]>
Commit | Line | Data |
---|---|---|
1557cefb A |
1 | <?php |
2 | ||
3 | /** | |
d65342e3 | 4 | * Get cURL callback function for CURLOPT_WRITEFUNCTION |
1557cefb | 5 | * |
d65342e3 A |
6 | * @param string $charset to extract from the downloaded page (reference) |
7 | * @param string $title to extract from the downloaded page (reference) | |
8 | * @param string $curlGetInfo Optionnaly overrides curl_getinfo function | |
1557cefb | 9 | * |
d65342e3 | 10 | * @return Closure |
1557cefb | 11 | */ |
d65342e3 | 12 | function get_curl_download_callback(&$charset, &$title, $curlGetInfo = 'curl_getinfo') |
1557cefb | 13 | { |
d65342e3 A |
14 | /** |
15 | * cURL callback function for CURLOPT_WRITEFUNCTION (called during the download). | |
16 | * | |
17 | * While downloading the remote page, we check that the HTTP code is 200 and content type is 'html/text' | |
18 | * Then we extract the title and the charset and stop the download when it's done. | |
19 | * | |
20 | * @param resource $ch cURL resource | |
21 | * @param string $data chunk of data being downloaded | |
22 | * | |
23 | * @return int|bool length of $data or false if we need to stop the download | |
24 | */ | |
25 | return function(&$ch, $data) use ($curlGetInfo, &$charset, &$title) { | |
26 | $responseCode = $curlGetInfo($ch, CURLINFO_RESPONSE_CODE); | |
27 | if (!empty($responseCode) && $responseCode != 200) { | |
28 | return false; | |
29 | } | |
30 | $contentType = $curlGetInfo($ch, CURLINFO_CONTENT_TYPE); | |
31 | if (!empty($contentType) && strpos($contentType, 'text/html') === false) { | |
32 | return false; | |
33 | } | |
34 | if (empty($charset)) { | |
35 | $charset = header_extract_charset($contentType); | |
36 | } | |
37 | if (empty($charset)) { | |
38 | $charset = html_extract_charset($data); | |
39 | } | |
40 | if (empty($title)) { | |
41 | $title = html_extract_title($data); | |
42 | } | |
43 | // We got everything we want, stop the download. | |
44 | if (!empty($responseCode) && !empty($contentType) && !empty($charset) && !empty($title)) { | |
45 | return false; | |
46 | } | |
47 | ||
48 | return strlen($data); | |
49 | }; | |
1557cefb A |
50 | } |
51 | ||
52 | /** | |
d65342e3 | 53 | * Extract title from an HTML document. |
1557cefb | 54 | * |
d65342e3 | 55 | * @param string $html HTML content where to look for a title. |
1557cefb | 56 | * |
d65342e3 | 57 | * @return bool|string Extracted title if found, false otherwise. |
1557cefb | 58 | */ |
d65342e3 | 59 | function html_extract_title($html) |
1557cefb | 60 | { |
d65342e3 A |
61 | if (preg_match('!<title.*?>(.*?)</title>!is', $html, $matches)) { |
62 | return trim(str_replace("\n", '', $matches[1])); | |
1557cefb | 63 | } |
d65342e3 | 64 | return false; |
1557cefb A |
65 | } |
66 | ||
67 | /** | |
d65342e3 | 68 | * Extract charset from HTTP header if it's defined. |
1557cefb | 69 | * |
d65342e3 | 70 | * @param string $header HTTP header Content-Type line. |
1557cefb A |
71 | * |
72 | * @return bool|string Charset string if found (lowercase), false otherwise. | |
73 | */ | |
d65342e3 | 74 | function header_extract_charset($header) |
1557cefb | 75 | { |
d65342e3 A |
76 | preg_match('/charset="?([^; ]+)/i', $header, $match); |
77 | if (! empty($match[1])) { | |
78 | return strtolower(trim($match[1])); | |
1557cefb A |
79 | } |
80 | ||
81 | return false; | |
82 | } | |
83 | ||
84 | /** | |
85 | * Extract charset HTML content (tag <meta charset>). | |
86 | * | |
87 | * @param string $html HTML content where to look for charset. | |
88 | * | |
89 | * @return bool|string Charset string if found, false otherwise. | |
90 | */ | |
91 | function html_extract_charset($html) | |
92 | { | |
93 | // Get encoding specified in HTML header. | |
ce7b0b64 | 94 | preg_match('#<meta .*charset=["\']?([^";\'>/]+)["\']? */?>#Usi', $html, $enc); |
1557cefb A |
95 | if (!empty($enc[1])) { |
96 | return strtolower($enc[1]); | |
97 | } | |
98 | ||
99 | return false; | |
100 | } | |
141a86c5 A |
101 | |
102 | /** | |
103 | * Count private links in given linklist. | |
104 | * | |
7af9a418 | 105 | * @param array|Countable $links Linklist. |
141a86c5 A |
106 | * |
107 | * @return int Number of private links. | |
108 | */ | |
109 | function count_private($links) | |
110 | { | |
111 | $cpt = 0; | |
112 | foreach ($links as $link) { | |
ee6f4b64 V |
113 | if ($link['private']) { |
114 | $cpt += 1; | |
115 | } | |
141a86c5 | 116 | } |
9ccca401 | 117 | |
141a86c5 A |
118 | return $cpt; |
119 | } | |
9ccca401 A |
120 | |
121 | /** | |
122 | * In a string, converts URLs to clickable links. | |
123 | * | |
124 | * @param string $text input string. | |
125 | * @param string $redirector if a redirector is set, use it to gerenate links. | |
126 | * | |
127 | * @return string returns $text with all links converted to HTML links. | |
128 | * | |
129 | * @see Function inspired from http://www.php.net/manual/en/function.preg-replace.php#85722 | |
130 | */ | |
131 | function text2clickable($text, $redirector = '') | |
132 | { | |
133 | $regex = '!(((?:https?|ftp|file)://|apt:|magnet:)\S+[[:alnum:]]/?)!si'; | |
134 | ||
135 | if (empty($redirector)) { | |
136 | return preg_replace($regex, '<a href="$1">$1</a>', $text); | |
137 | } | |
138 | // Redirector is set, urlencode the final URL. | |
139 | return preg_replace_callback( | |
140 | $regex, | |
141 | function ($matches) use ($redirector) { | |
142 | return '<a href="' . $redirector . urlencode($matches[1]) .'">'. $matches[1] .'</a>'; | |
143 | }, | |
144 | $text | |
145 | ); | |
146 | } | |
147 | ||
148 | /** | |
149 | * Auto-link hashtags. | |
150 | * | |
151 | * @param string $description Given description. | |
152 | * @param string $indexUrl Root URL. | |
153 | * | |
154 | * @return string Description with auto-linked hashtags. | |
155 | */ | |
156 | function hashtag_autolink($description, $indexUrl = '') | |
157 | { | |
158 | /* | |
159 | * To support unicode: http://stackoverflow.com/a/35498078/1484919 | |
160 | * \p{Pc} - to match underscore | |
161 | * \p{N} - numeric character in any script | |
162 | * \p{L} - letter from any language | |
163 | * \p{Mn} - any non marking space (accents, umlauts, etc) | |
164 | */ | |
165 | $regex = '/(^|\s)#([\p{Pc}\p{N}\p{L}\p{Mn}]+)/mui'; | |
166 | $replacement = '$1<a href="'. $indexUrl .'?addtag=$2" title="Hashtag $2">#$2</a>'; | |
167 | return preg_replace($regex, $replacement, $description); | |
168 | } | |
169 | ||
170 | /** | |
171 | * This function inserts where relevant so that multiple spaces are properly displayed in HTML | |
172 | * even in the absence of <pre> (This is used in description to keep text formatting). | |
173 | * | |
174 | * @param string $text input text. | |
175 | * | |
176 | * @return string formatted text. | |
177 | */ | |
178 | function space2nbsp($text) | |
179 | { | |
180 | return preg_replace('/(^| ) /m', '$1 ', $text); | |
181 | } | |
182 | ||
183 | /** | |
184 | * Format Shaarli's description | |
185 | * | |
186 | * @param string $description shaare's description. | |
187 | * @param string $redirector if a redirector is set, use it to gerenate links. | |
7af9a418 | 188 | * @param string $indexUrl URL to Shaarli's index. |
9ccca401 A |
189 | * |
190 | * @return string formatted description. | |
191 | */ | |
192 | function format_description($description, $redirector = '', $indexUrl = '') { | |
193 | return nl2br(space2nbsp(hashtag_autolink(text2clickable($description, $redirector), $indexUrl))); | |
194 | } | |
d592daea A |
195 | |
196 | /** | |
197 | * Generate a small hash for a link. | |
198 | * | |
199 | * @param DateTime $date Link creation date. | |
200 | * @param int $id Link ID. | |
201 | * | |
202 | * @return string the small hash generated from link data. | |
203 | */ | |
204 | function link_small_hash($date, $id) | |
205 | { | |
206 | return smallHash($date->format(LinkDB::LINK_DATE_FORMAT) . $id); | |
207 | } |