diff --git a/src/Content/OEmbed.php b/src/Content/OEmbed.php index 445c52ced0..9af01b76da 100644 --- a/src/Content/OEmbed.php +++ b/src/Content/OEmbed.php @@ -22,10 +22,9 @@ namespace Friendica\Content; use DOMDocument; -use DOMNode; -use DOMText; use DOMXPath; use Exception; +use Friendica\Content\Text\BBCode; use Friendica\Core\Cache\Enum\Duration; use Friendica\Core\Hook; use Friendica\Core\Renderer; @@ -49,32 +48,15 @@ use Friendica\Util\Strings; */ class OEmbed { - /** - * Callback for fetching URL, checking allowance and returning formatted HTML - * - * @param array $matches - * @return string Formatted HTML - */ - public static function replaceCallback(array $matches): string - { - $embedurl = $matches[1]; - $j = self::fetchURL($embedurl, !self::isAllowedURL($embedurl)); - $s = self::formatObject($j); - - return $s; - } - /** * Get data from an URL to embed its content. * - * @param string $embedurl The URL from which the data should be fetched. - * @param bool $no_rich_type If set to true rich type content won't be fetched. - * @param bool $use_parseurl Use the "ParseUrl" functionality to add additional data + * @param string $embedurl The URL from which the data should be fetched. * * @return \Friendica\Object\OEmbed * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ - public static function fetchURL(string $embedurl, bool $no_rich_type = false, bool $use_parseurl = true): \Friendica\Object\OEmbed + private static function fetchURL(string $embedurl): \Friendica\Object\OEmbed { $embedurl = trim($embedurl, '\'"'); @@ -119,7 +101,7 @@ class OEmbed $href = str_replace(['http://www.youtube.com/', 'http://player.vimeo.com/'], ['https://www.youtube.com/', 'https://player.vimeo.com/'], $href); $result = DI::httpClient()->fetchFull($href . '&maxwidth=' . $a->getThemeInfoValue('videowidth')); - if ($result->getReturnCode() === 200) { + if ($result->isSuccess()) { $json_string = $result->getBodyString(); break; } @@ -157,57 +139,55 @@ class OEmbed } // Improve the OEmbed data with data from OpenGraph, Twitter cards and other sources - if ($use_parseurl) { - $data = ParseUrl::getSiteinfoCached($embedurl, false); + $data = ParseUrl::getSiteinfoCached($embedurl); - if (($oembed->type == 'error') && empty($data['title']) && empty($data['text'])) { - return $oembed; - } + if (($oembed->type == 'error') && empty($data['title']) && empty($data['text'])) { + return $oembed; + } - if ($no_rich_type || ($oembed->type == 'error')) { - $oembed->html = ''; - $oembed->type = $data['type']; + if (!self::isAllowedURL($embedurl) || ($oembed->type == 'error')) { + $oembed->html = ''; + $oembed->type = $data['type']; - if ($oembed->type == 'photo') { - if (!empty($data['images'])) { - $oembed->url = $data['images'][0]['src']; - $oembed->width = $data['images'][0]['width']; - $oembed->height = $data['images'][0]['height']; - } else { - $oembed->type = 'link'; - } + if ($oembed->type == 'photo') { + if (!empty($data['images'])) { + $oembed->url = $data['images'][0]['src']; + $oembed->width = $data['images'][0]['width']; + $oembed->height = $data['images'][0]['height']; + } else { + $oembed->type = 'link'; } } + } - if (!empty($data['title'])) { - $oembed->title = $data['title']; - } + if (!empty($data['title'])) { + $oembed->title = $data['title']; + } - if (!empty($data['text'])) { - $oembed->description = $data['text']; - } + if (!empty($data['text'])) { + $oembed->description = $data['text']; + } - if (!empty($data['publisher_name'])) { - $oembed->provider_name = $data['publisher_name']; - } + if (!empty($data['publisher_name'])) { + $oembed->provider_name = $data['publisher_name']; + } - if (!empty($data['publisher_url'])) { - $oembed->provider_url = $data['publisher_url']; - } + if (!empty($data['publisher_url'])) { + $oembed->provider_url = $data['publisher_url']; + } - if (!empty($data['author_name'])) { - $oembed->author_name = $data['author_name']; - } + if (!empty($data['author_name'])) { + $oembed->author_name = $data['author_name']; + } - if (!empty($data['author_url'])) { - $oembed->author_url = $data['author_url']; - } + if (!empty($data['author_url'])) { + $oembed->author_url = $data['author_url']; + } - if (!empty($data['images']) && ($oembed->type != 'photo')) { - $oembed->thumbnail_url = $data['images'][0]['src']; - $oembed->thumbnail_width = $data['images'][0]['width']; - $oembed->thumbnail_height = $data['images'][0]['height']; - } + if (!empty($data['images']) && ($oembed->type != 'photo')) { + $oembed->thumbnail_url = $data['images'][0]['src']; + $oembed->thumbnail_width = $data['images'][0]['width']; + $oembed->thumbnail_height = $data['images'][0]['height']; } Hook::callAll('oembed_fetch_url', $embedurl, $oembed); @@ -219,9 +199,10 @@ class OEmbed * Returns a formatted string from OEmbed object * * @param \Friendica\Object\OEmbed $oembed + * @param int $uriid * @return string */ - private static function formatObject(\Friendica\Object\OEmbed $oembed): string + private static function formatObject(\Friendica\Object\OEmbed $oembed, int $uriid): string { $ret = '
'; +$test = Proxy::proxifyHtml($ret, $uriid); return str_replace("\n", "", $ret); } @@ -308,51 +298,19 @@ class OEmbed * Converts BBCode to HTML code * * @param string $text + * @param int $uriid * @return string */ - public static function BBCode2HTML(string $text): string + public static function BBCode2HTML(string $text, int $uriid): string { - if (DI::config()->get('system', 'no_oembed')) { - return preg_replace("/\[embed\](.+?)\[\/embed\]/is", "" . DI::l10n()->t('Embedding disabled') . " : $1", $text); - } - return preg_replace_callback("/\[embed\](.+?)\[\/embed\]/is", [self::class, 'replaceCallback'], $text); - } - - /** - * Find - * and replace it with [embed]url[/embed] - * - * @param string $text - * @return string - */ - public static function HTML2BBCode(string $text): string - { - // start parser only if 'oembed' is in text - if (strpos($text, 'oembed')) { - // convert non ascii chars to html entities - $html_text = mb_convert_encoding($text, 'HTML-ENTITIES', mb_detect_encoding($text)); - - // If it doesn't parse at all, just return the text. - $dom = new DOMDocument(); - if (!@$dom->loadHTML($html_text)) { - return $text; - } - $xpath = new DOMXPath($dom); - - $xattr = self::buildXPath('class', 'oembed'); - $entries = $xpath->query("//div[$xattr]"); - - $xattr = "@rel='oembed'"; //oe_build_xpath("rel","oembed"); - foreach ($entries as $e) { - $href = $xpath->evaluate("a[$xattr]/@href", $e)->item(0)->nodeValue; - if (!is_null($href)) { - $e->parentNode->replaceChild(new DOMText('[embed]' . $href . '[/embed]'), $e); - } - } - return self::getInnerHTML($dom->getElementsByTagName('body')->item(0)); - } else { + if (!preg_match_all("/\[embed\](.+?)\[\/embed\]/is", $text, $matches, PREG_SET_ORDER)) { return $text; } + foreach ($matches as $match) { + $data = self::fetchURL($match[1]); + $text = str_replace($match[0], self::formatObject($data, $uriid), $text); + } + return $text; } /** @@ -373,26 +331,25 @@ class OEmbed return false; } - $str_allowed = DI::config()->get('system', 'allowed_oembed', ''); - if (empty($str_allowed)) { + $allowed = DI::config()->get('system', 'allowed_oembed', ''); + if (empty($allowed)) { return false; } - $allowed = explode(',', $str_allowed); - - return Network::isDomainMatch($domain, $allowed); + return Network::isDomainMatch($domain, explode(',', $allowed)); } /** * Returns a formatted HTML code from given URL and sets optional title * * @param string $url URL to fetch - * @param string $title Optional title (default: what comes from OEmbed object) + * @param string $title title (default: what comes from OEmbed object) + * @param int $uriid * @return string Formatted HTML */ - public static function getHTML(string $url, string $title = ''): string + public static function getHTML(string $url, string $title, int $uriid): string { - $o = self::fetchURL($url, !self::isAllowedURL($url)); + $o = self::fetchURL($url); if (!is_object($o) || property_exists($o, 'type') && $o->type == 'error') { throw new Exception('OEmbed failed for URL: ' . $url); @@ -402,74 +359,8 @@ class OEmbed $o->title = $title; } - $html = self::formatObject($o); + $html = self::formatObject($o, $uriid); return $html; } - - /** - * Generates the iframe HTML for an oembed attachment. - * - * Width and height are given by the remote, and are regularly too small for - * the generated iframe. - * - * The width is entirely discarded for the actual width of the post, while fixed - * height is used as a starting point before the inevitable resizing. - * - * Since the iframe is automatically resized on load, there are no need for ugly - * and impractical scrollbars. - * - * @todo This function is currently unused until someoneā¢ adds support for a separate OEmbed domain - * - * @param string $src Original remote URL to embed - * @param string $width - * @param string $height - * @return string Formatted HTML - * - * @throws \Friendica\Network\HTTPException\InternalServerErrorException - * @see oembed_format_object() - */ - private static function iframe(string $src, string $width, string $height): string - { - if (!$height || strstr($height, '%')) { - $height = '200'; - } - $width = '100%'; - - $src = DI::baseUrl() . '/oembed/' . Strings::base64UrlEncode($src); - return ''; - } - - /** - * Generates attribute search XPath string - * - * Generates an XPath query to select elements whose provided attribute contains - * the provided value in a space-separated list. - * - * @param string $attr Name of the attribute to search - * @param string $value Value to search in a space-separated list - * @return string - */ - private static function buildXPath(string $attr, $value): string - { - // https://www.westhoffswelt.de/blog/2009/6/9/select-html-elements-with-more-than-one-css-class-using-xpath - return "contains(normalize-space(@$attr), ' $value ') or substring(normalize-space(@$attr), 1, string-length('$value') + 1) = '$value ' or substring(normalize-space(@$attr), string-length(@$attr) - string-length('$value')) = ' $value' or @$attr = '$value'"; - } - - /** - * Returns the inner XML string of a provided DOMNode - * - * @param DOMNode $node - * @return string - */ - private static function getInnerHTML(DOMNode $node): string - { - $innerHTML = ''; - $children = $node->childNodes; - foreach ($children as $child) { - $innerHTML .= $child->ownerDocument->saveXML($child); - } - return $innerHTML; - } - } diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php index f5bba8ce42..877e3e2e55 100644 --- a/src/Content/Text/BBCode.php +++ b/src/Content/Text/BBCode.php @@ -310,7 +310,7 @@ class BBCode return trim($text); } - private static function proxyUrl(string $image, int $simplehtml = self::INTERNAL, int $uriid = 0, string $size = ''): string + public static function proxyUrl(string $image, int $simplehtml = self::INTERNAL, int $uriid = 0, string $size = ''): string { // Only send proxied pictures to API and for internal display if (!in_array($simplehtml, [self::INTERNAL, self::MASTODON_API, self::TWITTER_API])) { @@ -453,7 +453,7 @@ class BBCode $return = ''; try { if ($tryoembed && OEmbed::isAllowedURL($data['url'])) { - $return = OEmbed::getHTML($data['url'], $data['title']); + $return = OEmbed::getHTML($data['url'], $data['title'], $uriid); } else { throw new Exception('OEmbed is disabled for this attachment.'); } @@ -1358,12 +1358,12 @@ class BBCode * $match[1] = $url * $match[2] = $title or absent */ - $try_oembed_callback = function (array $match) { + $try_oembed_callback = function (array $match) use ($uriid) { $url = $match[1]; $title = $match[2] ?? ''; try { - $return = OEmbed::getHTML($url, $title); + $return = OEmbed::getHTML($url, $title, $uriid); } catch (Exception $ex) { $return = $match[0]; } @@ -1810,7 +1810,7 @@ class BBCode } // oembed tag - $text = OEmbed::BBCode2HTML($text); + $text = OEmbed::BBCode2HTML($text, $uriid); // Avoid triple linefeeds through oembed $text = str_replace("