From: ArthurHoaro Date: Wed, 2 May 2018 16:28:09 +0000 (+0200) Subject: Merge pull request #1133 from ArthurHoaro/hotfix/title-dl X-Git-Tag: v0.10.0~28 X-Git-Url: https://git.immae.eu/?a=commitdiff_plain;h=3e35fc10e5f94ff7f49c416dfc4b77a0a15fbbfc;hp=3c0e27eec72f7bb00aea81540f978c4b4a1ac87f;p=github%2Fshaarli%2FShaarli.git Merge pull request #1133 from ArthurHoaro/hotfix/title-dl Title retrieval fixes --- diff --git a/application/LinkUtils.php b/application/LinkUtils.php index 3705f7e9..4df5c0ca 100644 --- a/application/LinkUtils.php +++ b/application/LinkUtils.php @@ -11,6 +11,7 @@ */ function get_curl_download_callback(&$charset, &$title, $curlGetInfo = 'curl_getinfo') { + $isRedirected = false; /** * cURL callback function for CURLOPT_WRITEFUNCTION (called during the download). * @@ -22,16 +23,24 @@ function get_curl_download_callback(&$charset, &$title, $curlGetInfo = 'curl_get * * @return int|bool length of $data or false if we need to stop the download */ - return function(&$ch, $data) use ($curlGetInfo, &$charset, &$title) { + return function(&$ch, $data) use ($curlGetInfo, &$charset, &$title, &$isRedirected) { $responseCode = $curlGetInfo($ch, CURLINFO_RESPONSE_CODE); - if (!empty($responseCode) && $responseCode != 200) { + if (!empty($responseCode) && in_array($responseCode, [301, 302])) { + $isRedirected = true; + return strlen($data); + } + if (!empty($responseCode) && $responseCode !== 200) { return false; } - $contentType = $curlGetInfo($ch, CURLINFO_CONTENT_TYPE); + // 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($charset)) { + if (!empty($contentType) && empty($charset)) { $charset = header_extract_charset($contentType); } if (empty($charset)) { diff --git a/index.php b/index.php index dbc2bb3b..2fe3f821 100644 --- a/index.php +++ b/index.php @@ -1376,8 +1376,8 @@ function renderPage($conf, $pluginManager, $LINKSDB, $history, $sessionManager, // The callback will fill $charset and $title with data from the downloaded page. get_http_response( $url, - $conf->get('general.download_max_size', 4194304), $conf->get('general.download_timeout', 30), + $conf->get('general.download_max_size', 4194304), get_curl_download_callback($charset, $title) ); if (! empty($title) && strtolower($charset) != 'utf-8') {