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 // Support quotes in double quoted content, and the other way around
72 $content = 'content=(["\'])((?:(?!\1).)*)\1';
73 // Try to retrieve OpenGraph tag.
74 $ogRegex = '#<meta[^>]+(?:' . $properties . ')=(?:' . $orCondition . ')[^>]*' . $content . '.*?>#';
75 // If the attributes are not in the order property => content (e.g. Github)
76 // New regex to keep this readable... more or less.
77 $ogRegexReverse = '#<meta[^>]+' . $content . '[^>]+(?:' . $properties . ')=(?:' . $orCondition . ').*?>#';
80 preg_match($ogRegex, $html, $matches) > 0
81 || preg_match($ogRegexReverse, $html, $matches) > 0
90 * In a string, converts URLs to clickable bookmarks.
92 * @param string $text input string.
94 * @return string returns $text with all bookmarks converted to HTML bookmarks.
96 * @see Function inspired from http://www.php.net/manual/en/function.preg-replace.php#85722
98 function text2clickable($text)
100 $regex = '!(((?:https?|ftp|file)://|apt:|magnet:)\S+[a-z0-9\(\)]/?)!si';
101 return preg_replace($regex, '<a href="$1">$1</a>', $text);
105 * Auto-link hashtags.
107 * @param string $description Given description.
108 * @param string $indexUrl Root URL.
110 * @return string Description with auto-linked hashtags.
112 function hashtag_autolink($description, $indexUrl = '')
115 * To support unicode: http://stackoverflow.com/a/35498078/1484919
116 * \p{Pc} - to match underscore
117 * \p{N} - numeric character in any script
118 * \p{L} - letter from any language
119 * \p{Mn} - any non marking space (accents, umlauts, etc)
121 $regex = '/(^|\s)#([\p{Pc}\p{N}\p{L}\p{Mn}]+)/mui';
122 $replacement = '$1<a href="' . $indexUrl . './add-tag/$2" title="Hashtag $2">#$2</a>';
123 return preg_replace($regex, $replacement, $description);
127 * This function inserts where relevant so that multiple spaces are properly displayed in HTML
128 * even in the absence of <pre> (This is used in description to keep text formatting).
130 * @param string $text input text.
132 * @return string formatted text.
134 function space2nbsp($text)
136 return preg_replace('/(^| ) /m', '$1 ', $text);
140 * Format Shaarli's description
142 * @param string $description shaare's description.
143 * @param string $indexUrl URL to Shaarli's index.
144 * @param bool $autolink Turn on/off automatic linkifications of URLs and hashtags
146 * @return string formatted description.
148 function format_description($description, $indexUrl = '', $autolink = true)
151 $description = hashtag_autolink(text2clickable($description), $indexUrl);
154 return nl2br(space2nbsp($description));
158 * Generate a small hash for a link.
160 * @param DateTime $date Link creation date.
161 * @param int $id Link ID.
163 * @return string the small hash generated from link data.
165 function link_small_hash($date, $id)
167 return smallHash($date->format(Bookmark
::LINK_DATE_FORMAT
) . $id);
171 * Returns whether or not the link is an internal note.
172 * Its URL starts by `?` because it's actually a permalink.
174 * @param string $linkUrl
176 * @return bool true if internal note, false otherwise.
178 function is_note($linkUrl)
180 return isset($linkUrl[0]) && $linkUrl[0] === '?';
184 * Extract an array of tags from a given tag string, with provided separator.
186 * @param string|null $tags String containing a list of tags separated by $separator.
187 * @param string $separator Shaarli's default: ' ' (whitespace)
189 * @return array List of tags
191 function tags_str2array(?string $tags, string $separator): array
193 // For whitespaces, we use the special \s regex character
194 $separator = $separator === ' ' ? '\s' : $separator;
196 return preg_split('/\s*' . $separator . '+\s*/', trim($tags) ?? '', -1, PREG_SPLIT_NO_EMPTY
);
200 * Return a tag string with provided separator from a list of tags.
201 * Note that given array is clean up by tags_filter().
203 * @param array|null $tags List of tags
204 * @param string $separator
208 function tags_array2str(?array $tags, string $separator): string
210 return implode($separator, tags_filter($tags, $separator));
214 * Clean an array of tags: trim + remove empty entries
216 * @param array|null $tags List of tags
217 * @param string $separator
221 function tags_filter(?array $tags, string $separator): array
223 $trimDefault = " \t\n\r\0\x0B";
224 return array_values(array_filter(array_map(function (string $entry) use ($separator, $trimDefault): string {
225 return trim($entry, $trimDefault . $separator);