3 use Shaarli\Bookmark\Bookmark
;
6 * Extract title from an HTML document.
8 * @param string $html HTML content where to look for a title.
10 * @return bool|string Extracted title if found, false otherwise.
12 function html_extract_title($html)
14 if (preg_match('!<title.*?>(.*?)</title>!is', $html, $matches)) {
15 return trim(str_replace("\n", '', $matches[1]));
21 * Extract charset from HTTP header if it's defined.
23 * @param string $header HTTP header Content-Type line.
25 * @return bool|string Charset string if found (lowercase), false otherwise.
27 function header_extract_charset($header)
29 preg_match('/charset=["\']?([^; "\']+)/i', $header, $match);
30 if (! empty($match[1])) {
31 return strtolower(trim($match[1]));
38 * Extract charset HTML content (tag <meta charset>).
40 * @param string $html HTML content where to look for charset.
42 * @return bool|string Charset string if found, false otherwise.
44 function html_extract_charset($html)
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]);
56 * Extract meta tag from HTML content in either:
57 * - OpenGraph: <meta property="og:[tag]" ...>
58 * - Meta tag: <meta name="[tag]" ...>
60 * @param string $tag Name of the tag to retrieve.
61 * @param string $html HTML content where to look for charset.
63 * @return bool|string Charset string if found, false otherwise.
65 function html_extract_tag($tag, $html)
67 $propertiesKey = ['property', 'name', 'itemprop'];
68 $properties = implode('|', $propertiesKey);
69 // We need a OR here to accept either 'property=og:noquote' or 'property="og:unrelated og:my-tag"'
70 $orCondition = '["\']?(?:og:)?'. $tag .'["\']?|["\'][^\'"]*?(?:og:)?' . $tag . '[^\'"]*?[\'"]';
71 // Try to retrieve OpenGraph image.
72 $ogRegex = '#<meta[^>]+(?:'. $properties .')=(?:'. $orCondition .')[^>]*content=["\'](.*?)["\'].*?>#';
73 // If the attributes are not in the order property => content (e.g. Github)
74 // New regex to keep this readable... more or less.
75 $ogRegexReverse = '#<meta[^>]+content=["\'](.*?)["\'][^>]+(?:'. $properties .')=(?:'. $orCondition .').*?>#';
77 if (preg_match($ogRegex, $html, $matches) > 0
78 || preg_match($ogRegexReverse, $html, $matches) > 0
87 * In a string, converts URLs to clickable bookmarks.
89 * @param string $text input string.
91 * @return string returns $text with all bookmarks converted to HTML bookmarks.
93 * @see Function inspired from http://www.php.net/manual/en/function.preg-replace.php#85722
95 function text2clickable($text)
97 $regex = '!(((?:https?|ftp|file)://|apt:|magnet:)\S+[a-z0-9\(\)]/?)!si';
98 return preg_replace($regex, '<a href="$1">$1</a>', $text);
102 * Auto-link hashtags.
104 * @param string $description Given description.
105 * @param string $indexUrl Root URL.
107 * @return string Description with auto-linked hashtags.
109 function hashtag_autolink($description, $indexUrl = '')
112 * To support unicode: http://stackoverflow.com/a/35498078/1484919
113 * \p{Pc} - to match underscore
114 * \p{N} - numeric character in any script
115 * \p{L} - letter from any language
116 * \p{Mn} - any non marking space (accents, umlauts, etc)
118 $regex = '/(^|\s)#([\p{Pc}\p{N}\p{L}\p{Mn}]+)/mui';
119 $replacement = '$1<a href="'. $indexUrl .'./add-tag/$2" title="Hashtag $2">#$2</a>';
120 return preg_replace($regex, $replacement, $description);
124 * This function inserts where relevant so that multiple spaces are properly displayed in HTML
125 * even in the absence of <pre> (This is used in description to keep text formatting).
127 * @param string $text input text.
129 * @return string formatted text.
131 function space2nbsp($text)
133 return preg_replace('/(^| ) /m', '$1 ', $text);
137 * Format Shaarli's description
139 * @param string $description shaare's description.
140 * @param string $indexUrl URL to Shaarli's index.
142 * @return string formatted description.
144 function format_description($description, $indexUrl = '')
146 return nl2br(space2nbsp(hashtag_autolink(text2clickable($description), $indexUrl)));
150 * Generate a small hash for a link.
152 * @param DateTime $date Link creation date.
153 * @param int $id Link ID.
155 * @return string the small hash generated from link data.
157 function link_small_hash($date, $id)
159 return smallHash($date->format(Bookmark
::LINK_DATE_FORMAT
) . $id);
163 * Returns whether or not the link is an internal note.
164 * Its URL starts by `?` because it's actually a permalink.
166 * @param string $linkUrl
168 * @return bool true if internal note, false otherwise.
170 function is_note($linkUrl)
172 return isset($linkUrl[0]) && $linkUrl[0] === '?';