MyShaarli/application/bookmark/LinkUtils.php
2020-07-23 21:19:21 +02:00

277 lines
8.8 KiB
PHP

<?php
use Shaarli\Bookmark\Bookmark;
/**
* Get cURL callback function for CURLOPT_WRITEFUNCTION
*
* @param string $charset to extract from the downloaded page (reference)
* @param string $title to extract from the downloaded page (reference)
* @param string $description to extract from the downloaded page (reference)
* @param string $keywords to extract from the downloaded page (reference)
* @param bool $retrieveDescription Automatically tries to retrieve description and keywords from HTML content
* @param string $curlGetInfo Optionally overrides curl_getinfo function
*
* @return Closure
*/
function get_curl_download_callback(
&$charset,
&$title,
&$description,
&$keywords,
$retrieveDescription,
$curlGetInfo = 'curl_getinfo'
) {
$isRedirected = false;
$currentChunk = 0;
$foundChunk = null;
/**
* cURL callback function for CURLOPT_WRITEFUNCTION (called during the download).
*
* While downloading the remote page, we check that the HTTP code is 200 and content type is 'html/text'
* Then we extract the title and the charset and stop the download when it's done.
*
* @param resource $ch cURL resource
* @param string $data chunk of data being downloaded
*
* @return int|bool length of $data or false if we need to stop the download
*/
return function (&$ch, $data) use (
$retrieveDescription,
$curlGetInfo,
&$charset,
&$title,
&$description,
&$keywords,
&$isRedirected,
&$currentChunk,
&$foundChunk
) {
$currentChunk++;
$responseCode = $curlGetInfo($ch, CURLINFO_RESPONSE_CODE);
if (!empty($responseCode) && in_array($responseCode, [301, 302])) {
$isRedirected = true;
return strlen($data);
}
if (!empty($responseCode) && $responseCode !== 200) {
return false;
}
// After a redirection, the content type will keep the previous request value
// until it finds the next content-type header.
if (! $isRedirected || strpos(strtolower($data), 'content-type') !== false) {
$contentType = $curlGetInfo($ch, CURLINFO_CONTENT_TYPE);
}
if (!empty($contentType) && strpos($contentType, 'text/html') === false) {
return false;
}
if (!empty($contentType) && empty($charset)) {
$charset = header_extract_charset($contentType);
}
if (empty($charset)) {
$charset = html_extract_charset($data);
}
if (empty($title)) {
$title = html_extract_title($data);
$foundChunk = ! empty($title) ? $currentChunk : $foundChunk;
}
if ($retrieveDescription && empty($description)) {
$description = html_extract_tag('description', $data);
$foundChunk = ! empty($description) ? $currentChunk : $foundChunk;
}
if ($retrieveDescription && empty($keywords)) {
$keywords = html_extract_tag('keywords', $data);
if (! empty($keywords)) {
$foundChunk = $currentChunk;
// Keywords use the format tag1, tag2 multiple words, tag
// So we format them to match Shaarli's separator and glue multiple words with '-'
$keywords = implode(' ', array_map(function($keyword) {
return implode('-', preg_split('/\s+/', trim($keyword)));
}, explode(',', $keywords)));
}
}
// We got everything we want, stop the download.
// If we already found either the title, description or keywords,
// it's highly unlikely that we'll found the other metas further than
// in the same chunk of data or the next one. So we also stop the download after that.
if ((!empty($responseCode) && !empty($contentType) && !empty($charset)) && $foundChunk !== null
&& (! $retrieveDescription
|| $foundChunk < $currentChunk
|| (!empty($title) && !empty($description) && !empty($keywords))
)
) {
return false;
}
return strlen($data);
};
}
/**
* Extract title from an HTML document.
*
* @param string $html HTML content where to look for a title.
*
* @return bool|string Extracted title if found, false otherwise.
*/
function html_extract_title($html)
{
if (preg_match('!<title.*?>(.*?)</title>!is', $html, $matches)) {
return trim(str_replace("\n", '', $matches[1]));
}
return false;
}
/**
* Extract charset from HTTP header if it's defined.
*
* @param string $header HTTP header Content-Type line.
*
* @return bool|string Charset string if found (lowercase), false otherwise.
*/
function header_extract_charset($header)
{
preg_match('/charset="?([^; ]+)/i', $header, $match);
if (! empty($match[1])) {
return strtolower(trim($match[1]));
}
return false;
}
/**
* Extract charset HTML content (tag <meta charset>).
*
* @param string $html HTML content where to look for charset.
*
* @return bool|string Charset string if found, false otherwise.
*/
function html_extract_charset($html)
{
// Get encoding specified in HTML header.
preg_match('#<meta .*charset=["\']?([^";\'>/]+)["\']? */?>#Usi', $html, $enc);
if (!empty($enc[1])) {
return strtolower($enc[1]);
}
return false;
}
/**
* Extract meta tag from HTML content in either:
* - OpenGraph: <meta property="og:[tag]" ...>
* - Meta tag: <meta name="[tag]" ...>
*
* @param string $tag Name of the tag to retrieve.
* @param string $html HTML content where to look for charset.
*
* @return bool|string Charset string if found, false otherwise.
*/
function html_extract_tag($tag, $html)
{
$propertiesKey = ['property', 'name', 'itemprop'];
$properties = implode('|', $propertiesKey);
// Try to retrieve OpenGraph image.
$ogRegex = '#<meta[^>]+(?:'. $properties .')=["\']?(?:og:)?'. $tag .'["\'\s][^>]*content=["\']?(.*?)["\'/>]#';
// If the attributes are not in the order property => content (e.g. Github)
// New regex to keep this readable... more or less.
$ogRegexReverse = '#<meta[^>]+content=["\']([^"\']+)[^>]+(?:'. $properties .')=["\']?(?:og)?:'. $tag .'["\'\s/>]#';
if (preg_match($ogRegex, $html, $matches) > 0
|| preg_match($ogRegexReverse, $html, $matches) > 0
) {
return $matches[1];
}
return false;
}
/**
* In a string, converts URLs to clickable bookmarks.
*
* @param string $text input string.
*
* @return string returns $text with all bookmarks converted to HTML bookmarks.
*
* @see Function inspired from http://www.php.net/manual/en/function.preg-replace.php#85722
*/
function text2clickable($text)
{
$regex = '!(((?:https?|ftp|file)://|apt:|magnet:)\S+[a-z0-9\(\)]/?)!si';
return preg_replace($regex, '<a href="$1">$1</a>', $text);
}
/**
* Auto-link hashtags.
*
* @param string $description Given description.
* @param string $indexUrl Root URL.
*
* @return string Description with auto-linked hashtags.
*/
function hashtag_autolink($description, $indexUrl = '')
{
/*
* To support unicode: http://stackoverflow.com/a/35498078/1484919
* \p{Pc} - to match underscore
* \p{N} - numeric character in any script
* \p{L} - letter from any language
* \p{Mn} - any non marking space (accents, umlauts, etc)
*/
$regex = '/(^|\s)#([\p{Pc}\p{N}\p{L}\p{Mn}]+)/mui';
$replacement = '$1<a href="'. $indexUrl .'./add-tag/$2" title="Hashtag $2">#$2</a>';
return preg_replace($regex, $replacement, $description);
}
/**
* This function inserts &nbsp; where relevant so that multiple spaces are properly displayed in HTML
* even in the absence of <pre> (This is used in description to keep text formatting).
*
* @param string $text input text.
*
* @return string formatted text.
*/
function space2nbsp($text)
{
return preg_replace('/(^| ) /m', '$1&nbsp;', $text);
}
/**
* Format Shaarli's description
*
* @param string $description shaare's description.
* @param string $indexUrl URL to Shaarli's index.
* @return string formatted description.
*/
function format_description($description, $indexUrl = '')
{
return nl2br(space2nbsp(hashtag_autolink(text2clickable($description), $indexUrl)));
}
/**
* Generate a small hash for a link.
*
* @param DateTime $date Link creation date.
* @param int $id Link ID.
*
* @return string the small hash generated from link data.
*/
function link_small_hash($date, $id)
{
return smallHash($date->format(Bookmark::LINK_DATE_FORMAT) . $id);
}
/**
* Returns whether or not the link is an internal note.
* Its URL starts by `?` because it's actually a permalink.
*
* @param string $linkUrl
*
* @return bool true if internal note, false otherwise.
*/
function is_note($linkUrl)
{
return isset($linkUrl[0]) && $linkUrl[0] === '?';
}