pdf-print / mpdf / mpdfi / pdf_parser.php

pdf_parser.php in PDF & Print by BestWebSoft – WordPress Posts and Pages PDF Generator Plugin 1.5, at mpdf/mpdfi/pdf_parser.php

690 lines 19.4 KB

Raw Download Zip

1	<?php
2	//
3	// FPDI - Version 1.2
4	//
5	// Copyright 2004-2007 Setasign - Jan Slabon
6	//
7	// Licensed under the Apache License, Version 2.0 (the "License");
8	// you may not use this file except in compliance with the License.
9	// You may obtain a copy of the License at
10	//
11	// http://www.apache.org/licenses/LICENSE-2.0
12	//
13	// Unless required by applicable law or agreed to in writing, software
14	// distributed under the License is distributed on an "AS IS" BASIS,
15	// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16	// See the License for the specific language governing permissions and
17	// limitations under the License.
18	//
19
20	if (!defined ('PDF_TYPE_NULL'))
21	define ('PDF_TYPE_NULL', 0);
22	if (!defined ('PDF_TYPE_NUMERIC'))
23	define ('PDF_TYPE_NUMERIC', 1);
24	if (!defined ('PDF_TYPE_TOKEN'))
25	define ('PDF_TYPE_TOKEN', 2);
26	if (!defined ('PDF_TYPE_HEX'))
27	define ('PDF_TYPE_HEX', 3);
28	if (!defined ('PDF_TYPE_STRING'))
29	define ('PDF_TYPE_STRING', 4);
30	if (!defined ('PDF_TYPE_DICTIONARY'))
31	define ('PDF_TYPE_DICTIONARY', 5);
32	if (!defined ('PDF_TYPE_ARRAY'))
33	define ('PDF_TYPE_ARRAY', 6);
34	if (!defined ('PDF_TYPE_OBJDEC'))
35	define ('PDF_TYPE_OBJDEC', 7);
36	if (!defined ('PDF_TYPE_OBJREF'))
37	define ('PDF_TYPE_OBJREF', 8);
38	if (!defined ('PDF_TYPE_OBJECT'))
39	define ('PDF_TYPE_OBJECT', 9);
40	if (!defined ('PDF_TYPE_STREAM'))
41	define ('PDF_TYPE_STREAM', 10);
42
43
44	class pdf_parser {
45
46	/**
47	* Filename
48	* @var string
49	*/
50	var $filename;
51
52	/**
53	* File resource
54	* @var resource
55	*/
56	var $f;
57
58	/**
59	* PDF Context
60	* @var object pdf_context-Instance
61	*/
62	var $c;
63
64	/**
65	* xref-Data
66	* @var array
67	*/
68	var $xref;
69
70	/**
71	* root-Object
72	* @var array
73	*/
74	var $root;
75
76	// mPDF 4.0 Added flag to show success on loading file
77	var $success;
78	var $errormsg;
79
80	/**
81	* Constructor
82	*
83	* @param string $filename Source-Filename
84	*/
85	function pdf_parser($filename) {
86	$this->filename = $filename;
87	// mPDF 4.0
88	$this->success = true;
89
90	$this->f = @fopen($this->filename, "rb");
91
92	if (!$this->f) {
93	$this->success = false;
94	$this->errormsg = sprintf("Cannot open %s !", $filename);
95	return false;
96	}
97	// mPDF 5.0 Removed pass by reference =&
98	$this->c = new pdf_context($this->f);
99	// Read xref-Data
100	$offset = $this->pdf_find_xref();
101	if ($offset===false) {
102	$this->success = false;
103	$this->errormsg = sprintf("Cannot open %s !", $filename);
104	return false;
105	}
106	$this->pdf_read_xref($this->xref, $offset);
107	if ($this->success == false) { return false; }
108
109	// Check for Encryption
110	$this->getEncryption();
111	if ($this->success == false) { return false; }
112
113	// Read root
114	$this->pdf_read_root();
115	if ($this->success == false) { return false; }
116	}
117
118	/**
119	* Close the opened file
120	*/
121	function closeFile() {
122	if (isset($this->f)) {
123	fclose($this->f);
124	unset($this->f);
125	}
126	}
127
128	/**
129	* Print Error and die
130	*
131	* @param string $msg Error-Message
132	*/
133	function error($msg) {
134	die("<b>PDF-Parser Error:</b> ".$msg);
135	}
136
137	/**
138	* Check Trailer for Encryption
139	*/
140	function getEncryption() {
141	if (isset($this->xref['trailer'][1]['/Encrypt'])) {
142	// mPDF 4.0
143	$this->success = false;
144	$this->errormsg = sprintf("File is encrypted!");
145	return false;
146	}
147	}
148
149	/**
150	* Find/Return /Root
151	*
152	* @return array
153	*/
154	function pdf_find_root() {
155	if ($this->xref['trailer'][1]['/Root'][0] != PDF_TYPE_OBJREF) {
156	// mPDF 4.0
157	$this->success = false;
158	$this->errormsg = sprintf("Wrong Type of Root-Element! Must be an indirect reference");
159	return false;
160	}
161	return $this->xref['trailer'][1]['/Root'];
162	}
163
164	/**
165	* Read the /Root
166	*/
167	function pdf_read_root() {
168	// read root
169	$root = $this->pdf_find_root();
170	if ($root ===false) {
171	$this->success = false;
172	return false;
173	}
174	$this->root = $this->pdf_resolve_object($this->c, $root);
175	}
176
177	/**
178	* Find the xref-Table
179	*/
180	function pdf_find_xref() {
181	fseek ($this->f, -min(filesize($this->filename),1500), SEEK_END);
182	$data = fread($this->f, 1500);
183
184	$pos = strlen($data) - strpos(strrev($data), strrev('startxref'));
185	$data = substr($data, $pos);
186
187	if (!preg_match('/\s(\d+).$/s', $data, $matches)) {
188	// mPDF 4.0
189	$this->success = false;
190	$this->errormsg = sprintf("Unable to find pointer to xref table");
191	return false;
192	}
193
194	return (int) $matches[1];
195	}
196
197	/**
198	* Read xref-table
199	*
200	* @param array $result Array of xref-table
201	* @param integer $offset of xref-table
202	* @param integer $start start-position in xref-table
203	* @param integer $end end-position in xref-table
204	*/
205	function pdf_read_xref(&$result, $offset, $start = null, $end = null) {
206	if (is_null ($start) \|\| is_null ($end)) {
207	fseek($this->f, $o_pos = $offset);
208	$data = trim(fgets($this->f,1024));
209
210	if (strlen($data) == 0)
211	$data = trim(fgets($this->f,1024));
212
213	if ($data !== 'xref') {
214	fseek($this->f, $o_pos);
215	$data = trim(_fgets($this->f, true));
216	if ($data !== 'xref') {
217	if (preg_match('/(.xref)(.)/m', $data, $m)) { // xref 0 128 - in one line
218	fseek($this->f, $o_pos+strlen($m[1]));
219	} elseif (preg_match('/(x\|r\|e\|f)+/', $data, $m)) { // correct invalid xref-pointer
220	$tmpOffset = $offset-4+strlen($m[0]);
221	$this->pdf_read_xref($result, $tmpOffset, $start, $end);
222	return;
223	} else {
224	// mPDF 4.0
225	$this->success = false;
226	$this->errormsg = sprintf("Unable to find xref table - Maybe a Problem with 'auto_detect_line_endings'");
227	return;
228	}
229	}
230	}
231
232	$o_pos = ftell($this->f);
233	$data = explode(' ', trim(fgets($this->f,1024)));
234	if (count($data) != 2) {
235	fseek($this->f, $o_pos);
236	$data = explode(' ', trim(_fgets($this->f, true)));
237
238	if (count($data) != 2) {
239	if (count($data) > 2) { // no lineending
240	$n_pos = $o_pos+strlen($data[0])+strlen($data[1])+2;
241	fseek($this->f, $n_pos);
242	} else {
243	// mPDF 4.0
244	$this->success = false;
245	$this->errormsg = sprintf("Unexpected header in xref table");
246	return;
247	}
248	}
249	}
250	$start = $data[0];
251	$end = $start + $data[1];
252	}
253
254	if (!isset($result['xref_location'])) {
255	$result['xref_location'] = $offset;
256	}
257
258	if (!isset($result['max_object']) \|\| $end > $result['max_object']) {
259	$result['max_object'] = $end;
260	}
261
262	for (; $start < $end; $start++) {
263	$data = ltrim(fread($this->f, 20)); // Spezifications says: 20 bytes including newlines
264	$offset = substr($data, 0, 10);
265	$generation = substr($data, 11, 5);
266
267	if (!isset ($result['xref'][$start][(int) $generation])) {
268	$result['xref'][$start][(int) $generation] = (int) $offset;
269	}
270	}
271
272	$o_pos = ftell($this->f);
273	$data = fgets($this->f,1024);
274	if (strlen(trim($data)) == 0)
275	$data = fgets($this->f, 1024);
276
277	if (preg_match("/trailer/",$data)) {
278	if (preg_match("/(.trailer[ \n\r])/",$data,$m)) {
279	fseek($this->f, $o_pos+strlen($m[1]));
280	}
281
282	// mPDF 5.0 Removed pass by reference =&
283	$c = new pdf_context($this->f);
284	$trailer = $this->pdf_read_value($c);
285
286	if (isset($trailer[1]['/Prev'])) {
287	$this->pdf_read_xref($result, $trailer[1]['/Prev'][1]);
288	$result['trailer'][1] = array_merge($result['trailer'][1], $trailer[1]);
289	} else {
290	$result['trailer'] = $trailer;
291	}
292	} else {
293	$data = explode(' ', trim($data));
294
295	if (count($data) != 2) {
296	fseek($this->f, $o_pos);
297	$data = explode(' ', trim (_fgets ($this->f, true)));
298
299	if (count($data) != 2) {
300	// mPDF 4.0
301	$this->success = false;
302	$this->errormsg = sprintf("Unexpected data in xref table");
303	return;
304	}
305	}
306
307	$this->pdf_read_xref($result, null, (int) $data[0], (int) $data[0] + (int) $data[1]);
308	}
309	}
310
311
312	/**
313	* Reads an Value
314	*
315	* @param object $c pdf_context
316	* @param string $token a Token
317	* @return mixed
318	*/
319	function pdf_read_value(&$c, $token = null) {
320	if (is_null($token)) {
321	$token = $this->pdf_read_token($c);
322	}
323
324	if ($token === false) {
325	return false;
326	}
327
328	switch ($token) {
329	case '<':
330	// This is a hex string.
331	// Read the value, then the terminator
332
333	$pos = $c->offset;
334
335	while(1) {
336
337	$match = strpos ($c->buffer, '>', $pos);
338
339	// If you can't find it, try
340	// reading more data from the stream
341
342	if ($match === false) {
343	if (!$c->increase_length()) {
344	return false;
345	} else {
346	continue;
347	}
348	}
349
350	$result = substr ($c->buffer, $c->offset, $match - $c->offset);
351	$c->offset = $match+1;
352
353	return array (PDF_TYPE_HEX, $result);
354	}
355
356	break;
357	case '<<':
358	// This is a dictionary.
359
360	$result = array();
361
362	// Recurse into this function until we reach
363	// the end of the dictionary.
364	while (($key = $this->pdf_read_token($c)) !== '>>') {
365	if ($key === false) {
366	return false;
367	}
368
369	if (($value = $this->pdf_read_value($c)) === false) {
370	return false;
371	}
372	$result[$key] = $value;
373	}
374
375	return array (PDF_TYPE_DICTIONARY, $result);
376
377	case '[':
378	// This is an array.
379
380	$result = array();
381
382	// Recurse into this function until we reach
383	// the end of the array.
384	while (($token = $this->pdf_read_token($c)) !== ']') {
385	if ($token === false) {
386	return false;
387	}
388
389	if (($value = $this->pdf_read_value($c, $token)) === false) {
390	return false;
391	}
392
393	$result[] = $value;
394	}
395
396	return array (PDF_TYPE_ARRAY, $result);
397
398	case '(' :
399	// This is a string
400
401	$pos = $c->offset;
402
403	while(1) {
404
405	// Start by finding the next closed
406	// parenthesis
407
408	$match = strpos ($c->buffer, ')', $pos);
409
410	// If you can't find it, try
411	// reading more data from the stream
412
413	if ($match === false) {
414	if (!$c->increase_length()) {
415	return false;
416	} else {
417	continue;
418	}
419	}
420
421	// Make sure that there is no backslash
422	// before the parenthesis. If there is,
423	// move on. Otherwise, return the string.
424	$esc = preg_match('/([\\\\]+)$/', $tmpresult = substr($c->buffer, $c->offset, $match - $c->offset), $m);
425
426	if ($esc === 0 \|\| strlen($m[1]) % 2 == 0) {
427	$result = $tmpresult;
428	$c->offset = $match + 1;
429	return array (PDF_TYPE_STRING, $result);
430	} else {
431	$pos = $match + 1;
432
433	if ($pos > $c->offset + $c->length) {
434	$c->increase_length();
435	}
436	}
437	}
438
439	case "stream":
440	$o_pos = ftell($c->file)-strlen($c->buffer);
441	$o_offset = $c->offset;
442
443	$c->reset($startpos = $o_pos + $o_offset);
444
445	$e = 0; // ensure line breaks in front of the stream
446	if ($c->buffer[0] == chr(10) \|\| $c->buffer[0] == chr(13))
447	$e++;
448	if ($c->buffer[1] == chr(10) && $c->buffer[0] != chr(10))
449	$e++;
450
451	if ($this->actual_obj[1][1]['/Length'][0] == PDF_TYPE_OBJREF) {
452	// mPDF 5.0 Removed pass by reference =&
453	$tmp_c = new pdf_context($this->f);
454	$tmp_length = $this->pdf_resolve_object($tmp_c,$this->actual_obj[1][1]['/Length']);
455	$length = $tmp_length[1][1];
456	} else {
457	$length = $this->actual_obj[1][1]['/Length'][1];
458	}
459
460	if ($length > 0) {
461	$c->reset($startpos+$e,$length);
462	$v = $c->buffer;
463	} else {
464	$v = '';
465	}
466	$c->reset($startpos+$e+$length+9); // 9 = strlen("endstream")
467
468	return array(PDF_TYPE_STREAM, $v);
469
470	default :
471	if (is_numeric ($token)) {
472	// A numeric token. Make sure that
473	// it is not part of something else.
474	if (($tok2 = $this->pdf_read_token ($c)) !== false) {
475	if (is_numeric ($tok2)) {
476
477	// Two numeric tokens in a row.
478	// In this case, we're probably in
479	// front of either an object reference
480	// or an object specification.
481	// Determine the case and return the data
482	if (($tok3 = $this->pdf_read_token ($c)) !== false) {
483	switch ($tok3) {
484	case 'obj' :
485	return array (PDF_TYPE_OBJDEC, (int) $token, (int) $tok2);
486	case 'R' :
487	return array (PDF_TYPE_OBJREF, (int) $token, (int) $tok2);
488	}
489	// If we get to this point, that numeric value up
490	// there was just a numeric value. Push the extra
491	// tokens back into the stack and return the value.
492	array_push ($c->stack, $tok3);
493	}
494	}
495
496	array_push ($c->stack, $tok2);
497	}
498
499	return array (PDF_TYPE_NUMERIC, $token);
500	} else {
501
502	// Just a token. Return it.
503	return array (PDF_TYPE_TOKEN, $token);
504	}
505
506	}
507	}
508
509	/**
510	* Resolve an object
511	*
512	* @param object $c pdf_context
513	* @param array $obj_spec The object-data
514	* @param boolean $encapsulate Must set to true, cause the parsing and fpdi use this method only without this para
515	*/
516	function pdf_resolve_object(&$c, $obj_spec, $encapsulate = true) {
517	// Exit if we get invalid data
518	if (!is_array($obj_spec)) {
519	return false;
520	}
521
522	if ($obj_spec[0] == PDF_TYPE_OBJREF) {
523
524	// This is a reference, resolve it
525	if (isset($this->xref['xref'][$obj_spec[1]][$obj_spec[2]])) {
526
527	// Save current file position
528	// This is needed if you want to resolve
529	// references while you're reading another object
530	// (e.g.: if you need to determine the length
531	// of a stream)
532
533	$old_pos = ftell($c->file);
534
535	// Reposition the file pointer and
536	// load the object header.
537
538	$c->reset($this->xref['xref'][$obj_spec[1]][$obj_spec[2]]);
539
540	$header = $this->pdf_read_value($c,null,true);
541
542	if ($header[0] != PDF_TYPE_OBJDEC \|\| $header[1] != $obj_spec[1] \|\| $header[2] != $obj_spec[2]) {
543	// mPDF 4.0
544	$this->success = false;
545	$this->errormsg = sprintf("Unable to find object ({$obj_spec[1]}, {$obj_spec[2]}) at expected location");
546	return false;
547	}
548
549	// If we're being asked to store all the information
550	// about the object, we add the object ID and generation
551	// number for later use
552	$this->actual_obj =& $result;
553	if ($encapsulate) {
554	$result = array (
555	PDF_TYPE_OBJECT,
556	'obj' => $obj_spec[1],
557	'gen' => $obj_spec[2]
558	);
559	} else {
560	$result = array();
561	}
562
563	// Now simply read the object data until
564	// we encounter an end-of-object marker
565	while(1) {
566	$value = $this->pdf_read_value($c);
567	if ($value === false \|\| count($result) > 4) {
568	// in this case the parser coudn't find an endobj so we break here
569	break;
570	}
571
572	if ($value[0] == PDF_TYPE_TOKEN && $value[1] === 'endobj') {
573	break;
574	}
575
576	$result[] = $value;
577	}
578
579	$c->reset($old_pos);
580
581	if (isset($result[2][0]) && $result[2][0] == PDF_TYPE_STREAM) {
582	$result[0] = PDF_TYPE_STREAM;
583	}
584
585	return $result;
586	}
587	} else {
588	return $obj_spec;
589	}
590	}
591
592
593
594	/**
595	* Reads a token from the file
596	*
597	* @param object $c pdf_context
598	* @return mixed
599	*/
600	function pdf_read_token(&$c)
601	{
602	// If there is a token available
603	// on the stack, pop it out and
604	// return it.
605
606	if (count($c->stack)) {
607	return array_pop($c->stack);
608	}
609
610	// Strip away any whitespace
611
612	do {
613	if (!$c->ensure_content()) {
614	return false;
615	}
616	$c->offset += _strspn($c->buffer, " \n\r\t", $c->offset);
617	} while ($c->offset >= $c->length - 1);
618
619	// Get the first character in the stream
620
621	$char = $c->buffer[$c->offset++];
622
623	switch ($char) {
624
625	case '[' :
626	case ']' :
627	case '(' :
628	case ')' :
629
630	// This is either an array or literal string
631	// delimiter, Return it
632
633	return $char;
634
635	case '<' :
636	case '>' :
637
638	// This could either be a hex string or
639	// dictionary delimiter. Determine the
640	// appropriate case and return the token
641
642	if ($c->buffer[$c->offset] == $char) {
643	if (!$c->ensure_content()) {
644	return false;
645	}
646	$c->offset++;
647	return $char . $char;
648	} else {
649	return $char;
650	}
651
652	default :
653
654	// This is "another" type of token (probably
655	// a dictionary entry or a numeric value)
656	// Find the end and return it.
657
658	if (!$c->ensure_content()) {
659	return false;
660	}
661
662	while(1) {
663
664	// Determine the length of the token
665
666	$pos = _strcspn($c->buffer, " []<>()\r\n\t/", $c->offset);
667	if ($c->offset + $pos <= $c->length - 1) {
668	break;
669	} else {
670	// If the script reaches this point,
671	// the token may span beyond the end
672	// of the current buffer. Therefore,
673	// we increase the size of the buffer
674	// and try again--just to be safe.
675
676	$c->increase_length();
677	}
678	}
679
680	$result = substr($c->buffer, $c->offset - 1, $pos + 1);
681
682	$c->offset += $pos;
683	return $result;
684	}
685	}
686
687
688	}
689
690	?>