/framework/vendor/zend/Zend/Pdf/Resource/Font/CidFont.php
PHP | 490 lines | 190 code | 54 blank | 246 comment | 32 complexity | e40cda733c9fa1c9c5c7d3f1ff380272 MD5 | raw file
1<?php 2/** 3 * Zend Framework 4 * 5 * LICENSE 6 * 7 * This source file is subject to the new BSD license that is bundled 8 * with this package in the file LICENSE.txt. 9 * It is also available through the world-wide-web at this URL: 10 * http://framework.zend.com/license/new-bsd 11 * If you did not receive a copy of the license and are unable to 12 * obtain it through the world-wide-web, please send an email 13 * to license@zend.com so we can send you a copy immediately. 14 * 15 * @category Zend 16 * @package Zend_Pdf 17 * @subpackage Fonts 18 * @copyright Copyright (c) 2005-2010 Zend Technologies USA Inc. (http://www.zend.com) 19 * @license http://framework.zend.com/license/new-bsd New BSD License 20 * @version $Id: CidFont.php 20096 2010-01-06 02:05:09Z bkarwin $ 21 */ 22 23/** Internally used classes */ 24require_once 'Zend/Pdf/Element/Array.php'; 25require_once 'Zend/Pdf/Element/Dictionary.php'; 26require_once 'Zend/Pdf/Element/Name.php'; 27require_once 'Zend/Pdf/Element/Numeric.php'; 28require_once 'Zend/Pdf/Element/String.php'; 29 30 31/** Zend_Pdf_Resource_Font */ 32require_once 'Zend/Pdf/Resource/Font.php'; 33 34/** 35 * Adobe PDF CIDFont font object implementation 36 * 37 * A CIDFont program contains glyph descriptions that are accessed using a CID as 38 * the character selector. There are two types of CIDFont. A Type 0 CIDFont contains 39 * glyph descriptions based on AdobeĆ¢€™s Type 1 font format, whereas those in a 40 * Type 2 CIDFont are based on the TrueType font format. 41 * 42 * A CIDFont dictionary is a PDF object that contains information about a CIDFont program. 43 * Although its Type value is Font, a CIDFont is not actually a font. It does not have an Encoding 44 * entry, it cannot be listed in the Font subdictionary of a resource dictionary, and it cannot be 45 * used as the operand of the Tf operator. It is used only as a descendant of a Type 0 font. 46 * The CMap in the Type 0 font is what defines the encoding that maps character codes to CIDs 47 * in the CIDFont. 48 * 49 * Font objects should be normally be obtained from the factory methods 50 * {@link Zend_Pdf_Font::fontWithName} and {@link Zend_Pdf_Font::fontWithPath}. 51 * 52 * @package Zend_Pdf 53 * @subpackage Fonts 54 * @copyright Copyright (c) 2005-2010 Zend Technologies USA Inc. (http://www.zend.com) 55 * @license http://framework.zend.com/license/new-bsd New BSD License 56 */ 57abstract class Zend_Pdf_Resource_Font_CidFont extends Zend_Pdf_Resource_Font 58{ 59 /** 60 * Object representing the font's cmap (character to glyph map). 61 * @var Zend_Pdf_Cmap 62 */ 63 protected $_cmap = null; 64 65 /** 66 * Array containing the widths of each character that have entries in used character map. 67 * 68 * @var array 69 */ 70 protected $_charWidths = null; 71 72 /** 73 * Width for characters missed in the font 74 * 75 * @var integer 76 */ 77 protected $_missingCharWidth = 0; 78 79 80 /** 81 * Object constructor 82 * 83 * @param Zend_Pdf_FileParser_Font_OpenType $fontParser Font parser object 84 * containing OpenType file. 85 * @param integer $embeddingOptions Options for font embedding. 86 * @throws Zend_Pdf_Exception 87 */ 88 public function __construct(Zend_Pdf_FileParser_Font_OpenType $fontParser) 89 { 90 parent::__construct(); 91 92 $fontParser->parse(); 93 94 95 /* Object properties */ 96 97 $this->_fontNames = $fontParser->names; 98 99 $this->_isBold = $fontParser->isBold; 100 $this->_isItalic = $fontParser->isItalic; 101 $this->_isMonospaced = $fontParser->isMonospaced; 102 103 $this->_underlinePosition = $fontParser->underlinePosition; 104 $this->_underlineThickness = $fontParser->underlineThickness; 105 $this->_strikePosition = $fontParser->strikePosition; 106 $this->_strikeThickness = $fontParser->strikeThickness; 107 108 $this->_unitsPerEm = $fontParser->unitsPerEm; 109 110 $this->_ascent = $fontParser->ascent; 111 $this->_descent = $fontParser->descent; 112 $this->_lineGap = $fontParser->lineGap; 113 114 115 $this->_cmap = $fontParser->cmap; 116 117 118 /* Resource dictionary */ 119 120 $baseFont = $this->getFontName(Zend_Pdf_Font::NAME_POSTSCRIPT, 'en', 'UTF-8'); 121 $this->_resource->BaseFont = new Zend_Pdf_Element_Name($baseFont); 122 123 124 /** 125 * Prepare widths array. 126 */ 127 /* Constract characters widths array using font CMap and glyphs widths array */ 128 $glyphWidths = $fontParser->glyphWidths; 129 $charGlyphs = $this->_cmap->getCoveredCharactersGlyphs(); 130 $charWidths = array(); 131 foreach ($charGlyphs as $charCode => $glyph) { 132 $charWidths[$charCode] = $glyphWidths[$glyph]; 133 } 134 $this->_charWidths = $charWidths; 135 $this->_missingCharWidth = $glyphWidths[0]; 136 137 /* Width array optimization. Step1: extract default value */ 138 $widthFrequencies = array_count_values($charWidths); 139 $defaultWidth = null; 140 $defaultWidthFrequency = -1; 141 foreach ($widthFrequencies as $width => $frequency) { 142 if ($frequency > $defaultWidthFrequency) { 143 $defaultWidth = $width; 144 $defaultWidthFrequency = $frequency; 145 } 146 } 147 148 // Store default value in the font dictionary 149 $this->_resource->DW = new Zend_Pdf_Element_Numeric($this->toEmSpace($defaultWidth)); 150 151 // Remove characters which corresponds to default width from the widths array 152 $defWidthChars = array_keys($charWidths, $defaultWidth); 153 foreach ($defWidthChars as $charCode) { 154 unset($charWidths[$charCode]); 155 } 156 157 // Order cheracter widths aray by character codes 158 ksort($charWidths, SORT_NUMERIC); 159 160 /* Width array optimization. Step2: Compact character codes sequences */ 161 $lastCharCode = -1; 162 $widthsSequences = array(); 163 foreach ($charWidths as $charCode => $width) { 164 if ($lastCharCode == -1) { 165 $charCodesSequense = array(); 166 $sequenceStartCode = $charCode; 167 } else if ($charCode != $lastCharCode + 1) { 168 // New chracters sequence detected 169 $widthsSequences[$sequenceStartCode] = $charCodesSequense; 170 $charCodesSequense = array(); 171 $sequenceStartCode = $charCode; 172 } 173 $charCodesSequense[] = $width; 174 $lastCharCode = $charCode; 175 } 176 // Save last sequence, if widths array is not empty (it may happens for monospaced fonts) 177 if (count($charWidths) != 0) { 178 $widthsSequences[$sequenceStartCode] = $charCodesSequense; 179 } 180 181 $pdfCharsWidths = array(); 182 foreach ($widthsSequences as $startCode => $widthsSequence) { 183 /* Width array optimization. Step3: Compact widths sequences */ 184 $pdfWidths = array(); 185 $lastWidth = -1; 186 $widthsInSequence = 0; 187 foreach ($widthsSequence as $width) { 188 if ($lastWidth != $width) { 189 // New width is detected 190 if ($widthsInSequence != 0) { 191 // Previous width value was a part of the widths sequence. Save it as 'c_1st c_last w'. 192 $pdfCharsWidths[] = new Zend_Pdf_Element_Numeric($startCode); // First character code 193 $pdfCharsWidths[] = new Zend_Pdf_Element_Numeric($startCode + $widthsInSequence - 1); // Last character code 194 $pdfCharsWidths[] = new Zend_Pdf_Element_Numeric($this->toEmSpace($lastWidth)); // Width 195 196 // Reset widths sequence 197 $startCode = $startCode + $widthsInSequence; 198 $widthsInSequence = 0; 199 } 200 201 // Collect new width 202 $pdfWidths[] = new Zend_Pdf_Element_Numeric($this->toEmSpace($width)); 203 204 $lastWidth = $width; 205 } else { 206 // Width is equal to previous 207 if (count($pdfWidths) != 0) { 208 // We already have some widths collected 209 // So, we've just detected new widths sequence 210 211 // Remove last element from widths list, since it's a part of widths sequence 212 array_pop($pdfWidths); 213 214 // and write the rest if it's not empty 215 if (count($pdfWidths) != 0) { 216 // Save it as 'c_1st [w1 w2 ... wn]'. 217 $pdfCharsWidths[] = new Zend_Pdf_Element_Numeric($startCode); // First character code 218 $pdfCharsWidths[] = new Zend_Pdf_Element_Array($pdfWidths); // Widths array 219 220 // Reset widths collection 221 $startCode += count($pdfWidths); 222 $pdfWidths = array(); 223 } 224 225 $widthsInSequence = 2; 226 } else { 227 // Continue widths sequence 228 $widthsInSequence++; 229 } 230 } 231 } 232 233 // Check if we have widths collection or widths sequence to wite it down 234 if (count($pdfWidths) != 0) { 235 // We have some widths collected 236 // Save it as 'c_1st [w1 w2 ... wn]'. 237 $pdfCharsWidths[] = new Zend_Pdf_Element_Numeric($startCode); // First character code 238 $pdfCharsWidths[] = new Zend_Pdf_Element_Array($pdfWidths); // Widths array 239 } else if ($widthsInSequence != 0){ 240 // We have widths sequence 241 // Save it as 'c_1st c_last w'. 242 $pdfCharsWidths[] = new Zend_Pdf_Element_Numeric($startCode); // First character code 243 $pdfCharsWidths[] = new Zend_Pdf_Element_Numeric($startCode + $widthsInSequence - 1); // Last character code 244 $pdfCharsWidths[] = new Zend_Pdf_Element_Numeric($this->toEmSpace($lastWidth)); // Width 245 } 246 } 247 248 /* Create the Zend_Pdf_Element_Array object and add it to the font's 249 * object factory and resource dictionary. 250 */ 251 $widthsArrayElement = new Zend_Pdf_Element_Array($pdfCharsWidths); 252 $widthsObject = $this->_objectFactory->newObject($widthsArrayElement); 253 $this->_resource->W = $widthsObject; 254 255 256 /* CIDSystemInfo dictionary */ 257 $cidSystemInfo = new Zend_Pdf_Element_Dictionary(); 258 $cidSystemInfo->Registry = new Zend_Pdf_Element_String('Adobe'); 259 $cidSystemInfo->Ordering = new Zend_Pdf_Element_String('UCS'); 260 $cidSystemInfo->Supplement = new Zend_Pdf_Element_Numeric(0); 261 $cidSystemInfoObject = $this->_objectFactory->newObject($cidSystemInfo); 262 $this->_resource->CIDSystemInfo = $cidSystemInfoObject; 263 } 264 265 266 267 /** 268 * Returns an array of glyph numbers corresponding to the Unicode characters. 269 * 270 * If a particular character doesn't exist in this font, the special 'missing 271 * character glyph' will be substituted. 272 * 273 * See also {@link glyphNumberForCharacter()}. 274 * 275 * @param array $characterCodes Array of Unicode character codes (code points). 276 * @return array Array of glyph numbers. 277 */ 278 public function glyphNumbersForCharacters($characterCodes) 279 { 280 /** 281 * CIDFont object is not actually a font. It does not have an Encoding entry, 282 * it cannot be listed in the Font subdictionary of a resource dictionary, and 283 * it cannot be used as the operand of the Tf operator. 284 * 285 * Throw an exception. 286 */ 287 require_once 'Zend/Pdf/Exception.php'; 288 throw new Zend_Pdf_Exception('CIDFont PDF objects could not be used as the operand of the text drawing operators'); 289 } 290 291 /** 292 * Returns the glyph number corresponding to the Unicode character. 293 * 294 * If a particular character doesn't exist in this font, the special 'missing 295 * character glyph' will be substituted. 296 * 297 * See also {@link glyphNumbersForCharacters()} which is optimized for bulk 298 * operations. 299 * 300 * @param integer $characterCode Unicode character code (code point). 301 * @return integer Glyph number. 302 */ 303 public function glyphNumberForCharacter($characterCode) 304 { 305 /** 306 * CIDFont object is not actually a font. It does not have an Encoding entry, 307 * it cannot be listed in the Font subdictionary of a resource dictionary, and 308 * it cannot be used as the operand of the Tf operator. 309 * 310 * Throw an exception. 311 */ 312 require_once 'Zend/Pdf/Exception.php'; 313 throw new Zend_Pdf_Exception('CIDFont PDF objects could not be used as the operand of the text drawing operators'); 314 } 315 316 317 /** 318 * Returns a number between 0 and 1 inclusive that indicates the percentage 319 * of characters in the string which are covered by glyphs in this font. 320 * 321 * Since no one font will contain glyphs for the entire Unicode character 322 * range, this method can be used to help locate a suitable font when the 323 * actual contents of the string are not known. 324 * 325 * Note that some fonts lie about the characters they support. Additionally, 326 * fonts don't usually contain glyphs for control characters such as tabs 327 * and line breaks, so it is rare that you will get back a full 1.0 score. 328 * The resulting value should be considered informational only. 329 * 330 * @param string $string 331 * @param string $charEncoding (optional) Character encoding of source text. 332 * If omitted, uses 'current locale'. 333 * @return float 334 */ 335 public function getCoveredPercentage($string, $charEncoding = '') 336 { 337 /* Convert the string to UTF-16BE encoding so we can match the string's 338 * character codes to those found in the cmap. 339 */ 340 if ($charEncoding != 'UTF-16BE') { 341 $string = iconv($charEncoding, 'UTF-16BE', $string); 342 } 343 344 $charCount = iconv_strlen($string, 'UTF-16BE'); 345 if ($charCount == 0) { 346 return 0; 347 } 348 349 /* Calculate the score by doing a lookup for each character. 350 */ 351 $score = 0; 352 $maxIndex = strlen($string); 353 for ($i = 0; $i < $maxIndex; $i++) { 354 /** 355 * @todo Properly handle characters encoded as surrogate pairs. 356 */ 357 $charCode = (ord($string[$i]) << 8) | ord($string[++$i]); 358 /* This could probably be optimized a bit with a binary search... 359 */ 360 if (isset($this->_charWidths[$charCode])) { 361 $score++; 362 } 363 } 364 return $score / $charCount; 365 } 366 367 /** 368 * Returns the widths of the Chars. 369 * 370 * The widths are expressed in the font's glyph space. You are responsible 371 * for converting to user space as necessary. See {@link unitsPerEm()}. 372 * 373 * See also {@link widthForChar()}. 374 * 375 * @param array &$glyphNumbers Array of glyph numbers. 376 * @return array Array of glyph widths (integers). 377 */ 378 public function widthsForChars($charCodes) 379 { 380 $widths = array(); 381 foreach ($charCodes as $key => $charCode) { 382 if (!isset($this->_charWidths[$charCode])) { 383 $widths[$key] = $this->_missingCharWidth; 384 } else { 385 $widths[$key] = $this->_charWidths[$charCode]; 386 } 387 } 388 return $widths; 389 } 390 391 /** 392 * Returns the width of the character. 393 * 394 * Like {@link widthsForChars()} but used for one char at a time. 395 * 396 * @param integer $charCode 397 * @return integer 398 */ 399 public function widthForChar($charCode) 400 { 401 if (!isset($this->_charWidths[$charCode])) { 402 return $this->_missingCharWidth; 403 } 404 return $this->_charWidths[$charCode]; 405 } 406 407 /** 408 * Returns the widths of the glyphs. 409 * 410 * @param array &$glyphNumbers Array of glyph numbers. 411 * @return array Array of glyph widths (integers). 412 * @throws Zend_Pdf_Exception 413 */ 414 public function widthsForGlyphs($glyphNumbers) 415 { 416 /** 417 * CIDFont object is not actually a font. It does not have an Encoding entry, 418 * it cannot be listed in the Font subdictionary of a resource dictionary, and 419 * it cannot be used as the operand of the Tf operator. 420 * 421 * Throw an exception. 422 */ 423 require_once 'Zend/Pdf/Exception.php'; 424 throw new Zend_Pdf_Exception('CIDFont PDF objects could not be used as the operand of the text drawing operators'); 425 } 426 427 /** 428 * Returns the width of the glyph. 429 * 430 * Like {@link widthsForGlyphs()} but used for one glyph at a time. 431 * 432 * @param integer $glyphNumber 433 * @return integer 434 * @throws Zend_Pdf_Exception 435 */ 436 public function widthForGlyph($glyphNumber) 437 { 438 /** 439 * CIDFont object is not actually a font. It does not have an Encoding entry, 440 * it cannot be listed in the Font subdictionary of a resource dictionary, and 441 * it cannot be used as the operand of the Tf operator. 442 * 443 * Throw an exception. 444 */ 445 require_once 'Zend/Pdf/Exception.php'; 446 throw new Zend_Pdf_Exception('CIDFont PDF objects could not be used as the operand of the text drawing operators'); 447 } 448 449 /** 450 * Convert string to the font encoding. 451 * 452 * @param string $string 453 * @param string $charEncoding Character encoding of source text. 454 * @return string 455 * @throws Zend_Pdf_Exception 456 * */ 457 public function encodeString($string, $charEncoding) 458 { 459 /** 460 * CIDFont object is not actually a font. It does not have an Encoding entry, 461 * it cannot be listed in the Font subdictionary of a resource dictionary, and 462 * it cannot be used as the operand of the Tf operator. 463 * 464 * Throw an exception. 465 */ 466 require_once 'Zend/Pdf/Exception.php'; 467 throw new Zend_Pdf_Exception('CIDFont PDF objects could not be used as the operand of the text drawing operators'); 468 } 469 470 /** 471 * Convert string from the font encoding. 472 * 473 * @param string $string 474 * @param string $charEncoding Character encoding of resulting text. 475 * @return string 476 * @throws Zend_Pdf_Exception 477 */ 478 public function decodeString($string, $charEncoding) 479 { 480 /** 481 * CIDFont object is not actually a font. It does not have an Encoding entry, 482 * it cannot be listed in the Font subdictionary of a resource dictionary, and 483 * it cannot be used as the operand of the Tf operator. 484 * 485 * Throw an exception. 486 */ 487 require_once 'Zend/Pdf/Exception.php'; 488 throw new Zend_Pdf_Exception('CIDFont PDF objects could not be used as the operand of the text drawing operators'); 489 } 490}