alxarafe /
alixar
| Total Complexity | 539 |
| Total Lines | 2648 |
| Duplicated Lines | 0 % |
| Changes | 0 | ||
Complex classes like TCPDF_FONTS 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 TCPDF_FONTS, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 49 | class TCPDF_FONTS |
||
| 50 | { |
||
| 51 | /** |
||
| 52 | * Static cache used for speed up uniord performances |
||
| 53 | * @protected |
||
| 54 | */ |
||
| 55 | protected static $cache_uniord = array(); |
||
| 56 | |||
| 57 | /** |
||
| 58 | * Convert and add the selected TrueType or Type1 font to the fonts folder (that must be writeable). |
||
| 59 | * @param string $fontfile Font file (full path). |
||
| 60 | * @param string $fonttype Font type. Leave empty for autodetect mode. Valid values are: TrueTypeUnicode, TrueType, Type1, CID0JP = CID-0 Japanese, CID0KR = CID-0 Korean, CID0CS = CID-0 Chinese Simplified, CID0CT = CID-0 Chinese Traditional. |
||
| 61 | * @param string $enc Name of the encoding table to use. Leave empty for default mode. Omit this parameter for TrueType Unicode and symbolic fonts like Symbol or ZapfDingBats. |
||
| 62 | * @param int $flags Unsigned 32-bit integer containing flags specifying various characteristics of the font (PDF32000:2008 - 9.8.2 Font Descriptor Flags): +1 for fixed font; +4 for symbol or +32 for non-symbol; +64 for italic. Fixed and Italic mode are generally autodetected so you have to set it to 32 = non-symbolic font (default) or 4 = symbolic font. |
||
| 63 | * @param string $outpath Output path for generated font files (must be writeable by the web server). Leave empty for default font folder. |
||
| 64 | * @param int $platid Platform ID for CMAP table to extract (when building a Unicode font for Windows this value should be 3, for Macintosh should be 1). |
||
| 65 | * @param int $encid Encoding ID for CMAP table to extract (when building a Unicode font for Windows this value should be 1, for Macintosh should be 0). When Platform ID is 3, legal values for Encoding ID are: 0=Symbol, 1=Unicode, 2=ShiftJIS, 3=PRC, 4=Big5, 5=Wansung, 6=Johab, 7=Reserved, 8=Reserved, 9=Reserved, 10=UCS-4. |
||
| 66 | * @param boolean $addcbbox If true includes the character bounding box information on the php font file. |
||
| 67 | * @param boolean $link If true link to system font instead of copying the font data (not transportable) - Note: do not work with Type1 fonts. |
||
| 68 | * @return string|false TCPDF font name or boolean false in case of error. |
||
| 69 | * @author Nicola Asuni |
||
| 70 | * @since 5.9.123 (2010-09-30) |
||
| 71 | * @public static |
||
| 72 | */ |
||
| 73 | public static function addTTFfont($fontfile, $fonttype = '', $enc = '', $flags = 32, $outpath = '', $platid = 3, $encid = 1, $addcbbox = false, $link = false) |
||
| 74 | { |
||
| 75 | if (!TCPDF_STATIC::file_exists($fontfile)) { |
||
| 76 | // Could not find file |
||
| 77 | return false; |
||
| 78 | } |
||
| 79 | // font metrics |
||
| 80 | $fmetric = array(); |
||
| 81 | // build new font name for TCPDF compatibility |
||
| 82 | $font_path_parts = pathinfo($fontfile); |
||
| 83 | if (!isset($font_path_parts['filename'])) { |
||
| 84 | $font_path_parts['filename'] = substr($font_path_parts['basename'], 0, -(strlen($font_path_parts['extension']) + 1)); |
||
| 85 | } |
||
| 86 | $font_name = strtolower($font_path_parts['filename']); |
||
| 87 | $font_name = preg_replace('/[^a-z0-9_]/', '', $font_name); |
||
| 88 | $search = array('bold', 'oblique', 'italic', 'regular'); |
||
| 89 | $replace = array('b', 'i', 'i', ''); |
||
| 90 | $font_name = str_replace($search, $replace, $font_name); |
||
| 91 | if (empty($font_name)) { |
||
| 92 | // set generic name |
||
| 93 | $font_name = 'tcpdffont'; |
||
| 94 | } |
||
| 95 | // set output path |
||
| 96 | if (empty($outpath)) { |
||
| 97 | $outpath = self::_getfontpath(); |
||
| 98 | } |
||
| 99 | // check if this font already exist |
||
| 100 | if (@TCPDF_STATIC::file_exists($outpath . $font_name . '.php')) { |
||
| 101 | // this font already exist (delete it from fonts folder to rebuild it) |
||
| 102 | return $font_name; |
||
| 103 | } |
||
| 104 | $fmetric['file'] = $font_name; |
||
| 105 | $fmetric['ctg'] = $font_name . '.ctg.z'; |
||
| 106 | // get font data |
||
| 107 | $font = file_get_contents($fontfile); |
||
| 108 | $fmetric['originalsize'] = strlen($font); |
||
| 109 | // autodetect font type |
||
| 110 | if (empty($fonttype)) { |
||
| 111 | if (TCPDF_STATIC::_getULONG($font, 0) == 0x10000) { |
||
| 112 | // True Type (Unicode or not) |
||
| 113 | $fonttype = 'TrueTypeUnicode'; |
||
| 114 | } elseif (substr($font, 0, 4) == 'OTTO') { |
||
| 115 | // Open Type (Unicode or not) |
||
| 116 | //Unsupported font format: OpenType with CFF data |
||
| 117 | return false; |
||
| 118 | } else { |
||
| 119 | // Type 1 |
||
| 120 | $fonttype = 'Type1'; |
||
| 121 | } |
||
| 122 | } |
||
| 123 | // set font type |
||
| 124 | switch ($fonttype) { |
||
| 125 | case 'CID0CT': |
||
| 126 | case 'CID0CS': |
||
| 127 | case 'CID0KR': |
||
| 128 | case 'CID0JP': { |
||
| 129 | $fmetric['type'] = 'cidfont0'; |
||
| 130 | break; |
||
| 131 | } |
||
| 132 | case 'Type1': { |
||
| 133 | $fmetric['type'] = 'Type1'; |
||
| 134 | if (empty($enc) and (($flags & 4) == 0)) { |
||
| 135 | $enc = 'cp1252'; |
||
| 136 | } |
||
| 137 | break; |
||
| 138 | } |
||
| 139 | case 'TrueType': { |
||
| 140 | $fmetric['type'] = 'TrueType'; |
||
| 141 | break; |
||
| 142 | } |
||
| 143 | case 'TrueTypeUnicode': |
||
| 144 | default: { |
||
| 145 | $fmetric['type'] = 'TrueTypeUnicode'; |
||
| 146 | break; |
||
| 147 | } |
||
| 148 | } |
||
| 149 | // set encoding maps (if any) |
||
| 150 | $fmetric['enc'] = preg_replace('/[^A-Za-z0-9_\-]/', '', $enc); |
||
| 151 | $fmetric['diff'] = ''; |
||
| 152 | if (($fmetric['type'] == 'TrueType') or ($fmetric['type'] == 'Type1')) { |
||
| 153 | if (!empty($enc) and ($enc != 'cp1252') and isset(TCPDF_FONT_DATA::$encmap[$enc])) { |
||
| 154 | // build differences from reference encoding |
||
| 155 | $enc_ref = TCPDF_FONT_DATA::$encmap['cp1252']; |
||
| 156 | $enc_target = TCPDF_FONT_DATA::$encmap[$enc]; |
||
| 157 | $last = 0; |
||
| 158 | for ($i = 32; $i <= 255; ++$i) { |
||
| 159 | if ($enc_target[$i] != $enc_ref[$i]) { |
||
| 160 | if ($i != ($last + 1)) { |
||
| 161 | $fmetric['diff'] .= $i . ' '; |
||
| 162 | } |
||
| 163 | $last = $i; |
||
| 164 | $fmetric['diff'] .= '/' . $enc_target[$i] . ' '; |
||
| 165 | } |
||
| 166 | } |
||
| 167 | } |
||
| 168 | } |
||
| 169 | // parse the font by type |
||
| 170 | if ($fmetric['type'] == 'Type1') { |
||
| 171 | // ---------- TYPE 1 ---------- |
||
| 172 | // read first segment |
||
| 173 | $a = unpack('Cmarker/Ctype/Vsize', substr($font, 0, 6)); |
||
| 174 | if ($a['marker'] != 128) { |
||
| 175 | // Font file is not a valid binary Type1 |
||
| 176 | return false; |
||
| 177 | } |
||
| 178 | $fmetric['size1'] = $a['size']; |
||
| 179 | $data = substr($font, 6, $fmetric['size1']); |
||
| 180 | // read second segment |
||
| 181 | $a = unpack('Cmarker/Ctype/Vsize', substr($font, (6 + $fmetric['size1']), 6)); |
||
| 182 | if ($a['marker'] != 128) { |
||
| 183 | // Font file is not a valid binary Type1 |
||
| 184 | return false; |
||
| 185 | } |
||
| 186 | $fmetric['size2'] = $a['size']; |
||
| 187 | $encrypted = substr($font, (12 + $fmetric['size1']), $fmetric['size2']); |
||
| 188 | $data .= $encrypted; |
||
| 189 | // store compressed font |
||
| 190 | $fmetric['file'] .= '.z'; |
||
| 191 | $fp = TCPDF_STATIC::fopenLocal($outpath . $fmetric['file'], 'wb'); |
||
| 192 | fwrite($fp, gzcompress($data)); |
||
| 193 | fclose($fp); |
||
| 194 | // get font info |
||
| 195 | $fmetric['Flags'] = $flags; |
||
| 196 | preg_match('#/FullName[\s]*\(([^\)]*)#', $font, $matches); |
||
| 197 | $fmetric['name'] = preg_replace('/[^a-zA-Z0-9_\-]/', '', $matches[1]); |
||
| 198 | preg_match('#/FontBBox[\s]*{([^}]*)#', $font, $matches); |
||
| 199 | $fmetric['bbox'] = trim($matches[1]); |
||
| 200 | $bv = explode(' ', $fmetric['bbox']); |
||
| 201 | $fmetric['Ascent'] = intval($bv[3]); |
||
| 202 | $fmetric['Descent'] = intval($bv[1]); |
||
| 203 | preg_match('#/ItalicAngle[\s]*([0-9\+\-]*)#', $font, $matches); |
||
| 204 | $fmetric['italicAngle'] = intval($matches[1]); |
||
| 205 | if ($fmetric['italicAngle'] != 0) { |
||
| 206 | $fmetric['Flags'] |= 64; |
||
| 207 | } |
||
| 208 | preg_match('#/UnderlinePosition[\s]*([0-9\+\-]*)#', $font, $matches); |
||
| 209 | $fmetric['underlinePosition'] = intval($matches[1]); |
||
| 210 | preg_match('#/UnderlineThickness[\s]*([0-9\+\-]*)#', $font, $matches); |
||
| 211 | $fmetric['underlineThickness'] = intval($matches[1]); |
||
| 212 | preg_match('#/isFixedPitch[\s]*([^\s]*)#', $font, $matches); |
||
| 213 | if ($matches[1] == 'true') { |
||
| 214 | $fmetric['Flags'] |= 1; |
||
| 215 | } |
||
| 216 | // get internal map |
||
| 217 | $imap = array(); |
||
| 218 | if (preg_match_all('#dup[\s]([0-9]+)[\s]*/([^\s]*)[\s]put#sU', $font, $fmap, PREG_SET_ORDER) > 0) { |
||
| 219 | foreach ($fmap as $v) { |
||
| 220 | $imap[$v[2]] = $v[1]; |
||
| 221 | } |
||
| 222 | } |
||
| 223 | // decrypt eexec encrypted part |
||
| 224 | $r = 55665; // eexec encryption constant |
||
| 225 | $c1 = 52845; |
||
| 226 | $c2 = 22719; |
||
| 227 | $elen = strlen($encrypted); |
||
| 228 | $eplain = ''; |
||
| 229 | for ($i = 0; $i < $elen; ++$i) { |
||
| 230 | $chr = ord($encrypted[$i]); |
||
| 231 | $eplain .= chr($chr ^ ($r >> 8)); |
||
| 232 | $r = ((($chr + $r) * $c1 + $c2) % 65536); |
||
| 233 | } |
||
| 234 | if (preg_match('#/ForceBold[\s]*([^\s]*)#', $eplain, $matches) > 0) { |
||
| 235 | if ($matches[1] == 'true') { |
||
| 236 | $fmetric['Flags'] |= 0x40000; |
||
| 237 | } |
||
| 238 | } |
||
| 239 | if (preg_match('#/StdVW[\s]*\[([^\]]*)#', $eplain, $matches) > 0) { |
||
| 240 | $fmetric['StemV'] = intval($matches[1]); |
||
| 241 | } else { |
||
| 242 | $fmetric['StemV'] = 70; |
||
| 243 | } |
||
| 244 | if (preg_match('#/StdHW[\s]*\[([^\]]*)#', $eplain, $matches) > 0) { |
||
| 245 | $fmetric['StemH'] = intval($matches[1]); |
||
| 246 | } else { |
||
| 247 | $fmetric['StemH'] = 30; |
||
| 248 | } |
||
| 249 | if (preg_match('#/BlueValues[\s]*\[([^\]]*)#', $eplain, $matches) > 0) { |
||
| 250 | $bv = explode(' ', $matches[1]); |
||
| 251 | if (count($bv) >= 6) { |
||
| 252 | $v1 = intval($bv[2]); |
||
| 253 | $v2 = intval($bv[4]); |
||
| 254 | if ($v1 <= $v2) { |
||
| 255 | $fmetric['XHeight'] = $v1; |
||
| 256 | $fmetric['CapHeight'] = $v2; |
||
| 257 | } else { |
||
| 258 | $fmetric['XHeight'] = $v2; |
||
| 259 | $fmetric['CapHeight'] = $v1; |
||
| 260 | } |
||
| 261 | } else { |
||
| 262 | $fmetric['XHeight'] = 450; |
||
| 263 | $fmetric['CapHeight'] = 700; |
||
| 264 | } |
||
| 265 | } else { |
||
| 266 | $fmetric['XHeight'] = 450; |
||
| 267 | $fmetric['CapHeight'] = 700; |
||
| 268 | } |
||
| 269 | // get the number of random bytes at the beginning of charstrings |
||
| 270 | if (preg_match('#/lenIV[\s]*([0-9]*)#', $eplain, $matches) > 0) { |
||
| 271 | $lenIV = intval($matches[1]); |
||
| 272 | } else { |
||
| 273 | $lenIV = 4; |
||
| 274 | } |
||
| 275 | $fmetric['Leading'] = 0; |
||
| 276 | // get charstring data |
||
| 277 | $eplain = substr($eplain, (strpos($eplain, '/CharStrings') + 1)); |
||
| 278 | preg_match_all('#/([A-Za-z0-9\.]*)[\s][0-9]+[\s]RD[\s](.*)[\s]ND#sU', $eplain, $matches, PREG_SET_ORDER); |
||
| 279 | if (!empty($enc) and isset(TCPDF_FONT_DATA::$encmap[$enc])) { |
||
| 280 | $enc_map = TCPDF_FONT_DATA::$encmap[$enc]; |
||
| 281 | } else { |
||
| 282 | $enc_map = false; |
||
| 283 | } |
||
| 284 | $fmetric['cw'] = ''; |
||
| 285 | $fmetric['MaxWidth'] = 0; |
||
| 286 | $cwidths = array(); |
||
| 287 | foreach ($matches as $k => $v) { |
||
| 288 | $cid = 0; |
||
| 289 | if (isset($imap[$v[1]])) { |
||
| 290 | $cid = $imap[$v[1]]; |
||
| 291 | } elseif ($enc_map !== false) { |
||
| 292 | $cid = array_search($v[1], $enc_map); |
||
| 293 | if ($cid === false) { |
||
| 294 | $cid = 0; |
||
| 295 | } elseif ($cid > 1000) { |
||
| 296 | $cid -= 1000; |
||
| 297 | } |
||
| 298 | } |
||
| 299 | // decrypt charstring encrypted part |
||
| 300 | $r = 4330; // charstring encryption constant |
||
| 301 | $c1 = 52845; |
||
| 302 | $c2 = 22719; |
||
| 303 | $cd = $v[2]; |
||
| 304 | $clen = strlen($cd); |
||
| 305 | $ccom = array(); |
||
| 306 | for ($i = 0; $i < $clen; ++$i) { |
||
| 307 | $chr = ord($cd[$i]); |
||
| 308 | $ccom[] = ($chr ^ ($r >> 8)); |
||
| 309 | $r = ((($chr + $r) * $c1 + $c2) % 65536); |
||
| 310 | } |
||
| 311 | // decode numbers |
||
| 312 | $cdec = array(); |
||
| 313 | $ck = 0; |
||
| 314 | $i = $lenIV; |
||
| 315 | while ($i < $clen) { |
||
| 316 | if ($ccom[$i] < 32) { |
||
| 317 | $cdec[$ck] = $ccom[$i]; |
||
| 318 | if (($ck > 0) and ($cdec[$ck] == 13)) { |
||
| 319 | // hsbw command: update width |
||
| 320 | $cwidths[$cid] = $cdec[($ck - 1)]; |
||
| 321 | } |
||
| 322 | ++$i; |
||
| 323 | } elseif (($ccom[$i] >= 32) and ($ccom[$i] <= 246)) { |
||
| 324 | $cdec[$ck] = ($ccom[$i] - 139); |
||
| 325 | ++$i; |
||
| 326 | } elseif (($ccom[$i] >= 247) and ($ccom[$i] <= 250)) { |
||
| 327 | $cdec[$ck] = ((($ccom[$i] - 247) * 256) + $ccom[($i + 1)] + 108); |
||
| 328 | $i += 2; |
||
| 329 | } elseif (($ccom[$i] >= 251) and ($ccom[$i] <= 254)) { |
||
| 330 | $cdec[$ck] = ((-($ccom[$i] - 251) * 256) - $ccom[($i + 1)] - 108); |
||
| 331 | $i += 2; |
||
| 332 | } elseif ($ccom[$i] == 255) { |
||
| 333 | $sval = chr($ccom[($i + 1)]) . chr($ccom[($i + 2)]) . chr($ccom[($i + 3)]) . chr($ccom[($i + 4)]); |
||
| 334 | $vsval = unpack('li', $sval); |
||
| 335 | $cdec[$ck] = $vsval['i']; |
||
| 336 | $i += 5; |
||
| 337 | } |
||
| 338 | ++$ck; |
||
| 339 | } |
||
| 340 | } // end for each matches |
||
| 341 | $fmetric['MissingWidth'] = $cwidths[0]; |
||
| 342 | $fmetric['MaxWidth'] = $fmetric['MissingWidth']; |
||
| 343 | $fmetric['AvgWidth'] = 0; |
||
| 344 | // set chars widths |
||
| 345 | for ($cid = 0; $cid <= 255; ++$cid) { |
||
| 346 | if (isset($cwidths[$cid])) { |
||
| 347 | if ($cwidths[$cid] > $fmetric['MaxWidth']) { |
||
| 348 | $fmetric['MaxWidth'] = $cwidths[$cid]; |
||
| 349 | } |
||
| 350 | $fmetric['AvgWidth'] += $cwidths[$cid]; |
||
| 351 | $fmetric['cw'] .= ',' . $cid . '=>' . $cwidths[$cid]; |
||
| 352 | } else { |
||
| 353 | $fmetric['cw'] .= ',' . $cid . '=>' . $fmetric['MissingWidth']; |
||
| 354 | } |
||
| 355 | } |
||
| 356 | $fmetric['AvgWidth'] = round($fmetric['AvgWidth'] / count($cwidths)); |
||
| 357 | } else { |
||
| 358 | // ---------- TRUE TYPE ---------- |
||
| 359 | $offset = 0; // offset position of the font data |
||
| 360 | if (TCPDF_STATIC::_getULONG($font, $offset) != 0x10000) { |
||
| 361 | // sfnt version must be 0x00010000 for TrueType version 1.0. |
||
| 362 | return false; |
||
| 363 | } |
||
| 364 | if ($fmetric['type'] != 'cidfont0') { |
||
| 365 | if ($link) { |
||
| 366 | // creates a symbolic link to the existing font |
||
| 367 | symlink($fontfile, $outpath . $fmetric['file']); |
||
| 368 | } else { |
||
| 369 | // store compressed font |
||
| 370 | $fmetric['file'] .= '.z'; |
||
| 371 | $fp = TCPDF_STATIC::fopenLocal($outpath . $fmetric['file'], 'wb'); |
||
| 372 | fwrite($fp, gzcompress($font)); |
||
| 373 | fclose($fp); |
||
| 374 | } |
||
| 375 | } |
||
| 376 | $offset += 4; |
||
| 377 | // get number of tables |
||
| 378 | $numTables = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 379 | $offset += 2; |
||
| 380 | // skip searchRange, entrySelector and rangeShift |
||
| 381 | $offset += 6; |
||
| 382 | // tables array |
||
| 383 | $table = array(); |
||
| 384 | // ---------- get tables ---------- |
||
| 385 | for ($i = 0; $i < $numTables; ++$i) { |
||
| 386 | // get table info |
||
| 387 | $tag = substr($font, $offset, 4); |
||
| 388 | $offset += 4; |
||
| 389 | $table[$tag] = array(); |
||
| 390 | $table[$tag]['checkSum'] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 391 | $offset += 4; |
||
| 392 | $table[$tag]['offset'] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 393 | $offset += 4; |
||
| 394 | $table[$tag]['length'] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 395 | $offset += 4; |
||
| 396 | } |
||
| 397 | // check magicNumber |
||
| 398 | $offset = $table['head']['offset'] + 12; |
||
| 399 | if (TCPDF_STATIC::_getULONG($font, $offset) != 0x5F0F3CF5) { |
||
| 400 | // magicNumber must be 0x5F0F3CF5 |
||
| 401 | return false; |
||
| 402 | } |
||
| 403 | $offset += 4; |
||
| 404 | $offset += 2; // skip flags |
||
| 405 | // get FUnits |
||
| 406 | $fmetric['unitsPerEm'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 407 | $offset += 2; |
||
| 408 | // units ratio constant |
||
| 409 | $urk = (1000 / $fmetric['unitsPerEm']); |
||
| 410 | $offset += 16; // skip created, modified |
||
| 411 | $xMin = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 412 | $offset += 2; |
||
| 413 | $yMin = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 414 | $offset += 2; |
||
| 415 | $xMax = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 416 | $offset += 2; |
||
| 417 | $yMax = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 418 | $offset += 2; |
||
| 419 | $fmetric['bbox'] = '' . $xMin . ' ' . $yMin . ' ' . $xMax . ' ' . $yMax . ''; |
||
| 420 | $macStyle = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 421 | $offset += 2; |
||
| 422 | // PDF font flags |
||
| 423 | $fmetric['Flags'] = $flags; |
||
| 424 | if (($macStyle & 2) == 2) { |
||
| 425 | // italic flag |
||
| 426 | $fmetric['Flags'] |= 64; |
||
| 427 | } |
||
| 428 | // get offset mode (indexToLocFormat : 0 = short, 1 = long) |
||
| 429 | $offset = $table['head']['offset'] + 50; |
||
| 430 | $short_offset = (TCPDF_STATIC::_getSHORT($font, $offset) == 0); |
||
| 431 | $offset += 2; |
||
| 432 | // get the offsets to the locations of the glyphs in the font, relative to the beginning of the glyphData table |
||
| 433 | $indexToLoc = array(); |
||
| 434 | $offset = $table['loca']['offset']; |
||
| 435 | if ($short_offset) { |
||
| 436 | // short version |
||
| 437 | $tot_num_glyphs = floor($table['loca']['length'] / 2); // numGlyphs + 1 |
||
| 438 | for ($i = 0; $i < $tot_num_glyphs; ++$i) { |
||
| 439 | $indexToLoc[$i] = TCPDF_STATIC::_getUSHORT($font, $offset) * 2; |
||
| 440 | if (isset($indexToLoc[($i - 1)]) && ($indexToLoc[$i] == $indexToLoc[($i - 1)])) { |
||
| 441 | // the last glyph didn't have an outline |
||
| 442 | unset($indexToLoc[($i - 1)]); |
||
| 443 | } |
||
| 444 | $offset += 2; |
||
| 445 | } |
||
| 446 | } else { |
||
| 447 | // long version |
||
| 448 | $tot_num_glyphs = floor($table['loca']['length'] / 4); // numGlyphs + 1 |
||
| 449 | for ($i = 0; $i < $tot_num_glyphs; ++$i) { |
||
| 450 | $indexToLoc[$i] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 451 | if (isset($indexToLoc[($i - 1)]) && ($indexToLoc[$i] == $indexToLoc[($i - 1)])) { |
||
| 452 | // the last glyph didn't have an outline |
||
| 453 | unset($indexToLoc[($i - 1)]); |
||
| 454 | } |
||
| 455 | $offset += 4; |
||
| 456 | } |
||
| 457 | } |
||
| 458 | // get glyphs indexes of chars from cmap table |
||
| 459 | $offset = $table['cmap']['offset'] + 2; |
||
| 460 | $numEncodingTables = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 461 | $offset += 2; |
||
| 462 | $encodingTables = array(); |
||
| 463 | for ($i = 0; $i < $numEncodingTables; ++$i) { |
||
| 464 | $encodingTables[$i]['platformID'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 465 | $offset += 2; |
||
| 466 | $encodingTables[$i]['encodingID'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 467 | $offset += 2; |
||
| 468 | $encodingTables[$i]['offset'] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 469 | $offset += 4; |
||
| 470 | } |
||
| 471 | // ---------- get os/2 metrics ---------- |
||
| 472 | $offset = $table['OS/2']['offset']; |
||
| 473 | $offset += 2; // skip version |
||
| 474 | // xAvgCharWidth |
||
| 475 | $fmetric['AvgWidth'] = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 476 | $offset += 2; |
||
| 477 | // usWeightClass |
||
| 478 | $usWeightClass = round(TCPDF_STATIC::_getUFWORD($font, $offset) * $urk); |
||
| 479 | // estimate StemV and StemH (400 = usWeightClass for Normal - Regular font) |
||
| 480 | $fmetric['StemV'] = round((70 * $usWeightClass) / 400); |
||
| 481 | $fmetric['StemH'] = round((30 * $usWeightClass) / 400); |
||
| 482 | $offset += 2; |
||
| 483 | $offset += 2; // usWidthClass |
||
| 484 | $fsType = TCPDF_STATIC::_getSHORT($font, $offset); |
||
| 485 | $offset += 2; |
||
| 486 | if ($fsType == 2) { |
||
| 487 | // This Font cannot be modified, embedded or exchanged in any manner without first obtaining permission of the legal owner. |
||
| 488 | return false; |
||
| 489 | } |
||
| 490 | // ---------- get font name ---------- |
||
| 491 | $fmetric['name'] = ''; |
||
| 492 | $offset = $table['name']['offset']; |
||
| 493 | $offset += 2; // skip Format selector (=0). |
||
| 494 | // Number of NameRecords that follow n. |
||
| 495 | $numNameRecords = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 496 | $offset += 2; |
||
| 497 | // Offset to start of string storage (from start of table). |
||
| 498 | $stringStorageOffset = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 499 | $offset += 2; |
||
| 500 | for ($i = 0; $i < $numNameRecords; ++$i) { |
||
| 501 | $offset += 6; // skip Platform ID, Platform-specific encoding ID, Language ID. |
||
| 502 | // Name ID. |
||
| 503 | $nameID = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 504 | $offset += 2; |
||
| 505 | if ($nameID == 6) { |
||
| 506 | // String length (in bytes). |
||
| 507 | $stringLength = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 508 | $offset += 2; |
||
| 509 | // String offset from start of storage area (in bytes). |
||
| 510 | $stringOffset = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 511 | $offset += 2; |
||
| 512 | $offset = ($table['name']['offset'] + $stringStorageOffset + $stringOffset); |
||
| 513 | $fmetric['name'] = substr($font, $offset, $stringLength); |
||
| 514 | $fmetric['name'] = preg_replace('/[^a-zA-Z0-9_\-]/', '', $fmetric['name']); |
||
| 515 | break; |
||
| 516 | } else { |
||
| 517 | $offset += 4; // skip String length, String offset |
||
| 518 | } |
||
| 519 | } |
||
| 520 | if (empty($fmetric['name'])) { |
||
| 521 | $fmetric['name'] = $font_name; |
||
| 522 | } |
||
| 523 | // ---------- get post data ---------- |
||
| 524 | $offset = $table['post']['offset']; |
||
| 525 | $offset += 4; // skip Format Type |
||
| 526 | $fmetric['italicAngle'] = TCPDF_STATIC::_getFIXED($font, $offset); |
||
| 527 | $offset += 4; |
||
| 528 | $fmetric['underlinePosition'] = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 529 | $offset += 2; |
||
| 530 | $fmetric['underlineThickness'] = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 531 | $offset += 2; |
||
| 532 | $isFixedPitch = (TCPDF_STATIC::_getULONG($font, $offset) == 0) ? false : true; |
||
| 533 | $offset += 2; |
||
| 534 | if ($isFixedPitch) { |
||
| 535 | $fmetric['Flags'] |= 1; |
||
| 536 | } |
||
| 537 | // ---------- get hhea data ---------- |
||
| 538 | $offset = $table['hhea']['offset']; |
||
| 539 | $offset += 4; // skip Table version number |
||
| 540 | // Ascender |
||
| 541 | $fmetric['Ascent'] = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 542 | $offset += 2; |
||
| 543 | // Descender |
||
| 544 | $fmetric['Descent'] = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 545 | $offset += 2; |
||
| 546 | // LineGap |
||
| 547 | $fmetric['Leading'] = round(TCPDF_STATIC::_getFWORD($font, $offset) * $urk); |
||
| 548 | $offset += 2; |
||
| 549 | // advanceWidthMax |
||
| 550 | $fmetric['MaxWidth'] = round(TCPDF_STATIC::_getUFWORD($font, $offset) * $urk); |
||
| 551 | $offset += 2; |
||
| 552 | $offset += 22; // skip some values |
||
| 553 | // get the number of hMetric entries in hmtx table |
||
| 554 | $numberOfHMetrics = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 555 | // ---------- get maxp data ---------- |
||
| 556 | $offset = $table['maxp']['offset']; |
||
| 557 | $offset += 4; // skip Table version number |
||
| 558 | // get the the number of glyphs in the font. |
||
| 559 | $numGlyphs = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 560 | // ---------- get CIDToGIDMap ---------- |
||
| 561 | $ctg = array(); |
||
| 562 | $c = 0; |
||
| 563 | foreach ($encodingTables as $enctable) { |
||
| 564 | // get only specified Platform ID and Encoding ID |
||
| 565 | if (($enctable['platformID'] == $platid) and ($enctable['encodingID'] == $encid)) { |
||
| 566 | $offset = $table['cmap']['offset'] + $enctable['offset']; |
||
| 567 | $format = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 568 | $offset += 2; |
||
| 569 | switch ($format) { |
||
| 570 | case 0: { // Format 0: Byte encoding table |
||
| 571 | $offset += 4; // skip length and version/language |
||
| 572 | for ($c = 0; $c < 256; ++$c) { |
||
| 573 | $g = TCPDF_STATIC::_getBYTE($font, $offset); |
||
| 574 | $ctg[$c] = $g; |
||
| 575 | ++$offset; |
||
| 576 | } |
||
| 577 | break; |
||
| 578 | } |
||
| 579 | case 2: { // Format 2: High-byte mapping through table |
||
| 580 | $offset += 4; // skip length and version/language |
||
| 581 | $numSubHeaders = 0; |
||
| 582 | for ($i = 0; $i < 256; ++$i) { |
||
| 583 | // Array that maps high bytes to subHeaders: value is subHeader index * 8. |
||
| 584 | $subHeaderKeys[$i] = (TCPDF_STATIC::_getUSHORT($font, $offset) / 8); |
||
| 585 | $offset += 2; |
||
| 586 | if ($numSubHeaders < $subHeaderKeys[$i]) { |
||
| 587 | $numSubHeaders = $subHeaderKeys[$i]; |
||
| 588 | } |
||
| 589 | } |
||
| 590 | // the number of subHeaders is equal to the max of subHeaderKeys + 1 |
||
| 591 | ++$numSubHeaders; |
||
| 592 | // read subHeader structures |
||
| 593 | $subHeaders = array(); |
||
| 594 | $numGlyphIndexArray = 0; |
||
| 595 | for ($k = 0; $k < $numSubHeaders; ++$k) { |
||
| 596 | $subHeaders[$k]['firstCode'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 597 | $offset += 2; |
||
| 598 | $subHeaders[$k]['entryCount'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 599 | $offset += 2; |
||
| 600 | $subHeaders[$k]['idDelta'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 601 | $offset += 2; |
||
| 602 | $subHeaders[$k]['idRangeOffset'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 603 | $offset += 2; |
||
| 604 | $subHeaders[$k]['idRangeOffset'] -= (2 + (($numSubHeaders - $k - 1) * 8)); |
||
| 605 | $subHeaders[$k]['idRangeOffset'] /= 2; |
||
| 606 | $numGlyphIndexArray += $subHeaders[$k]['entryCount']; |
||
| 607 | } |
||
| 608 | for ($k = 0; $k < $numGlyphIndexArray; ++$k) { |
||
| 609 | $glyphIndexArray[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 610 | $offset += 2; |
||
| 611 | } |
||
| 612 | for ($i = 0; $i < 256; ++$i) { |
||
| 613 | $k = $subHeaderKeys[$i]; |
||
| 614 | if ($k == 0) { |
||
| 615 | // one byte code |
||
| 616 | $c = $i; |
||
| 617 | $g = $glyphIndexArray[0]; |
||
| 618 | $ctg[$c] = $g; |
||
| 619 | } else { |
||
| 620 | // two bytes code |
||
| 621 | $start_byte = $subHeaders[$k]['firstCode']; |
||
| 622 | $end_byte = $start_byte + $subHeaders[$k]['entryCount']; |
||
| 623 | for ($j = $start_byte; $j < $end_byte; ++$j) { |
||
| 624 | // combine high and low bytes |
||
| 625 | $c = (($i << 8) + $j); |
||
| 626 | $idRangeOffset = ($subHeaders[$k]['idRangeOffset'] + $j - $subHeaders[$k]['firstCode']); |
||
| 627 | $g = ($glyphIndexArray[$idRangeOffset] + $subHeaders[$k]['idDelta']) % 65536; |
||
| 628 | if ($g < 0) { |
||
| 629 | $g = 0; |
||
| 630 | } |
||
| 631 | $ctg[$c] = $g; |
||
| 632 | } |
||
| 633 | } |
||
| 634 | } |
||
| 635 | break; |
||
| 636 | } |
||
| 637 | case 4: { // Format 4: Segment mapping to delta values |
||
| 638 | $length = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 639 | $offset += 2; |
||
| 640 | $offset += 2; // skip version/language |
||
| 641 | $segCount = floor(TCPDF_STATIC::_getUSHORT($font, $offset) / 2); |
||
| 642 | $offset += 2; |
||
| 643 | $offset += 6; // skip searchRange, entrySelector, rangeShift |
||
| 644 | $endCount = array(); // array of end character codes for each segment |
||
| 645 | for ($k = 0; $k < $segCount; ++$k) { |
||
| 646 | $endCount[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 647 | $offset += 2; |
||
| 648 | } |
||
| 649 | $offset += 2; // skip reservedPad |
||
| 650 | $startCount = array(); // array of start character codes for each segment |
||
| 651 | for ($k = 0; $k < $segCount; ++$k) { |
||
| 652 | $startCount[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 653 | $offset += 2; |
||
| 654 | } |
||
| 655 | $idDelta = array(); // delta for all character codes in segment |
||
| 656 | for ($k = 0; $k < $segCount; ++$k) { |
||
| 657 | $idDelta[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 658 | $offset += 2; |
||
| 659 | } |
||
| 660 | $idRangeOffset = array(); // Offsets into glyphIdArray or 0 |
||
| 661 | for ($k = 0; $k < $segCount; ++$k) { |
||
| 662 | $idRangeOffset[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 663 | $offset += 2; |
||
| 664 | } |
||
| 665 | $gidlen = (floor($length / 2) - 8 - (4 * $segCount)); |
||
| 666 | $glyphIdArray = array(); // glyph index array |
||
| 667 | for ($k = 0; $k < $gidlen; ++$k) { |
||
| 668 | $glyphIdArray[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 669 | $offset += 2; |
||
| 670 | } |
||
| 671 | for ($k = 0; $k < $segCount - 1; ++$k) { |
||
| 672 | for ($c = $startCount[$k]; $c <= $endCount[$k]; ++$c) { |
||
| 673 | if ($idRangeOffset[$k] == 0) { |
||
| 674 | $g = ($idDelta[$k] + $c) % 65536; |
||
| 675 | } else { |
||
| 676 | $gid = (floor($idRangeOffset[$k] / 2) + ($c - $startCount[$k]) - ($segCount - $k)); |
||
| 677 | $g = ($glyphIdArray[$gid] + $idDelta[$k]) % 65536; |
||
| 678 | } |
||
| 679 | if ($g < 0) { |
||
| 680 | $g = 0; |
||
| 681 | } |
||
| 682 | $ctg[$c] = $g; |
||
| 683 | } |
||
| 684 | } |
||
| 685 | break; |
||
| 686 | } |
||
| 687 | case 6: { // Format 6: Trimmed table mapping |
||
| 688 | $offset += 4; // skip length and version/language |
||
| 689 | $firstCode = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 690 | $offset += 2; |
||
| 691 | $entryCount = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 692 | $offset += 2; |
||
| 693 | for ($k = 0; $k < $entryCount; ++$k) { |
||
| 694 | $c = ($k + $firstCode); |
||
| 695 | $g = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 696 | $offset += 2; |
||
| 697 | $ctg[$c] = $g; |
||
| 698 | } |
||
| 699 | break; |
||
| 700 | } |
||
| 701 | case 8: { // Format 8: Mixed 16-bit and 32-bit coverage |
||
| 702 | $offset += 10; // skip reserved, length and version/language |
||
| 703 | for ($k = 0; $k < 8192; ++$k) { |
||
| 704 | $is32[$k] = TCPDF_STATIC::_getBYTE($font, $offset); |
||
| 705 | ++$offset; |
||
| 706 | } |
||
| 707 | $nGroups = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 708 | $offset += 4; |
||
| 709 | for ($i = 0; $i < $nGroups; ++$i) { |
||
| 710 | $startCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 711 | $offset += 4; |
||
| 712 | $endCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 713 | $offset += 4; |
||
| 714 | $startGlyphID = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 715 | $offset += 4; |
||
| 716 | for ($k = $startCharCode; $k <= $endCharCode; ++$k) { |
||
| 717 | $is32idx = floor($c / 8); |
||
| 718 | if ((isset($is32[$is32idx])) and (($is32[$is32idx] & (1 << (7 - ($c % 8)))) == 0)) { |
||
| 719 | $c = $k; |
||
| 720 | } else { |
||
| 721 | // 32 bit format |
||
| 722 | // convert to decimal (http://www.unicode.org/faq//utf_bom.html#utf16-4) |
||
| 723 | //LEAD_OFFSET = (0xD800 - (0x10000 >> 10)) = 55232 |
||
| 724 | //SURROGATE_OFFSET = (0x10000 - (0xD800 << 10) - 0xDC00) = -56613888 |
||
| 725 | $c = ((55232 + ($k >> 10)) << 10) + (0xDC00 + ($k & 0x3FF)) - 56613888; |
||
| 726 | } |
||
| 727 | $ctg[$c] = 0; |
||
| 728 | ++$startGlyphID; |
||
| 729 | } |
||
| 730 | } |
||
| 731 | break; |
||
| 732 | } |
||
| 733 | case 10: { // Format 10: Trimmed array |
||
| 734 | $offset += 10; // skip reserved, length and version/language |
||
| 735 | $startCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 736 | $offset += 4; |
||
| 737 | $numChars = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 738 | $offset += 4; |
||
| 739 | for ($k = 0; $k < $numChars; ++$k) { |
||
| 740 | $c = ($k + $startCharCode); |
||
| 741 | $g = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 742 | $ctg[$c] = $g; |
||
| 743 | $offset += 2; |
||
| 744 | } |
||
| 745 | break; |
||
| 746 | } |
||
| 747 | case 12: { // Format 12: Segmented coverage |
||
| 748 | $offset += 10; // skip length and version/language |
||
| 749 | $nGroups = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 750 | $offset += 4; |
||
| 751 | for ($k = 0; $k < $nGroups; ++$k) { |
||
| 752 | $startCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 753 | $offset += 4; |
||
| 754 | $endCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 755 | $offset += 4; |
||
| 756 | $startGlyphCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 757 | $offset += 4; |
||
| 758 | for ($c = $startCharCode; $c <= $endCharCode; ++$c) { |
||
| 759 | $ctg[$c] = $startGlyphCode; |
||
| 760 | ++$startGlyphCode; |
||
| 761 | } |
||
| 762 | } |
||
| 763 | break; |
||
| 764 | } |
||
| 765 | case 13: { // Format 13: Many-to-one range mappings |
||
| 766 | // to be implemented ... |
||
| 767 | break; |
||
| 768 | } |
||
| 769 | case 14: { // Format 14: Unicode Variation Sequences |
||
| 770 | // to be implemented ... |
||
| 771 | break; |
||
| 772 | } |
||
| 773 | } |
||
| 774 | } |
||
| 775 | } |
||
| 776 | if (!isset($ctg[0])) { |
||
| 777 | $ctg[0] = 0; |
||
| 778 | } |
||
| 779 | // get xHeight (height of x) |
||
| 780 | $offset = ($table['glyf']['offset'] + $indexToLoc[$ctg[120]] + 4); |
||
| 781 | $yMin = TCPDF_STATIC::_getFWORD($font, $offset); |
||
| 782 | $offset += 4; |
||
| 783 | $yMax = TCPDF_STATIC::_getFWORD($font, $offset); |
||
| 784 | $offset += 2; |
||
| 785 | $fmetric['XHeight'] = round(($yMax - $yMin) * $urk); |
||
| 786 | // get CapHeight (height of H) |
||
| 787 | $offset = ($table['glyf']['offset'] + $indexToLoc[$ctg[72]] + 4); |
||
| 788 | $yMin = TCPDF_STATIC::_getFWORD($font, $offset); |
||
| 789 | $offset += 4; |
||
| 790 | $yMax = TCPDF_STATIC::_getFWORD($font, $offset); |
||
| 791 | $offset += 2; |
||
| 792 | $fmetric['CapHeight'] = round(($yMax - $yMin) * $urk); |
||
| 793 | // ceate widths array |
||
| 794 | $cw = array(); |
||
| 795 | $offset = $table['hmtx']['offset']; |
||
| 796 | for ($i = 0 ; $i < $numberOfHMetrics; ++$i) { |
||
| 797 | $cw[$i] = round(TCPDF_STATIC::_getUFWORD($font, $offset) * $urk); |
||
| 798 | $offset += 4; // skip lsb |
||
| 799 | } |
||
| 800 | if ($numberOfHMetrics < $numGlyphs) { |
||
| 801 | // fill missing widths with the last value |
||
| 802 | $cw = array_pad($cw, $numGlyphs, $cw[($numberOfHMetrics - 1)]); |
||
| 803 | } |
||
| 804 | $fmetric['MissingWidth'] = $cw[0]; |
||
| 805 | $fmetric['cw'] = ''; |
||
| 806 | $fmetric['cbbox'] = ''; |
||
| 807 | for ($cid = 0; $cid <= 65535; ++$cid) { |
||
| 808 | if (isset($ctg[$cid])) { |
||
| 809 | if (isset($cw[$ctg[$cid]])) { |
||
| 810 | $fmetric['cw'] .= ',' . $cid . '=>' . $cw[$ctg[$cid]]; |
||
| 811 | } |
||
| 812 | if ($addcbbox and isset($indexToLoc[$ctg[$cid]])) { |
||
| 813 | $offset = ($table['glyf']['offset'] + $indexToLoc[$ctg[$cid]]); |
||
| 814 | $xMin = round(TCPDF_STATIC::_getFWORD($font, $offset + 2) * $urk); |
||
| 815 | $yMin = round(TCPDF_STATIC::_getFWORD($font, $offset + 4) * $urk); |
||
| 816 | $xMax = round(TCPDF_STATIC::_getFWORD($font, $offset + 6) * $urk); |
||
| 817 | $yMax = round(TCPDF_STATIC::_getFWORD($font, $offset + 8) * $urk); |
||
| 818 | $fmetric['cbbox'] .= ',' . $cid . '=>array(' . $xMin . ',' . $yMin . ',' . $xMax . ',' . $yMax . ')'; |
||
| 819 | } |
||
| 820 | } |
||
| 821 | } |
||
| 822 | } // end of true type |
||
| 823 | if (($fmetric['type'] == 'TrueTypeUnicode') and (count($ctg) == 256)) { |
||
| 824 | $fmetric['type'] = 'TrueType'; |
||
| 825 | } |
||
| 826 | // ---------- create php font file ---------- |
||
| 827 | $pfile = '<' . '?' . 'php' . "\n"; |
||
| 828 | $pfile .= '// TCPDF FONT FILE DESCRIPTION' . "\n"; |
||
| 829 | $pfile .= '$type=\'' . $fmetric['type'] . '\';' . "\n"; |
||
| 830 | $pfile .= '$name=\'' . $fmetric['name'] . '\';' . "\n"; |
||
| 831 | $pfile .= '$up=' . $fmetric['underlinePosition'] . ';' . "\n"; |
||
| 832 | $pfile .= '$ut=' . $fmetric['underlineThickness'] . ';' . "\n"; |
||
| 833 | if ($fmetric['MissingWidth'] > 0) { |
||
| 834 | $pfile .= '$dw=' . $fmetric['MissingWidth'] . ';' . "\n"; |
||
| 835 | } else { |
||
| 836 | $pfile .= '$dw=' . $fmetric['AvgWidth'] . ';' . "\n"; |
||
| 837 | } |
||
| 838 | $pfile .= '$diff=\'' . $fmetric['diff'] . '\';' . "\n"; |
||
| 839 | if ($fmetric['type'] == 'Type1') { |
||
| 840 | // Type 1 |
||
| 841 | $pfile .= '$enc=\'' . $fmetric['enc'] . '\';' . "\n"; |
||
| 842 | $pfile .= '$file=\'' . $fmetric['file'] . '\';' . "\n"; |
||
| 843 | $pfile .= '$size1=' . $fmetric['size1'] . ';' . "\n"; |
||
| 844 | $pfile .= '$size2=' . $fmetric['size2'] . ';' . "\n"; |
||
| 845 | } else { |
||
| 846 | $pfile .= '$originalsize=' . $fmetric['originalsize'] . ';' . "\n"; |
||
| 847 | if ($fmetric['type'] == 'cidfont0') { |
||
| 848 | // CID-0 |
||
| 849 | switch ($fonttype) { |
||
| 850 | case 'CID0JP': { |
||
| 851 | $pfile .= '// Japanese' . "\n"; |
||
| 852 | $pfile .= '$enc=\'UniJIS-UTF16-H\';' . "\n"; |
||
| 853 | $pfile .= '$cidinfo=array(\'Registry\'=>\'Adobe\', \'Ordering\'=>\'Japan1\',\'Supplement\'=>5);' . "\n"; |
||
| 854 | $pfile .= 'include(dirname(__FILE__).\'/uni2cid_aj16.php\');' . "\n"; |
||
| 855 | break; |
||
| 856 | } |
||
| 857 | case 'CID0KR': { |
||
| 858 | $pfile .= '// Korean' . "\n"; |
||
| 859 | $pfile .= '$enc=\'UniKS-UTF16-H\';' . "\n"; |
||
| 860 | $pfile .= '$cidinfo=array(\'Registry\'=>\'Adobe\', \'Ordering\'=>\'Korea1\',\'Supplement\'=>0);' . "\n"; |
||
| 861 | $pfile .= 'include(dirname(__FILE__).\'/uni2cid_ak12.php\');' . "\n"; |
||
| 862 | break; |
||
| 863 | } |
||
| 864 | case 'CID0CS': { |
||
| 865 | $pfile .= '// Chinese Simplified' . "\n"; |
||
| 866 | $pfile .= '$enc=\'UniGB-UTF16-H\';' . "\n"; |
||
| 867 | $pfile .= '$cidinfo=array(\'Registry\'=>\'Adobe\', \'Ordering\'=>\'GB1\',\'Supplement\'=>2);' . "\n"; |
||
| 868 | $pfile .= 'include(dirname(__FILE__).\'/uni2cid_ag15.php\');' . "\n"; |
||
| 869 | break; |
||
| 870 | } |
||
| 871 | case 'CID0CT': |
||
| 872 | default: { |
||
| 873 | $pfile .= '// Chinese Traditional' . "\n"; |
||
| 874 | $pfile .= '$enc=\'UniCNS-UTF16-H\';' . "\n"; |
||
| 875 | $pfile .= '$cidinfo=array(\'Registry\'=>\'Adobe\', \'Ordering\'=>\'CNS1\',\'Supplement\'=>0);' . "\n"; |
||
| 876 | $pfile .= 'include(dirname(__FILE__).\'/uni2cid_aj16.php\');' . "\n"; |
||
| 877 | break; |
||
| 878 | } |
||
| 879 | } |
||
| 880 | } else { |
||
| 881 | // TrueType |
||
| 882 | $pfile .= '$enc=\'' . $fmetric['enc'] . '\';' . "\n"; |
||
| 883 | $pfile .= '$file=\'' . $fmetric['file'] . '\';' . "\n"; |
||
| 884 | $pfile .= '$ctg=\'' . $fmetric['ctg'] . '\';' . "\n"; |
||
| 885 | // create CIDToGIDMap |
||
| 886 | $cidtogidmap = str_pad('', 131072, "\x00"); // (256 * 256 * 2) = 131072 |
||
| 887 | foreach ($ctg as $cid => $gid) { |
||
| 888 | $cidtogidmap = self::updateCIDtoGIDmap($cidtogidmap, $cid, $ctg[$cid]); |
||
| 889 | } |
||
| 890 | // store compressed CIDToGIDMap |
||
| 891 | $fp = TCPDF_STATIC::fopenLocal($outpath . $fmetric['ctg'], 'wb'); |
||
| 892 | fwrite($fp, gzcompress($cidtogidmap)); |
||
| 893 | fclose($fp); |
||
| 894 | } |
||
| 895 | } |
||
| 896 | $pfile .= '$desc=array('; |
||
| 897 | $pfile .= '\'Flags\'=>' . $fmetric['Flags'] . ','; |
||
| 898 | $pfile .= '\'FontBBox\'=>\'[' . $fmetric['bbox'] . ']\','; |
||
| 899 | $pfile .= '\'ItalicAngle\'=>' . $fmetric['italicAngle'] . ','; |
||
| 900 | $pfile .= '\'Ascent\'=>' . $fmetric['Ascent'] . ','; |
||
| 901 | $pfile .= '\'Descent\'=>' . $fmetric['Descent'] . ','; |
||
| 902 | $pfile .= '\'Leading\'=>' . $fmetric['Leading'] . ','; |
||
| 903 | $pfile .= '\'CapHeight\'=>' . $fmetric['CapHeight'] . ','; |
||
| 904 | $pfile .= '\'XHeight\'=>' . $fmetric['XHeight'] . ','; |
||
| 905 | $pfile .= '\'StemV\'=>' . $fmetric['StemV'] . ','; |
||
| 906 | $pfile .= '\'StemH\'=>' . $fmetric['StemH'] . ','; |
||
| 907 | $pfile .= '\'AvgWidth\'=>' . $fmetric['AvgWidth'] . ','; |
||
| 908 | $pfile .= '\'MaxWidth\'=>' . $fmetric['MaxWidth'] . ','; |
||
| 909 | $pfile .= '\'MissingWidth\'=>' . $fmetric['MissingWidth'] . ''; |
||
| 910 | $pfile .= ');' . "\n"; |
||
| 911 | if (!empty($fmetric['cbbox'])) { |
||
| 912 | $pfile .= '$cbbox=array(' . substr($fmetric['cbbox'], 1) . ');' . "\n"; |
||
| 913 | } |
||
| 914 | $pfile .= '$cw=array(' . substr($fmetric['cw'], 1) . ');' . "\n"; |
||
| 915 | $pfile .= '// --- EOF ---' . "\n"; |
||
| 916 | // store file |
||
| 917 | $fp = TCPDF_STATIC::fopenLocal($outpath . $font_name . '.php', 'w'); |
||
| 918 | fwrite($fp, $pfile); |
||
| 919 | fclose($fp); |
||
| 920 | // return TCPDF font name |
||
| 921 | return $font_name; |
||
| 922 | } |
||
| 923 | |||
| 924 | /** |
||
| 925 | * Returs the checksum of a TTF table. |
||
| 926 | * @param string $table table to check |
||
| 927 | * @param int $length length of table in bytes |
||
| 928 | * @return int checksum |
||
| 929 | * @author Nicola Asuni |
||
| 930 | * @since 5.2.000 (2010-06-02) |
||
| 931 | * @public static |
||
| 932 | */ |
||
| 933 | public static function _getTTFtableChecksum($table, $length) |
||
| 934 | { |
||
| 935 | $sum = 0; |
||
| 936 | $tlen = ($length / 4); |
||
| 937 | $offset = 0; |
||
| 938 | for ($i = 0; $i < $tlen; ++$i) { |
||
| 939 | $v = unpack('Ni', substr($table, $offset, 4)); |
||
| 940 | $sum += $v['i']; |
||
| 941 | $offset += 4; |
||
| 942 | } |
||
| 943 | $sum = unpack('Ni', pack('N', $sum)); |
||
| 944 | return $sum['i']; |
||
| 945 | } |
||
| 946 | |||
| 947 | /** |
||
| 948 | * Returns a subset of the TrueType font data without the unused glyphs. |
||
| 949 | * @param string $font TrueType font data. |
||
| 950 | * @param array $subsetchars Array of used characters (the glyphs to keep). |
||
| 951 | * @return string A subset of TrueType font data without the unused glyphs. |
||
| 952 | * @author Nicola Asuni |
||
| 953 | * @since 5.2.000 (2010-06-02) |
||
| 954 | * @public static |
||
| 955 | */ |
||
| 956 | public static function _getTrueTypeFontSubset($font, $subsetchars) |
||
| 957 | { |
||
| 958 | ksort($subsetchars); |
||
| 959 | $offset = 0; // offset position of the font data |
||
| 960 | if (TCPDF_STATIC::_getULONG($font, $offset) != 0x10000) { |
||
| 961 | // sfnt version must be 0x00010000 for TrueType version 1.0. |
||
| 962 | return $font; |
||
| 963 | } |
||
| 964 | $c = 0; |
||
| 965 | $offset += 4; |
||
| 966 | // get number of tables |
||
| 967 | $numTables = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 968 | $offset += 2; |
||
| 969 | // skip searchRange, entrySelector and rangeShift |
||
| 970 | $offset += 6; |
||
| 971 | // tables array |
||
| 972 | $table = array(); |
||
| 973 | // for each table |
||
| 974 | for ($i = 0; $i < $numTables; ++$i) { |
||
| 975 | // get table info |
||
| 976 | $tag = substr($font, $offset, 4); |
||
| 977 | $offset += 4; |
||
| 978 | $table[$tag] = array(); |
||
| 979 | $table[$tag]['checkSum'] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 980 | $offset += 4; |
||
| 981 | $table[$tag]['offset'] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 982 | $offset += 4; |
||
| 983 | $table[$tag]['length'] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 984 | $offset += 4; |
||
| 985 | } |
||
| 986 | // check magicNumber |
||
| 987 | $offset = $table['head']['offset'] + 12; |
||
| 988 | if (TCPDF_STATIC::_getULONG($font, $offset) != 0x5F0F3CF5) { |
||
| 989 | // magicNumber must be 0x5F0F3CF5 |
||
| 990 | return $font; |
||
| 991 | } |
||
| 992 | $offset += 4; |
||
| 993 | // get offset mode (indexToLocFormat : 0 = short, 1 = long) |
||
| 994 | $offset = $table['head']['offset'] + 50; |
||
| 995 | $short_offset = (TCPDF_STATIC::_getSHORT($font, $offset) == 0); |
||
| 996 | $offset += 2; |
||
| 997 | // get the offsets to the locations of the glyphs in the font, relative to the beginning of the glyphData table |
||
| 998 | $indexToLoc = array(); |
||
| 999 | $offset = $table['loca']['offset']; |
||
| 1000 | if ($short_offset) { |
||
| 1001 | // short version |
||
| 1002 | $tot_num_glyphs = floor($table['loca']['length'] / 2); // numGlyphs + 1 |
||
| 1003 | for ($i = 0; $i < $tot_num_glyphs; ++$i) { |
||
| 1004 | $indexToLoc[$i] = TCPDF_STATIC::_getUSHORT($font, $offset) * 2; |
||
| 1005 | $offset += 2; |
||
| 1006 | } |
||
| 1007 | } else { |
||
| 1008 | // long version |
||
| 1009 | $tot_num_glyphs = ($table['loca']['length'] / 4); // numGlyphs + 1 |
||
| 1010 | for ($i = 0; $i < $tot_num_glyphs; ++$i) { |
||
| 1011 | $indexToLoc[$i] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1012 | $offset += 4; |
||
| 1013 | } |
||
| 1014 | } |
||
| 1015 | // get glyphs indexes of chars from cmap table |
||
| 1016 | $subsetglyphs = array(); // glyph IDs on key |
||
| 1017 | $subsetglyphs[0] = true; // character codes that do not correspond to any glyph in the font should be mapped to glyph index 0 |
||
| 1018 | $offset = $table['cmap']['offset'] + 2; |
||
| 1019 | $numEncodingTables = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1020 | $offset += 2; |
||
| 1021 | $encodingTables = array(); |
||
| 1022 | for ($i = 0; $i < $numEncodingTables; ++$i) { |
||
| 1023 | $encodingTables[$i]['platformID'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1024 | $offset += 2; |
||
| 1025 | $encodingTables[$i]['encodingID'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1026 | $offset += 2; |
||
| 1027 | $encodingTables[$i]['offset'] = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1028 | $offset += 4; |
||
| 1029 | } |
||
| 1030 | foreach ($encodingTables as $enctable) { |
||
| 1031 | // get all platforms and encodings |
||
| 1032 | $offset = $table['cmap']['offset'] + $enctable['offset']; |
||
| 1033 | $format = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1034 | $offset += 2; |
||
| 1035 | switch ($format) { |
||
| 1036 | case 0: { // Format 0: Byte encoding table |
||
| 1037 | $offset += 4; // skip length and version/language |
||
| 1038 | for ($c = 0; $c < 256; ++$c) { |
||
| 1039 | if (isset($subsetchars[$c])) { |
||
| 1040 | $g = TCPDF_STATIC::_getBYTE($font, $offset); |
||
| 1041 | $subsetglyphs[$g] = true; |
||
| 1042 | } |
||
| 1043 | ++$offset; |
||
| 1044 | } |
||
| 1045 | break; |
||
| 1046 | } |
||
| 1047 | case 2: { // Format 2: High-byte mapping through table |
||
| 1048 | $offset += 4; // skip length and version/language |
||
| 1049 | $numSubHeaders = 0; |
||
| 1050 | for ($i = 0; $i < 256; ++$i) { |
||
| 1051 | // Array that maps high bytes to subHeaders: value is subHeader index * 8. |
||
| 1052 | $subHeaderKeys[$i] = (TCPDF_STATIC::_getUSHORT($font, $offset) / 8); |
||
| 1053 | $offset += 2; |
||
| 1054 | if ($numSubHeaders < $subHeaderKeys[$i]) { |
||
| 1055 | $numSubHeaders = $subHeaderKeys[$i]; |
||
| 1056 | } |
||
| 1057 | } |
||
| 1058 | // the number of subHeaders is equal to the max of subHeaderKeys + 1 |
||
| 1059 | ++$numSubHeaders; |
||
| 1060 | // read subHeader structures |
||
| 1061 | $subHeaders = array(); |
||
| 1062 | $numGlyphIndexArray = 0; |
||
| 1063 | for ($k = 0; $k < $numSubHeaders; ++$k) { |
||
| 1064 | $subHeaders[$k]['firstCode'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1065 | $offset += 2; |
||
| 1066 | $subHeaders[$k]['entryCount'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1067 | $offset += 2; |
||
| 1068 | $subHeaders[$k]['idDelta'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1069 | $offset += 2; |
||
| 1070 | $subHeaders[$k]['idRangeOffset'] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1071 | $offset += 2; |
||
| 1072 | $subHeaders[$k]['idRangeOffset'] -= (2 + (($numSubHeaders - $k - 1) * 8)); |
||
| 1073 | $subHeaders[$k]['idRangeOffset'] /= 2; |
||
| 1074 | $numGlyphIndexArray += $subHeaders[$k]['entryCount']; |
||
| 1075 | } |
||
| 1076 | for ($k = 0; $k < $numGlyphIndexArray; ++$k) { |
||
| 1077 | $glyphIndexArray[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1078 | $offset += 2; |
||
| 1079 | } |
||
| 1080 | for ($i = 0; $i < 256; ++$i) { |
||
| 1081 | $k = $subHeaderKeys[$i]; |
||
| 1082 | if ($k == 0) { |
||
| 1083 | // one byte code |
||
| 1084 | $c = $i; |
||
| 1085 | if (isset($subsetchars[$c])) { |
||
| 1086 | $g = $glyphIndexArray[0]; |
||
| 1087 | $subsetglyphs[$g] = true; |
||
| 1088 | } |
||
| 1089 | } else { |
||
| 1090 | // two bytes code |
||
| 1091 | $start_byte = $subHeaders[$k]['firstCode']; |
||
| 1092 | $end_byte = $start_byte + $subHeaders[$k]['entryCount']; |
||
| 1093 | for ($j = $start_byte; $j < $end_byte; ++$j) { |
||
| 1094 | // combine high and low bytes |
||
| 1095 | $c = (($i << 8) + $j); |
||
| 1096 | if (isset($subsetchars[$c])) { |
||
| 1097 | $idRangeOffset = ($subHeaders[$k]['idRangeOffset'] + $j - $subHeaders[$k]['firstCode']); |
||
| 1098 | $g = ($glyphIndexArray[$idRangeOffset] + $subHeaders[$k]['idDelta']) % 65536; |
||
| 1099 | if ($g < 0) { |
||
| 1100 | $g = 0; |
||
| 1101 | } |
||
| 1102 | $subsetglyphs[$g] = true; |
||
| 1103 | } |
||
| 1104 | } |
||
| 1105 | } |
||
| 1106 | } |
||
| 1107 | break; |
||
| 1108 | } |
||
| 1109 | case 4: { // Format 4: Segment mapping to delta values |
||
| 1110 | $length = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1111 | $offset += 2; |
||
| 1112 | $offset += 2; // skip version/language |
||
| 1113 | $segCount = floor(TCPDF_STATIC::_getUSHORT($font, $offset) / 2); |
||
| 1114 | $offset += 2; |
||
| 1115 | $offset += 6; // skip searchRange, entrySelector, rangeShift |
||
| 1116 | $endCount = array(); // array of end character codes for each segment |
||
| 1117 | for ($k = 0; $k < $segCount; ++$k) { |
||
| 1118 | $endCount[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1119 | $offset += 2; |
||
| 1120 | } |
||
| 1121 | $offset += 2; // skip reservedPad |
||
| 1122 | $startCount = array(); // array of start character codes for each segment |
||
| 1123 | for ($k = 0; $k < $segCount; ++$k) { |
||
| 1124 | $startCount[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1125 | $offset += 2; |
||
| 1126 | } |
||
| 1127 | $idDelta = array(); // delta for all character codes in segment |
||
| 1128 | for ($k = 0; $k < $segCount; ++$k) { |
||
| 1129 | $idDelta[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1130 | $offset += 2; |
||
| 1131 | } |
||
| 1132 | $idRangeOffset = array(); // Offsets into glyphIdArray or 0 |
||
| 1133 | for ($k = 0; $k < $segCount; ++$k) { |
||
| 1134 | $idRangeOffset[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1135 | $offset += 2; |
||
| 1136 | } |
||
| 1137 | $gidlen = (floor($length / 2) - 8 - (4 * $segCount)); |
||
| 1138 | $glyphIdArray = array(); // glyph index array |
||
| 1139 | for ($k = 0; $k < $gidlen; ++$k) { |
||
| 1140 | $glyphIdArray[$k] = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1141 | $offset += 2; |
||
| 1142 | } |
||
| 1143 | for ($k = 0; $k < $segCount; ++$k) { |
||
| 1144 | for ($c = $startCount[$k]; $c <= $endCount[$k]; ++$c) { |
||
| 1145 | if (isset($subsetchars[$c])) { |
||
| 1146 | if ($idRangeOffset[$k] == 0) { |
||
| 1147 | $g = ($idDelta[$k] + $c) % 65536; |
||
| 1148 | } else { |
||
| 1149 | $gid = (floor($idRangeOffset[$k] / 2) + ($c - $startCount[$k]) - ($segCount - $k)); |
||
| 1150 | $g = ($glyphIdArray[$gid] + $idDelta[$k]) % 65536; |
||
| 1151 | } |
||
| 1152 | if ($g < 0) { |
||
| 1153 | $g = 0; |
||
| 1154 | } |
||
| 1155 | $subsetglyphs[$g] = true; |
||
| 1156 | } |
||
| 1157 | } |
||
| 1158 | } |
||
| 1159 | break; |
||
| 1160 | } |
||
| 1161 | case 6: { // Format 6: Trimmed table mapping |
||
| 1162 | $offset += 4; // skip length and version/language |
||
| 1163 | $firstCode = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1164 | $offset += 2; |
||
| 1165 | $entryCount = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1166 | $offset += 2; |
||
| 1167 | for ($k = 0; $k < $entryCount; ++$k) { |
||
| 1168 | $c = ($k + $firstCode); |
||
| 1169 | if (isset($subsetchars[$c])) { |
||
| 1170 | $g = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1171 | $subsetglyphs[$g] = true; |
||
| 1172 | } |
||
| 1173 | $offset += 2; |
||
| 1174 | } |
||
| 1175 | break; |
||
| 1176 | } |
||
| 1177 | case 8: { // Format 8: Mixed 16-bit and 32-bit coverage |
||
| 1178 | $offset += 10; // skip reserved, length and version/language |
||
| 1179 | for ($k = 0; $k < 8192; ++$k) { |
||
| 1180 | $is32[$k] = TCPDF_STATIC::_getBYTE($font, $offset); |
||
| 1181 | ++$offset; |
||
| 1182 | } |
||
| 1183 | $nGroups = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1184 | $offset += 4; |
||
| 1185 | for ($i = 0; $i < $nGroups; ++$i) { |
||
| 1186 | $startCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1187 | $offset += 4; |
||
| 1188 | $endCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1189 | $offset += 4; |
||
| 1190 | $startGlyphID = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1191 | $offset += 4; |
||
| 1192 | for ($k = $startCharCode; $k <= $endCharCode; ++$k) { |
||
| 1193 | $is32idx = floor($c / 8); |
||
| 1194 | if ((isset($is32[$is32idx])) and (($is32[$is32idx] & (1 << (7 - ($c % 8)))) == 0)) { |
||
| 1195 | $c = $k; |
||
| 1196 | } else { |
||
| 1197 | // 32 bit format |
||
| 1198 | // convert to decimal (http://www.unicode.org/faq//utf_bom.html#utf16-4) |
||
| 1199 | //LEAD_OFFSET = (0xD800 - (0x10000 >> 10)) = 55232 |
||
| 1200 | //SURROGATE_OFFSET = (0x10000 - (0xD800 << 10) - 0xDC00) = -56613888 |
||
| 1201 | $c = ((55232 + ($k >> 10)) << 10) + (0xDC00 + ($k & 0x3FF)) - 56613888; |
||
| 1202 | } |
||
| 1203 | if (isset($subsetchars[$c])) { |
||
| 1204 | $subsetglyphs[$startGlyphID] = true; |
||
| 1205 | } |
||
| 1206 | ++$startGlyphID; |
||
| 1207 | } |
||
| 1208 | } |
||
| 1209 | break; |
||
| 1210 | } |
||
| 1211 | case 10: { // Format 10: Trimmed array |
||
| 1212 | $offset += 10; // skip reserved, length and version/language |
||
| 1213 | $startCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1214 | $offset += 4; |
||
| 1215 | $numChars = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1216 | $offset += 4; |
||
| 1217 | for ($k = 0; $k < $numChars; ++$k) { |
||
| 1218 | $c = ($k + $startCharCode); |
||
| 1219 | if (isset($subsetchars[$c])) { |
||
| 1220 | $g = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1221 | $subsetglyphs[$g] = true; |
||
| 1222 | } |
||
| 1223 | $offset += 2; |
||
| 1224 | } |
||
| 1225 | break; |
||
| 1226 | } |
||
| 1227 | case 12: { // Format 12: Segmented coverage |
||
| 1228 | $offset += 10; // skip length and version/language |
||
| 1229 | $nGroups = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1230 | $offset += 4; |
||
| 1231 | for ($k = 0; $k < $nGroups; ++$k) { |
||
| 1232 | $startCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1233 | $offset += 4; |
||
| 1234 | $endCharCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1235 | $offset += 4; |
||
| 1236 | $startGlyphCode = TCPDF_STATIC::_getULONG($font, $offset); |
||
| 1237 | $offset += 4; |
||
| 1238 | for ($c = $startCharCode; $c <= $endCharCode; ++$c) { |
||
| 1239 | if (isset($subsetchars[$c])) { |
||
| 1240 | $subsetglyphs[$startGlyphCode] = true; |
||
| 1241 | } |
||
| 1242 | ++$startGlyphCode; |
||
| 1243 | } |
||
| 1244 | } |
||
| 1245 | break; |
||
| 1246 | } |
||
| 1247 | case 13: { // Format 13: Many-to-one range mappings |
||
| 1248 | // to be implemented ... |
||
| 1249 | break; |
||
| 1250 | } |
||
| 1251 | case 14: { // Format 14: Unicode Variation Sequences |
||
| 1252 | // to be implemented ... |
||
| 1253 | break; |
||
| 1254 | } |
||
| 1255 | } |
||
| 1256 | } |
||
| 1257 | // include all parts of composite glyphs |
||
| 1258 | $new_sga = $subsetglyphs; |
||
| 1259 | while (!empty($new_sga)) { |
||
| 1260 | $sga = $new_sga; |
||
| 1261 | $new_sga = array(); |
||
| 1262 | foreach ($sga as $key => $val) { |
||
| 1263 | if (isset($indexToLoc[$key])) { |
||
| 1264 | $offset = ($table['glyf']['offset'] + $indexToLoc[$key]); |
||
| 1265 | $numberOfContours = TCPDF_STATIC::_getSHORT($font, $offset); |
||
| 1266 | $offset += 2; |
||
| 1267 | if ($numberOfContours < 0) { // composite glyph |
||
| 1268 | $offset += 8; // skip xMin, yMin, xMax, yMax |
||
| 1269 | do { |
||
| 1270 | $flags = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1271 | $offset += 2; |
||
| 1272 | $glyphIndex = TCPDF_STATIC::_getUSHORT($font, $offset); |
||
| 1273 | $offset += 2; |
||
| 1274 | if (!isset($subsetglyphs[$glyphIndex])) { |
||
| 1275 | // add missing glyphs |
||
| 1276 | $new_sga[$glyphIndex] = true; |
||
| 1277 | } |
||
| 1278 | // skip some bytes by case |
||
| 1279 | if ($flags & 1) { |
||
| 1280 | $offset += 4; |
||
| 1281 | } else { |
||
| 1282 | $offset += 2; |
||
| 1283 | } |
||
| 1284 | if ($flags & 8) { |
||
| 1285 | $offset += 2; |
||
| 1286 | } elseif ($flags & 64) { |
||
| 1287 | $offset += 4; |
||
| 1288 | } elseif ($flags & 128) { |
||
| 1289 | $offset += 8; |
||
| 1290 | } |
||
| 1291 | } while ($flags & 32); |
||
| 1292 | } |
||
| 1293 | } |
||
| 1294 | } |
||
| 1295 | $subsetglyphs += $new_sga; |
||
| 1296 | } |
||
| 1297 | // sort glyphs by key (and remove duplicates) |
||
| 1298 | ksort($subsetglyphs); |
||
| 1299 | // build new glyf and loca tables |
||
| 1300 | $glyf = ''; |
||
| 1301 | $loca = ''; |
||
| 1302 | $offset = 0; |
||
| 1303 | $glyf_offset = $table['glyf']['offset']; |
||
| 1304 | for ($i = 0; $i < $tot_num_glyphs; ++$i) { |
||
| 1305 | if (isset($subsetglyphs[$i])) { |
||
| 1306 | $length = ($indexToLoc[($i + 1)] - $indexToLoc[$i]); |
||
| 1307 | $glyf .= substr($font, ($glyf_offset + $indexToLoc[$i]), $length); |
||
| 1308 | } else { |
||
| 1309 | $length = 0; |
||
| 1310 | } |
||
| 1311 | if ($short_offset) { |
||
| 1312 | $loca .= pack('n', floor($offset / 2)); |
||
| 1313 | } else { |
||
| 1314 | $loca .= pack('N', $offset); |
||
| 1315 | } |
||
| 1316 | $offset += $length; |
||
| 1317 | } |
||
| 1318 | // array of table names to preserve (loca and glyf tables will be added later) |
||
| 1319 | // the cmap table is not needed and shall not be present, since the mapping from character codes to glyph descriptions is provided separately |
||
| 1320 | $table_names = array ('head', 'hhea', 'hmtx', 'maxp', 'cvt ', 'fpgm', 'prep'); // minimum required table names |
||
| 1321 | // get the tables to preserve |
||
| 1322 | $offset = 12; |
||
| 1323 | foreach ($table as $tag => $val) { |
||
| 1324 | if (in_array($tag, $table_names)) { |
||
| 1325 | $table[$tag]['data'] = substr($font, $table[$tag]['offset'], $table[$tag]['length']); |
||
| 1326 | if ($tag == 'head') { |
||
| 1327 | // set the checkSumAdjustment to 0 |
||
| 1328 | $table[$tag]['data'] = substr($table[$tag]['data'], 0, 8) . "\x0\x0\x0\x0" . substr($table[$tag]['data'], 12); |
||
| 1329 | } |
||
| 1330 | $table[$tag]['offset'] = $offset; |
||
| 1331 | $offset += $table[$tag]['length']; |
||
| 1332 | $numPad = ($offset + 3 & ~3) - $offset; |
||
| 1333 | if($numPad > 0) { |
||
| 1334 | $table[$tag]['data'] .= str_repeat("\x0", $numPad); |
||
| 1335 | $offset += $numPad; |
||
| 1336 | } |
||
| 1337 | // check sum is not changed (so keep the following line commented) |
||
| 1338 | //$table[$tag]['checkSum'] = self::_getTTFtableChecksum($table[$tag]['data'], $table[$tag]['length'] + $numPad); |
||
| 1339 | } else { |
||
| 1340 | unset($table[$tag]); |
||
| 1341 | } |
||
| 1342 | } |
||
| 1343 | // add loca |
||
| 1344 | $table['loca'] = array(); |
||
| 1345 | $table['loca']['data'] = $loca; |
||
| 1346 | $table['loca']['length'] = strlen($loca); |
||
| 1347 | $table['loca']['offset'] = $offset; |
||
| 1348 | $offset += $table['loca']['length']; |
||
| 1349 | $numPad = ($offset + 3 & ~3) - $offset; |
||
| 1350 | if($numPad > 0) { |
||
| 1351 | $table['loca']['data'] .= str_repeat("\x0", $numPad); |
||
| 1352 | $offset += $numPad; |
||
| 1353 | } |
||
| 1354 | $table['loca']['checkSum'] = self::_getTTFtableChecksum($table['loca']['data'], $table['loca']['length'] + $numPad); |
||
| 1355 | // add glyf |
||
| 1356 | $table['glyf'] = array(); |
||
| 1357 | $table['glyf']['data'] = $glyf; |
||
| 1358 | $table['glyf']['length'] = strlen($glyf); |
||
| 1359 | $table['glyf']['offset'] = $offset; |
||
| 1360 | $offset += $table['glyf']['length']; |
||
| 1361 | $numPad = ($offset + 3 & ~3) - $offset; |
||
| 1362 | if($numPad > 0) { |
||
| 1363 | $table['glyf']['data'] .= str_repeat("\x0", $numPad); |
||
| 1364 | $offset += $numPad; |
||
| 1365 | } |
||
| 1366 | $table['glyf']['checkSum'] = self::_getTTFtableChecksum($table['glyf']['data'], $table['glyf']['length'] + $numPad); |
||
| 1367 | // rebuild font |
||
| 1368 | $font = ''; |
||
| 1369 | $font .= pack('N', 0x10000); // sfnt version |
||
| 1370 | $numTables = count($table); |
||
| 1371 | $font .= pack('n', $numTables); // numTables |
||
| 1372 | $entrySelector = floor(log($numTables, 2)); |
||
| 1373 | $searchRange = pow(2, $entrySelector) * 16; |
||
| 1374 | $rangeShift = ($numTables * 16) - $searchRange; |
||
| 1375 | $font .= pack('n', $searchRange); // searchRange |
||
| 1376 | $font .= pack('n', $entrySelector); // entrySelector |
||
| 1377 | $font .= pack('n', $rangeShift); // rangeShift |
||
| 1378 | $offset = ($numTables * 16); |
||
| 1379 | foreach ($table as $tag => $data) { |
||
| 1380 | $font .= $tag; // tag |
||
| 1381 | $font .= pack('N', $data['checkSum']); // checkSum |
||
| 1382 | $font .= pack('N', ($data['offset'] + $offset)); // offset |
||
| 1383 | $font .= pack('N', $data['length']); // length |
||
| 1384 | } |
||
| 1385 | foreach ($table as $data) { |
||
| 1386 | $font .= $data['data']; |
||
| 1387 | } |
||
| 1388 | // set checkSumAdjustment on head table |
||
| 1389 | $checkSumAdjustment = 0xB1B0AFBA - self::_getTTFtableChecksum($font, strlen($font)); |
||
| 1390 | $font = substr($font, 0, $table['head']['offset'] + $offset + 8) . pack('N', $checkSumAdjustment) . substr($font, $table['head']['offset'] + $offset + 12); |
||
| 1391 | return $font; |
||
| 1392 | } |
||
| 1393 | |||
| 1394 | /** |
||
| 1395 | * Outputs font widths |
||
| 1396 | * @param array $font font data |
||
| 1397 | * @param int $cidoffset offset for CID values |
||
| 1398 | * @return string PDF command string for font widths |
||
| 1399 | * @author Nicola Asuni |
||
| 1400 | * @since 4.4.000 (2008-12-07) |
||
| 1401 | * @public static |
||
| 1402 | */ |
||
| 1403 | public static function _putfontwidths($font, $cidoffset = 0) |
||
| 1404 | { |
||
| 1405 | ksort($font['cw']); |
||
| 1406 | $rangeid = 0; |
||
| 1407 | $range = array(); |
||
| 1408 | $prevcid = -2; |
||
| 1409 | $prevwidth = -1; |
||
| 1410 | $interval = false; |
||
| 1411 | // for each character |
||
| 1412 | foreach ($font['cw'] as $cid => $width) { |
||
| 1413 | $cid -= $cidoffset; |
||
| 1414 | if ($font['subset'] and (!isset($font['subsetchars'][$cid]))) { |
||
| 1415 | // ignore the unused characters (font subsetting) |
||
| 1416 | continue; |
||
| 1417 | } |
||
| 1418 | if ($width != $font['dw']) { |
||
| 1419 | if ($cid == ($prevcid + 1)) { |
||
| 1420 | // consecutive CID |
||
| 1421 | if ($width == $prevwidth) { |
||
| 1422 | if ($width == $range[$rangeid][0]) { |
||
| 1423 | $range[$rangeid][] = $width; |
||
| 1424 | } else { |
||
| 1425 | array_pop($range[$rangeid]); |
||
| 1426 | // new range |
||
| 1427 | $rangeid = $prevcid; |
||
| 1428 | $range[$rangeid] = array(); |
||
| 1429 | $range[$rangeid][] = $prevwidth; |
||
| 1430 | $range[$rangeid][] = $width; |
||
| 1431 | } |
||
| 1432 | $interval = true; |
||
| 1433 | $range[$rangeid]['interval'] = true; |
||
| 1434 | } else { |
||
| 1435 | if ($interval) { |
||
| 1436 | // new range |
||
| 1437 | $rangeid = $cid; |
||
| 1438 | $range[$rangeid] = array(); |
||
| 1439 | $range[$rangeid][] = $width; |
||
| 1440 | } else { |
||
| 1441 | $range[$rangeid][] = $width; |
||
| 1442 | } |
||
| 1443 | $interval = false; |
||
| 1444 | } |
||
| 1445 | } else { |
||
| 1446 | // new range |
||
| 1447 | $rangeid = $cid; |
||
| 1448 | $range[$rangeid] = array(); |
||
| 1449 | $range[$rangeid][] = $width; |
||
| 1450 | $interval = false; |
||
| 1451 | } |
||
| 1452 | $prevcid = $cid; |
||
| 1453 | $prevwidth = $width; |
||
| 1454 | } |
||
| 1455 | } |
||
| 1456 | // optimize ranges |
||
| 1457 | $prevk = -1; |
||
| 1458 | $nextk = -1; |
||
| 1459 | $prevint = false; |
||
| 1460 | foreach ($range as $k => $ws) { |
||
| 1461 | $cws = count($ws); |
||
| 1462 | if (($k == $nextk) and (!$prevint) and ((!isset($ws['interval'])) or ($cws < 4))) { |
||
| 1463 | if (isset($range[$k]['interval'])) { |
||
| 1464 | unset($range[$k]['interval']); |
||
| 1465 | } |
||
| 1466 | $range[$prevk] = array_merge($range[$prevk], $range[$k]); |
||
| 1467 | unset($range[$k]); |
||
| 1468 | } else { |
||
| 1469 | $prevk = $k; |
||
| 1470 | } |
||
| 1471 | $nextk = $k + $cws; |
||
| 1472 | if (isset($ws['interval'])) { |
||
| 1473 | if ($cws > 3) { |
||
| 1474 | $prevint = true; |
||
| 1475 | } else { |
||
| 1476 | $prevint = false; |
||
| 1477 | } |
||
| 1478 | if (isset($range[$k]['interval'])) { |
||
| 1479 | unset($range[$k]['interval']); |
||
| 1480 | } |
||
| 1481 | --$nextk; |
||
| 1482 | } else { |
||
| 1483 | $prevint = false; |
||
| 1484 | } |
||
| 1485 | } |
||
| 1486 | // output data |
||
| 1487 | $w = ''; |
||
| 1488 | foreach ($range as $k => $ws) { |
||
| 1489 | if (count(array_count_values($ws)) == 1) { |
||
| 1490 | // interval mode is more compact |
||
| 1491 | $w .= ' ' . $k . ' ' . ($k + count($ws) - 1) . ' ' . $ws[0]; |
||
| 1492 | } else { |
||
| 1493 | // range mode |
||
| 1494 | $w .= ' ' . $k . ' [ ' . implode(' ', $ws) . ' ]'; |
||
| 1495 | } |
||
| 1496 | } |
||
| 1497 | return '/W [' . $w . ' ]'; |
||
| 1498 | } |
||
| 1499 | |||
| 1500 | |||
| 1501 | |||
| 1502 | |||
| 1503 | /** |
||
| 1504 | * Update the CIDToGIDMap string with a new value. |
||
| 1505 | * @param string $map CIDToGIDMap. |
||
| 1506 | * @param int $cid CID value. |
||
| 1507 | * @param int $gid GID value. |
||
| 1508 | * @return string CIDToGIDMap. |
||
| 1509 | * @author Nicola Asuni |
||
| 1510 | * @since 5.9.123 (2011-09-29) |
||
| 1511 | * @public static |
||
| 1512 | */ |
||
| 1513 | public static function updateCIDtoGIDmap($map, $cid, $gid) |
||
| 1514 | { |
||
| 1515 | if (($cid >= 0) and ($cid <= 0xFFFF) and ($gid >= 0)) { |
||
| 1516 | if ($gid > 0xFFFF) { |
||
| 1517 | $gid -= 0x10000; |
||
| 1518 | } |
||
| 1519 | $map[($cid * 2)] = chr($gid >> 8); |
||
| 1520 | $map[(($cid * 2) + 1)] = chr($gid & 0xFF); |
||
| 1521 | } |
||
| 1522 | return $map; |
||
| 1523 | } |
||
| 1524 | |||
| 1525 | /** |
||
| 1526 | * Return fonts path |
||
| 1527 | * @return string |
||
| 1528 | * @public static |
||
| 1529 | */ |
||
| 1530 | public static function _getfontpath() |
||
| 1531 | { |
||
| 1532 | if (!defined('K_PATH_FONTS') and is_dir($fdir = realpath(dirname(__FILE__) . '/../fonts'))) { |
||
| 1533 | if (substr($fdir, -1) != '/') { |
||
| 1534 | $fdir .= '/'; |
||
| 1535 | } |
||
| 1536 | define('K_PATH_FONTS', $fdir); |
||
| 1537 | } |
||
| 1538 | return defined('K_PATH_FONTS') ? K_PATH_FONTS : ''; |
||
| 1539 | } |
||
| 1540 | |||
| 1541 | |||
| 1542 | |||
| 1543 | /** |
||
| 1544 | * Return font full path |
||
| 1545 | * @param string $file Font file name. |
||
| 1546 | * @param string $fontdir Font directory (set to false fto search on default directories) |
||
| 1547 | * @return string Font full path or empty string |
||
| 1548 | * @author Nicola Asuni |
||
| 1549 | * @since 6.0.025 |
||
| 1550 | * @public static |
||
| 1551 | */ |
||
| 1552 | public static function getFontFullPath($file, $fontdir = false) |
||
| 1553 | { |
||
| 1554 | $fontfile = ''; |
||
| 1555 | // search files on various directories |
||
| 1556 | if (($fontdir !== false) and @TCPDF_STATIC::file_exists($fontdir . $file)) { |
||
| 1557 | $fontfile = $fontdir . $file; |
||
| 1558 | } elseif (@TCPDF_STATIC::file_exists(self::_getfontpath() . $file)) { |
||
| 1559 | $fontfile = self::_getfontpath() . $file; |
||
| 1560 | } elseif (@TCPDF_STATIC::file_exists($file)) { |
||
| 1561 | $fontfile = $file; |
||
| 1562 | } |
||
| 1563 | return $fontfile; |
||
| 1564 | } |
||
| 1565 | |||
| 1566 | |||
| 1567 | |||
| 1568 | |||
| 1569 | /** |
||
| 1570 | * Get a reference font size. |
||
| 1571 | * @param string $size String containing font size value. |
||
| 1572 | * @param float $refsize Reference font size in points. |
||
| 1573 | * @return float value in points |
||
| 1574 | * @public static |
||
| 1575 | */ |
||
| 1576 | public static function getFontRefSize($size, $refsize = 12) |
||
| 1577 | { |
||
| 1578 | switch ($size) { |
||
| 1579 | case 'xx-small': { |
||
| 1580 | $size = ($refsize - 4); |
||
| 1581 | break; |
||
| 1582 | } |
||
| 1583 | case 'x-small': { |
||
| 1584 | $size = ($refsize - 3); |
||
| 1585 | break; |
||
| 1586 | } |
||
| 1587 | case 'small': { |
||
| 1588 | $size = ($refsize - 2); |
||
| 1589 | break; |
||
| 1590 | } |
||
| 1591 | case 'medium': { |
||
| 1592 | $size = $refsize; |
||
| 1593 | break; |
||
| 1594 | } |
||
| 1595 | case 'large': { |
||
| 1596 | $size = ($refsize + 2); |
||
| 1597 | break; |
||
| 1598 | } |
||
| 1599 | case 'x-large': { |
||
| 1600 | $size = ($refsize + 4); |
||
| 1601 | break; |
||
| 1602 | } |
||
| 1603 | case 'xx-large': { |
||
| 1604 | $size = ($refsize + 6); |
||
| 1605 | break; |
||
| 1606 | } |
||
| 1607 | case 'smaller': { |
||
| 1608 | $size = ($refsize - 3); |
||
| 1609 | break; |
||
| 1610 | } |
||
| 1611 | case 'larger': { |
||
| 1612 | $size = ($refsize + 3); |
||
| 1613 | break; |
||
| 1614 | } |
||
| 1615 | } |
||
| 1616 | return $size; |
||
| 1617 | } |
||
| 1618 | |||
| 1619 | |||
| 1620 | |||
| 1621 | |||
| 1622 | |||
| 1623 | |||
| 1624 | |||
| 1625 | |||
| 1626 | |||
| 1627 | |||
| 1628 | |||
| 1629 | |||
| 1630 | |||
| 1631 | |||
| 1632 | |||
| 1633 | |||
| 1634 | |||
| 1635 | |||
| 1636 | |||
| 1637 | |||
| 1638 | |||
| 1639 | |||
| 1640 | |||
| 1641 | |||
| 1642 | |||
| 1643 | |||
| 1644 | |||
| 1645 | |||
| 1646 | |||
| 1647 | |||
| 1648 | |||
| 1649 | |||
| 1650 | |||
| 1651 | |||
| 1652 | |||
| 1653 | |||
| 1654 | |||
| 1655 | |||
| 1656 | |||
| 1657 | |||
| 1658 | // ==================================================================================================================== |
||
| 1659 | // REIMPLEMENTED |
||
| 1660 | // ==================================================================================================================== |
||
| 1661 | |||
| 1662 | |||
| 1663 | |||
| 1664 | |||
| 1665 | |||
| 1666 | |||
| 1667 | |||
| 1668 | |||
| 1669 | /** |
||
| 1670 | * Returns the unicode caracter specified by the value |
||
| 1671 | * @param int $c UTF-8 value |
||
| 1672 | * @param boolean $unicode True if we are in unicode mode, false otherwise. |
||
| 1673 | * @return string Returns the specified character. |
||
| 1674 | * @since 2.3.000 (2008-03-05) |
||
| 1675 | * @public static |
||
| 1676 | */ |
||
| 1677 | public static function unichr($c, $unicode = true) |
||
| 1678 | { |
||
| 1679 | $c = intval($c); |
||
| 1680 | if (!$unicode) { |
||
| 1681 | return chr($c); |
||
| 1682 | } elseif ($c <= 0x7F) { |
||
| 1683 | // one byte |
||
| 1684 | return chr($c); |
||
| 1685 | } elseif ($c <= 0x7FF) { |
||
| 1686 | // two bytes |
||
| 1687 | return chr(0xC0 | $c >> 6) . chr(0x80 | $c & 0x3F); |
||
| 1688 | } elseif ($c <= 0xFFFF) { |
||
| 1689 | // three bytes |
||
| 1690 | return chr(0xE0 | $c >> 12) . chr(0x80 | $c >> 6 & 0x3F) . chr(0x80 | $c & 0x3F); |
||
| 1691 | } elseif ($c <= 0x10FFFF) { |
||
| 1692 | // four bytes |
||
| 1693 | return chr(0xF0 | $c >> 18) . chr(0x80 | $c >> 12 & 0x3F) . chr(0x80 | $c >> 6 & 0x3F) . chr(0x80 | $c & 0x3F); |
||
| 1694 | } else { |
||
| 1695 | return ''; |
||
| 1696 | } |
||
| 1697 | } |
||
| 1698 | |||
| 1699 | /** |
||
| 1700 | * Returns the unicode caracter specified by UTF-8 value |
||
| 1701 | * @param int $c UTF-8 value |
||
| 1702 | * @return string Returns the specified character. |
||
| 1703 | * @public static |
||
| 1704 | */ |
||
| 1705 | public static function unichrUnicode($c) |
||
| 1706 | { |
||
| 1707 | return self::unichr($c, true); |
||
| 1708 | } |
||
| 1709 | |||
| 1710 | /** |
||
| 1711 | * Returns the unicode caracter specified by ASCII value |
||
| 1712 | * @param int $c UTF-8 value |
||
| 1713 | * @return string Returns the specified character. |
||
| 1714 | * @public static |
||
| 1715 | */ |
||
| 1716 | public static function unichrASCII($c) |
||
| 1717 | { |
||
| 1718 | return self::unichr($c, false); |
||
| 1719 | } |
||
| 1720 | |||
| 1721 | /** |
||
| 1722 | * Converts array of UTF-8 characters to UTF16-BE string.<br> |
||
| 1723 | * Based on: http://www.faqs.org/rfcs/rfc2781.html |
||
| 1724 | * <pre> |
||
| 1725 | * Encoding UTF-16: |
||
| 1726 | * |
||
| 1727 | * Encoding of a single character from an ISO 10646 character value to |
||
| 1728 | * UTF-16 proceeds as follows. Let U be the character number, no greater |
||
| 1729 | * than 0x10FFFF. |
||
| 1730 | * |
||
| 1731 | * 1) If U < 0x10000, encode U as a 16-bit unsigned integer and |
||
| 1732 | * terminate. |
||
| 1733 | * |
||
| 1734 | * 2) Let U' = U - 0x10000. Because U is less than or equal to 0x10FFFF, |
||
| 1735 | * U' must be less than or equal to 0xFFFFF. That is, U' can be |
||
| 1736 | * represented in 20 bits. |
||
| 1737 | * |
||
| 1738 | * 3) Initialize two 16-bit unsigned integers, W1 and W2, to 0xD800 and |
||
| 1739 | * 0xDC00, respectively. These integers each have 10 bits free to |
||
| 1740 | * encode the character value, for a total of 20 bits. |
||
| 1741 | * |
||
| 1742 | * 4) Assign the 10 high-order bits of the 20-bit U' to the 10 low-order |
||
| 1743 | * bits of W1 and the 10 low-order bits of U' to the 10 low-order |
||
| 1744 | * bits of W2. Terminate. |
||
| 1745 | * |
||
| 1746 | * Graphically, steps 2 through 4 look like: |
||
| 1747 | * U' = yyyyyyyyyyxxxxxxxxxx |
||
| 1748 | * W1 = 110110yyyyyyyyyy |
||
| 1749 | * W2 = 110111xxxxxxxxxx |
||
| 1750 | * </pre> |
||
| 1751 | * @param array $unicode array containing UTF-8 unicode values |
||
| 1752 | * @param boolean $setbom if true set the Byte Order Mark (BOM = 0xFEFF) |
||
| 1753 | * @return string |
||
| 1754 | * @protected |
||
| 1755 | * @author Nicola Asuni |
||
| 1756 | * @since 2.1.000 (2008-01-08) |
||
| 1757 | * @public static |
||
| 1758 | */ |
||
| 1759 | public static function arrUTF8ToUTF16BE($unicode, $setbom = false) |
||
| 1760 | { |
||
| 1761 | $outstr = ''; // string to be returned |
||
| 1762 | if ($setbom) { |
||
| 1763 | $outstr .= "\xFE\xFF"; // Byte Order Mark (BOM) |
||
| 1764 | } |
||
| 1765 | foreach ($unicode as $char) { |
||
| 1766 | if ($char == 0x200b) { |
||
| 1767 | // skip Unicode Character 'ZERO WIDTH SPACE' (DEC:8203, U+200B) |
||
| 1768 | } elseif ($char == 0xFFFD) { |
||
| 1769 | $outstr .= "\xFF\xFD"; // replacement character |
||
| 1770 | } elseif ($char < 0x10000) { |
||
| 1771 | $outstr .= chr($char >> 0x08); |
||
| 1772 | $outstr .= chr($char & 0xFF); |
||
| 1773 | } else { |
||
| 1774 | $char -= 0x10000; |
||
| 1775 | $w1 = 0xD800 | ($char >> 0x0a); |
||
| 1776 | $w2 = 0xDC00 | ($char & 0x3FF); |
||
| 1777 | $outstr .= chr($w1 >> 0x08); |
||
| 1778 | $outstr .= chr($w1 & 0xFF); |
||
| 1779 | $outstr .= chr($w2 >> 0x08); |
||
| 1780 | $outstr .= chr($w2 & 0xFF); |
||
| 1781 | } |
||
| 1782 | } |
||
| 1783 | return $outstr; |
||
| 1784 | } |
||
| 1785 | |||
| 1786 | /** |
||
| 1787 | * Convert an array of UTF8 values to array of unicode characters |
||
| 1788 | * @param array $ta The input array of UTF8 values. |
||
| 1789 | * @param boolean $isunicode True for Unicode mode, false otherwise. |
||
| 1790 | * @return array Return array of unicode characters |
||
| 1791 | * @since 4.5.037 (2009-04-07) |
||
| 1792 | * @public static |
||
| 1793 | */ |
||
| 1794 | public static function UTF8ArrayToUniArray($ta, $isunicode = true) |
||
| 1795 | { |
||
| 1796 | if ($isunicode) { |
||
| 1797 | return array_map(get_called_class() . '::unichrUnicode', $ta); |
||
| 1798 | } |
||
| 1799 | return array_map(get_called_class() . '::unichrASCII', $ta); |
||
| 1800 | } |
||
| 1801 | |||
| 1802 | /** |
||
| 1803 | * Extract a slice of the $strarr array and return it as string. |
||
| 1804 | * @param string[] $strarr The input array of characters. |
||
| 1805 | * @param int $start the starting element of $strarr. |
||
| 1806 | * @param int $end first element that will not be returned. |
||
| 1807 | * @param boolean $unicode True if we are in unicode mode, false otherwise. |
||
| 1808 | * @return string Return part of a string |
||
| 1809 | * @public static |
||
| 1810 | */ |
||
| 1811 | public static function UTF8ArrSubString($strarr, $start = '', $end = '', $unicode = true) |
||
| 1812 | { |
||
| 1813 | if (strlen($start) == 0) { |
||
| 1814 | $start = 0; |
||
| 1815 | } |
||
| 1816 | if (strlen($end) == 0) { |
||
| 1817 | $end = count($strarr); |
||
| 1818 | } |
||
| 1819 | $string = ''; |
||
| 1820 | for ($i = $start; $i < $end; ++$i) { |
||
| 1821 | $string .= self::unichr($strarr[$i], $unicode); |
||
| 1822 | } |
||
| 1823 | return $string; |
||
| 1824 | } |
||
| 1825 | |||
| 1826 | /** |
||
| 1827 | * Extract a slice of the $uniarr array and return it as string. |
||
| 1828 | * @param string[] $uniarr The input array of characters. |
||
| 1829 | * @param int $start the starting element of $strarr. |
||
| 1830 | * @param int $end first element that will not be returned. |
||
| 1831 | * @return string Return part of a string |
||
| 1832 | * @since 4.5.037 (2009-04-07) |
||
| 1833 | * @public static |
||
| 1834 | */ |
||
| 1835 | public static function UniArrSubString($uniarr, $start = '', $end = '') |
||
| 1836 | { |
||
| 1837 | if (strlen($start) == 0) { |
||
| 1838 | $start = 0; |
||
| 1839 | } |
||
| 1840 | if (strlen($end) == 0) { |
||
| 1841 | $end = count($uniarr); |
||
| 1842 | } |
||
| 1843 | $string = ''; |
||
| 1844 | for ($i = $start; $i < $end; ++$i) { |
||
| 1845 | $string .= $uniarr[$i]; |
||
| 1846 | } |
||
| 1847 | return $string; |
||
| 1848 | } |
||
| 1849 | |||
| 1850 | /** |
||
| 1851 | * Converts UTF-8 characters array to array of Latin1 characters array<br> |
||
| 1852 | * @param array $unicode array containing UTF-8 unicode values |
||
| 1853 | * @return array |
||
| 1854 | * @author Nicola Asuni |
||
| 1855 | * @since 4.8.023 (2010-01-15) |
||
| 1856 | * @public static |
||
| 1857 | */ |
||
| 1858 | public static function UTF8ArrToLatin1Arr($unicode) |
||
| 1859 | { |
||
| 1860 | $outarr = array(); // array to be returned |
||
| 1861 | foreach ($unicode as $char) { |
||
| 1862 | if ($char < 256) { |
||
| 1863 | $outarr[] = $char; |
||
| 1864 | } elseif (array_key_exists($char, TCPDF_FONT_DATA::$uni_utf8tolatin)) { |
||
| 1865 | // map from UTF-8 |
||
| 1866 | $outarr[] = TCPDF_FONT_DATA::$uni_utf8tolatin[$char]; |
||
| 1867 | } elseif ($char == 0xFFFD) { |
||
| 1868 | // skip |
||
| 1869 | } else { |
||
| 1870 | $outarr[] = 63; // '?' character |
||
| 1871 | } |
||
| 1872 | } |
||
| 1873 | return $outarr; |
||
| 1874 | } |
||
| 1875 | |||
| 1876 | /** |
||
| 1877 | * Converts UTF-8 characters array to Latin1 string<br> |
||
| 1878 | * @param array $unicode array containing UTF-8 unicode values |
||
| 1879 | * @return string |
||
| 1880 | * @author Nicola Asuni |
||
| 1881 | * @since 4.8.023 (2010-01-15) |
||
| 1882 | * @public static |
||
| 1883 | */ |
||
| 1884 | public static function UTF8ArrToLatin1($unicode) |
||
| 1885 | { |
||
| 1886 | $outstr = ''; // string to be returned |
||
| 1887 | foreach ($unicode as $char) { |
||
| 1888 | if ($char < 256) { |
||
| 1889 | $outstr .= chr($char); |
||
| 1890 | } elseif (array_key_exists($char, TCPDF_FONT_DATA::$uni_utf8tolatin)) { |
||
| 1891 | // map from UTF-8 |
||
| 1892 | $outstr .= chr(TCPDF_FONT_DATA::$uni_utf8tolatin[$char]); |
||
| 1893 | } elseif ($char == 0xFFFD) { |
||
| 1894 | // skip |
||
| 1895 | } else { |
||
| 1896 | $outstr .= '?'; |
||
| 1897 | } |
||
| 1898 | } |
||
| 1899 | return $outstr; |
||
| 1900 | } |
||
| 1901 | |||
| 1902 | /** |
||
| 1903 | * Converts UTF-8 character to integer value.<br> |
||
| 1904 | * Uses the getUniord() method if the value is not cached. |
||
| 1905 | * @param string $uch character string to process. |
||
| 1906 | * @return int Unicode value |
||
| 1907 | * @public static |
||
| 1908 | */ |
||
| 1909 | public static function uniord($uch) |
||
| 1910 | { |
||
| 1911 | if (!isset(self::$cache_uniord[$uch])) { |
||
| 1912 | self::$cache_uniord[$uch] = self::getUniord($uch); |
||
| 1913 | } |
||
| 1914 | return self::$cache_uniord[$uch]; |
||
| 1915 | } |
||
| 1916 | |||
| 1917 | /** |
||
| 1918 | * Converts UTF-8 character to integer value.<br> |
||
| 1919 | * Invalid byte sequences will be replaced with 0xFFFD (replacement character)<br> |
||
| 1920 | * Based on: http://www.faqs.org/rfcs/rfc3629.html |
||
| 1921 | * <pre> |
||
| 1922 | * Char. number range | UTF-8 octet sequence |
||
| 1923 | * (hexadecimal) | (binary) |
||
| 1924 | * --------------------+----------------------------------------------- |
||
| 1925 | * 0000 0000-0000 007F | 0xxxxxxx |
||
| 1926 | * 0000 0080-0000 07FF | 110xxxxx 10xxxxxx |
||
| 1927 | * 0000 0800-0000 FFFF | 1110xxxx 10xxxxxx 10xxxxxx |
||
| 1928 | * 0001 0000-0010 FFFF | 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx |
||
| 1929 | * --------------------------------------------------------------------- |
||
| 1930 | * |
||
| 1931 | * ABFN notation: |
||
| 1932 | * --------------------------------------------------------------------- |
||
| 1933 | * UTF8-octets = *( UTF8-char ) |
||
| 1934 | * UTF8-char = UTF8-1 / UTF8-2 / UTF8-3 / UTF8-4 |
||
| 1935 | * UTF8-1 = %x00-7F |
||
| 1936 | * UTF8-2 = %xC2-DF UTF8-tail |
||
| 1937 | * |
||
| 1938 | * UTF8-3 = %xE0 %xA0-BF UTF8-tail / %xE1-EC 2( UTF8-tail ) / |
||
| 1939 | * %xED %x80-9F UTF8-tail / %xEE-EF 2( UTF8-tail ) |
||
| 1940 | * UTF8-4 = %xF0 %x90-BF 2( UTF8-tail ) / %xF1-F3 3( UTF8-tail ) / |
||
| 1941 | * %xF4 %x80-8F 2( UTF8-tail ) |
||
| 1942 | * UTF8-tail = %x80-BF |
||
| 1943 | * --------------------------------------------------------------------- |
||
| 1944 | * </pre> |
||
| 1945 | * @param string $uch character string to process. |
||
| 1946 | * @return int Unicode value |
||
| 1947 | * @author Nicola Asuni |
||
| 1948 | * @public static |
||
| 1949 | */ |
||
| 1950 | public static function getUniord($uch) |
||
| 1951 | { |
||
| 1952 | if (function_exists('mb_convert_encoding')) { |
||
| 1953 | list(, $char) = @unpack('N', mb_convert_encoding($uch, 'UCS-4BE', 'UTF-8')); |
||
| 1954 | if ($char >= 0) { |
||
| 1955 | return $char; |
||
| 1956 | } |
||
| 1957 | } |
||
| 1958 | $bytes = array(); // array containing single character byte sequences |
||
| 1959 | $countbytes = 0; |
||
| 1960 | $numbytes = 1; // number of octetc needed to represent the UTF-8 character |
||
| 1961 | $length = strlen($uch); |
||
| 1962 | for ($i = 0; $i < $length; ++$i) { |
||
| 1963 | $char = ord($uch[$i]); // get one string character at time |
||
| 1964 | if ($countbytes == 0) { // get starting octect |
||
| 1965 | if ($char <= 0x7F) { |
||
| 1966 | return $char; // use the character "as is" because is ASCII |
||
| 1967 | } elseif (($char >> 0x05) == 0x06) { // 2 bytes character (0x06 = 110 BIN) |
||
| 1968 | $bytes[] = ($char - 0xC0) << 0x06; |
||
| 1969 | ++$countbytes; |
||
| 1970 | $numbytes = 2; |
||
| 1971 | } elseif (($char >> 0x04) == 0x0E) { // 3 bytes character (0x0E = 1110 BIN) |
||
| 1972 | $bytes[] = ($char - 0xE0) << 0x0C; |
||
| 1973 | ++$countbytes; |
||
| 1974 | $numbytes = 3; |
||
| 1975 | } elseif (($char >> 0x03) == 0x1E) { // 4 bytes character (0x1E = 11110 BIN) |
||
| 1976 | $bytes[] = ($char - 0xF0) << 0x12; |
||
| 1977 | ++$countbytes; |
||
| 1978 | $numbytes = 4; |
||
| 1979 | } else { |
||
| 1980 | // use replacement character for other invalid sequences |
||
| 1981 | return 0xFFFD; |
||
| 1982 | } |
||
| 1983 | } elseif (($char >> 0x06) == 0x02) { // bytes 2, 3 and 4 must start with 0x02 = 10 BIN |
||
| 1984 | $bytes[] = $char - 0x80; |
||
| 1985 | ++$countbytes; |
||
| 1986 | if ($countbytes == $numbytes) { |
||
| 1987 | // compose UTF-8 bytes to a single unicode value |
||
| 1988 | $char = $bytes[0]; |
||
| 1989 | for ($j = 1; $j < $numbytes; ++$j) { |
||
| 1990 | $char += ($bytes[$j] << (($numbytes - $j - 1) * 0x06)); |
||
| 1991 | } |
||
| 1992 | if ((($char >= 0xD800) and ($char <= 0xDFFF)) or ($char >= 0x10FFFF)) { |
||
| 1993 | // The definition of UTF-8 prohibits encoding character numbers between |
||
| 1994 | // U+D800 and U+DFFF, which are reserved for use with the UTF-16 |
||
| 1995 | // encoding form (as surrogate pairs) and do not directly represent |
||
| 1996 | // characters. |
||
| 1997 | return 0xFFFD; // use replacement character |
||
| 1998 | } else { |
||
| 1999 | return $char; |
||
| 2000 | } |
||
| 2001 | } |
||
| 2002 | } else { |
||
| 2003 | // use replacement character for other invalid sequences |
||
| 2004 | return 0xFFFD; |
||
| 2005 | } |
||
| 2006 | } |
||
| 2007 | return 0xFFFD; |
||
| 2008 | } |
||
| 2009 | |||
| 2010 | /** |
||
| 2011 | * Converts UTF-8 strings to codepoints array.<br> |
||
| 2012 | * Invalid byte sequences will be replaced with 0xFFFD (replacement character)<br> |
||
| 2013 | * @param string $str string to process. |
||
| 2014 | * @param boolean $isunicode True when the documetn is in Unicode mode, false otherwise. |
||
| 2015 | * @param array $currentfont Reference to current font array. |
||
| 2016 | * @return array containing codepoints (UTF-8 characters values) |
||
| 2017 | * @author Nicola Asuni |
||
| 2018 | * @public static |
||
| 2019 | */ |
||
| 2020 | public static function UTF8StringToArray($str, $isunicode, &$currentfont) |
||
| 2021 | { |
||
| 2022 | $str = is_null($str) ? '' : $str; |
||
| 2023 | if ($isunicode) { |
||
| 2024 | // requires PCRE unicode support turned on |
||
| 2025 | $chars = TCPDF_STATIC::pregSplit('//', 'u', $str, -1, PREG_SPLIT_NO_EMPTY); |
||
| 2026 | $carr = array_map(get_called_class() . '::uniord', $chars); |
||
| 2027 | } else { |
||
| 2028 | $chars = str_split($str); |
||
| 2029 | $carr = array_map('ord', $chars); |
||
| 2030 | } |
||
| 2031 | if (is_array($currentfont['subsetchars']) && is_array($carr)) { |
||
| 2032 | $currentfont['subsetchars'] += array_fill_keys($carr, true); |
||
| 2033 | } else { |
||
| 2034 | $currentfont['subsetchars'] = array_merge($currentfont['subsetchars'], $carr); |
||
| 2035 | } |
||
| 2036 | return $carr; |
||
| 2037 | } |
||
| 2038 | |||
| 2039 | /** |
||
| 2040 | * Converts UTF-8 strings to Latin1 when using the standard 14 core fonts.<br> |
||
| 2041 | * @param string $str string to process. |
||
| 2042 | * @param boolean $isunicode True when the documetn is in Unicode mode, false otherwise. |
||
| 2043 | * @param array $currentfont Reference to current font array. |
||
| 2044 | * @return string |
||
| 2045 | * @since 3.2.000 (2008-06-23) |
||
| 2046 | * @public static |
||
| 2047 | */ |
||
| 2048 | public static function UTF8ToLatin1($str, $isunicode, &$currentfont) |
||
| 2049 | { |
||
| 2050 | $unicode = self::UTF8StringToArray($str, $isunicode, $currentfont); // array containing UTF-8 unicode values |
||
| 2051 | return self::UTF8ArrToLatin1($unicode); |
||
| 2052 | } |
||
| 2053 | |||
| 2054 | /** |
||
| 2055 | * Converts UTF-8 strings to UTF16-BE.<br> |
||
| 2056 | * @param string $str string to process. |
||
| 2057 | * @param boolean $setbom if true set the Byte Order Mark (BOM = 0xFEFF) |
||
| 2058 | * @param boolean $isunicode True when the documetn is in Unicode mode, false otherwise. |
||
| 2059 | * @param array $currentfont Reference to current font array. |
||
| 2060 | * @return string |
||
| 2061 | * @author Nicola Asuni |
||
| 2062 | * @since 1.53.0.TC005 (2005-01-05) |
||
| 2063 | * @public static |
||
| 2064 | */ |
||
| 2065 | public static function UTF8ToUTF16BE($str, $setbom, $isunicode, &$currentfont) |
||
| 2066 | { |
||
| 2067 | if (!$isunicode) { |
||
| 2068 | return $str; // string is not in unicode |
||
| 2069 | } |
||
| 2070 | $unicode = self::UTF8StringToArray($str, $isunicode, $currentfont); // array containing UTF-8 unicode values |
||
| 2071 | return self::arrUTF8ToUTF16BE($unicode, $setbom); |
||
| 2072 | } |
||
| 2073 | |||
| 2074 | /** |
||
| 2075 | * Reverse the RLT substrings using the Bidirectional Algorithm (http://unicode.org/reports/tr9/). |
||
| 2076 | * @param string $str string to manipulate. |
||
| 2077 | * @param bool $setbom if true set the Byte Order Mark (BOM = 0xFEFF) |
||
| 2078 | * @param bool $forcertl if true forces RTL text direction |
||
| 2079 | * @param boolean $isunicode True if the document is in Unicode mode, false otherwise. |
||
| 2080 | * @param array $currentfont Reference to current font array. |
||
| 2081 | * @return string |
||
| 2082 | * @author Nicola Asuni |
||
| 2083 | * @since 2.1.000 (2008-01-08) |
||
| 2084 | * @public static |
||
| 2085 | */ |
||
| 2086 | public static function utf8StrRev($str, $setbom, $forcertl, $isunicode, &$currentfont) |
||
| 2087 | { |
||
| 2088 | return self::utf8StrArrRev(self::UTF8StringToArray($str, $isunicode, $currentfont), $str, $setbom, $forcertl, $isunicode, $currentfont); |
||
| 2089 | } |
||
| 2090 | |||
| 2091 | /** |
||
| 2092 | * Reverse the RLT substrings array using the Bidirectional Algorithm (http://unicode.org/reports/tr9/). |
||
| 2093 | * @param array $arr array of unicode values. |
||
| 2094 | * @param string $str string to manipulate (or empty value). |
||
| 2095 | * @param bool $setbom if true set the Byte Order Mark (BOM = 0xFEFF) |
||
| 2096 | * @param bool $forcertl if true forces RTL text direction |
||
| 2097 | * @param boolean $isunicode True if the document is in Unicode mode, false otherwise. |
||
| 2098 | * @param array $currentfont Reference to current font array. |
||
| 2099 | * @return string |
||
| 2100 | * @author Nicola Asuni |
||
| 2101 | * @since 4.9.000 (2010-03-27) |
||
| 2102 | * @public static |
||
| 2103 | */ |
||
| 2104 | public static function utf8StrArrRev($arr, $str, $setbom, $forcertl, $isunicode, &$currentfont) |
||
| 2105 | { |
||
| 2106 | return self::arrUTF8ToUTF16BE(self::utf8Bidi($arr, $str, $forcertl, $isunicode, $currentfont), $setbom); |
||
| 2107 | } |
||
| 2108 | |||
| 2109 | /** |
||
| 2110 | * Reverse the RLT substrings using the Bidirectional Algorithm (http://unicode.org/reports/tr9/). |
||
| 2111 | * @param array $ta array of characters composing the string. |
||
| 2112 | * @param string $str string to process |
||
| 2113 | * @param bool $forcertl if 'R' forces RTL, if 'L' forces LTR |
||
| 2114 | * @param boolean $isunicode True if the document is in Unicode mode, false otherwise. |
||
| 2115 | * @param array $currentfont Reference to current font array. |
||
| 2116 | * @return array of unicode chars |
||
| 2117 | * @author Nicola Asuni |
||
| 2118 | * @since 2.4.000 (2008-03-06) |
||
| 2119 | * @public static |
||
| 2120 | */ |
||
| 2121 | public static function utf8Bidi($ta, $str, $forcertl, $isunicode, &$currentfont) |
||
| 2122 | { |
||
| 2123 | // paragraph embedding level |
||
| 2124 | $pel = 0; |
||
| 2125 | // max level |
||
| 2126 | $maxlevel = 0; |
||
| 2127 | if (TCPDF_STATIC::empty_string($str)) { |
||
| 2128 | // create string from array |
||
| 2129 | $str = self::UTF8ArrSubString($ta, '', '', $isunicode); |
||
| 2130 | } |
||
| 2131 | // check if string contains arabic text |
||
| 2132 | if (preg_match(TCPDF_FONT_DATA::$uni_RE_PATTERN_ARABIC, $str)) { |
||
| 2133 | $arabic = true; |
||
| 2134 | } else { |
||
| 2135 | $arabic = false; |
||
| 2136 | } |
||
| 2137 | // check if string contains RTL text |
||
| 2138 | if (!($forcertl or $arabic or preg_match(TCPDF_FONT_DATA::$uni_RE_PATTERN_RTL, $str))) { |
||
| 2139 | return $ta; |
||
| 2140 | } |
||
| 2141 | |||
| 2142 | // get number of chars |
||
| 2143 | $numchars = count($ta); |
||
| 2144 | |||
| 2145 | if ($forcertl == 'R') { |
||
| 2146 | $pel = 1; |
||
| 2147 | } elseif ($forcertl == 'L') { |
||
| 2148 | $pel = 0; |
||
| 2149 | } else { |
||
| 2150 | // P2. In each paragraph, find the first character of type L, AL, or R. |
||
| 2151 | // P3. If a character is found in P2 and it is of type AL or R, then set the paragraph embedding level to one; otherwise, set it to zero. |
||
| 2152 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2153 | $type = TCPDF_FONT_DATA::$uni_type[$ta[$i]]; |
||
| 2154 | if ($type == 'L') { |
||
| 2155 | $pel = 0; |
||
| 2156 | break; |
||
| 2157 | } elseif (($type == 'AL') or ($type == 'R')) { |
||
| 2158 | $pel = 1; |
||
| 2159 | break; |
||
| 2160 | } |
||
| 2161 | } |
||
| 2162 | } |
||
| 2163 | |||
| 2164 | // Current Embedding Level |
||
| 2165 | $cel = $pel; |
||
| 2166 | // directional override status |
||
| 2167 | $dos = 'N'; |
||
| 2168 | $remember = array(); |
||
| 2169 | // start-of-level-run |
||
| 2170 | $sor = $pel % 2 ? 'R' : 'L'; |
||
| 2171 | $eor = $sor; |
||
| 2172 | |||
| 2173 | // Array of characters data |
||
| 2174 | $chardata = array(); |
||
| 2175 | |||
| 2176 | // X1. Begin by setting the current embedding level to the paragraph embedding level. Set the directional override status to neutral. Process each character iteratively, applying rules X2 through X9. Only embedding levels from 0 to 61 are valid in this phase. |
||
| 2177 | // In the resolution of levels in rules I1 and I2, the maximum embedding level of 62 can be reached. |
||
| 2178 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2179 | if ($ta[$i] == TCPDF_FONT_DATA::$uni_RLE) { |
||
| 2180 | // X2. With each RLE, compute the least greater odd embedding level. |
||
| 2181 | // a. If this new level would be valid, then this embedding code is valid. Remember (push) the current embedding level and override status. Reset the current level to this new level, and reset the override status to neutral. |
||
| 2182 | // b. If the new level would not be valid, then this code is invalid. Do not change the current level or override status. |
||
| 2183 | $next_level = $cel + ($cel % 2) + 1; |
||
| 2184 | if ($next_level < 62) { |
||
| 2185 | $remember[] = array('num' => TCPDF_FONT_DATA::$uni_RLE, 'cel' => $cel, 'dos' => $dos); |
||
| 2186 | $cel = $next_level; |
||
| 2187 | $dos = 'N'; |
||
| 2188 | $sor = $eor; |
||
| 2189 | $eor = $cel % 2 ? 'R' : 'L'; |
||
| 2190 | } |
||
| 2191 | } elseif ($ta[$i] == TCPDF_FONT_DATA::$uni_LRE) { |
||
| 2192 | // X3. With each LRE, compute the least greater even embedding level. |
||
| 2193 | // a. If this new level would be valid, then this embedding code is valid. Remember (push) the current embedding level and override status. Reset the current level to this new level, and reset the override status to neutral. |
||
| 2194 | // b. If the new level would not be valid, then this code is invalid. Do not change the current level or override status. |
||
| 2195 | $next_level = $cel + 2 - ($cel % 2); |
||
| 2196 | if ($next_level < 62) { |
||
| 2197 | $remember[] = array('num' => TCPDF_FONT_DATA::$uni_LRE, 'cel' => $cel, 'dos' => $dos); |
||
| 2198 | $cel = $next_level; |
||
| 2199 | $dos = 'N'; |
||
| 2200 | $sor = $eor; |
||
| 2201 | $eor = $cel % 2 ? 'R' : 'L'; |
||
| 2202 | } |
||
| 2203 | } elseif ($ta[$i] == TCPDF_FONT_DATA::$uni_RLO) { |
||
| 2204 | // X4. With each RLO, compute the least greater odd embedding level. |
||
| 2205 | // a. If this new level would be valid, then this embedding code is valid. Remember (push) the current embedding level and override status. Reset the current level to this new level, and reset the override status to right-to-left. |
||
| 2206 | // b. If the new level would not be valid, then this code is invalid. Do not change the current level or override status. |
||
| 2207 | $next_level = $cel + ($cel % 2) + 1; |
||
| 2208 | if ($next_level < 62) { |
||
| 2209 | $remember[] = array('num' => TCPDF_FONT_DATA::$uni_RLO, 'cel' => $cel, 'dos' => $dos); |
||
| 2210 | $cel = $next_level; |
||
| 2211 | $dos = 'R'; |
||
| 2212 | $sor = $eor; |
||
| 2213 | $eor = $cel % 2 ? 'R' : 'L'; |
||
| 2214 | } |
||
| 2215 | } elseif ($ta[$i] == TCPDF_FONT_DATA::$uni_LRO) { |
||
| 2216 | // X5. With each LRO, compute the least greater even embedding level. |
||
| 2217 | // a. If this new level would be valid, then this embedding code is valid. Remember (push) the current embedding level and override status. Reset the current level to this new level, and reset the override status to left-to-right. |
||
| 2218 | // b. If the new level would not be valid, then this code is invalid. Do not change the current level or override status. |
||
| 2219 | $next_level = $cel + 2 - ($cel % 2); |
||
| 2220 | if ($next_level < 62) { |
||
| 2221 | $remember[] = array('num' => TCPDF_FONT_DATA::$uni_LRO, 'cel' => $cel, 'dos' => $dos); |
||
| 2222 | $cel = $next_level; |
||
| 2223 | $dos = 'L'; |
||
| 2224 | $sor = $eor; |
||
| 2225 | $eor = $cel % 2 ? 'R' : 'L'; |
||
| 2226 | } |
||
| 2227 | } elseif ($ta[$i] == TCPDF_FONT_DATA::$uni_PDF) { |
||
| 2228 | // X7. With each PDF, determine the matching embedding or override code. If there was a valid matching code, restore (pop) the last remembered (pushed) embedding level and directional override. |
||
| 2229 | if (count($remember)) { |
||
| 2230 | $last = count($remember) - 1; |
||
| 2231 | if ( |
||
| 2232 | ($remember[$last]['num'] == TCPDF_FONT_DATA::$uni_RLE) or |
||
| 2233 | ($remember[$last]['num'] == TCPDF_FONT_DATA::$uni_LRE) or |
||
| 2234 | ($remember[$last]['num'] == TCPDF_FONT_DATA::$uni_RLO) or |
||
| 2235 | ($remember[$last]['num'] == TCPDF_FONT_DATA::$uni_LRO) |
||
| 2236 | ) { |
||
| 2237 | $match = array_pop($remember); |
||
| 2238 | $cel = $match['cel']; |
||
| 2239 | $dos = $match['dos']; |
||
| 2240 | $sor = $eor; |
||
| 2241 | $eor = ($cel > $match['cel'] ? $cel : $match['cel']) % 2 ? 'R' : 'L'; |
||
| 2242 | } |
||
| 2243 | } |
||
| 2244 | } elseif ( |
||
| 2245 | ($ta[$i] != TCPDF_FONT_DATA::$uni_RLE) and |
||
| 2246 | ($ta[$i] != TCPDF_FONT_DATA::$uni_LRE) and |
||
| 2247 | ($ta[$i] != TCPDF_FONT_DATA::$uni_RLO) and |
||
| 2248 | ($ta[$i] != TCPDF_FONT_DATA::$uni_LRO) and |
||
| 2249 | ($ta[$i] != TCPDF_FONT_DATA::$uni_PDF) |
||
| 2250 | ) { |
||
| 2251 | // X6. For all types besides RLE, LRE, RLO, LRO, and PDF: |
||
| 2252 | // a. Set the level of the current character to the current embedding level. |
||
| 2253 | // b. Whenever the directional override status is not neutral, reset the current character type to the directional override status. |
||
| 2254 | if ($dos != 'N') { |
||
| 2255 | $chardir = $dos; |
||
| 2256 | } else { |
||
| 2257 | if (isset(TCPDF_FONT_DATA::$uni_type[$ta[$i]])) { |
||
| 2258 | $chardir = TCPDF_FONT_DATA::$uni_type[$ta[$i]]; |
||
| 2259 | } else { |
||
| 2260 | $chardir = 'L'; |
||
| 2261 | } |
||
| 2262 | } |
||
| 2263 | // stores string characters and other information |
||
| 2264 | $chardata[] = array('char' => $ta[$i], 'level' => $cel, 'type' => $chardir, 'sor' => $sor, 'eor' => $eor); |
||
| 2265 | } |
||
| 2266 | } // end for each char |
||
| 2267 | |||
| 2268 | // X8. All explicit directional embeddings and overrides are completely terminated at the end of each paragraph. Paragraph separators are not included in the embedding. |
||
| 2269 | // X9. Remove all RLE, LRE, RLO, LRO, PDF, and BN codes. |
||
| 2270 | // X10. The remaining rules are applied to each run of characters at the same level. For each run, determine the start-of-level-run (sor) and end-of-level-run (eor) type, either L or R. This depends on the higher of the two levels on either side of the boundary (at the start or end of the paragraph, the level of the 'other' run is the base embedding level). If the higher level is odd, the type is R; otherwise, it is L. |
||
| 2271 | |||
| 2272 | // 3.3.3 Resolving Weak Types |
||
| 2273 | // Weak types are now resolved one level run at a time. At level run boundaries where the type of the character on the other side of the boundary is required, the type assigned to sor or eor is used. |
||
| 2274 | // Nonspacing marks are now resolved based on the previous characters. |
||
| 2275 | $numchars = count($chardata); |
||
| 2276 | |||
| 2277 | // W1. Examine each nonspacing mark (NSM) in the level run, and change the type of the NSM to the type of the previous character. If the NSM is at the start of the level run, it will get the type of sor. |
||
| 2278 | $prevlevel = -1; // track level changes |
||
| 2279 | $levcount = 0; // counts consecutive chars at the same level |
||
| 2280 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2281 | if ($chardata[$i]['type'] == 'NSM') { |
||
| 2282 | if ($levcount) { |
||
| 2283 | $chardata[$i]['type'] = $chardata[$i]['sor']; |
||
| 2284 | } elseif ($i > 0) { |
||
| 2285 | $chardata[$i]['type'] = $chardata[($i - 1)]['type']; |
||
| 2286 | } |
||
| 2287 | } |
||
| 2288 | if ($chardata[$i]['level'] != $prevlevel) { |
||
| 2289 | $levcount = 0; |
||
| 2290 | } else { |
||
| 2291 | ++$levcount; |
||
| 2292 | } |
||
| 2293 | $prevlevel = $chardata[$i]['level']; |
||
| 2294 | } |
||
| 2295 | |||
| 2296 | // W2. Search backward from each instance of a European number until the first strong type (R, L, AL, or sor) is found. If an AL is found, change the type of the European number to Arabic number. |
||
| 2297 | $prevlevel = -1; |
||
| 2298 | $levcount = 0; |
||
| 2299 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2300 | if ($chardata[$i]['char'] == 'EN') { |
||
| 2301 | for ($j = $levcount; $j >= 0; $j--) { |
||
| 2302 | if ($chardata[$j]['type'] == 'AL') { |
||
| 2303 | $chardata[$i]['type'] = 'AN'; |
||
| 2304 | } elseif (($chardata[$j]['type'] == 'L') or ($chardata[$j]['type'] == 'R')) { |
||
| 2305 | break; |
||
| 2306 | } |
||
| 2307 | } |
||
| 2308 | } |
||
| 2309 | if ($chardata[$i]['level'] != $prevlevel) { |
||
| 2310 | $levcount = 0; |
||
| 2311 | } else { |
||
| 2312 | ++$levcount; |
||
| 2313 | } |
||
| 2314 | $prevlevel = $chardata[$i]['level']; |
||
| 2315 | } |
||
| 2316 | |||
| 2317 | // W3. Change all ALs to R. |
||
| 2318 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2319 | if ($chardata[$i]['type'] == 'AL') { |
||
| 2320 | $chardata[$i]['type'] = 'R'; |
||
| 2321 | } |
||
| 2322 | } |
||
| 2323 | |||
| 2324 | // W4. A single European separator between two European numbers changes to a European number. A single common separator between two numbers of the same type changes to that type. |
||
| 2325 | $prevlevel = -1; |
||
| 2326 | $levcount = 0; |
||
| 2327 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2328 | if (($levcount > 0) and (($i + 1) < $numchars) and ($chardata[($i + 1)]['level'] == $prevlevel)) { |
||
| 2329 | if (($chardata[$i]['type'] == 'ES') and ($chardata[($i - 1)]['type'] == 'EN') and ($chardata[($i + 1)]['type'] == 'EN')) { |
||
| 2330 | $chardata[$i]['type'] = 'EN'; |
||
| 2331 | } elseif (($chardata[$i]['type'] == 'CS') and ($chardata[($i - 1)]['type'] == 'EN') and ($chardata[($i + 1)]['type'] == 'EN')) { |
||
| 2332 | $chardata[$i]['type'] = 'EN'; |
||
| 2333 | } elseif (($chardata[$i]['type'] == 'CS') and ($chardata[($i - 1)]['type'] == 'AN') and ($chardata[($i + 1)]['type'] == 'AN')) { |
||
| 2334 | $chardata[$i]['type'] = 'AN'; |
||
| 2335 | } |
||
| 2336 | } |
||
| 2337 | if ($chardata[$i]['level'] != $prevlevel) { |
||
| 2338 | $levcount = 0; |
||
| 2339 | } else { |
||
| 2340 | ++$levcount; |
||
| 2341 | } |
||
| 2342 | $prevlevel = $chardata[$i]['level']; |
||
| 2343 | } |
||
| 2344 | |||
| 2345 | // W5. A sequence of European terminators adjacent to European numbers changes to all European numbers. |
||
| 2346 | $prevlevel = -1; |
||
| 2347 | $levcount = 0; |
||
| 2348 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2349 | if ($chardata[$i]['type'] == 'ET') { |
||
| 2350 | if (($levcount > 0) and ($chardata[($i - 1)]['type'] == 'EN')) { |
||
| 2351 | $chardata[$i]['type'] = 'EN'; |
||
| 2352 | } else { |
||
| 2353 | $j = $i + 1; |
||
| 2354 | while (($j < $numchars) and ($chardata[$j]['level'] == $prevlevel)) { |
||
| 2355 | if ($chardata[$j]['type'] == 'EN') { |
||
| 2356 | $chardata[$i]['type'] = 'EN'; |
||
| 2357 | break; |
||
| 2358 | } elseif ($chardata[$j]['type'] != 'ET') { |
||
| 2359 | break; |
||
| 2360 | } |
||
| 2361 | ++$j; |
||
| 2362 | } |
||
| 2363 | } |
||
| 2364 | } |
||
| 2365 | if ($chardata[$i]['level'] != $prevlevel) { |
||
| 2366 | $levcount = 0; |
||
| 2367 | } else { |
||
| 2368 | ++$levcount; |
||
| 2369 | } |
||
| 2370 | $prevlevel = $chardata[$i]['level']; |
||
| 2371 | } |
||
| 2372 | |||
| 2373 | // W6. Otherwise, separators and terminators change to Other Neutral. |
||
| 2374 | $prevlevel = -1; |
||
| 2375 | $levcount = 0; |
||
| 2376 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2377 | if (($chardata[$i]['type'] == 'ET') or ($chardata[$i]['type'] == 'ES') or ($chardata[$i]['type'] == 'CS')) { |
||
| 2378 | $chardata[$i]['type'] = 'ON'; |
||
| 2379 | } |
||
| 2380 | if ($chardata[$i]['level'] != $prevlevel) { |
||
| 2381 | $levcount = 0; |
||
| 2382 | } else { |
||
| 2383 | ++$levcount; |
||
| 2384 | } |
||
| 2385 | $prevlevel = $chardata[$i]['level']; |
||
| 2386 | } |
||
| 2387 | |||
| 2388 | //W7. Search backward from each instance of a European number until the first strong type (R, L, or sor) is found. If an L is found, then change the type of the European number to L. |
||
| 2389 | $prevlevel = -1; |
||
| 2390 | $levcount = 0; |
||
| 2391 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2392 | if ($chardata[$i]['char'] == 'EN') { |
||
| 2393 | for ($j = $levcount; $j >= 0; $j--) { |
||
| 2394 | if ($chardata[$j]['type'] == 'L') { |
||
| 2395 | $chardata[$i]['type'] = 'L'; |
||
| 2396 | } elseif ($chardata[$j]['type'] == 'R') { |
||
| 2397 | break; |
||
| 2398 | } |
||
| 2399 | } |
||
| 2400 | } |
||
| 2401 | if ($chardata[$i]['level'] != $prevlevel) { |
||
| 2402 | $levcount = 0; |
||
| 2403 | } else { |
||
| 2404 | ++$levcount; |
||
| 2405 | } |
||
| 2406 | $prevlevel = $chardata[$i]['level']; |
||
| 2407 | } |
||
| 2408 | |||
| 2409 | // N1. A sequence of neutrals takes the direction of the surrounding strong text if the text on both sides has the same direction. European and Arabic numbers act as if they were R in terms of their influence on neutrals. Start-of-level-run (sor) and end-of-level-run (eor) are used at level run boundaries. |
||
| 2410 | $prevlevel = -1; |
||
| 2411 | $levcount = 0; |
||
| 2412 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2413 | if (($levcount > 0) and (($i + 1) < $numchars) and ($chardata[($i + 1)]['level'] == $prevlevel)) { |
||
| 2414 | if (($chardata[$i]['type'] == 'N') and ($chardata[($i - 1)]['type'] == 'L') and ($chardata[($i + 1)]['type'] == 'L')) { |
||
| 2415 | $chardata[$i]['type'] = 'L'; |
||
| 2416 | } elseif ( |
||
| 2417 | ($chardata[$i]['type'] == 'N') and |
||
| 2418 | (($chardata[($i - 1)]['type'] == 'R') or ($chardata[($i - 1)]['type'] == 'EN') or ($chardata[($i - 1)]['type'] == 'AN')) and |
||
| 2419 | (($chardata[($i + 1)]['type'] == 'R') or ($chardata[($i + 1)]['type'] == 'EN') or ($chardata[($i + 1)]['type'] == 'AN')) |
||
| 2420 | ) { |
||
| 2421 | $chardata[$i]['type'] = 'R'; |
||
| 2422 | } elseif ($chardata[$i]['type'] == 'N') { |
||
| 2423 | // N2. Any remaining neutrals take the embedding direction |
||
| 2424 | $chardata[$i]['type'] = $chardata[$i]['sor']; |
||
| 2425 | } |
||
| 2426 | } elseif (($levcount == 0) and (($i + 1) < $numchars) and ($chardata[($i + 1)]['level'] == $prevlevel)) { |
||
| 2427 | // first char |
||
| 2428 | if (($chardata[$i]['type'] == 'N') and ($chardata[$i]['sor'] == 'L') and ($chardata[($i + 1)]['type'] == 'L')) { |
||
| 2429 | $chardata[$i]['type'] = 'L'; |
||
| 2430 | } elseif ( |
||
| 2431 | ($chardata[$i]['type'] == 'N') and |
||
| 2432 | (($chardata[$i]['sor'] == 'R') or ($chardata[$i]['sor'] == 'EN') or ($chardata[$i]['sor'] == 'AN')) and |
||
| 2433 | (($chardata[($i + 1)]['type'] == 'R') or ($chardata[($i + 1)]['type'] == 'EN') or ($chardata[($i + 1)]['type'] == 'AN')) |
||
| 2434 | ) { |
||
| 2435 | $chardata[$i]['type'] = 'R'; |
||
| 2436 | } elseif ($chardata[$i]['type'] == 'N') { |
||
| 2437 | // N2. Any remaining neutrals take the embedding direction |
||
| 2438 | $chardata[$i]['type'] = $chardata[$i]['sor']; |
||
| 2439 | } |
||
| 2440 | } elseif (($levcount > 0) and ((($i + 1) == $numchars) or (($i + 1) < $numchars) and ($chardata[($i + 1)]['level'] != $prevlevel))) { |
||
| 2441 | //last char |
||
| 2442 | if (($chardata[$i]['type'] == 'N') and ($chardata[($i - 1)]['type'] == 'L') and ($chardata[$i]['eor'] == 'L')) { |
||
| 2443 | $chardata[$i]['type'] = 'L'; |
||
| 2444 | } elseif ( |
||
| 2445 | ($chardata[$i]['type'] == 'N') and |
||
| 2446 | (($chardata[($i - 1)]['type'] == 'R') or ($chardata[($i - 1)]['type'] == 'EN') or ($chardata[($i - 1)]['type'] == 'AN')) and |
||
| 2447 | (($chardata[$i]['eor'] == 'R') or ($chardata[$i]['eor'] == 'EN') or ($chardata[$i]['eor'] == 'AN')) |
||
| 2448 | ) { |
||
| 2449 | $chardata[$i]['type'] = 'R'; |
||
| 2450 | } elseif ($chardata[$i]['type'] == 'N') { |
||
| 2451 | // N2. Any remaining neutrals take the embedding direction |
||
| 2452 | $chardata[$i]['type'] = $chardata[$i]['sor']; |
||
| 2453 | } |
||
| 2454 | } elseif ($chardata[$i]['type'] == 'N') { |
||
| 2455 | // N2. Any remaining neutrals take the embedding direction |
||
| 2456 | $chardata[$i]['type'] = $chardata[$i]['sor']; |
||
| 2457 | } |
||
| 2458 | if ($chardata[$i]['level'] != $prevlevel) { |
||
| 2459 | $levcount = 0; |
||
| 2460 | } else { |
||
| 2461 | ++$levcount; |
||
| 2462 | } |
||
| 2463 | $prevlevel = $chardata[$i]['level']; |
||
| 2464 | } |
||
| 2465 | |||
| 2466 | // I1. For all characters with an even (left-to-right) embedding direction, those of type R go up one level and those of type AN or EN go up two levels. |
||
| 2467 | // I2. For all characters with an odd (right-to-left) embedding direction, those of type L, EN or AN go up one level. |
||
| 2468 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2469 | $odd = $chardata[$i]['level'] % 2; |
||
| 2470 | if ($odd) { |
||
| 2471 | if (($chardata[$i]['type'] == 'L') or ($chardata[$i]['type'] == 'AN') or ($chardata[$i]['type'] == 'EN')) { |
||
| 2472 | $chardata[$i]['level'] += 1; |
||
| 2473 | } |
||
| 2474 | } else { |
||
| 2475 | if ($chardata[$i]['type'] == 'R') { |
||
| 2476 | $chardata[$i]['level'] += 1; |
||
| 2477 | } elseif (($chardata[$i]['type'] == 'AN') or ($chardata[$i]['type'] == 'EN')) { |
||
| 2478 | $chardata[$i]['level'] += 2; |
||
| 2479 | } |
||
| 2480 | } |
||
| 2481 | $maxlevel = max($chardata[$i]['level'], $maxlevel); |
||
| 2482 | } |
||
| 2483 | |||
| 2484 | // L1. On each line, reset the embedding level of the following characters to the paragraph embedding level: |
||
| 2485 | // 1. Segment separators, |
||
| 2486 | // 2. Paragraph separators, |
||
| 2487 | // 3. Any sequence of whitespace characters preceding a segment separator or paragraph separator, and |
||
| 2488 | // 4. Any sequence of white space characters at the end of the line. |
||
| 2489 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2490 | if (($chardata[$i]['type'] == 'B') or ($chardata[$i]['type'] == 'S')) { |
||
| 2491 | $chardata[$i]['level'] = $pel; |
||
| 2492 | } elseif ($chardata[$i]['type'] == 'WS') { |
||
| 2493 | $j = $i + 1; |
||
| 2494 | while ($j < $numchars) { |
||
| 2495 | if ( |
||
| 2496 | (($chardata[$j]['type'] == 'B') or ($chardata[$j]['type'] == 'S')) or |
||
| 2497 | (($j == ($numchars - 1)) and ($chardata[$j]['type'] == 'WS')) |
||
| 2498 | ) { |
||
| 2499 | $chardata[$i]['level'] = $pel; |
||
| 2500 | break; |
||
| 2501 | } elseif ($chardata[$j]['type'] != 'WS') { |
||
| 2502 | break; |
||
| 2503 | } |
||
| 2504 | ++$j; |
||
| 2505 | } |
||
| 2506 | } |
||
| 2507 | } |
||
| 2508 | |||
| 2509 | // Arabic Shaping |
||
| 2510 | // Cursively connected scripts, such as Arabic or Syriac, require the selection of positional character shapes that depend on adjacent characters. Shaping is logically applied after the Bidirectional Algorithm is used and is limited to characters within the same directional run. |
||
| 2511 | if ($arabic) { |
||
| 2512 | $endedletter = array(1569,1570,1571,1572,1573,1575,1577,1583,1584,1585,1586,1608,1688); |
||
| 2513 | $alfletter = array(1570,1571,1573,1575); |
||
| 2514 | $chardata2 = $chardata; |
||
| 2515 | $laaletter = false; |
||
| 2516 | $charAL = array(); |
||
| 2517 | $x = 0; |
||
| 2518 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2519 | if ((TCPDF_FONT_DATA::$uni_type[$chardata[$i]['char']] == 'AL') or ($chardata[$i]['char'] == 32) or ($chardata[$i]['char'] == 8204)) { |
||
| 2520 | $charAL[$x] = $chardata[$i]; |
||
| 2521 | $charAL[$x]['i'] = $i; |
||
| 2522 | $chardata[$i]['x'] = $x; |
||
| 2523 | ++$x; |
||
| 2524 | } |
||
| 2525 | } |
||
| 2526 | $numAL = $x; |
||
| 2527 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2528 | $thischar = $chardata[$i]; |
||
| 2529 | if ($i > 0) { |
||
| 2530 | $prevchar = $chardata[($i - 1)]; |
||
| 2531 | } else { |
||
| 2532 | $prevchar = false; |
||
| 2533 | } |
||
| 2534 | if (($i + 1) < $numchars) { |
||
| 2535 | $nextchar = $chardata[($i + 1)]; |
||
| 2536 | } else { |
||
| 2537 | $nextchar = false; |
||
| 2538 | } |
||
| 2539 | if (TCPDF_FONT_DATA::$uni_type[$thischar['char']] == 'AL') { |
||
| 2540 | $x = $thischar['x']; |
||
| 2541 | if ($x > 0) { |
||
| 2542 | $prevchar = $charAL[($x - 1)]; |
||
| 2543 | } else { |
||
| 2544 | $prevchar = false; |
||
| 2545 | } |
||
| 2546 | if (($x + 1) < $numAL) { |
||
| 2547 | $nextchar = $charAL[($x + 1)]; |
||
| 2548 | } else { |
||
| 2549 | $nextchar = false; |
||
| 2550 | } |
||
| 2551 | // if laa letter |
||
| 2552 | if (($prevchar !== false) and ($prevchar['char'] == 1604) and (in_array($thischar['char'], $alfletter))) { |
||
| 2553 | $arabicarr = TCPDF_FONT_DATA::$uni_laa_array; |
||
| 2554 | $laaletter = true; |
||
| 2555 | if ($x > 1) { |
||
| 2556 | $prevchar = $charAL[($x - 2)]; |
||
| 2557 | } else { |
||
| 2558 | $prevchar = false; |
||
| 2559 | } |
||
| 2560 | } else { |
||
| 2561 | $arabicarr = TCPDF_FONT_DATA::$uni_arabicsubst; |
||
| 2562 | $laaletter = false; |
||
| 2563 | } |
||
| 2564 | if ( |
||
| 2565 | ($prevchar !== false) and ($nextchar !== false) and |
||
| 2566 | ((TCPDF_FONT_DATA::$uni_type[$prevchar['char']] == 'AL') or (TCPDF_FONT_DATA::$uni_type[$prevchar['char']] == 'NSM')) and |
||
| 2567 | ((TCPDF_FONT_DATA::$uni_type[$nextchar['char']] == 'AL') or (TCPDF_FONT_DATA::$uni_type[$nextchar['char']] == 'NSM')) and |
||
| 2568 | ($prevchar['type'] == $thischar['type']) and |
||
| 2569 | ($nextchar['type'] == $thischar['type']) and |
||
| 2570 | ($nextchar['char'] != 1567) |
||
| 2571 | ) { |
||
| 2572 | if (in_array($prevchar['char'], $endedletter)) { |
||
| 2573 | if (isset($arabicarr[$thischar['char']][2])) { |
||
| 2574 | // initial |
||
| 2575 | $chardata2[$i]['char'] = $arabicarr[$thischar['char']][2]; |
||
| 2576 | } |
||
| 2577 | } else { |
||
| 2578 | if (isset($arabicarr[$thischar['char']][3])) { |
||
| 2579 | // medial |
||
| 2580 | $chardata2[$i]['char'] = $arabicarr[$thischar['char']][3]; |
||
| 2581 | } |
||
| 2582 | } |
||
| 2583 | } elseif ( |
||
| 2584 | ($nextchar !== false) and |
||
| 2585 | ((TCPDF_FONT_DATA::$uni_type[$nextchar['char']] == 'AL') or (TCPDF_FONT_DATA::$uni_type[$nextchar['char']] == 'NSM')) and |
||
| 2586 | ($nextchar['type'] == $thischar['type']) and |
||
| 2587 | ($nextchar['char'] != 1567) |
||
| 2588 | ) { |
||
| 2589 | if (isset($arabicarr[$chardata[$i]['char']][2])) { |
||
| 2590 | // initial |
||
| 2591 | $chardata2[$i]['char'] = $arabicarr[$thischar['char']][2]; |
||
| 2592 | } |
||
| 2593 | } elseif ( |
||
| 2594 | (($prevchar !== false) and |
||
| 2595 | ((TCPDF_FONT_DATA::$uni_type[$prevchar['char']] == 'AL') or (TCPDF_FONT_DATA::$uni_type[$prevchar['char']] == 'NSM')) and |
||
| 2596 | ($prevchar['type'] == $thischar['type'])) or |
||
| 2597 | (($nextchar !== false) and ($nextchar['char'] == 1567)) |
||
| 2598 | ) { |
||
| 2599 | // final |
||
| 2600 | if ( |
||
| 2601 | ($i > 1) and ($thischar['char'] == 1607) and |
||
| 2602 | ($chardata[$i - 1]['char'] == 1604) and |
||
| 2603 | ($chardata[$i - 2]['char'] == 1604) |
||
| 2604 | ) { |
||
| 2605 | //Allah Word |
||
| 2606 | // mark characters to delete with false |
||
| 2607 | $chardata2[$i - 2]['char'] = false; |
||
| 2608 | $chardata2[$i - 1]['char'] = false; |
||
| 2609 | $chardata2[$i]['char'] = 65010; |
||
| 2610 | } else { |
||
| 2611 | if (($prevchar !== false) and in_array($prevchar['char'], $endedletter)) { |
||
| 2612 | if (isset($arabicarr[$thischar['char']][0])) { |
||
| 2613 | // isolated |
||
| 2614 | $chardata2[$i]['char'] = $arabicarr[$thischar['char']][0]; |
||
| 2615 | } |
||
| 2616 | } else { |
||
| 2617 | if (isset($arabicarr[$thischar['char']][1])) { |
||
| 2618 | // final |
||
| 2619 | $chardata2[$i]['char'] = $arabicarr[$thischar['char']][1]; |
||
| 2620 | } |
||
| 2621 | } |
||
| 2622 | } |
||
| 2623 | } elseif (isset($arabicarr[$thischar['char']][0])) { |
||
| 2624 | // isolated |
||
| 2625 | $chardata2[$i]['char'] = $arabicarr[$thischar['char']][0]; |
||
| 2626 | } |
||
| 2627 | // if laa letter |
||
| 2628 | if ($laaletter) { |
||
| 2629 | // mark characters to delete with false |
||
| 2630 | $chardata2[($charAL[($x - 1)]['i'])]['char'] = false; |
||
| 2631 | } |
||
| 2632 | } // end if AL (Arabic Letter) |
||
| 2633 | } // end for each char |
||
| 2634 | /* |
||
| 2635 | * Combining characters that can occur with Arabic Shadda (0651 HEX, 1617 DEC) are replaced. |
||
| 2636 | * Putting the combining mark and shadda in the same glyph allows us to avoid the two marks overlapping each other in an illegible manner. |
||
| 2637 | */ |
||
| 2638 | for ($i = 0; $i < ($numchars - 1); ++$i) { |
||
| 2639 | if (($chardata2[$i]['char'] == 1617) and (isset(TCPDF_FONT_DATA::$uni_diacritics[($chardata2[$i + 1]['char'])]))) { |
||
| 2640 | // check if the subtitution font is defined on current font |
||
| 2641 | if (isset($currentfont['cw'][(TCPDF_FONT_DATA::$uni_diacritics[($chardata2[$i + 1]['char'])])])) { |
||
| 2642 | $chardata2[$i]['char'] = false; |
||
| 2643 | $chardata2[$i + 1]['char'] = TCPDF_FONT_DATA::$uni_diacritics[($chardata2[$i + 1]['char'])]; |
||
| 2644 | } |
||
| 2645 | } |
||
| 2646 | } |
||
| 2647 | // remove marked characters |
||
| 2648 | foreach ($chardata2 as $key => $value) { |
||
| 2649 | if ($value['char'] === false) { |
||
| 2650 | unset($chardata2[$key]); |
||
| 2651 | } |
||
| 2652 | } |
||
| 2653 | $chardata = array_values($chardata2); |
||
| 2654 | $numchars = count($chardata); |
||
| 2655 | unset($chardata2); |
||
| 2656 | unset($arabicarr); |
||
| 2657 | unset($laaletter); |
||
| 2658 | unset($charAL); |
||
| 2659 | } |
||
| 2660 | |||
| 2661 | // L2. From the highest level found in the text to the lowest odd level on each line, including intermediate levels not actually present in the text, reverse any contiguous sequence of characters that are at that level or higher. |
||
| 2662 | for ($j = $maxlevel; $j > 0; $j--) { |
||
| 2663 | $ordarray = array(); |
||
| 2664 | $revarr = array(); |
||
| 2665 | $onlevel = false; |
||
| 2666 | for ($i = 0; $i < $numchars; ++$i) { |
||
| 2667 | if ($chardata[$i]['level'] >= $j) { |
||
| 2668 | $onlevel = true; |
||
| 2669 | if (isset(TCPDF_FONT_DATA::$uni_mirror[$chardata[$i]['char']])) { |
||
| 2670 | // L4. A character is depicted by a mirrored glyph if and only if (a) the resolved directionality of that character is R, and (b) the Bidi_Mirrored property value of that character is true. |
||
| 2671 | $chardata[$i]['char'] = TCPDF_FONT_DATA::$uni_mirror[$chardata[$i]['char']]; |
||
| 2672 | } |
||
| 2673 | $revarr[] = $chardata[$i]; |
||
| 2674 | } else { |
||
| 2675 | if ($onlevel) { |
||
| 2676 | $revarr = array_reverse($revarr); |
||
| 2677 | $ordarray = array_merge($ordarray, $revarr); |
||
| 2678 | $revarr = array(); |
||
| 2679 | $onlevel = false; |
||
| 2680 | } |
||
| 2681 | $ordarray[] = $chardata[$i]; |
||
| 2682 | } |
||
| 2683 | } |
||
| 2684 | if ($onlevel) { |
||
| 2685 | $revarr = array_reverse($revarr); |
||
| 2686 | $ordarray = array_merge($ordarray, $revarr); |
||
| 2687 | } |
||
| 2688 | $chardata = $ordarray; |
||
| 2689 | } |
||
| 2690 | $ordarray = array(); |
||
| 2691 | foreach ($chardata as $cd) { |
||
| 2692 | $ordarray[] = $cd['char']; |
||
| 2693 | // store char values for subsetting |
||
| 2694 | $currentfont['subsetchars'][$cd['char']] = true; |
||
| 2695 | } |
||
| 2696 | return $ordarray; |
||
| 2697 | } |
||
| 2703 |
