wordpress-seo / src / images / Application / image-content-extractor.php

image-content-extractor.php in Yoast SEO – Advanced SEO with real-time guidance and built-in AI 28.5, at src/images/Application/image-content-extractor.php

178 lines 5.2 KB

Raw Download Zip

1	<?php
2
3	namespace Yoast\WP\SEO\Images\Application;
4
5	use DOMDocument;
6	use WP_HTML_Tag_Processor;
7
8	/**
9	* The image content extractor.
10	*/
11	class Image_Content_Extractor {
12
13	/**
14	* Gathers all images from content.
15	*
16	* @param string $content The content.
17	*
18	* @return int[] An associated array of image IDs, keyed by their URLs.
19	*/
20	public function gather_images( $content ) {
21
22	/**
23	* Filter 'wpseo_force_creating_and_using_attachment_indexables' - Filters if we should use attachment indexables to find all content images. Instead of scanning the content.
24	*
25	* The default value is false.
26	*
27	* @since 21.1
28	*/
29	$should_not_parse_content = \apply_filters( 'wpseo_force_creating_and_using_attachment_indexables', false );
30	/**
31	* Filter 'wpseo_force_skip_image_content_parsing' - Filters if we should force skip scanning the content to parse images.
32	* This filter can be used if the regex gives a faster result than scanning the code.
33	*
34	* The default value is false.
35	*
36	* @since 21.1
37	*/
38	$should_not_parse_content = \apply_filters( 'wpseo_force_skip_image_content_parsing', $should_not_parse_content );
39
40	if ( ! $should_not_parse_content && \class_exists( WP_HTML_Tag_Processor::class ) ) {
41	return $this->gather_images_wp( $content );
42	}
43
44	if ( ! $should_not_parse_content && \class_exists( DOMDocument::class ) ) {
45
46	return $this->gather_images_DOMDocument( $content );
47	}
48
49	if ( \strpos( $content, 'src' ) === false ) {
50	// Nothing to do.
51	return [];
52	}
53
54	$images = [];
55	$regexp = '<img\s[^>]src=("??)([^" >]?)\\1[^>]*>';
56	// Used modifiers iU to match case insensitive and make greedy quantifiers lazy.
57	if ( \preg_match_all( "/$regexp/iU", $content, $matches, \PREG_SET_ORDER ) ) {
58	foreach ( $matches as $match ) {
59	$images[ $match[2] ] = 0;
60	}
61	}
62
63	return $images;
64	}
65
66	/**
67	* Gathers all images from content with WP's WP_HTML_Tag_Processor() and returns them along with their IDs, if
68	* possible.
69	*
70	* @param string $content The content.
71	*
72	* @return int[] An associated array of image IDs, keyed by their URL.
73	*/
74	protected function gather_images_wp( $content ) {
75	$processor = new WP_HTML_Tag_Processor( $content );
76	$images = [];
77
78	$query = [
79	'tag_name' => 'img',
80	];
81
82	/**
83	* Filter 'wpseo_image_attribute_containing_id' - Allows filtering what attribute will be used to extract image IDs from.
84	*
85	* Defaults to "class", which is where WP natively stores the image IDs, in a `wp-image-<ID>` format.
86	*
87	* @api string The attribute to be used to extract image IDs from.
88	*/
89	$attribute = \apply_filters( 'wpseo_image_attribute_containing_id', 'class' );
90	while ( $processor->next_tag( $query ) ) {
91	$src_raw = $processor->get_attribute( 'src' );
92	if ( ! $src_raw ) {
93	continue;
94	}
95
96	$src = \htmlentities( $src_raw, ( \ENT_QUOTES \| \ENT_SUBSTITUTE \| \ENT_HTML401 ), \get_bloginfo( 'charset' ) );
97	$classes = $processor->get_attribute( $attribute );
98	$id = $this->extract_id_of_classes( $classes );
99
100	$images[ $src ] = $id;
101	}
102
103	return $images;
104	}
105
106	/**
107	* Gathers all images from content with DOMDocument() and returns them along with their IDs, if possible.
108	*
109	* @param string $content The content.
110	*
111	* @return int[] An associated array of image IDs, keyed by their URL.
112	*/
113	protected function gather_images_domdocument( $content ) {
114	$images = [];
115	$charset = \get_bloginfo( 'charset' );
116
117	/**
118	* Filter 'wpseo_image_attribute_containing_id' - Allows filtering what attribute will be used to extract image IDs from.
119	*
120	* Defaults to "class", which is where WP natively stores the image IDs, in a `wp-image-<ID>` format.
121	*
122	* @api string The attribute to be used to extract image IDs from.
123	*/
124	$attribute = \apply_filters( 'wpseo_image_attribute_containing_id', 'class' );
125
126	\libxml_use_internal_errors( true );
127	$post_dom = new DOMDocument();
128	$post_dom->loadHTML( '<?xml encoding="' . $charset . '">' . $content );
129	\libxml_clear_errors();
130
131	foreach ( $post_dom->getElementsByTagName( 'img' ) as $img ) {
132	$src = \htmlentities( $img->getAttribute( 'src' ), ( \ENT_QUOTES \| \ENT_SUBSTITUTE \| \ENT_HTML401 ), $charset );
133	$classes = $img->getAttribute( $attribute );
134
135	$id = $this->extract_id_of_classes( $classes );
136
137	$images[ $src ] = $id;
138	}
139
140	return $images;
141	}
142
143	/**
144	* Extracts image ID out of the image's classes.
145	*
146	* @param string $classes The classes assigned to the image.
147	*
148	* @return int The ID that's extracted from the classes.
149	*/
150	protected function extract_id_of_classes( $classes ) {
151	if ( ! $classes ) {
152	return 0;
153	}
154
155	/**
156	* Filter 'wpseo_extract_id_pattern' - Allows filtering the regex patern to be used to extract image IDs from class/attribute names.
157	*
158	* Defaults to the pattern that extracts image IDs from core's `wp-image-<ID>` native format in image classes.
159	*
160	* @api string The regex pattern to be used to extract image IDs from class names. Empty string if the whole class/attribute should be returned.
161	*/
162	$pattern = \apply_filters( 'wpseo_extract_id_pattern', '/(?<!\S)wp-image-(\d+)(?!\S)/i' );
163
164	if ( $pattern === '' ) {
165	return (int) $classes;
166	}
167
168	$matches = [];
169
170	if ( \preg_match( $pattern, $classes, $matches ) ) {
171
172	return (int) $matches[1];
173	}
174
175	return 0;
176	}
177	}
178