]>
Commit | Line | Data |
---|---|---|
1 | <?php | |
2 | ||
3 | use Shaarli\Bookmark\Bookmark; | |
4 | ||
5 | /** | |
6 | * Extract title from an HTML document. | |
7 | * | |
8 | * @param string $html HTML content where to look for a title. | |
9 | * | |
10 | * @return bool|string Extracted title if found, false otherwise. | |
11 | */ | |
12 | function html_extract_title($html) | |
13 | { | |
14 | if (preg_match('!<title.*?>(.*?)</title>!is', $html, $matches)) { | |
15 | return trim(str_replace("\n", '', $matches[1])); | |
16 | } | |
17 | return false; | |
18 | } | |
19 | ||
20 | /** | |
21 | * Extract charset from HTTP header if it's defined. | |
22 | * | |
23 | * @param string $header HTTP header Content-Type line. | |
24 | * | |
25 | * @return bool|string Charset string if found (lowercase), false otherwise. | |
26 | */ | |
27 | function header_extract_charset($header) | |
28 | { | |
29 | preg_match('/charset="?([^; ]+)/i', $header, $match); | |
30 | if (! empty($match[1])) { | |
31 | return strtolower(trim($match[1])); | |
32 | } | |
33 | ||
34 | return false; | |
35 | } | |
36 | ||
37 | /** | |
38 | * Extract charset HTML content (tag <meta charset>). | |
39 | * | |
40 | * @param string $html HTML content where to look for charset. | |
41 | * | |
42 | * @return bool|string Charset string if found, false otherwise. | |
43 | */ | |
44 | function html_extract_charset($html) | |
45 | { | |
46 | // Get encoding specified in HTML header. | |
47 | preg_match('#<meta .*charset=["\']?([^";\'>/]+)["\']? */?>#Usi', $html, $enc); | |
48 | if (!empty($enc[1])) { | |
49 | return strtolower($enc[1]); | |
50 | } | |
51 | ||
52 | return false; | |
53 | } | |
54 | ||
55 | /** | |
56 | * Extract meta tag from HTML content in either: | |
57 | * - OpenGraph: <meta property="og:[tag]" ...> | |
58 | * - Meta tag: <meta name="[tag]" ...> | |
59 | * | |
60 | * @param string $tag Name of the tag to retrieve. | |
61 | * @param string $html HTML content where to look for charset. | |
62 | * | |
63 | * @return bool|string Charset string if found, false otherwise. | |
64 | */ | |
65 | function html_extract_tag($tag, $html) | |
66 | { | |
67 | $propertiesKey = ['property', 'name', 'itemprop']; | |
68 | $properties = implode('|', $propertiesKey); | |
69 | // Try to retrieve OpenGraph image. | |
70 | $ogRegex = '#<meta[^>]+(?:'. $properties .')=["\']?(?:og:)?'. $tag .'["\'\s][^>]*content=["\']?(.*?)["\'/>]#'; | |
71 | // If the attributes are not in the order property => content (e.g. Github) | |
72 | // New regex to keep this readable... more or less. | |
73 | $ogRegexReverse = '#<meta[^>]+content=["\']([^"\']+)[^>]+(?:'. $properties .')=["\']?(?:og)?:'. $tag .'["\'\s/>]#'; | |
74 | ||
75 | if (preg_match($ogRegex, $html, $matches) > 0 | |
76 | || preg_match($ogRegexReverse, $html, $matches) > 0 | |
77 | ) { | |
78 | return $matches[1]; | |
79 | } | |
80 | ||
81 | return false; | |
82 | } | |
83 | ||
84 | /** | |
85 | * In a string, converts URLs to clickable bookmarks. | |
86 | * | |
87 | * @param string $text input string. | |
88 | * | |
89 | * @return string returns $text with all bookmarks converted to HTML bookmarks. | |
90 | * | |
91 | * @see Function inspired from http://www.php.net/manual/en/function.preg-replace.php#85722 | |
92 | */ | |
93 | function text2clickable($text) | |
94 | { | |
95 | $regex = '!(((?:https?|ftp|file)://|apt:|magnet:)\S+[a-z0-9\(\)]/?)!si'; | |
96 | return preg_replace($regex, '<a href="$1">$1</a>', $text); | |
97 | } | |
98 | ||
99 | /** | |
100 | * Auto-link hashtags. | |
101 | * | |
102 | * @param string $description Given description. | |
103 | * @param string $indexUrl Root URL. | |
104 | * | |
105 | * @return string Description with auto-linked hashtags. | |
106 | */ | |
107 | function hashtag_autolink($description, $indexUrl = '') | |
108 | { | |
109 | /* | |
110 | * To support unicode: http://stackoverflow.com/a/35498078/1484919 | |
111 | * \p{Pc} - to match underscore | |
112 | * \p{N} - numeric character in any script | |
113 | * \p{L} - letter from any language | |
114 | * \p{Mn} - any non marking space (accents, umlauts, etc) | |
115 | */ | |
116 | $regex = '/(^|\s)#([\p{Pc}\p{N}\p{L}\p{Mn}]+)/mui'; | |
117 | $replacement = '$1<a href="'. $indexUrl .'./add-tag/$2" title="Hashtag $2">#$2</a>'; | |
118 | return preg_replace($regex, $replacement, $description); | |
119 | } | |
120 | ||
121 | /** | |
122 | * This function inserts where relevant so that multiple spaces are properly displayed in HTML | |
123 | * even in the absence of <pre> (This is used in description to keep text formatting). | |
124 | * | |
125 | * @param string $text input text. | |
126 | * | |
127 | * @return string formatted text. | |
128 | */ | |
129 | function space2nbsp($text) | |
130 | { | |
131 | return preg_replace('/(^| ) /m', '$1 ', $text); | |
132 | } | |
133 | ||
134 | /** | |
135 | * Format Shaarli's description | |
136 | * | |
137 | * @param string $description shaare's description. | |
138 | * @param string $indexUrl URL to Shaarli's index. | |
139 | ||
140 | * @return string formatted description. | |
141 | */ | |
142 | function format_description($description, $indexUrl = '') | |
143 | { | |
144 | return nl2br(space2nbsp(hashtag_autolink(text2clickable($description), $indexUrl))); | |
145 | } | |
146 | ||
147 | /** | |
148 | * Generate a small hash for a link. | |
149 | * | |
150 | * @param DateTime $date Link creation date. | |
151 | * @param int $id Link ID. | |
152 | * | |
153 | * @return string the small hash generated from link data. | |
154 | */ | |
155 | function link_small_hash($date, $id) | |
156 | { | |
157 | return smallHash($date->format(Bookmark::LINK_DATE_FORMAT) . $id); | |
158 | } | |
159 | ||
160 | /** | |
161 | * Returns whether or not the link is an internal note. | |
162 | * Its URL starts by `?` because it's actually a permalink. | |
163 | * | |
164 | * @param string $linkUrl | |
165 | * | |
166 | * @return bool true if internal note, false otherwise. | |
167 | */ | |
168 | function is_note($linkUrl) | |
169 | { | |
170 | return isset($linkUrl[0]) && $linkUrl[0] === '?'; | |
171 | } |