X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=mod%2Fparse_url.php;h=6b393932ebc8aa73e176670d6889329a70320fb7;hb=b2d685482928363ce86c3c0519c8ff39d0af43ca;hp=a6b7cd5024c9aa929120ebb7b87b03901297f848;hpb=e9226eaf45e70bcd5a9a9f66b6b922dbc15c47ba;p=friendica.git diff --git a/mod/parse_url.php b/mod/parse_url.php index a6b7cd5024..6b393932eb 100644 --- a/mod/parse_url.php +++ b/mod/parse_url.php @@ -1,153 +1,197 @@ - * - * - * - * - *

Shiny Trinket

- * - *

Shiny trinkets are shiny.

- * - * @endverbatim -*/ - -use \Friendica\ParseUrl; - -require_once("include/items.php"); - -function parse_url_content(&$a) { +/** + * @file mod/parse_url.php + * @brief The parse_url module + * + * This module does parse an url for embeddable content (audio, video, image files or link) + * information and does format this information to BBCode + * + * @see ParseUrl::getSiteinfo() for more information about scraping embeddable content + */ + +use Friendica\App; +use Friendica\Core\Hook; +use Friendica\Core\Logger; +use Friendica\Core\System; +use Friendica\Util\Network; +use Friendica\Util\ParseUrl; +use Friendica\Util\Strings; + +function parse_url_content(App $a) +{ $text = null; - $str_tags = ""; + $str_tags = ''; + $format = ''; + $ret= ['success' => false, 'contentType' => '']; - $textmode = false; - - if (local_user() && (!feature_enabled(local_user(), "richtext"))) { - $textmode = true; - } + $br = "\n"; - $br = (($textmode) ? "\n" : "
"); - - if (x($_GET,"binurl")) { - $url = trim(hex2bin($_GET["binurl"])); + if (!empty($_GET['binurl']) && Strings::isHex($_GET['binurl'])) { + $url = trim(hex2bin($_GET['binurl'])); + } elseif (!empty($_GET['url'])) { + $url = trim($_GET['url']); + // fallback in case no url is valid } else { - $url = trim($_GET["url"]); + Logger::info('No url given'); + exit(); } - if ($_GET["title"]) { - $title = strip_tags(trim($_GET["title"])); + if (!empty($_GET['title'])) { + $title = strip_tags(trim($_GET['title'])); } - if ($_GET["description"]) { - $text = strip_tags(trim($_GET["description"])); + if (!empty($_GET['description'])) { + $text = strip_tags(trim($_GET['description'])); } - if ($_GET["tags"]) { - $arr_tags = ParseUrl::convertTagsToArray($_GET["tags"]); + if (!empty($_GET['tags'])) { + $arr_tags = ParseUrl::convertTagsToArray($_GET['tags']); if (count($arr_tags)) { - $str_tags = $br . implode(" ", $arr_tags) . $br; + $str_tags = $br . implode(' ', $arr_tags) . $br; } } + if (isset($_GET['format']) && $_GET['format'] == 'json') { + $format = 'json'; + } + // Add url scheme if it is missing $arrurl = parse_url($url); - if (!x($arrurl, "scheme")) { - if (x($arrurl, "host")) { - $url = "http:".$url; + if (empty($arrurl['scheme'])) { + if (!empty($arrurl['host'])) { + $url = 'http:' . $url; } else { - $url = "http://".$url; + $url = 'http://' . $url; } } - logger("prse_url: " . $url); + Logger::log($url); - // If the URL is a image, video or audio file format the URL with the corresponding - // BBCode media tag - $redirects = 0; + // Check if the URL is an image, video or audio file. If so format + // the URL with the corresponding BBCode media tag // Fetch the header of the URL - $result = z_fetch_url($url, false, $redirects, array("novalidate" => true, "nobody" => true)); - if($result["success"]) { + $curlResponse = Network::curl($url, false, ['novalidate' => true, 'nobody' => true]); + + if ($curlResponse->isSuccess()) { // Convert the header fields into an array - $hdrs = array(); - $h = explode("\n", $result["header"]); + $hdrs = []; + $h = explode("\n", $curlResponse->getHeader()); foreach ($h as $l) { - list($k,$v) = array_map("trim", explode(":", trim($l), 2)); - $hdrs[$k] = $v; + $header = array_map('trim', explode(':', trim($l), 2)); + if (count($header) == 2) { + list($k, $v) = $header; + $hdrs[$k] = $v; + } } - if (array_key_exists("Content-Type", $hdrs)) { - $type = $hdrs["Content-Type"]; + $type = null; + $content_type = ''; + $bbcode = ''; + if (array_key_exists('Content-Type', $hdrs)) { + $type = $hdrs['Content-Type']; } if ($type) { - if(stripos($type, "image/") !== false) { - echo $br . "[img]" . $url . "[/img]" . $br; - killme(); + if (stripos($type, 'image/') !== false) { + $content_type = 'image'; + $bbcode = $br . '[img]' . $url . '[/img]' . $br; } - if (stripos($type, "video/") !== false) { - echo $br . "[video]" . $url . "[/video]" . $br; - killme(); + if (stripos($type, 'video/') !== false) { + $content_type = 'video'; + $bbcode = $br . '[video]' . $url . '[/video]' . $br; } - if (stripos($type, "audio/") !== false) { - echo $br . "[audio]" . $url . "[/audio]" . $br; - killme(); + if (stripos($type, 'audio/') !== false) { + $content_type = 'audio'; + $bbcode = $br . '[audio]' . $url . '[/audio]' . $br; } } - } + if (!empty($content_type)) { + if ($format == 'json') { + $ret['contentType'] = $content_type; + $ret['data'] = ['url' => $url]; + $ret['success'] = true; + System::jsonExit($ret); + } - if ($textmode) { - $template = "[bookmark=%s]%s[/bookmark]%s"; - } else { - $template = "%s%s"; + echo $bbcode; + exit(); + } } - $arr = array("url" => $url, "text" => ""); - call_hooks("parse_link", $arr); + $template = '[bookmark=%s]%s[/bookmark]%s'; - if (strlen($arr["text"])) { - echo $arr["text"]; - killme(); - } + $arr = ['url' => $url, 'text' => '']; + Hook::callAll('parse_link', $arr); - if ($url && $title && $text) { + if (strlen($arr['text'])) { + echo $arr['text']; + exit(); + } - $title = str_replace(array("\r","\n"),array("",""),$title); + // If there is already some content information submitted we don't + // need to parse the url for content. + if (!empty($url) && !empty($title) && !empty($text)) { + $title = str_replace(["\r", "\n"], ['', ''], $title); - if ($textmode) { - $text = "[quote]" . trim($text) . "[/quote]" . $br; - } else { - $text = "
" . htmlspecialchars(trim($text)) . "

"; - $title = htmlspecialchars($title); - } + $text = '[quote]' . trim($text) . '[/quote]' . $br; $result = sprintf($template, $url, ($title) ? $title : $url, $text) . $str_tags; - logger("parse_url (unparsed): returns: " . $result); + Logger::log('(unparsed): returns: ' . $result); echo $result; - killme(); + exit(); } - // Fetch the information from the webpage + // Fetch the information directly from the webpage $siteinfo = ParseUrl::getSiteinfo($url); - unset($siteinfo["keywords"]); + unset($siteinfo['keywords']); - // Format it as BBCode attachment - $info = add_page_info_data($siteinfo); + // Bypass attachment if parse url for a comment + if (!empty($_GET['noAttachment'])) { + echo $br . '[url=' . $url . ']' . $siteinfo['title'] . '[/url]'; + exit(); + } - if (!$textmode) { - // Replace ' with ’ - not perfect - but the richtext editor has problems otherwise - $info = str_replace(array("'"), array("’"), $info); + if ($format == 'json') { + $ret['data'] = $siteinfo; + $ret['contentType'] = 'attachment'; + $ret['success'] = true; + + System::jsonExit($ret); } + // Format it as BBCode attachment + $info = add_page_info_data($siteinfo); + echo $info; - killme(); + exit(); +} + +/** + * @brief Legacy function to call ParseUrl::getSiteinfoCached + * + * Note: We have moved the function to ParseUrl.php. This function is only for + * legacy support and will be remove in the future + * + * @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 $do_oembed The false option is used by the function fetch_oembed() + * to avoid endless loops + * + * @return array which contains needed data for embedding + * + * @throws \Friendica\Network\HTTPException\InternalServerErrorException + * @see ParseUrl::getSiteinfoCached() + * + * @deprecated since version 3.6 use ParseUrl::getSiteinfoCached instead + */ +function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) +{ + $siteinfo = ParseUrl::getSiteinfoCached($url, $no_guessing, $do_oembed); + return $siteinfo; }