X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FUtil%2FParseUrl.php;h=9a3de88cf2ac789d5866edf28d41e35bdf6c412e;hb=01b7a3976ff1a00700c429db43fcbdec7158abca;hp=2c134542a0392e279dd6f64f4008ec7cf8e52c74;hpb=62c5e5d861d715d467962ea8601fc9d058d2d765;p=friendica.git
diff --git a/src/Util/ParseUrl.php b/src/Util/ParseUrl.php
index 2c134542a0..9a3de88cf2 100644
--- a/src/Util/ParseUrl.php
+++ b/src/Util/ParseUrl.php
@@ -8,12 +8,11 @@ namespace Friendica\Util;
use DOMDocument;
use DOMXPath;
use Friendica\Content\OEmbed;
-use Friendica\Core\Addon;
+use Friendica\Core\Hook;
+use Friendica\Core\Logger;
use Friendica\Database\DBA;
use Friendica\Object\Image;
-require_once 'include/dba.php';
-
/**
* @brief Class with methods for extracting certain content from an url
*/
@@ -23,10 +22,10 @@ class ParseUrl
* @brief Search for chached embeddable data of an url otherwise fetch it
*
* @param string $url The url of the page which should be scraped
- * @param bool $no_guessing If true the parse doens't search for
- * preview pictures
- * @param bool $do_oembed The false option is used by the function fetch_oembed()
- * to avoid endless loops
+ * @param bool $no_guessing If true the parse doens't search for
+ * preview pictures
+ * @param bool $do_oembed The false option is used by the function fetch_oembed()
+ * to avoid endless loops
*
* @return array which contains needed data for embedding
* string 'url' => The url of the parsed page
@@ -38,7 +37,8 @@ class ParseUrl
* array'images' = Array of preview pictures
* string 'keywords' => The tags which belong to the content
*
- * @see ParseUrl::getSiteinfo() for more information about scraping
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
+ * @see ParseUrl::getSiteinfo() for more information about scraping
* embeddable content
*/
public static function getSiteinfoCached($url, $no_guessing = false, $do_oembed = true)
@@ -48,7 +48,7 @@ class ParseUrl
}
$parsed_url = DBA::selectFirst('parsed_url', ['content'],
- ['url' => normalise_link($url), 'guessing' => !$no_guessing, 'oembed' => $do_oembed]
+ ['url' => Strings::normaliseLink($url), 'guessing' => !$no_guessing, 'oembed' => $do_oembed]
);
if (!empty($parsed_url['content'])) {
$data = unserialize($parsed_url['content']);
@@ -60,7 +60,7 @@ class ParseUrl
DBA::insert(
'parsed_url',
[
- 'url' => normalise_link($url), 'guessing' => !$no_guessing,
+ 'url' => Strings::normaliseLink($url), 'guessing' => !$no_guessing,
'oembed' => $do_oembed, 'content' => serialize($data),
'created' => DateTimeFormat::utcNow()
],
@@ -69,6 +69,7 @@ class ParseUrl
return $data;
}
+
/**
* @brief Parse a page for embeddable content information
*
@@ -80,11 +81,11 @@ class ParseUrl
* \
*
* @param string $url The url of the page which should be scraped
- * @param bool $no_guessing If true the parse doens't search for
- * preview pictures
- * @param bool $do_oembed The false option is used by the function fetch_oembed()
- * to avoid endless loops
- * @param int $count Internal counter to avoid endless loops
+ * @param bool $no_guessing If true the parse doens't search for
+ * preview pictures
+ * @param bool $do_oembed The false option is used by the function fetch_oembed()
+ * to avoid endless loops
+ * @param int $count Internal counter to avoid endless loops
*
* @return array which contains needed data for embedding
* string 'url' => The url of the parsed page
@@ -96,7 +97,8 @@ class ParseUrl
* array'images' = Array of preview pictures
* string 'keywords' => The tags which belong to the content
*
- * @todo https://developers.google.com/+/plugins/snippet/
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
+ * @todo https://developers.google.com/+/plugins/snippet/
* @verbatim
*
*
@@ -111,8 +113,6 @@ class ParseUrl
*/
public static function getSiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1)
{
- $a = get_app();
-
$siteinfo = [];
// Check if the URL does contain a scheme
@@ -123,7 +123,7 @@ class ParseUrl
}
if ($count > 10) {
- logger('Endless loop detected for ' . $url, LOGGER_DEBUG);
+ Logger::log('Endless loop detected for ' . $url, Logger::DEBUG);
return $siteinfo;
}
@@ -135,23 +135,23 @@ class ParseUrl
$siteinfo['url'] = $url;
$siteinfo['type'] = 'link';
- $data = Network::curl($url);
- if (!$data['success']) {
+ $curlResult = Network::curl($url);
+ if (!$curlResult->isSuccess()) {
return $siteinfo;
}
// If the file is too large then exit
- if ($data['info']['download_content_length'] > 1000000) {
+ if (defaults($curlResult->getInfo(), 'download_content_length', 0) > 1000000) {
return $siteinfo;
}
// If it isn't a HTML file then exit
- if (($data['info']['content_type'] != '') && !strstr(strtolower($data['info']['content_type']), 'html')) {
+ if (($curlResult->getContentType() != '') && !strstr(strtolower($curlResult->getContentType()), 'html')) {
return $siteinfo;
}
- $header = $data['header'];
- $body = $data['body'];
+ $header = $curlResult->getHeader();
+ $body = $curlResult->getBody();
if ($do_oembed) {
$oembed_data = OEmbed::fetchURL($url);
@@ -161,7 +161,8 @@ class ParseUrl
$siteinfo['type'] = $oembed_data->type;
}
- if (($oembed_data->type == 'link') && ($siteinfo['type'] != 'photo')) {
+ // See https://github.com/friendica/friendica/pull/5763#discussion_r217913178
+ if ($siteinfo['type'] != 'photo') {
if (isset($oembed_data->title)) {
$siteinfo['title'] = trim($oembed_data->title);
}
@@ -181,12 +182,11 @@ class ParseUrl
$charset = trim(trim(trim(array_pop($matches)), ';,'));
}
- if ($charset == '') {
- $charset = 'utf-8';
- }
+ if ($charset && strtoupper($charset) != 'UTF-8') {
+ // See https://github.com/friendica/friendica/issues/5470#issuecomment-418351211
+ $charset = str_ireplace('latin-1', 'latin1', $charset);
- if (($charset != '') && (strtoupper($charset) != 'UTF-8')) {
- logger('detected charset ' . $charset, LOGGER_DEBUG);
+ Logger::log('detected charset ' . $charset, Logger::DEBUG);
$body = iconv($charset, 'UTF-8//TRANSLIT', $body);
}
@@ -271,11 +271,9 @@ class ParseUrl
$siteinfo['image'] = $meta_tag['content'];
break;
case 'twitter:card':
- // Obsolete card type
- if ($meta_tag['content'] == 'photo') {
- $siteinfo['type'] = 'summary_large_image';
- } else {
- $siteinfo['type'] = $meta_tag['content'];
+ // Detect photo pages
+ if ($meta_tag['content'] == 'summary_large_image') {
+ $siteinfo['type'] = 'photo';
}
break;
case 'twitter:description':
@@ -299,10 +297,6 @@ class ParseUrl
}
}
- if ($siteinfo['type'] == 'summary' || $siteinfo['type'] == 'summary_large_image') {
- $siteinfo['type'] = 'link';
- }
-
if (isset($keywords)) {
$siteinfo['keywords'] = [];
foreach ($keywords as $keyword) {
@@ -338,7 +332,12 @@ class ParseUrl
}
}
- if ((@$siteinfo['image'] == '') && !$no_guessing) {
+ // Prevent to have a photo type without an image
+ if ((empty($siteinfo['image']) || !empty($siteinfo['text'])) && ($siteinfo['type'] == 'photo')) {
+ $siteinfo['type'] = 'link';
+ }
+
+ if (empty($siteinfo['image']) && !$no_guessing) {
$list = $xpath->query('//img[@src]');
foreach ($list as $node) {
$img_tag = [];
@@ -421,9 +420,9 @@ class ParseUrl
}
}
- logger('Siteinfo for ' . $url . ' ' . print_r($siteinfo, true), LOGGER_DEBUG);
+ Logger::log('Siteinfo for ' . $url . ' ' . print_r($siteinfo, true), Logger::DEBUG);
- Addon::callHooks('getsiteinfo', $siteinfo);
+ Hook::callAll('getsiteinfo', $siteinfo);
return $siteinfo;
}
@@ -487,21 +486,23 @@ class ParseUrl
$complete = $schemearr["scheme"]."://".$schemearr["host"];
- if (@$schemearr["port"] != "") {
+ if (!empty($schemearr["port"])) {
$complete .= ":".$schemearr["port"];
}
- if (strpos($urlarr["path"], "/") !== 0) {
- $complete .= "/";
- }
+ if (!empty($urlarr["path"])) {
+ if (strpos($urlarr["path"], "/") !== 0) {
+ $complete .= "/";
+ }
- $complete .= $urlarr["path"];
+ $complete .= $urlarr["path"];
+ }
- if (@$urlarr["query"] != "") {
+ if (!empty($urlarr["query"])) {
$complete .= "?".$urlarr["query"];
}
- if (@$urlarr["fragment"] != "") {
+ if (!empty($urlarr["fragment"])) {
$complete .= "#".$urlarr["fragment"];
}