mxchat-basic / includes / pdf-parser / src / Smalot / PdfParser / Document.php

Document.php in MxChat – AI Chatbot & Content Generation for WordPress 3.2.12, at includes/pdf-parser/src/Smalot/PdfParser/Document.php

461 lines 15.1 KB

Raw Download Zip

1	<?php
2
3	/**
4	* @file
5	* This file is part of the PdfParser library.
6	*
7	* @author Sébastien MALOT <sebastien@malot.fr>
8	*
9	* @date 2017-01-03
10	*
11	* @license LGPLv3
12	*
13	* @url <https://github.com/smalot/pdfparser>
14	*
15	* PdfParser is a pdf library written in PHP, extraction oriented.
16	* Copyright (C) 2017 - Sébastien MALOT <sebastien@malot.fr>
17	*
18	* This program is free software: you can redistribute it and/or modify
19	* it under the terms of the GNU Lesser General Public License as published by
20	* the Free Software Foundation, either version 3 of the License, or
21	* (at your option) any later version.
22	*
23	* This program is distributed in the hope that it will be useful,
24	* but WITHOUT ANY WARRANTY; without even the implied warranty of
25	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
26	* GNU Lesser General Public License for more details.
27	*
28	* You should have received a copy of the GNU Lesser General Public License
29	* along with this program.
30	* If not, see <http://www.pdfparser.org/sites/default/LICENSE.txt>.
31	*/
32
33	namespace Smalot\PdfParser;
34
35	use Smalot\PdfParser\Encoding\PDFDocEncoding;
36
37	/**
38	* Technical references :
39	* - http://www.mactech.com/articles/mactech/Vol.15/15.09/PDFIntro/index.html
40	* - http://framework.zend.com/issues/secure/attachment/12512/Pdf.php
41	* - http://www.php.net/manual/en/ref.pdf.php#74211
42	* - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/ISOLatin1Encoding.pm
43	* - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/ISOLatin9Encoding.pm
44	* - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/StandardEncoding.pm
45	* - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/WinAnsiEncoding.pm
46	*
47	* Class Document
48	*/
49	class Document
50	{
51	/**
52	* @var PDFObject[]
53	*/
54	protected $objects = [];
55
56	/**
57	* @var array
58	*/
59	protected $dictionary = [];
60
61	/**
62	* @var Header
63	*/
64	protected $trailer;
65
66	/**
67	* @var array<mixed>
68	*/
69	protected $metadata = [];
70
71	/**
72	* @var array
73	*/
74	protected $details;
75
76	public function __construct()
77	{
78	$this->trailer = new Header([], $this);
79	}
80
81	public function init()
82	{
83	$this->buildDictionary();
84
85	$this->buildDetails();
86
87	// Propagate init to objects.
88	foreach ($this->objects as $object) {
89	$object->getHeader()->init();
90	$object->init();
91	}
92	}
93
94	/**
95	* Build dictionary based on type header field.
96	*/
97	protected function buildDictionary()
98	{
99	// Build dictionary.
100	$this->dictionary = [];
101
102	foreach ($this->objects as $id => $object) {
103	// Cache objects by type and subtype
104	$type = $object->getHeader()->get('Type')->getContent();
105
106	if (null != $type) {
107	if (!isset($this->dictionary[$type])) {
108	$this->dictionary[$type] = [
109	'all' => [],
110	'subtype' => [],
111	];
112	}
113
114	$this->dictionary[$type]['all'][$id] = $object;
115
116	$subtype = $object->getHeader()->get('Subtype')->getContent();
117	if (null != $subtype) {
118	if (!isset($this->dictionary[$type]['subtype'][$subtype])) {
119	$this->dictionary[$type]['subtype'][$subtype] = [];
120	}
121	$this->dictionary[$type]['subtype'][$subtype][$id] = $object;
122	}
123	}
124	}
125	}
126
127	/**
128	* Build details array.
129	*/
130	protected function buildDetails()
131	{
132	// Build details array.
133	$details = [];
134
135	// Extract document info
136	if ($this->trailer->has('Info')) {
137	/** @var PDFObject $info */
138	$info = $this->trailer->get('Info');
139	// This could be an ElementMissing object, so we need to check for
140	// the getHeader method first.
141	if (null !== $info && method_exists($info, 'getHeader')) {
142	$details = $info->getHeader()->getDetails();
143	}
144	}
145
146	// Retrieve the page count
147	try {
148	$pages = $this->getPages();
149	$details['Pages'] = \count($pages);
150	} catch (\Exception $e) {
151	$details['Pages'] = 0;
152	}
153
154	// Decode and repair encoded document properties
155	foreach ($details as $key => $value) {
156	if (\is_string($value)) {
157	// If the string is already UTF-8 encoded, that means we only
158	// need to repair Adobe's ham-fisted insertion of line-feeds
159	// every ~127 characters, which doesn't seem to be multi-byte
160	// safe
161	if (mb_check_encoding($value, 'UTF-8')) {
162	// Remove literal backslash + line-feed "\\r"
163	$value = str_replace("\x5c\x0d", '', $value);
164
165	// Remove backslash plus bytes written into high part of
166	// multibyte unicode character
167	while (preg_match("/\x5c\x5c\xe0([\xb4-\xb8])(.)/", $value, $match)) {
168	$diff = (\ord($match[1]) - 182) * 64;
169	$newbyte = PDFDocEncoding::convertPDFDoc2UTF8(\chr(\ord($match[2]) + $diff));
170	$value = preg_replace("/\x5c\x5c\xe0".$match[1].$match[2].'/', $newbyte, $value);
171	}
172
173	// Remove bytes written into low part of multibyte unicode
174	// character
175	while (preg_match("/(.)\x9c\xe0([\xb3-\xb7])/", $value, $match)) {
176	$diff = \ord($match[2]) - 181;
177	$newbyte = \chr(\ord($match[1]) + $diff);
178	$value = preg_replace('/'.$match[1]."\x9c\xe0".$match[2].'/', $newbyte, $value);
179	}
180
181	// Remove this byte string that Adobe occasionally adds
182	// between two single byte characters in a unicode string
183	$value = str_replace("\xe5\xb0\x8d", '', $value);
184
185	$details[$key] = $value;
186	} else {
187	// If the string is just PDFDocEncoding, remove any line-feeds
188	// and decode the whole thing.
189	$value = str_replace("\\\r", '', $value);
190	$details[$key] = PDFDocEncoding::convertPDFDoc2UTF8($value);
191	}
192	}
193	}
194
195	$details = array_merge($details, $this->metadata);
196
197	$this->details = $details;
198	}
199
200	/**
201	* Extract XMP Metadata
202	*/
203	public function extractXMPMetadata(string $content): void
204	{
205	$xml = xml_parser_create();
206	xml_parser_set_option($xml, \XML_OPTION_SKIP_WHITE, 1);
207
208	if (1 === xml_parse_into_struct($xml, $content, $values, $index)) {
209	/*
210	* short overview about the following code parts:
211	*
212	* The output of xml_parse_into_struct is a single dimensional array (= $values), and the $stack is a last-on,
213	* first-off array of pointers to positions in $metadata, while iterating through it, that potentially turn the
214	* results into a more intuitive multi-dimensional array. When an "open" XML tag is encountered,
215	* we save the current $metadata context in the $stack, then create a child array of $metadata and
216	* make that the current $metadata context. When a "close" XML tag is encountered, the operations are
217	* reversed: the most recently added $metadata context from $stack (IOW, the parent of the current
218	* element) is set as the current $metadata context.
219	*/
220	$metadata = [];
221	$stack = [];
222	foreach ($values as $val) {
223	// Standardize to lowercase
224	$val['tag'] = strtolower($val['tag']);
225
226	// Ignore structural x: and rdf: XML elements
227	if (0 === strpos($val['tag'], 'x:')) {
228	continue;
229	} elseif (0 === strpos($val['tag'], 'rdf:') && 'rdf:li' != $val['tag']) {
230	continue;
231	}
232
233	switch ($val['type']) {
234	case 'open':
235	// Create an array of list items
236	if ('rdf:li' == $val['tag']) {
237	$metadata[] = [];
238
239	// Move up one level in the stack
240	$stack[\count($stack)] = &$metadata;
241	$metadata = &$metadata[\count($metadata) - 1];
242	} else {
243	// Else create an array of named values
244	$metadata[$val['tag']] = [];
245
246	// Move up one level in the stack
247	$stack[\count($stack)] = &$metadata;
248	$metadata = &$metadata[$val['tag']];
249	}
250	break;
251
252	case 'complete':
253	if (isset($val['value'])) {
254	// Assign a value to this list item
255	if ('rdf:li' == $val['tag']) {
256	$metadata[] = $val['value'];
257
258	// Else assign a value to this property
259	} else {
260	$metadata[$val['tag']] = $val['value'];
261	}
262	}
263	break;
264
265	case 'close':
266	// If the value of this property is an array
267	if (\is_array($metadata)) {
268	// If the value is a single element array
269	// where the element is of type string, use
270	// the value of the first list item as the
271	// value for this property
272	if (1 == \count($metadata) && isset($metadata[0]) && \is_string($metadata[0])) {
273	$metadata = $metadata[0];
274	} elseif (0 == \count($metadata)) {
275	// if the value is an empty array, set
276	// the value of this property to the empty
277	// string
278	$metadata = '';
279	}
280	}
281
282	// Move down one level in the stack
283	$metadata = &$stack[\count($stack) - 1];
284	unset($stack[\count($stack) - 1]);
285	break;
286	}
287	}
288
289	// Only use this metadata if it's referring to a PDF
290	if (!isset($metadata['dc:format']) \|\| 'application/pdf' == $metadata['dc:format']) {
291	// According to the XMP specifications: 'Conflict resolution
292	// for separate packets that describe the same resource is
293	// beyond the scope of this document.' - Section 6.1
294	// Source: https://www.adobe.com/devnet/xmp.html
295	// Source: https://github.com/adobe/XMP-Toolkit-SDK/blob/main/docs/XMPSpecificationPart1.pdf
296	// So if there are multiple XMP blocks, just merge the values
297	// of each found block over top of the existing values
298	$this->metadata = array_merge($this->metadata, $metadata);
299	}
300	}
301	xml_parser_free($xml);
302	}
303
304	public function getDictionary(): array
305	{
306	return $this->dictionary;
307	}
308
309	/**
310	* @param PDFObject[] $objects
311	*/
312	public function setObjects($objects = [])
313	{
314	$this->objects = (array) $objects;
315
316	$this->init();
317	}
318
319	/**
320	* @return PDFObject[]
321	*/
322	public function getObjects()
323	{
324	return $this->objects;
325	}
326
327	/**
328	* @return PDFObject\|Font\|Page\|Element\|null
329	*/
330	public function getObjectById(string $id)
331	{
332	if (isset($this->objects[$id])) {
333	return $this->objects[$id];
334	}
335
336	return null;
337	}
338
339	public function hasObjectsByType(string $type, ?string $subtype = null): bool
340	{
341	return 0 < \count($this->getObjectsByType($type, $subtype));
342	}
343
344	public function getObjectsByType(string $type, ?string $subtype = null): array
345	{
346	if (!isset($this->dictionary[$type])) {
347	return [];
348	}
349
350	if (null != $subtype) {
351	if (!isset($this->dictionary[$type]['subtype'][$subtype])) {
352	return [];
353	}
354
355	return $this->dictionary[$type]['subtype'][$subtype];
356	}
357
358	return $this->dictionary[$type]['all'];
359	}
360
361	/**
362	* @return Font[]
363	*/
364	public function getFonts()
365	{
366	return $this->getObjectsByType('Font');
367	}
368
369	public function getFirstFont(): ?Font
370	{
371	$fonts = $this->getFonts();
372	if ([] === $fonts) {
373	return null;
374	}
375
376	return reset($fonts);
377	}
378
379	/**
380	* @return Page[]
381	*
382	* @throws \Exception
383	*/
384	public function getPages()
385	{
386	if ($this->hasObjectsByType('Catalog')) {
387	// Search for catalog to list pages.
388	$catalogues = $this->getObjectsByType('Catalog');
389	$catalogue = reset($catalogues);
390
391	/** @var Pages $object */
392	$object = $catalogue->get('Pages');
393	if (method_exists($object, 'getPages')) {
394	return $object->getPages(true);
395	}
396	}
397
398	if ($this->hasObjectsByType('Pages')) {
399	// Search for pages to list kids.
400	$pages = [];
401
402	/** @var Pages[] $objects */
403	$objects = $this->getObjectsByType('Pages');
404	foreach ($objects as $object) {
405	$pages = array_merge($pages, $object->getPages(true));
406	}
407
408	return $pages;
409	}
410
411	if ($this->hasObjectsByType('Page')) {
412	// Search for 'page' (unordered pages).
413	$pages = $this->getObjectsByType('Page');
414
415	return array_values($pages);
416	}
417
418	throw new \Exception('Missing catalog.');
419	}
420
421	public function getText(?int $pageLimit = null): string
422	{
423	$texts = [];
424	$pages = $this->getPages();
425
426	// Only use the first X number of pages if $pageLimit is set and numeric.
427	if (\is_int($pageLimit) && 0 < $pageLimit) {
428	$pages = \array_slice($pages, 0, $pageLimit);
429	}
430
431	foreach ($pages as $index => $page) {
432	/**
433	* In some cases, the $page variable may be null.
434	*/
435	if (null === $page) {
436	continue;
437	}
438	if ($text = trim($page->getText())) {
439	$texts[] = $text;
440	}
441	}
442
443	return implode("\n\n", $texts);
444	}
445
446	public function getTrailer(): Header
447	{
448	return $this->trailer;
449	}
450
451	public function setTrailer(Header $trailer)
452	{
453	$this->trailer = $trailer;
454	}
455
456	public function getDetails(): array
457	{
458	return $this->details;
459	}
460	}
461