUnitId ] ) && !$this->hasWrittenCounts ) { |
||
| 166 | |||
| 167 | // we need to update counts here |
||
| 168 | $this->updateCounts(); |
||
| 169 | $this->hasWrittenCounts = true; |
||
| 170 | |||
| 171 | $tag .= $this->getWordCountGroupForXliffV2( $this->counts[ 'raw_word_count' ], $this->counts[ 'eq_word_count' ] ); |
||
| 172 | $this->unitContainsMda = true; |
||
| 173 | } |
||
| 174 | } |
||
| 175 | |||
| 176 | // construct tag |
||
| 177 | $tag .= "<$name "; |
||
| 178 | |||
| 179 | $lastMrkState = null; |
||
| 180 | $stateProp = ''; |
||
| 181 | |||
| 182 | foreach ( $attr as $k => $v ) { |
||
| 183 | |||
| 184 | //if tag name is file, we must replace the target-language attribute |
||
| 185 | if ( $name === 'file' && $k === 'target-language' && !empty( $this->targetLang ) ) { |
||
| 186 | //replace Target language with job language provided from constructor |
||
| 187 | $tag .= "$k=\"$this->targetLang\" "; |
||
| 188 | } else { |
||
| 189 | $pos = 0; |
||
| 190 | if ( $this->currentTransUnitId and isset($this->transUnits[ $this->currentTransUnitId ])) { |
||
|
|
|||
| 191 | $pos = current( $this->transUnits[ $this->currentTransUnitId ] ); |
||
| 192 | } |
||
| 193 | |||
| 194 | if ( $name === $this->tuTagName and isset($this->segments[ $pos ]) and isset($this->segments[ $pos ][ 'sid' ]) ) { |
||
| 195 | |||
| 196 | $sid = $this->segments[ $pos ][ 'sid' ]; |
||
| 197 | |||
| 198 | // add `help-id` to xliff v.1* |
||
| 199 | // add `mtc:segment-id` to xliff v.2* |
||
| 200 | if ( $this->xliffVersion === 1 && strpos( $tag, 'help-id' ) === false ) { |
||
| 201 | if ( !empty( $sid ) ) { |
||
| 202 | $tag .= "help-id=\"$sid\" "; |
||
| 203 | } |
||
| 204 | } elseif ( $this->xliffVersion === 2 && strpos( $tag, 'mtc:segment-id' ) === false ) { |
||
| 205 | if ( !empty( $sid ) ) { |
||
| 206 | $tag .= "mtc:segment-id=\"$sid\" "; |
||
| 207 | } |
||
| 208 | } |
||
| 209 | |||
| 210 | } elseif ( 'segment' === $name && $this->xliffVersion === 2 ) { // add state to segment in Xliff v2 |
||
| 211 | list( $stateProp, $lastMrkState ) = $this->setTransUnitState( $this->segments[ $pos ], $stateProp, $lastMrkState ); |
||
| 212 | } |
||
| 213 | |||
| 214 | //normal tag flux, put attributes in it |
||
| 215 | $tag .= "$k=\"$v\" "; |
||
| 216 | |||
| 217 | // replace state for xliff v2 |
||
| 218 | if ( $stateProp ) { |
||
| 219 | $pattern = '/state=\"(.*)\"/i'; |
||
| 220 | $tag = preg_replace( $pattern, $stateProp, $tag ); |
||
| 221 | } |
||
| 222 | } |
||
| 223 | } |
||
| 224 | |||
| 225 | // add oasis xliff 20 namespace |
||
| 226 | if ( $this->xliffVersion === 2 && $name === 'xliff' && !array_key_exists( 'xmlns:mda', $attr ) ) { |
||
| 227 | $tag .= 'xmlns:mda="urn:oasis:names:tc:xliff:metadata:2.0"'; |
||
| 228 | } |
||
| 229 | |||
| 230 | // add MateCat specific namespace, we want maybe add non-XLIFF attributes |
||
| 231 | if ( $name === 'xliff' && !array_key_exists( 'xmlns:mtc', $attr ) ) { |
||
| 232 | $tag .= ' xmlns:mtc="https://www.matecat.com" '; |
||
| 233 | } |
||
| 234 | |||
| 235 | // trgLang |
||
| 236 | if ( $name === 'xliff' ) { |
||
| 237 | $tag = preg_replace( '/trgLang="(.*?)"/', 'trgLang="' . $this->targetLang . '"', $tag ); |
||
| 238 | } |
||
| 239 | |||
| 240 | //this logic helps detecting empty tags |
||
| 241 | //get current position of SAX pointer in all the stream of data is has read so far: |
||
| 242 | //it points at the end of current tag |
||
| 243 | $idx = xml_get_current_byte_index( $parser ); |
||
| 244 | |||
| 245 | //check whether the bounds of current tag are entirely in current buffer or the end of the current tag |
||
| 246 | //is outside current buffer (in the latter case, it's in next buffer to be read by the while loop); |
||
| 247 | //this check is necessary because we may have truncated a tag in half with current read, |
||
| 248 | //and the other half may be encountered in the next buffer it will be passed |
||
| 249 | if ( isset( $this->currentBuffer[ $idx - $this->offset ] ) ) { |
||
| 250 | //if this tag entire lenght fitted in the buffer, the last char must be the last |
||
| 251 | //symbol before the '>'; if it's an empty tag, it is assumed that it's a '/' |
||
| 252 | $lastChar = $this->currentBuffer[ $idx - $this->offset ]; |
||
| 253 | } else { |
||
| 254 | //if it's out, simple use the last character of the chunk |
||
| 255 | $lastChar = $this->currentBuffer[ $this->len - 1 ]; |
||
| 256 | } |
||
| 257 | |||
| 258 | //trim last space |
||
| 259 | $tag = rtrim( $tag ); |
||
| 260 | |||
| 261 | //detect empty tag |
||
| 262 | $this->isEmpty = ( $lastChar == '/' || $name == 'x' ); |
||
| 263 | if ( $this->isEmpty ) { |
||
| 264 | $tag .= '/'; |
||
| 265 | } |
||
| 266 | |||
| 267 | //add tag ending |
||
| 268 | $tag .= ">"; |
||
| 269 | |||
| 270 | //set a a Buffer for the segSource Source tag |
||
| 271 | if ( $this->bufferIsActive || in_array( $name, $this->nodesToCopy ) ) { // we are opening a critical CDATA section |
||
| 272 | |||
| 273 | //WARNING BECAUSE SOURCE AND SEG-SOURCE TAGS CAN BE EMPTY IN SOME CASES!!!!! |
||
| 274 | //so check for isEmpty also in conjunction with name |
||
| 275 | if ( $this->isEmpty && ( 'source' === $name || 'seg-source' === $name ) ) { |
||
| 276 | $this->postProcAndFlush( $this->outputFP, $tag ); |
||
| 277 | } else { |
||
| 278 | //these are NOT source/seg-source/value empty tags, THERE IS A CONTENT, write it in buffer |
||
| 279 | $this->bufferIsActive = true; |
||
| 280 | $this->CDATABuffer .= $tag; |
||
| 281 | } |
||
| 282 | } else { |
||
| 283 | $this->postProcAndFlush( $this->outputFP, $tag ); |
||
| 284 | } |
||
| 285 | } |
||
| 286 | |||
| 287 | // update segmentPositionInTu |
||
| 288 | |||
| 289 | if ( $this->xliffVersion === 1 && $this->inTU && $name === 'source' ) { |
||
| 290 | $this->segmentPositionInTu++; |
||
| 291 | } |
||
| 292 | |||
| 293 | if ( $this->xliffVersion === 2 && $this->inTU && $name === 'segment' ) { |
||
| 294 | $this->segmentPositionInTu++; |
||
| 295 | } |
||
| 296 | } |
||
| 297 | |||
| 298 | /** |
||
| 299 | * @inheritDoc |
||
| 300 | */ |
||
| 301 | protected function tagClose( $parser, $name ) { |
||
| 302 | $tag = ''; |
||
| 303 | |||
| 304 | /** |
||
| 305 | * if is a tag within <target> or |
||
| 306 | * if it is an empty tag, do not add closing tag because we have already closed it in |
||
| 307 | * |
||
| 308 | * self::tagOpen method |
||
| 309 | */ |
||
| 310 | if ( !$this->isEmpty && !( $this->inTarget && $name !== 'target' ) ) { |
||
| 311 | |||
| 312 | if ( !$this->inTarget ) { |
||
| 313 | $tag = "</$name>"; |
||
| 314 | } |
||
| 315 | |||
| 316 | if ( 'target' == $name ) { |
||
| 317 | |||
| 318 | if ( $this->currentTransUnitTranslate === 'no' ) { |
||
| 319 | // do nothing |
||
| 320 | } elseif ( isset( $this->transUnits[ $this->currentTransUnitId ] ) ) { |
||
| 321 | |||
| 322 | // get translation of current segment, by indirect indexing: id -> positional index -> segment |
||
| 323 | // actually there may be more that one segment to that ID if there are two mrk of the same source segment |
||
| 324 | |||
| 325 | $listOfSegmentsIds = $this->transUnits[ $this->currentTransUnitId ]; |
||
| 326 | |||
| 327 | // $currentSegmentId |
||
| 328 | if ( !empty( $listOfSegmentsIds ) ) { |
||
| 329 | $this->setCurrentSegmentArray( $listOfSegmentsIds ); |
||
| 330 | } |
||
| 331 | |||
| 332 | /* |
||
| 333 | * At the end of every cycle the segment grouping information is lost: unset( 'matecat|' . $this->currentId ) |
||
| 334 | * |
||
| 335 | * We need to take the info about the last segment parsed |
||
| 336 | * ( normally more than 1 db row because of mrk tags ) |
||
| 337 | * |
||
| 338 | * So, copy the current segment data group to an another structure to take the last one segment |
||
| 339 | * for the next tagOpen ( possible sdl:seg-defs ) |
||
| 340 | * |
||
| 341 | */ |
||
| 342 | |||
| 343 | $this->lastTransUnit = []; |
||
| 344 | |||
| 345 | $last_value = null; |
||
| 346 | $segmentsCount = count( $listOfSegmentsIds ); |
||
| 347 | for ( $i = 0; $i < $segmentsCount; $i++ ) { |
||
| 348 | $id = $listOfSegmentsIds[ $i ]; |
||
| 349 | if ( isset( $this->segments[ $id ] ) && ( $i == 0 || $last_value + 1 == $listOfSegmentsIds[ $i ] ) ) { |
||
| 350 | $last_value = $listOfSegmentsIds[ $i ]; |
||
| 351 | $this->lastTransUnit[] = $this->segments[ $id ]; |
||
| 352 | } |
||
| 353 | } |
||
| 354 | |||
| 355 | // init translation and state |
||
| 356 | $translation = ''; |
||
| 357 | $lastMrkState = null; |
||
| 358 | $stateProp = ''; |
||
| 359 | |||
| 360 | // we must reset the lastMrkId found because this is a new segment. |
||
| 361 | $lastMrkId = -1; |
||
| 362 | |||
| 363 | if ( $this->xliffVersion === 2 ) { |
||
| 364 | $seg = $this->segments[ $this->currentSegmentArray[ 'sid' ] ]; |
||
| 365 | |||
| 366 | // update counts |
||
| 367 | if ( !$this->hasWrittenCounts && !empty( $seg ) ) { |
||
| 368 | $this->updateSegmentCounts( $seg ); |
||
| 369 | } |
||
| 370 | |||
| 371 | // delete translations so the prepareSegment |
||
| 372 | // will put source content in target tag |
||
| 373 | if ( $this->sourceInTarget ) { |
||
| 374 | $seg[ 'translation' ] = ''; |
||
| 375 | $this->resetCounts(); |
||
| 376 | } |
||
| 377 | |||
| 378 | // append $translation |
||
| 379 | $translation = $this->prepareTranslation( $seg, $translation ); |
||
| 380 | |||
| 381 | list( $stateProp, $lastMrkState ) = $this->setTransUnitState( $seg, $stateProp, $lastMrkState ); |
||
| 382 | } else { |
||
| 383 | foreach ( $listOfSegmentsIds as $pos => $id ) { |
||
| 384 | |||
| 385 | /* |
||
| 386 | * This routine works to respect the positional orders of markers. |
||
| 387 | * In every cycle we check if the mrk of the segment is below or equal the last one. |
||
| 388 | * When this is true, means that the mrk id belongs to the next segment with the same internal_id |
||
| 389 | * so we MUST stop to apply markers and translations |
||
| 390 | * and stop to add eq_word_count |
||
| 391 | * |
||
| 392 | * Begin: |
||
| 393 | * pre-assign zero to the new mrk if this is the first one ( in this segment ) |
||
| 394 | * If it is null leave it NULL |
||
| 395 | */ |
||
| 396 | if ( (int)$this->segments[ $id ][ "mrk_id" ] < 0 && $this->segments[ $id ][ "mrk_id" ] !== null ) { |
||
| 397 | $this->segments[ $id ][ "mrk_id" ] = 0; |
||
| 398 | } |
||
| 399 | |||
| 400 | /* |
||
| 401 | * WARNING: |
||
| 402 | * For those seg-source that doesn't have a mrk ( having a mrk id === null ) |
||
| 403 | * ( null <= -1 ) === true |
||
| 404 | * so, cast to int |
||
| 405 | */ |
||
| 406 | if ( (int)$this->segments[ $id ][ "mrk_id" ] <= $lastMrkId ) { |
||
| 407 | break; |
||
| 408 | } |
||
| 409 | |||
| 410 | // set $this->currentSegment |
||
| 411 | $seg = $this->segments[ $id ]; |
||
| 412 | |||
| 413 | // update counts |
||
| 414 | if ( !empty( $seg ) ) { |
||
| 415 | $this->updateSegmentCounts( $seg ); |
||
| 416 | } |
||
| 417 | |||
| 418 | // delete translations so the prepareSegment |
||
| 419 | // will put source content in target tag |
||
| 420 | if ( $this->sourceInTarget ) { |
||
| 421 | $seg[ 'translation' ] = ''; |
||
| 422 | $this->resetCounts(); |
||
| 423 | } |
||
| 424 | |||
| 425 | // append $translation |
||
| 426 | $translation = $this->prepareTranslation( $seg, $translation ); |
||
| 427 | |||
| 428 | // for xliff 2 we need $this->transUnits[ $this->currentId ] [ $pos ] for populating metadata |
||
| 429 | |||
| 430 | unset( $this->transUnits[ $this->currentTransUnitId ] [ $pos ] ); |
||
| 431 | |||
| 432 | $lastMrkId = $this->segments[ $id ][ "mrk_id" ]; |
||
| 433 | |||
| 434 | list( $stateProp, $lastMrkState ) = $this->setTransUnitState( $seg, $stateProp, $lastMrkState ); |
||
| 435 | } |
||
| 436 | } |
||
| 437 | |||
| 438 | //append translation |
||
| 439 | $targetLang = ''; |
||
| 440 | if ( $this->xliffVersion === 1 ) { |
||
| 441 | $targetLang = ' xml:lang="' . $this->targetLang . '"'; |
||
| 442 | } |
||
| 443 | |||
| 444 | $tag = $this->buildTranslateTag( $targetLang, $stateProp, $translation, $this->counts[ 'raw_word_count' ], $this->counts[ 'eq_word_count' ] ); |
||
| 445 | } |
||
| 446 | |||
| 447 | // signal we are leaving a target |
||
| 448 | $this->targetWasWritten = true; |
||
| 449 | $this->inTarget = false; |
||
| 450 | $this->postProcAndFlush( $this->outputFP, $tag, $treatAsCDATA = true ); |
||
| 451 | } elseif ( in_array( $name, $this->nodesToCopy ) ) { // we are closing a critical CDATA section |
||
| 452 | |||
| 453 | $this->bufferIsActive = false; |
||
| 454 | |||
| 455 | // only for Xliff 2.* |
||
| 456 | // write here <mda:metaGroup> and <mda:meta> if already present in the <unit> |
||
| 457 | if ( 'mda:metadata' === $name && $this->unitContainsMda && $this->xliffVersion === 2 && !$this->hasWrittenCounts ) { |
||
| 458 | |||
| 459 | // we need to update counts here |
||
| 460 | $this->updateCounts(); |
||
| 461 | $this->hasWrittenCounts = true; |
||
| 462 | |||
| 463 | $tag = $this->CDATABuffer; |
||
| 464 | $tag .= $this->getWordCountGroupForXliffV2( $this->counts[ 'raw_word_count' ], $this->counts[ 'eq_word_count' ], false ); |
||
| 465 | $tag .= " </mda:metadata>"; |
||
| 466 | |||
| 467 | } else { |
||
| 468 | $tag = $this->CDATABuffer . "</$name>"; |
||
| 469 | } |
||
| 470 | |||
| 471 | $this->CDATABuffer = ""; |
||
| 472 | |||
| 473 | //flush to pointer |
||
| 474 | $this->postProcAndFlush( $this->outputFP, $tag ); |
||
| 475 | } elseif ( 'segment' === $name ) { |
||
| 476 | |||
| 477 | // only for Xliff 2.* |
||
| 478 | // if segment has no <target> add it BEFORE </segment> |
||
| 479 | if ( $this->xliffVersion === 2 && !$this->targetWasWritten ) { |
||
| 480 | |||
| 481 | $seg = $this->getCurrentSegment(); |
||
| 482 | |||
| 483 | // copy attr from <source> |
||
| 484 | $tag = '<target'; |
||
| 485 | foreach ( $this->sourceAttributes as $k => $v ) { |
||
| 486 | $tag .= " $k=\"$v\""; |
||
| 487 | } |
||
| 488 | |||
| 489 | $tag .= '>' . $seg[ 'translation' ] . '</target></segment>'; |
||
| 490 | } |
||
| 491 | |||
| 492 | $this->postProcAndFlush( $this->outputFP, $tag ); |
||
| 493 | |||
| 494 | // we are leaving <segment>, reset $segmentHasTarget |
||
| 495 | $this->targetWasWritten = false; |
||
| 496 | |||
| 497 | } elseif ( $name === 'trans-unit' ) { |
||
| 498 | |||
| 499 | // only for Xliff 1.* |
||
| 500 | // handling </trans-unit> closure |
||
| 501 | if ( !$this->targetWasWritten ) { |
||
| 502 | $seg = $this->getCurrentSegment(); |
||
| 503 | $lastMrkState = null; |
||
| 504 | $stateProp = ''; |
||
| 505 | $tag = ''; |
||
| 506 | |||
| 507 | // if there is translation available insert <target> BEFORE </trans-unit> |
||
| 508 | if ( isset( $seg[ 'translation' ] ) ) { |
||
| 509 | list( $stateProp, $lastMrkState ) = $this->setTransUnitState( $seg, $stateProp, $lastMrkState ); |
||
| 510 | $tag .= $this->createTargetTag( $seg[ 'translation' ], $stateProp ); |
||
| 511 | } |
||
| 512 | |||
| 513 | $tag .= '</trans-unit>'; |
||
| 514 | $this->postProcAndFlush( $this->outputFP, $tag ); |
||
| 515 | } else { |
||
| 516 | $this->postProcAndFlush( $this->outputFP, '</trans-unit>' ); |
||
| 517 | } |
||
| 518 | } elseif ( $this->bufferIsActive ) { // this is a tag ( <g | <mrk ) inside a seg or seg-source tag |
||
| 519 | $this->CDATABuffer .= "</$name>"; |
||
| 520 | // Do NOT Flush |
||
| 521 | } else { //generic tag closure do Nothing |
||
| 522 | // flush to pointer |
||
| 523 | $this->postProcAndFlush( $this->outputFP, $tag ); |
||
| 524 | } |
||
| 525 | } elseif ( $this->CDATABuffer === '<note/>' && $this->bufferIsActive === true ) { |
||
| 526 | $this->postProcAndFlush( $this->outputFP, '<note/>' ); |
||
| 527 | $this->bufferIsActive = false; |
||
| 528 | $this->CDATABuffer = ''; |
||
| 529 | $this->isEmpty = false; |
||
| 530 | } else { |
||
| 531 | //ok, nothing to be done; reset flag for next coming tag |
||
| 532 | $this->isEmpty = false; |
||
| 533 | } |
||
| 534 | |||
| 535 | // check if we are leaving a <trans-unit> (xliff v1.*) or <unit> (xliff v2.*) |
||
| 536 | if ( $this->tuTagName === $name ) { |
||
| 537 | $this->currentTransUnitTranslate = null; |
||
| 538 | $this->inTU = false; |
||
| 539 | $this->segmentPositionInTu = -1; |
||
| 540 | $this->unitContainsMda = false; |
||
| 541 | $this->hasWrittenCounts = false; |
||
| 542 | $this->sourceAttributes = []; |
||
| 543 | } |
||
| 544 | } |
||
| 545 | |||
| 546 | /** |
||
| 547 | * Set the current segment array (with segment id and trans-unit id) |
||
| 548 | * |
||
| 549 | * @param array $listOfSegmentsIds |
||
| 550 | */ |
||
| 551 | private function setCurrentSegmentArray( array $listOfSegmentsIds = [] ) { |
||
| 567 | ]; |
||
| 568 | } |
||
| 569 | } |
||
| 570 | } |
||
| 571 | |||
| 572 | /** |
||
| 573 | * Update counts |
||
| 574 | */ |
||
| 575 | private function updateCounts() { |
||
| 599 | } |
||
| 600 | |||
| 601 | /** |
||
| 602 | * @param array $seg |
||
| 603 | */ |
||
| 604 | private function updateSegmentCounts( array $seg = [] ) { |
||
| 605 | $this->counts[ 'raw_word_count' ] += $seg[ 'raw_word_count' ]; |
||
| 606 | $this->counts[ 'eq_word_count' ] += ( floor( $seg[ 'eq_word_count' ] * 100 ) / 100 ); |
||
| 607 | } |
||
| 608 | |||
| 609 | private function resetCounts() { |
||
| 612 | } |
||
| 613 | |||
| 614 | /** |
||
| 615 | * prepare segment tagging for xliff insertion |
||
| 616 | * |
||
| 617 | * @param array $seg |
||
| 618 | * @param string $transUnitTranslation |
||
| 619 | * |
||
| 620 | * @return string |
||
| 621 | */ |
||
| 622 | protected function prepareTranslation( $seg, $transUnitTranslation = "" ) { |
||
| 623 | $endTags = ""; |
||
| 624 | |||
| 625 | $segment = Strings::removeDangerousChars( $seg [ 'segment' ] ); |
||
| 626 | $translation = Strings::removeDangerousChars( $seg [ 'translation' ] ); |
||
| 627 | $dataRefMap = ( isset( $seg[ 'data_ref_map' ] ) && $seg[ 'data_ref_map' ] !== null ) ? Strings::jsonToArray( $seg[ 'data_ref_map' ] ) : []; |
||
| 628 | |||
| 629 | if ( is_null( $seg [ 'translation' ] ) || $seg [ 'translation' ] == '' ) { |
||
| 630 | $translation = $segment; |
||
| 631 | } else { |
||
| 632 | if ( $this->callback instanceof XliffReplacerCallbackInterface ) { |
||
| 633 | $error = (isset($seg['error'])) ? $seg['error'] : null; |
||
| 634 | if ( $this->callback->thereAreErrors( $seg[ 'sid' ], $segment, $translation, $dataRefMap, $error ) ) { |
||
| 635 | $translation = '|||UNTRANSLATED_CONTENT_START|||' . $segment . '|||UNTRANSLATED_CONTENT_END|||'; |
||
| 636 | } |
||
| 637 | } |
||
| 638 | } |
||
| 639 | |||
| 640 | // for xliff v2 we ignore the marks on purpose |
||
| 641 | if ( $this->xliffVersion === 2 ) { |
||
| 642 | return $translation; |
||
| 643 | } |
||
| 644 | |||
| 645 | if ( $seg[ 'mrk_id' ] !== null && $seg[ 'mrk_id' ] != '' ) { |
||
| 646 | if ( $this->targetLang === 'ja-JP' ) { |
||
| 647 | $seg[ 'mrk_succ_tags' ] = ltrim( $seg[ 'mrk_succ_tags' ] ); |
||
| 648 | } |
||
| 649 | |||
| 650 | $translation = "<mrk mid=\"" . $seg[ 'mrk_id' ] . "\" mtype=\"seg\">" . $seg[ 'mrk_prev_tags' ] . $translation . $seg[ 'mrk_succ_tags' ] . "</mrk>"; |
||
| 651 | } |
||
| 652 | |||
| 653 | $transUnitTranslation .= $seg[ 'prev_tags' ] . $translation . $endTags . $seg[ 'succ_tags' ]; |
||
| 654 | |||
| 655 | return $transUnitTranslation; |
||
| 656 | } |
||
| 657 | |||
| 658 | /** |
||
| 659 | * @param $targetLang |
||
| 660 | * @param $stateProp |
||
| 661 | * @param $translation |
||
| 662 | * @param $rawWordCount |
||
| 663 | * @param $eqWordCount |
||
| 664 | * |
||
| 665 | * @return string |
||
| 666 | */ |
||
| 667 | private function buildTranslateTag( $targetLang, $stateProp, $translation, $rawWordCount, $eqWordCount ) { |
||
| 668 | switch ( $this->xliffVersion ) { |
||
| 669 | case 1: |
||
| 670 | default: |
||
| 671 | $tag = "<target $targetLang $stateProp>$translation</target>"; |
||
| 672 | |||
| 673 | // if it's a Trados file don't append count group |
||
| 674 | if ( get_class( $this ) !== SdlXliffSAXTranslationReplacer::class ) { |
||
| 675 | $tag .= $this->getWordCountGroup( $rawWordCount, $eqWordCount ); |
||
| 676 | } |
||
| 677 | |||
| 678 | return $tag; |
||
| 679 | |||
| 680 | case 2: |
||
| 681 | return "<target>$translation</target>"; |
||
| 682 | } |
||
| 683 | } |
||
| 684 | |||
| 685 | /** |
||
| 686 | * @param $raw_word_count |
||
| 687 | * @param $eq_word_count |
||
| 688 | * |
||
| 689 | * @return string |
||
| 690 | */ |
||
| 691 | private function getWordCountGroup( $raw_word_count, $eq_word_count ) { |
||
| 692 | return "\n<count-group name=\"$this->currentTransUnitId\"><count count-type=\"x-matecat-raw\">$raw_word_count</count><count count-type=\"x-matecat-weighted\">$eq_word_count</count></count-group>"; |
||
| 693 | } |
||
| 694 | |||
| 695 | /** |
||
| 696 | * @return array |
||
| 697 | */ |
||
| 698 | private function getCurrentSegment() { |
||
| 699 | if ( $this->currentTransUnitTranslate === 'yes' && isset( $this->transUnits[ $this->currentTransUnitId ] ) ) { |
||
| 700 | $index = $this->transUnits[ $this->currentTransUnitId ][ $this->segmentPositionInTu ]; |
||
| 701 | |||
| 702 | if ( isset( $this->segments[ $index ] ) ) { |
||
| 703 | return $this->segments[ $index ]; |
||
| 704 | } |
||
| 705 | } |
||
| 706 | |||
| 707 | return []; |
||
| 708 | } |
||
| 709 | |||
| 710 | /** |
||
| 711 | * This function create a <target> |
||
| 712 | * |
||
| 713 | * @param $translation |
||
| 714 | * @param $stateProp |
||
| 715 | * |
||
| 716 | * @return string |
||
| 717 | */ |
||
| 718 | private function createTargetTag( $translation, $stateProp ) { |
||
| 719 | $targetLang = 'xml:lang="' . $this->targetLang . '"'; |
||
| 720 | |||
| 721 | return $this->buildTranslateTag( $targetLang, $stateProp, $translation, $this->counts[ 'raw_word_count' ], $this->counts[ 'eq_word_count' ] ); |
||
| 722 | } |
||
| 723 | |||
| 724 | /** |
||
| 725 | * @param $raw_word_count |
||
| 726 | * @param $eq_word_count |
||
| 727 | * @param bool $withMetadataTag |
||
| 728 | * |
||
| 729 | * @return string |
||
| 730 | */ |
||
| 731 | private function getWordCountGroupForXliffV2( $raw_word_count, $eq_word_count, $withMetadataTag = true ) { |
||
| 747 | </mda:metaGroup> |
||
| 748 | </mda:metadata>"; |
||
| 749 | } |
||
| 750 | |||
| 751 | /** |
||
| 752 | * @param $seg |
||
| 753 | * @param $state_prop |
||
| 754 | * @param $lastMrkState |
||
| 755 | * |
||
| 756 | * @return array |
||
| 757 | */ |
||
| 758 | private function setTransUnitState( $seg, $state_prop, $lastMrkState ) { |
||
| 759 | switch ( $seg[ 'status' ] ) { |
||
| 760 | |||
| 761 | case TranslationStatus::STATUS_FIXED: |
||
| 762 | case TranslationStatus::STATUS_APPROVED: |
||
| 763 | if ( $lastMrkState == null || $lastMrkState == TranslationStatus::STATUS_APPROVED ) { |
||
| 764 | $state_prop = ( $this->xliffVersion === 2 ) ? "state=\"reviewed\"" : "state=\"signed-off\""; |
||
| 765 | $lastMrkState = TranslationStatus::STATUS_APPROVED; |
||
| 766 | } |
||
| 767 | break; |
||
| 768 | |||
| 769 | case TranslationStatus::STATUS_TRANSLATED: |
||
| 770 | if ( $lastMrkState == null || $lastMrkState == TranslationStatus::STATUS_TRANSLATED || $lastMrkState == TranslationStatus::STATUS_APPROVED ) { |
||
| 771 | $state_prop = "state=\"translated\""; |
||
| 772 | $lastMrkState = TranslationStatus::STATUS_TRANSLATED; |
||
| 773 | } |
||
| 774 | break; |
||
| 775 | |||
| 776 | case TranslationStatus::STATUS_REJECTED: // if there is a mark REJECTED and there is not a DRAFT, all the trans-unit is REJECTED. In V2 there is no way to mark |
||
| 777 | case TranslationStatus::STATUS_REBUTTED: |
||
| 778 | if ( ( $lastMrkState == null ) || ( $lastMrkState != TranslationStatus::STATUS_NEW || $lastMrkState != TranslationStatus::STATUS_DRAFT ) ) { |
||
| 779 | $state_prop = ( $this->xliffVersion === 2 ) ? "state=\"initial\"" : "state=\"needs-review-translation\""; |
||
| 780 | $lastMrkState = TranslationStatus::STATUS_REJECTED; |
||
| 781 | } |
||
| 782 | break; |
||
| 783 | |||
| 784 | case TranslationStatus::STATUS_NEW: |
||
| 785 | if ( ( $lastMrkState == null ) || $lastMrkState != TranslationStatus::STATUS_DRAFT ) { |
||
| 786 | $state_prop = ( $this->xliffVersion === 2 ) ? "state=\"initial\"" : "state=\"new\""; |
||
| 787 | $lastMrkState = TranslationStatus::STATUS_NEW; |
||
| 788 | } |
||
| 789 | break; |
||
| 790 | |||
| 791 | case TranslationStatus::STATUS_DRAFT: |
||
| 792 | $state_prop = "state=\"needs-translation\""; |
||
| 793 | $lastMrkState = TranslationStatus::STATUS_DRAFT; |
||
| 794 | break; |
||
| 795 | default: |
||
| 796 | // this is the case when a segment is not showed in cattool, so the row in |
||
| 797 | // segment_translations does not exists and |
||
| 798 | // ---> $seg[ 'status' ] is NULL |
||
| 799 | if ( $lastMrkState == null ) { //this is the first MRK ID |
||
| 800 | $state_prop = "state=\"translated\""; |
||
| 801 | $lastMrkState = TranslationStatus::STATUS_TRANSLATED; |
||
| 802 | } else { |
||
| 803 | /* Do nothing and preserve the last state */ |
||
| 804 | } |
||
| 805 | break; |
||
| 806 | } |
||
| 807 | |||
| 808 | return [ $state_prop, $lastMrkState ]; |
||
| 809 | } |
||
| 810 | |||
| 811 | /** |
||
| 812 | * @inheritDoc |
||
| 813 | */ |
||
| 814 | protected function characterData( $parser, $data ) { |
||
| 820 | } |
||
| 821 | } |
||
| 822 | } |
||
| 823 |
Loading history...
PHP has two types of connecting operators (logical operators, and boolean operators):
and&&or||The difference between these is the order in which they are executed. In most cases, you would want to use a boolean operator like
&&, or||.Let’s take a look at a few examples:
Logical Operators are used for Control-Flow
One case where you explicitly want to use logical operators is for control-flow such as this:
Since
dieintroduces problems of its own, f.e. it makes our code hardly testable, and prevents any kind of more sophisticated error handling; you probably do not want to use this in real-world code. Unfortunately, logical operators cannot be combined withthrowat this point:These limitations lead to logical operators rarely being of use in current PHP code.