Document::extractXMPMetadata() - Code Metrics - Inspection of "Read XMP Metadata and add it to data returned by g..." - smalot/pdfparser - Measure and Improve Code Quality continuously with Scrutinizer

Test Failed

Pull Request — master (#606)

by Konrad

created 2023-07-04 06:12 UTC

Document::extractXMPMetadata() D

↳ Parent: Document

Complexity

Conditions	18
Paths	3

Size

Total Lines	91
Code Lines	38

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	34
CRAP Score	18.1728

Importance

Changes	5
Bugs	1	Features	1

Metric	Value
cc	18
eloc	38
c	5
b	1
f	1
nc	3
nop	1
dl	0
loc	91
ccs	34
cts	37
cp	0.9189
crap	18.1728
rs	4.8666

How to fix Long Method Complexity

<?php

/**
 * @file
 *          This file is part of the PdfParser library.
 *
 * @author  Sébastien MALOT <[email protected]>
 *
 * @date    2017-01-03
 *
 * @license LGPLv3
 *
 * @url     <https://github.com/smalot/pdfparser>
 *
 *  PdfParser is a pdf library written in PHP, extraction oriented.
 *  Copyright (C) 2017 - Sébastien MALOT <[email protected]>
 *
 *  This program is free software: you can redistribute it and/or modify
 *  it under the terms of the GNU Lesser General Public License as published by
 *  the Free Software Foundation, either version 3 of the License, or
 *  (at your option) any later version.
 *
 *  This program is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU Lesser General Public License for more details.
 *
 *  You should have received a copy of the GNU Lesser General Public License
 *  along with this program.
 *  If not, see <http://www.pdfparser.org/sites/default/LICENSE.txt>.
 */

namespace Smalot\PdfParser;

/**
 * Technical references :
 * - http://www.mactech.com/articles/mactech/Vol.15/15.09/PDFIntro/index.html
 * - http://framework.zend.com/issues/secure/attachment/12512/Pdf.php
 * - http://www.php.net/manual/en/ref.pdf.php#74211
 * - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/ISOLatin1Encoding.pm
 * - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/ISOLatin9Encoding.pm
 * - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/StandardEncoding.pm
 * - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/WinAnsiEncoding.pm
 *
 * Class Document
 */
class Document
{
    /**
     * @var PDFObject[]
     */
    protected $objects = [];

    /**
     * @var array
     */
    protected $dictionary = [];

    /**
     * @var Header
     */
    protected $trailer;

    /**
     * @var array<mixed>
     */
    protected $metadata = [];

    /**
     * @var array
     */
    protected $details;

    public function __construct()
    {
        $this->trailer = new Header([], $this);
    }

    public function init()
    {
        $this->buildDictionary();

        $this->buildDetails();

        // Propagate init to objects.
        foreach ($this->objects as $object) {
            $object->getHeader()->init();
            $object->init();
        }
    }

    /**
     * Build dictionary based on type header field.
     */
    protected function buildDictionary()
    {
        // Build dictionary.
        $this->dictionary = [];

        foreach ($this->objects as $id => $object) {
            // Cache objects by type and subtype
            $type = $object->getHeader()->get('Type')->getContent();

            if (null != $type) {
                if (!isset($this->dictionary[$type])) {
                    $this->dictionary[$type] = [
                        'all' => [],
                        'subtype' => [],
                    ];
                }

                $this->dictionary[$type]['all'][$id] = $object;

                $subtype = $object->getHeader()->get('Subtype')->getContent();
                if (null != $subtype) {
                    if (!isset($this->dictionary[$type]['subtype'][$subtype])) {
                        $this->dictionary[$type]['subtype'][$subtype] = [];
                    }
                    $this->dictionary[$type]['subtype'][$subtype][$id] = $object;
                }
            }
        }
    }

    /**
     * Build details array.
     */
    protected function buildDetails()
    {
        // Build details array.
        $details = [];

        // Extract document info
        if ($this->trailer->has('Info')) {
            /** @var PDFObject $info */
            $info = $this->trailer->get('Info');
            // This could be an ElementMissing object, so we need to check for
            // the getHeader method first.
            if (null !== $info && method_exists($info, 'getHeader')) {
                $details = $info->getHeader()->getDetails();
            }
        }

        // Retrieve the page count
        try {
            $pages = $this->getPages();
            $details['Pages'] = \count($pages);
        } catch (\Exception $e) {
            $details['Pages'] = 0;
        }

        $details = array_merge($details, $this->metadata);

        $this->details = $details;
    }

    /**
     * Extract XMP Metadata
     */
    public function extractXMPMetadata(string $content): void
    {
        $xml = xml_parser_create();
        xml_parser_set_option($xml, \XML_OPTION_SKIP_WHITE, 1);

        if (1 === xml_parse_into_struct($xml, $content, $values, $index)) {
            /*
             * short overview about the following code parts:
             *
             * The output of xml_parse_into_struct is a single dimensional array (= $values), and the $stack is a last-on,
             * first-off array of pointers to positions in $metadata, while iterating through it, that potentially turn the
             * results into a more intuitive multi-dimensional array. When an "open" XML tag is encountered,
             * we save the current $metadata context in the $stack, then create a child array of $metadata and
             * make that the current $metadata context. When a "close" XML tag is encountered, the operations are
             * reversed: the most recently added $metadata context from $stack (IOW, the parent of the current
             * element) is set as the current $metadata context.
             */
            $metadata = [];
            $stack = [];
            foreach ($values as $val) {
                // Standardize to lowercase
                $val['tag'] = strtolower($val['tag']);

                // Ignore structural x: and rdf: XML elements
                if (str_starts_with($val['tag'], 'x:')) {
                    continue;
                } elseif (str_starts_with($val['tag'], 'rdf:') && 'rdf:li' != $val['tag']) {
                    continue;
                }

                switch ($val['type']) {
                    case 'open':
                        // Create an array of list items
                        if ('rdf:li' == $val['tag']) {
                            $metadata[] = [];

                            // Move up one level in the stack
                            $stack[\count($stack)] = &$metadata;
                            $metadata = &$metadata[\count($metadata) - 1];
                        } else {
                            // Else create an array of named values
                            $metadata[$val['tag']] = [];

                            // Move up one level in the stack
                            $stack[\count($stack)] = &$metadata;
                            $metadata = &$metadata[$val['tag']];
                        }
                        break;

                    case 'complete':
                        if (isset($val['value'])) {
                            // Assign a value to this list item
                            if ('rdf:li' == $val['tag']) {
                                $metadata[] = $val['value'];

                                // Else assign a value to this property
                            } else {
                                $metadata[$val['tag']] = $val['value'];
                            }
                        }
                        break;

                    case 'close':
                        // If the value of this property is a single-
                        // element array where the element is of type
                        // string, use the value of the first list item
                        // as the value for this property
                        if (\is_array($metadata) && isset($metadata[0]) && 1 == \count($metadata) && \is_string($metadata[0])) {
                            $metadata = $metadata[0];
                        }

                        // Move down one level in the stack
                        $metadata = &$stack[\count($stack) - 1];
                        unset($stack[\count($stack) - 1]);
                        break;
                }
            }

            // Only use this metadata if it's referring to a PDF
            if (isset($metadata['dc:format']) && 'application/pdf' == $metadata['dc:format']) {
                // According to the XMP specifications: 'Conflict resolution
                // for separate packets that describe the same resource is
                // beyond the scope of this document.' - Section 6.1
                // Source: https://www.adobe.com/devnet/xmp.html
                // Source: https://github.com/adobe/XMP-Toolkit-SDK/blob/main/docs/XMPSpecificationPart1.pdf
                // So if there are multiple XMP blocks, just merge the values
                // of each found block over top of the existing values
                $this->metadata = array_merge($this->metadata, $metadata);
            }
        }
        xml_parser_free($xml);
    }

    public function getDictionary(): array
    {
        return $this->dictionary;
    }

    /**
     * @param PDFObject[] $objects
     */
    public function setObjects($objects = [])
    {
        $this->objects = (array) $objects;

        $this->init();
    }

    /**
     * @return PDFObject[]
     */
    public function getObjects()
    {
        return $this->objects;
    }

    /**
     * @return PDFObject|Font|Page|Element|null
     */
    public function getObjectById(string $id)
    {
        if (isset($this->objects[$id])) {
            return $this->objects[$id];
        }

        return null;
    }

    public function hasObjectsByType(string $type, string $subtype = null): bool
    {
        return 0 < \count($this->getObjectsByType($type, $subtype));
    }

    public function getObjectsByType(string $type, string $subtype = null): array
    {
        if (!isset($this->dictionary[$type])) {
            return [];
        }

        if (null != $subtype) {

            if (!isset($this->dictionary[$type]['subtype'][$subtype])) {
                return [];
            }

            return $this->dictionary[$type]['subtype'][$subtype];
        }

        return $this->dictionary[$type]['all'];
    }

    /**
     * @return Font[]
     */
    public function getFonts()
    {
        return $this->getObjectsByType('Font');
    }

    public function getFirstFont(): ?Font
    {
        $fonts = $this->getFonts();
        if ([] === $fonts) {
            return null;
        }

        return reset($fonts);
    }

    /**
     * @return Page[]
     *
     * @throws \Exception
     */
    public function getPages()
    {
        if ($this->hasObjectsByType('Catalog')) {
            // Search for catalog to list pages.
            $catalogues = $this->getObjectsByType('Catalog');
            $catalogue = reset($catalogues);

            /** @var Pages $object */
            $object = $catalogue->get('Pages');
            if (method_exists($object, 'getPages')) {
                return $object->getPages(true);
            }
        }

        if ($this->hasObjectsByType('Pages')) {
            // Search for pages to list kids.
            $pages = [];

            /** @var Pages[] $objects */
            $objects = $this->getObjectsByType('Pages');
            foreach ($objects as $object) {
                $pages = array_merge($pages, $object->getPages(true));
            }

            return $pages;
        }

        if ($this->hasObjectsByType('Page')) {
            // Search for 'page' (unordered pages).
            $pages = $this->getObjectsByType('Page');

            return array_values($pages);
        }

        throw new \Exception('Missing catalog.');
    }

    public function getText(int $pageLimit = null): string
    {
        $texts = [];
        $pages = $this->getPages();

        // Only use the first X number of pages if $pageLimit is set and numeric.
        if (\is_int($pageLimit) && 0 < $pageLimit) {
            $pages = \array_slice($pages, 0, $pageLimit);
        }

        foreach ($pages as $index => $page) {
            /**
             * In some cases, the $page variable may be null.
             */
            if (null === $page) {
                continue;
            }
            if ($text = trim($page->getText())) {
                $texts[] = $text;
            }
        }

        return implode("\n\n", $texts);
    }

    public function getTrailer(): Header
    {
        return $this->trailer;
    }

    public function setTrailer(Header $trailer)
    {
        $this->trailer = $trailer;
    }

    public function getDetails(): array
    {
        return $this->details;
    }
}


1		<?php
2
3		/**
4		* @file
5		* This file is part of the PdfParser library.
6		*
7		* @author Sébastien MALOT <[email protected]>
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 <[email protected]>
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		/**
36		* Technical references :
37		* - http://www.mactech.com/articles/mactech/Vol.15/15.09/PDFIntro/index.html
38		* - http://framework.zend.com/issues/secure/attachment/12512/Pdf.php
39		* - http://www.php.net/manual/en/ref.pdf.php#74211
40		* - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/ISOLatin1Encoding.pm
41		* - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/ISOLatin9Encoding.pm
42		* - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/StandardEncoding.pm
43		* - http://cpansearch.perl.org/src/JV/PostScript-Font-1.10.02/lib/PostScript/WinAnsiEncoding.pm
44		*
45		* Class Document
46		*/
47		class Document
48		{
49		/**
50		* @var PDFObject[]
51		*/
52		protected $objects = [];
53
54		/**
55		* @var array
56		*/
57		protected $dictionary = [];
58
59		/**
60		* @var Header
61		*/
62		protected $trailer;
63
64		/**
65		* @var array<mixed>
66		*/
67		protected $metadata = [];
68
69	72	/**
70		* @var array
71	72	*/
72	72	protected $details;
73
74	49	public function __construct()
75		{
76	49	$this->trailer = new Header([], $this);
77		}
78	49
79		public function init()
80		{
81	49	$this->buildDictionary();
82	49
83	49	$this->buildDetails();
84
85	49	// Propagate init to objects.
86		foreach ($this->objects as $object) {
87		$object->getHeader()->init();
88		$object->init();
89		}
90	49	}
91
92		/**
93	49	* Build dictionary based on type header field.
94		*/
95	49	protected function buildDictionary()
96		{
97	49	// Build dictionary.
98		$this->dictionary = [];
99	49
100	49	foreach ($this->objects as $id => $object) {
101	49	// Cache objects by type and subtype
102		$type = $object->getHeader()->get('Type')->getContent();
103
104		if (null != $type) {
105		if (!isset($this->dictionary[$type])) {
106		$this->dictionary[$type] = [
107	49	'all' => [],
108		'subtype' => [],
109	49	];
110	49	}
111	42
112	42	$this->dictionary[$type]['all'][$id] = $object;
113
114	42	$subtype = $object->getHeader()->get('Subtype')->getContent();
115		if (null != $subtype) {
116		if (!isset($this->dictionary[$type]['subtype'][$subtype])) {
117		$this->dictionary[$type]['subtype'][$subtype] = [];
118	49	}
119		$this->dictionary[$type]['subtype'][$subtype][$id] = $object;
120		}
121		}
122		}
123	49	}
124
125		/**
126	49	* Build details array.
127		*/
128		protected function buildDetails()
129	49	{
130		// Build details array.
131	40	$details = [];
132
133		// Extract document info
134	40	if ($this->trailer->has('Info')) {
135	40	/** @var PDFObject $info */
136		$info = $this->trailer->get('Info');
137		// This could be an ElementMissing object, so we need to check for
138		// the getHeader method first.
139		if (null !== $info && method_exists($info, 'getHeader')) {
140		$details = $info->getHeader()->getDetails();
141	49	}
142	48	}
143	2
144	2	// Retrieve the page count
145		try {
146		$pages = $this->getPages();
147	49	$details['Pages'] = \count($pages);
148	49	} catch (\Exception $e) {
149		$details['Pages'] = 0;
150	1	}
151
152	1	$details = array_merge($details, $this->metadata);
153
154		$this->details = $details;
155		}
156
157		/**
158	49	* Extract XMP Metadata
159		*/
160	49	public function extractXMPMetadata(string $content): void
161		{
162	49	$xml = xml_parser_create();
163	49	xml_parser_set_option($xml, \XML_OPTION_SKIP_WHITE, 1);
164
165		if (1 === xml_parse_into_struct($xml, $content, $values, $index)) {
166		/*
167		* short overview about the following code parts:
168	1	*
169		* The output of xml_parse_into_struct is a single dimensional array (= $values), and the $stack is a last-on,
170	1	* first-off array of pointers to positions in $metadata, while iterating through it, that potentially turn the
171		* results into a more intuitive multi-dimensional array. When an "open" XML tag is encountered,
172		* we save the current $metadata context in the $stack, then create a child array of $metadata and
173		* make that the current $metadata context. When a "close" XML tag is encountered, the operations are
174		* reversed: the most recently added $metadata context from $stack (IOW, the parent of the current
175		* element) is set as the current $metadata context.
176	46	*/
177		$metadata = [];
178	46	$stack = [];
179	46	foreach ($values as $val) {
180		// Standardize to lowercase
181		$val['tag'] = strtolower($val['tag']);
182	3
183		// Ignore structural x: and rdf: XML elements
184		if (str_starts_with($val['tag'], 'x:')) {
185	50	continue;
186		} elseif (str_starts_with($val['tag'], 'rdf:') && 'rdf:li' != $val['tag']) {
187	50	continue;
188		}
189
190	53	switch ($val['type']) {
191		case 'open':
192	53	// Create an array of list items
193	12	if ('rdf:li' == $val['tag']) {
194		$metadata[] = [];
195
196	48	// Move up one level in the stack
197		$stack[\count($stack)] = &$metadata;
198		$metadata = &$metadata[\count($metadata) - 1];
199		} else {
200		// Else create an array of named values
201		$metadata[$val['tag']] = [];
202
203		// Move up one level in the stack
204	48	$stack[\count($stack)] = &$metadata;
205		$metadata = &$metadata[$val['tag']];
206		}
207		break;
208
209		case 'complete':
210	27	if (isset($val['value'])) {
211		// Assign a value to this list item
212	27	if ('rdf:li' == $val['tag']) {
213		$metadata[] = $val['value'];
214
215	21	// Else assign a value to this property
216		} else {
217	21	$metadata[$val['tag']] = $val['value'];
218	21	}
219	3	}
220		break;
221
222	18	case 'close':
223		// If the value of this property is a single-
224		// element array where the element is of type
225		// string, use the value of the first list item
226		// as the value for this property
227		if (\is_array($metadata) && isset($metadata[0]) && 1 == \count($metadata) && \is_string($metadata[0])) {
228		$metadata = $metadata[0];
229		}
230	50
231		// Move down one level in the stack
232	50	$metadata = &$stack[\count($stack) - 1];
233		unset($stack[\count($stack) - 1]);
234	42	break;
235	42	}
236		}
237
238	42	// Only use this metadata if it's referring to a PDF
239	42	if (isset($metadata['dc:format']) && 'application/pdf' == $metadata['dc:format']) {
240	42	// According to the XMP specifications: 'Conflict resolution
241		// for separate packets that describe the same resource is
242		// beyond the scope of this document.' - Section 6.1
243		// Source: https://www.adobe.com/devnet/xmp.html
244	9	// Source: https://github.com/adobe/XMP-Toolkit-SDK/blob/main/docs/XMPSpecificationPart1.pdf
245		// So if there are multiple XMP blocks, just merge the values
246	1	// of each found block over top of the existing values
247		$this->metadata = array_merge($this->metadata, $metadata);
248		}
249	1	}
250	1	xml_parser_free($xml);
251	1	}
252
253		public function getDictionary(): array
254	1	{
255		return $this->dictionary;
256		}
257	9
258		/**
259	7	* @param PDFObject[] $objects
260		*/
261	7	public function setObjects($objects = [])
262		{
263		$this->objects = (array) $objects;
264	3
265		$this->init();
266		}
267	12
268		/**
269	12	* @return PDFObject[]
270	12	*/
271		public function getObjects()
272		{
273	12	return $this->objects;
274	1	}
275
276		/**
277	12	* @return PDFObject\|Font\|Page\|Element\|null
278		*/
279		public function getObjectById(string $id)
280		{
281	12	if (isset($this->objects[$id])) {
282		return $this->objects[$id];
283		}
284	12
285	12	return null;
286		}
287
288		public function hasObjectsByType(string $type, string $subtype = null): bool
289	12	{
290		return 0 < \count($this->getObjectsByType($type, $subtype));
291		}
292
293		public function getObjectsByType(string $type, string $subtype = null): array
294		{
295		if (!isset($this->dictionary[$type])) {
296		return [];
297	41	}
298
299	41	if (null != $subtype) {
		0 ignored issues – show Bug introduced 2021-06-13 00:39 UTC by Report Bug Copy Issue Report It seems like you are loosely comparing `$subtype` of type `null\|string` against `null`; this is ambiguous if the string can be empty. Consider using a strict comparison `!==` instead. Loading history...
300	41	if (!isset($this->dictionary[$type]['subtype'][$subtype])) {
301		return [];
302	12	}
303
304	12	return $this->dictionary[$type]['subtype'][$subtype];
305		}
306
307		return $this->dictionary[$type]['all'];
308		}
309
310		/**
311		* @return Font[]
312		*/
313		public function getFonts()
314		{
315		return $this->getObjectsByType('Font');
316		}
317
318		public function getFirstFont(): ?Font
319		{
320		$fonts = $this->getFonts();
321		if ([] === $fonts) {
322		return null;
323		}
324
325		return reset($fonts);
326		}
327
328		/**
329		* @return Page[]
330		*
331		* @throws \Exception
332		*/
333		public function getPages()
334		{
335		if ($this->hasObjectsByType('Catalog')) {
336		// Search for catalog to list pages.
337		$catalogues = $this->getObjectsByType('Catalog');
338		$catalogue = reset($catalogues);
339
340		/** @var Pages $object */
341		$object = $catalogue->get('Pages');
342		if (method_exists($object, 'getPages')) {
343		return $object->getPages(true);
344		}
345		}
346
347		if ($this->hasObjectsByType('Pages')) {
348		// Search for pages to list kids.
349		$pages = [];
350
351		/** @var Pages[] $objects */
352		$objects = $this->getObjectsByType('Pages');
353		foreach ($objects as $object) {
354		$pages = array_merge($pages, $object->getPages(true));
355		}
356
357		return $pages;
358		}
359
360		if ($this->hasObjectsByType('Page')) {
361		// Search for 'page' (unordered pages).
362		$pages = $this->getObjectsByType('Page');
363
364		return array_values($pages);
365		}
366
367		throw new \Exception('Missing catalog.');
368		}
369
370		public function getText(int $pageLimit = null): string
371		{
372		$texts = [];
373		$pages = $this->getPages();
374
375		// Only use the first X number of pages if $pageLimit is set and numeric.
376		if (\is_int($pageLimit) && 0 < $pageLimit) {
377		$pages = \array_slice($pages, 0, $pageLimit);
378		}
379
380		foreach ($pages as $index => $page) {
381		/**
382		* In some cases, the $page variable may be null.
383		*/
384		if (null === $page) {
385		continue;
386		}
387		if ($text = trim($page->getText())) {
388		$texts[] = $text;
389		}
390		}
391
392		return implode("\n\n", $texts);
393		}
394
395		public function getTrailer(): Header
396		{
397		return $this->trailer;
398		}
399
400		public function setTrailer(Header $trailer)
401		{
402		$this->trailer = $trailer;
403		}
404
405		public function getDetails(): array
406		{
407		return $this->details;
408		}
409		}
410

smalot / pdfparser

Pull Request — master (#606)

Document::extractXMPMetadata() D

Complexity

Size

Duplication

Code Coverage

Importance

How to fix Long Method Complexity

Long Method

Duplication Side-by-Side

Filter issues like