From f91ad2831962b59aea1a3c95f6beb39687f64845 Mon Sep 17 00:00:00 2001 From: Hypolite Petovan Date: Sun, 4 Feb 2018 18:22:48 -0500 Subject: [PATCH] Copy include/bbcode functions to src methods - Add Core\Network class --- src/Content/Text/BBCode.php | 1349 +++++++++++++++++++++++++++++++- src/Content/Text/Plaintext.php | 37 + src/Core/Network.php | 126 +++ 3 files changed, 1510 insertions(+), 2 deletions(-) create mode 100644 src/Core/Network.php diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php index 6feafdf85..974ef1464 100644 --- a/src/Content/Text/BBCode.php +++ b/src/Content/Text/BBCode.php @@ -4,15 +4,22 @@ */ namespace Friendica\Content\Text; -use Friendica\App; +use DOMDocument; +use Exception; +use Friendica\Content\OEmbed; +use Friendica\Content\Smilies; use Friendica\Content\Text\Plaintext; +use Friendica\Core\Addon; +use Friendica\Core\Cache; use Friendica\Core\Config; use Friendica\Core\L10n; use Friendica\Core\PConfig; use Friendica\Core\System; +use Friendica\Model\Contact; use Friendica\Object\Image; -use Friendica\Util\ParseUrl; +use Friendica\Util\Map; use Friendica\Util\Network; +use Friendica\Util\ParseUrl; require_once "include/bbcode.php"; require_once "include/html2plain.php"; @@ -636,4 +643,1342 @@ class BBCode return $body; } } + + /** + * Processes [attachment] tags + * + * Note: Can produce a [bookmark] tag in the returned string + * + * @brief Processes [attachment] tags + * @param string $return + * @param bool|int $simplehtml + * @param bool $tryoembed + * @return string + */ + private static function convertAttachment($return, $simplehtml = false, $tryoembed = true) + { + $data = self::getAttachmentData($return); + if (!$data) { + return $return; + } + + if (isset($data["title"])) { + $data["title"] = strip_tags($data["title"]); + $data["title"] = str_replace(["http://", "https://"], "", $data["title"]); + } + + if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) { + $data["preview"] = $data["image"]; + $data["image"] = ""; + } + + $return = ''; + if ($simplehtml == 7) { + $return = self::convertUrlForMastodon($data["url"]); + } elseif (($simplehtml != 4) && ($simplehtml != 0)) { + $return = sprintf('%s
', $data["url"], $data["title"]); + } else { + try { + if ($tryoembed && OEmbed::isAllowedURL($data['url'])) { + $return = OEmbed::getHTML($data['url'], $data['title']); + } else { + throw new Exception('OEmbed is disabled for this attachment.'); + } + } catch (Exception $e) { + if ($simplehtml != 4) { + $return = sprintf('
', $data["type"]); + } + + if ($data["image"] != "") { + $return .= sprintf('
', $data["url"], proxy_url($data["image"]), $data["title"]); + } elseif ($data["preview"] != "") { + $return .= sprintf('
', $data["url"], proxy_url($data["preview"]), $data["title"]); + } + + if (($data["type"] == "photo") && ($data["url"] != "") && ($data["image"] != "")) { + $return .= sprintf('', $data["url"], proxy_url($data["image"]), $data["title"]); + } else { + $return .= sprintf('

%s

', $data['url'], $data['title']); + } + + if ($data["description"] != "" && $data["description"] != $data["title"]) { + $return .= sprintf('
%s
', trim(bbcode($data["description"]))); + } + + if ($data["type"] == "link") { + $return .= sprintf('%s', $data['url'], parse_url($data['url'], PHP_URL_HOST)); + } + + if ($simplehtml != 4) { + $return .= '
'; + } + } + } + + return trim($data["text"] . ' ' . $return . ' ' . $data["after"]); + } + + public static function removeShareInformation($Text, $plaintext = false, $nolink = false) + { + $data = self::getAttachmentData($Text); + + if (!$data) { + return $Text; + } elseif ($nolink) { + return $data["text"] . $data["after"]; + } + + $title = htmlentities($data["title"], ENT_QUOTES, 'UTF-8', false); + $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false); + if ($plaintext || (($title != "") && strstr($text, $title))) { + $data["title"] = $data["url"]; + } elseif (($text != "") && strstr($title, $text)) { + $data["text"] = $data["title"]; + $data["title"] = $data["url"]; + } + + if (($data["text"] == "") && ($data["title"] != "") && ($data["url"] == "")) { + return $data["title"] . $data["after"]; + } + + // If the link already is included in the post, don't add it again + if (($data["url"] != "") && strpos($data["text"], $data["url"])) { + return $data["text"] . $data["after"]; + } + + $text = $data["text"]; + + if (($data["url"] != "") && ($data["title"] != "")) { + $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]"; + } elseif (($data["url"] != "")) { + $text .= "\n" . $data["url"]; + } + + return $text . "\n" . $data["after"]; + } + + private static function cleanCss($input) + { + $cleaned = ""; + + $input = strtolower($input); + + for ($i = 0; $i < strlen($input); $i++) { + $char = substr($input, $i, 1); + + if (($char >= "a") && ($char <= "z")) { + $cleaned .= $char; + } + + if (!(strpos(" #;:0123456789-_.%", $char) === false)) { + $cleaned .= $char; + } + } + + return $cleaned; + } + + /** + * Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function) + * + * @brief Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function) + * @param array $match Array with the matching values + * @return string reformatted link including HTML codes + */ + private static function convertUrlForMastodonCallback($match) + { + $url = $match[1]; + + if (isset($match[2]) && ($match[1] != $match[2])) { + return $match[0]; + } + + $parts = parse_url($url); + if (!isset($parts['scheme'])) { + return $match[0]; + } + + return self::convertUrlForMastodon($url); + } + + /** + * @brief Converts [url] BBCodes in a format that looks fine on Mastodon and GNU Social. + * @param string $url URL that is about to be reformatted + * @return string reformatted link including HTML codes + */ + private static function convertUrlForMastodon($url) + { + $parts = parse_url($url); + $scheme = $parts['scheme'] . '://'; + $styled_url = str_replace($scheme, '', $url); + + $html = '' . + ''; + + if (strlen($styled_url) > 30) { + $html .= '%s' . + ''; + + $ellipsis = substr($styled_url, 0, 30); + $rest = substr($styled_url, 30); + return sprintf($html, $url, $scheme, $ellipsis, $rest); + } else { + $html .= '%s'; + return sprintf($html, $url, $scheme, $styled_url); + } + } + + /* + * [noparse][i]italic[/i][/noparse] turns into + * [noparse][ i ]italic[ /i ][/noparse], + * to hide them from parser. + */ + private static function escapeNoparseCallback($match) + { + $whole_match = $match[0]; + $captured = $match[1]; + $spacefied = preg_replace("/\[(.*?)\]/", "[ $1 ]", $captured); + $new_str = str_replace($captured, $spacefied, $whole_match); + return $new_str; + } + + /* + * The previously spacefied [noparse][ i ]italic[ /i ][/noparse], + * now turns back and the [noparse] tags are trimed + * returning [i]italic[/i] + */ + private static function unescapeNoparseCallback($match) + { + $captured = $match[1]; + $unspacefied = preg_replace("/\[ (.*?)\ ]/", "[$1]", $captured); + return $unspacefied; + } + + /** + * Returns the bracket character positions of a set of opening and closing BBCode tags, optionally skipping first + * occurrences + * + * @param string $text Text to search + * @param string $name Tag name + * @param int $occurrences Number of first occurrences to skip + * @return boolean|array + */ + public static function getTagPosition($text, $name, $occurrences = 0) + { + if ($occurrences < 0) { + $occurrences = 0; + } + + $start_open = -1; + for ($i = 0; $i <= $occurrences; $i++) { + if ($start_open !== false) { + $start_open = strpos($text, '[' . $name, $start_open + 1); // allow [name= type tags + } + } + + if ($start_open === false) { + return false; + } + + $start_equal = strpos($text, '=', $start_open); + $start_close = strpos($text, ']', $start_open); + + if ($start_close === false) { + return false; + } + + $start_close++; + + $end_open = strpos($text, '[/' . $name . ']', $start_close); + + if ($end_open === false) { + return false; + } + + $res = [ + 'start' => [ + 'open' => $start_open, + 'close' => $start_close + ], + 'end' => [ + 'open' => $end_open, + 'close' => $end_open + strlen('[/' . $name . ']') + ], + ]; + + if ($start_equal !== false) { + $res['start']['equal'] = $start_equal + 1; + } + + return $res; + } + + /** + * Performs a preg_replace within the boundaries of all named BBCode tags in a text + * + * @param type $pattern Preg pattern string + * @param type $replace Preg replace string + * @param type $name BBCode tag name + * @param type $text Text to search + * @return string + */ + public static function pregReplaceInTag($pattern, $replace, $name, $text) + { + $occurence = 1; + $pos = get_bb_tag_pos($text, $name, $occurence); + while ($pos !== false && $occurence < 1000) { + $start = substr($text, 0, $pos['start']['open']); + $subject = substr($text, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']); + $end = substr($text, $pos['end']['close']); + if ($end === false) { + $end = ''; + } + + $subject = preg_replace($pattern, $replace, $subject); + $text = $start . $subject . $end; + + $occurence++; + $pos = get_bb_tag_pos($text, $name, $occurence); + } + + return $text; + } + + public static function extractImagesFromItemBody($body) + { + $saved_image = []; + $orig_body = $body; + $new_body = ''; + + $cnt = 0; + $img_start = strpos($orig_body, '[img'); + $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false); + $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false); + while (($img_st_close !== false) && ($img_end !== false)) { + $img_st_close++; // make it point to AFTER the closing bracket + $img_end += $img_start; + + if (!strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) { + // This is an embedded image + $saved_image[$cnt] = substr($orig_body, $img_start + $img_st_close, $img_end - ($img_start + $img_st_close)); + $new_body = $new_body . substr($orig_body, 0, $img_start) . '[$#saved_image' . $cnt . '#$]'; + + $cnt++; + } else { + $new_body = $new_body . substr($orig_body, 0, $img_end + strlen('[/img]')); + } + + $orig_body = substr($orig_body, $img_end + strlen('[/img]')); + + if ($orig_body === false) { + // in case the body ends on a closing image tag + $orig_body = ''; + } + + $img_start = strpos($orig_body, '[img'); + $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false); + $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false); + } + + $new_body = $new_body . $orig_body; + + return ['body' => $new_body, 'images' => $saved_image]; + } + + public static function interpolateSavedImagesIntoItemBody($body, array $images) + { + $newbody = $body; + + $cnt = 0; + foreach ($images as $image) { + // We're depending on the property of 'foreach' (specified on the PHP website) that + // it loops over the array starting from the first element and going sequentially + // to the last element + $newbody = str_replace('[$#saved_image' . $cnt . '#$]', + '' . L10n::t('Image/photo') . '', $newbody); + $cnt++; + } + + return $newbody; + } + + /** + * Processes [share] tags + * + * Note: Can produce a [bookmark] tag in the output + * + * @brief Processes [share] tags + * @param array $share preg_match_callback result array + * @param bool|int $simplehtml + * @return string + */ + public static function convertShare($share, $simplehtml) + { + $attributes = $share[2]; + + $author = ""; + preg_match("/author='(.*?)'/ism", $attributes, $matches); + if (x($matches, 1)) { + $author = html_entity_decode($matches[1], ENT_QUOTES, 'UTF-8'); + } + + preg_match('/author="(.*?)"/ism', $attributes, $matches); + if (x($matches, 1)) { + $author = $matches[1]; + } + + $profile = ""; + preg_match("/profile='(.*?)'/ism", $attributes, $matches); + if (x($matches, 1)) { + $profile = $matches[1]; + } + + preg_match('/profile="(.*?)"/ism', $attributes, $matches); + if (x($matches, 1)) { + $profile = $matches[1]; + } + + $avatar = ""; + preg_match("/avatar='(.*?)'/ism", $attributes, $matches); + if (x($matches, 1)) { + $avatar = $matches[1]; + } + + preg_match('/avatar="(.*?)"/ism', $attributes, $matches); + if (x($matches, 1)) { + $avatar = $matches[1]; + } + + $link = ""; + preg_match("/link='(.*?)'/ism", $attributes, $matches); + if (x($matches, 1)) { + $link = $matches[1]; + } + + preg_match('/link="(.*?)"/ism', $attributes, $matches); + if (x($matches, 1)) { + $link = $matches[1]; + } + + $posted = ""; + + preg_match("/posted='(.*?)'/ism", $attributes, $matches); + if (x($matches, 1)) { + $posted = $matches[1]; + } + + preg_match('/posted="(.*?)"/ism', $attributes, $matches); + if (x($matches, 1)) { + $posted = $matches[1]; + } + + // We only call this so that a previously unknown contact can be added. + // This is important for the function "get_contact_details_by_url". + // This function then can fetch an entry from the contact table. + Contact::getIdForURL($profile, 0); + + $data = Contact::getDetailsByURL($profile); + + if (x($data, "name") && x($data, "addr")) { + $userid_compact = $data["name"] . " (" . $data["addr"] . ")"; + } else { + $userid_compact = GetProfileUsername($profile, $author, true); + } + + if (x($data, "addr")) { + $userid = $data["addr"]; + } else { + $userid = GetProfileUsername($profile, $author, false); + } + + if (x($data, "name")) { + $author = $data["name"]; + } + + if (x($data, "micro")) { + $avatar = $data["micro"]; + } + + $preshare = trim($share[1]); + if ($preshare != "") { + $preshare .= "
"; + } + + switch ($simplehtml) { + case 1: + $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid . ":
»" . $share[3] . "«"; + break; + case 2: + $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ":
" . $share[3]; + break; + case 3: // Diaspora + $headline .= '' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . $userid . ':
'; + + $text = trim($share[1]); + + if ($text != "") { + $text .= "
"; + } + + if (stripos(normalise_link($link), 'http://twitter.com/') === 0) { + $text .= $headline . '
' . trim($share[3]) . "

"; + + if ($link != "") { + $text .= '
[l]'; + } + } else { + $text .= '
' . $link . ''; + } + + break; + case 4: + $headline .= '
' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8'); + $headline .= L10n::t('%2$s %3$s', $link, $userid, $posted); + $headline .= ":
"; + + $text = trim($share[1]); + + if ($text != "") { + $text .= "
"; + } + + $text .= $headline . '
' . trim($share[3]) . "

"; + + break; + case 5: + $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ":
" . $share[3]; + break; + case 6: // app.net + $text = $preshare . ">> @" . $userid_compact . ":
" . $share[3]; + break; + case 7: // statusnet/GNU Social + $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3]; + break; + case 8: // twitter + $text = $preshare . "RT @" . $userid_compact . ": " . $share[3]; + break; + case 9: // Google+/Facebook + $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ":
" . $share[3]; + + if ($link != "") { + $text .= "

" . $link; + } + break; + default: + // Transforms quoted tweets in rich attachments to avoid nested tweets + if (stripos(normalise_link($link), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($link)) { + try { + $oembed = OEmbed::getHTML($link, $preshare); + } catch (Exception $e) { + $oembed = sprintf('[bookmark=%s]%s[/bookmark]', $link, $preshare); + } + + $text = $preshare . $oembed; + } else { + $text = trim($share[1]) . "\n"; + + $avatar = proxy_url($avatar, false, PROXY_SIZE_THUMB); + + $tpl = get_markup_template('shared_content.tpl'); + $text .= replace_macros($tpl, [ + '$profile' => $profile, + '$avatar' => $avatar, + '$author' => $author, + '$link' => $link, + '$posted' => $posted, + '$content' => trim($share[3]) + ]); + } + break; + } + + return $text; + } + + private static function removePictureLinksCallback($match) + { + $text = Cache::get($match[1]); + + if (is_null($text)) { + $a = get_app(); + + $stamp1 = microtime(true); + + $ch = @curl_init($match[1]); + @curl_setopt($ch, CURLOPT_NOBODY, true); + @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true); + @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent()); + @curl_exec($ch); + $curl_info = @curl_getinfo($ch); + + $a->save_timestamp($stamp1, "network"); + + if (substr($curl_info["content_type"], 0, 6) == "image/") { + $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]"; + } else { + $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]"; + + // if its not a picture then look if its a page that contains a picture link + $body = Network::fetchUrl($match[1]); + + $doc = new DOMDocument(); + @$doc->loadHTML($body); + $xpath = new DomXPath($doc); + $list = $xpath->query("//meta[@name]"); + foreach ($list as $node) { + $attr = []; + + if ($node->attributes->length) { + foreach ($node->attributes as $attribute) { + $attr[$attribute->name] = $attribute->value; + } + } + + if (strtolower($attr["name"]) == "twitter:image") { + $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]"; + } + } + } + Cache::set($match[1], $text); + } + + return $text; + } + + private static function expandLinksCallback($match) + { + if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) { + return ($match[1] . "[url]" . $match[2] . "[/url]"); + } else { + return ($match[1] . $match[3] . " [url]" . $match[2] . "[/url]"); + } + } + + private static function cleanPictureLinksCallback($match) + { + $text = Cache::get($match[1]); + + if (is_null($text)) { + $a = get_app(); + + $stamp1 = microtime(true); + + $ch = @curl_init($match[1]); + @curl_setopt($ch, CURLOPT_NOBODY, true); + @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true); + @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent()); + @curl_exec($ch); + $curl_info = @curl_getinfo($ch); + + $a->save_timestamp($stamp1, "network"); + + // if its a link to a picture then embed this picture + if (substr($curl_info["content_type"], 0, 6) == "image/") { + $text = "[img]" . $match[1] . "[/img]"; + } else { + $text = "[img]" . $match[2] . "[/img]"; + + // if its not a picture then look if its a page that contains a picture link + $body = Network::fetchUrl($match[1]); + + $doc = new DOMDocument(); + @$doc->loadHTML($body); + $xpath = new DomXPath($doc); + $list = $xpath->query("//meta[@name]"); + foreach ($list as $node) { + $attr = []; + if ($node->attributes->length) { + foreach ($node->attributes as $attribute) { + $attr[$attribute->name] = $attribute->value; + } + } + + if (strtolower($attr["name"]) == "twitter:image") { + $text = "[img]" . $attr["content"] . "[/img]"; + } + } + } + Cache::set($match[1], $text); + } + + return $text; + } + + public static function cleanPictureLinks($text) + { + $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text); + return $return; + } + + private static function textHighlightCallback($match) + { + if (in_array(strtolower($match[1]), + ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby', + 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh']) + ) { + return text_highlight($match[2], strtolower($match[1])); + } + return $match[0]; + } + + /** + * @brief Converts a BBCode message to HTML message + * + * BBcode 2 HTML was written by WAY2WEB.net + * extended to work with Mistpark/Friendica - Mike Macgirvin + * + * Simple HTML values meaning: + * - 0: Friendica display + * - 1: Unused + * - 2: Used for Facebook, Google+, Windows Phone push, Friendica API + * - 3: Used before converting to Markdown in bb2diaspora.php + * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr + * - 5: Unused + * - 6: Used for Appnet + * - 7: Used for dfrn, OStatus + * - 8: Used for WP backlink text setting + * + * @param string $text + * @param bool $preserve_nl + * @param bool $try_oembed + * @param int $simple_html + * @param bool $for_plaintext + * @return string + */ + public static function convert($text, $preserve_nl = false, $try_oembed = true, $simple_html = false, $for_plaintext = false) + { + $a = get_app(); + + /* + * preg_match_callback function to replace potential Oembed tags with Oembed content + * + * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag] + * $match[1] = $url + * $match[2] = $title or absent + */ + $try_oembed_callback = function ($match) + { + $url = $match[1]; + $title = defaults($match, 2, null); + + try { + $return = OEmbed::getHTML($url, $title); + } catch (Exception $ex) { + $return = $match[0]; + } + + return $return; + }; + + // Hide all [noparse] contained bbtags by spacefying them + // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image? + + $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text); + $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text); + $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text); + + // Remove the abstract element. It is a non visible element. + $text = remove_abstract($text); + + // Move all spaces out of the tags + $text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $text); + $text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $text); + + // Extract the private images which use data urls since preg has issues with + // large data sizes. Stash them away while we do bbcode conversion, and then put them back + // in after we've done all the regex matching. We cannot use any preg functions to do this. + + $extracted = bb_extract_images($text); + $text = $extracted['body']; + $saved_image = $extracted['images']; + + // If we find any event code, turn it into an event. + // After we're finished processing the bbcode we'll + // replace all of the event code with a reformatted version. + + $ev = bbtoevent($text); + + // Replace any html brackets with HTML Entities to prevent executing HTML or script + // Don't use strip_tags here because it breaks [url] search by replacing & with amp + + $text = str_replace("<", "<", $text); + $text = str_replace(">", ">", $text); + + // remove some newlines before the general conversion + $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text); + $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text); + + $text = preg_replace("/\n\[code\]/ism", "[code]", $text); + $text = preg_replace("/\[\/code\]\n/ism", "[/code]", $text); + + // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems + if (!$try_oembed) { + $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text); + } + + // Check for [code] text here, before the linefeeds are messed with. + // The highlighter will unescape and re-escape the content. + if (strpos($text, '[code=') !== false) { + $text = preg_replace_callback("/\[code=(.*?)\](.*?)\[\/code\]/ism", 'self::textHighlightCallback', $text); + } + // Convert new line chars to html
tags + + // nlbr seems to be hopelessly messed up + // $Text = nl2br($Text); + + // We'll emulate it. + + $text = trim($text); + $text = str_replace("\r\n", "\n", $text); + + // removing multiplicated newlines + if (Config::get("system", "remove_multiplicated_lines")) { + $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n", + "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"]; + $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]", + "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"]; + do { + $oldtext = $text; + $text = str_replace($search, $replace, $text); + } while ($oldtext != $text); + } + + // Set up the parameters for a URL search string + $URLSearchString = "^\[\]"; + // Set up the parameters for a MAIL search string + $MAILSearchString = $URLSearchString; + + // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text + if (!$for_plaintext) { + // Autolink feature (thanks to http://code.seebz.net/p/autolink-php/) + // Currently disabled, since the function is too greedy + // $autolink_regex = "`([^\]\=\"']|^)(https?\://[^\s<]+[^\s<\.\)])`ism"; + $autolink_regex = "/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism"; + $text = preg_replace($autolink_regex, '$1[url]$2[/url]', $text); + if ($simple_html == 7) { + $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForMastodonCallback', $text); + $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForMastodonCallback', $text); + } + } else { + $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text); + $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text); + } + + + // Handle attached links or videos + $text = self::convertAttachment($text, $simple_html, $try_oembed); + + $text = str_replace(["\r","\n"], ['
', '
'], $text); + + if ($preserve_nl) { + $text = str_replace(["\n", "\r"], ['', ''], $text); + } + + // Remove all hashtag addresses + if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7])) { + $text = preg_replace("/([#@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $text); + } elseif ($simple_html == 3) { + // The ! is converted to @ since Diaspora only understands the @ + $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", + '@$3', + $text); + } elseif ($simple_html == 7) { + $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", + '$1$3', + $text); + } elseif (!$simple_html) { + $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", + '$1$3', + $text); + } + + // Bookmarks in red - will be converted to bookmarks in friendica + $text = preg_replace("/#\^\[url\]([$URLSearchString]*)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text); + $text = preg_replace("/#\^\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text); + $text = preg_replace("/#\[url\=[$URLSearchString]*\]\^\[\/url\]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i", + "[bookmark=$1]$2[/bookmark]", $text); + + if (in_array($simple_html, [2, 6, 7, 8, 9])) { + $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text); + //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text); + $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text); + } + + if ($simple_html == 5) { + $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text); + } + + // Perform URL Search + if ($try_oembed) { + $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text); + } + + if ($simple_html == 5) { + $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url]$1[/url]', $text); + } else { + $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text); + } + + // Handle Diaspora posts + $text = preg_replace_callback( + "&\[url=/posts/([^\[\]]*)\](.*)\[\/url\]&Usi", + function ($match) { + return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]"; + }, $text + ); + + // Server independent link to posts and comments + // See issue: https://github.com/diaspora/diaspora_federation/issues/75 + $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism"; + $text = preg_replace($expression, System::baseUrl()."/display/$1", $text); + + $text = preg_replace("/([#])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", + '$1$3', $text); + + $text = preg_replace("/\[url\=([$URLSearchString]*)\]#(.*?)\[\/url\]/ism", + '#$2', $text); + + $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '$1', $text); + $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$2', $text); + //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '$2', $Text); + + // Red compatibility, though the link can't be authenticated on Friendica + $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '$2', $text); + + + // we may need to restrict this further if it picks up too many strays + // link acct:user@host to a webfinger profile redirector + + $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', 'acct:$1@$2', $text); + + // Perform MAIL Search + $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '$1', $text); + $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '$2', $text); + + // leave open the posibility of [map=something] + // this is replaced in prepare_body() which has knowledge of the item location + + if (strpos($text, '[/map]') !== false) { + $text = preg_replace_callback( + "/\[map\](.*?)\[\/map\]/ism", + function ($match) { + // the extra space in the following line is intentional + // Whyyy? - @MrPetovan + return str_replace($match[0], '
' . Map::byLocation($match[1]) . '
', $match[0]); + }, + $text + ); + } + if (strpos($text, '[map=') !== false) { + $text = preg_replace_callback( + "/\[map=(.*?)\]/ism", + function ($match) { + // the extra space in the following line is intentional + // Whyyy? - @MrPetovan + return str_replace($match[0], '
' . Map::byCoordinates(str_replace('/', ' ', $match[1])) . '
', $match[0]); + }, + $text + ); + } + if (strpos($text, '[map]') !== false) { + $text = preg_replace("/\[map\]/", '
', $text); + } + + // Check for headers + $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '

$1

', $text); + $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '

$1

', $text); + $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '

$1

', $text); + $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '

$1

', $text); + $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '
$1
', $text); + $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '
$1
', $text); + + // Check for paragraph + $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '

$1

', $text); + + // Check for bold text + $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '$1', $text); + + // Check for Italics text + $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '$1', $text); + + // Check for Underline text + $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '$1', $text); + + // Check for strike-through text + $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '$1', $text); + + // Check for over-line text + $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '$1', $text); + + // Check for colored text + $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "$2", $text); + + // Check for sized text + // [size=50] --> font-size: 50px (with the unit). + $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "$2", $text); + $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text); + + // Check for centered text + $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "
$1
", $text); + + // Check for list text + $text = str_replace("[*]", "
  • ", $text); + + // Check for style sheet commands + $text = preg_replace_callback( + "(\[style=(.*?)\](.*?)\[\/style\])ism", + function ($match) { + return "" . $match[2] . ""; + }, + $text + ); + + // Check for CSS classes + $text = preg_replace_callback( + "(\[class=(.*?)\](.*?)\[\/class\])ism", + function ($match) { + return "" . $match[2] . ""; + }, + $text + ); + + // handle nested lists + $endlessloop = 0; + + while ((((strpos($text, "[/list]") !== false) && (strpos($text, "[list") !== false)) || + ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) || + ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) || + ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) { + $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '', $text); + $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '', $text); + $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '', $text); + $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '', $text); + $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '', $text); + $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '', $text); + $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '', $text); + $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '', $text); + $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '', $text); + $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '
  • $1
    • ', $text); + } + + $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '$1', $text); + $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '$1', $text); + $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '$1', $text); + $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '$1
    ', $text); + + $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '$1
    ', $text); + $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '$1
    ', $text); + + $text = str_replace('[hr]', '
    ', $text); + + // This is actually executed in prepare_body() + + $text = str_replace('[nosmile]', '', $text); + + // Check for font change text + $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "$2", $text); + + // Declare the format for [code] layout + + $CodeLayout = '$1'; + // Check for [code] text + $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text); + + // Declare the format for [spoiler] layout + $SpoilerLayout = '
    $1
    '; + + // Check for [spoiler] text + // handle nested quotes + $endlessloop = 0; + while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) { + $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text); + } + + // Check for [spoiler=Author] text + + $t_wrote = L10n::t('$1 wrote:'); + + // handle nested quotes + $endlessloop = 0; + while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) { + $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism", + "
    " . $t_wrote . "
    $2
    ", + $text); + } + + // Declare the format for [quote] layout + $QuoteLayout = '
    $1
    '; + + // Check for [quote] text + // handle nested quotes + $endlessloop = 0; + while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) { + $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text); + } + + // Check for [quote=Author] text + + $t_wrote = L10n::t('$1 wrote:'); + + // handle nested quotes + $endlessloop = 0; + while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) { + $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism", + "
    " . $t_wrote . "
    $2
    ", + $text); + } + + + // [img=widthxheight]image source[/img] + $text = preg_replace_callback( + "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", + function ($matches) { + if (strpos($matches[3], "data:image/") === 0) { + return $matches[0]; + } + + $matches[3] = proxy_url($matches[3]); + return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]"; + }, + $text + ); + + $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '', $text); + $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '', $text); + + // Images + // [img]pathtoimage[/img] + $text = preg_replace_callback( + "/\[img\](.*?)\[\/img\]/ism", + function ($matches) { + if (strpos($matches[1], "data:image/") === 0) { + return $matches[0]; + } + + $matches[1] = proxy_url($matches[1]); + return "[img]" . $matches[1] . "[/img]"; + }, + $text + ); + + $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '' . L10n::t('Image/photo') . '', $text); + $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '' . L10n::t('Image/photo') . '', $text); + + // Shared content + $text = preg_replace_callback("/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism", + function ($match) use ($simple_html) { + return bb_ShareAttributes($match, $simple_html); + }, $text); + + $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '
    ' . L10n::t('Encrypted content') . '
    ', $text); + $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '
    ' . L10n::t('Encrypted content') . '
    ', $text); + //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '
    ' . L10n::t('Encrypted content') . '
    ', $Text); + + // Try to Oembed + if ($try_oembed) { + $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4))\[\/video\]/ism", '', $text); + $text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3))\[\/audio\]/ism", '', $text); + + $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text); + $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text); + } else { + $text = preg_replace("/\[video\](.*?)\[\/video\]/", + '$1', $text); + $text = preg_replace("/\[audio\](.*?)\[\/audio\]/", + '$1', $text); + } + + // html5 video and audio + + + if ($try_oembed) { + $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '', $text); + } else { + $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '$1', $text); + } + + // Youtube extensions + if ($try_oembed) { + $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text); + $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text); + $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text); + } + + $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text); + $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text); + $text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text); + + if ($try_oembed) { + $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '', $text); + } else { + $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", + 'https://www.youtube.com/watch?v=$1', $text); + } + + if ($try_oembed) { + $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text); + $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text); + } + + $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text); + $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text); + + if ($try_oembed) { + $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '', $text); + } else { + $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", + 'https://vimeo.com/$1', $text); + } + + // oembed tag + $text = OEmbed::BBCode2HTML($text); + + // Avoid triple linefeeds through oembed + $text = str_replace("


    ", "

    ", $text); + + // If we found an event earlier, strip out all the event code and replace with a reformatted version. + // Replace the event-start section with the entire formatted event. The other bbcode is stripped. + // Summary (e.g. title) is required, earlier revisions only required description (in addition to + // start which is always required). Allow desc with a missing summary for compatibility. + + if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) { + $sub = format_event_html($ev, $simple_html); + + $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text); + $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text); + $text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism", $sub, $text); + $text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism", '', $text); + $text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism", '', $text); + $text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism", '', $text); + $text = preg_replace("/\[event\-id\](.*?)\[\/event\-id\]/ism", '', $text); + } + + // Replace non graphical smilies for external posts + if ($simple_html) { + $text = Smilies::replace($text, false, true); + } + + // Replace inline code blocks + $text = preg_replace_callback("|(?!]*>)([^<]*)(?!]*>)|ism", + function ($match) use ($simple_html) { + $return = '' . $match[1] . ''; + // Use for Diaspora inline code blocks + if ($simple_html === 3) { + $return = '' . $match[1] . ''; + } + return $return; + } + , $text); + + // Unhide all [noparse] contained bbtags unspacefying them + // and triming the [noparse] tag. + + $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::unescapeNoparseCallback', $text); + $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text); + $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text); + + + $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text); + $text = preg_replace('/\&\#039\;/', '\'', $text); + $text = preg_replace('/\"\;/', '"', $text); + + // fix any escaped ampersands that may have been converted into links + $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text); + + // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails) + $allowed_src_protocols = ['http', 'redir', 'cid']; + $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism', + '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . L10n::t('Invalid source protocol') . '">', $text); + + // sanitize href attributes (only whitelisted protocols URLs) + // default value for backward compatibility + $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', ['ftp', 'mailto', 'gopher', 'cid']); + + // Always allowed protocol even if config isn't set or not including it + $allowed_link_protocols[] = 'http'; + $allowed_link_protocols[] = 'redir/'; + + $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism'; + $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . L10n::t('Invalid link protocol') . '">', $text); + + if ($saved_image) { + $text = bb_replace_images($text, $saved_image); + } + + // Clean up the HTML by loading and saving the HTML with the DOM. + // Bad structured html can break a whole page. + // For performance reasons do it only with ativated item cache or at export. + if (!$try_oembed || (get_itemcachepath() != "")) { + $doc = new DOMDocument(); + $doc->preserveWhiteSpace = false; + + $text = mb_convert_encoding($text, 'HTML-ENTITIES', "UTF-8"); + + $doctype = ''; + $encoding = ''; + @$doc->loadHTML($encoding.$doctype."".$text.""); + $doc->encoding = 'UTF-8'; + $text = $doc->saveHTML(); + $text = str_replace(["", "", $doctype, $encoding], ["", "", "", ""], $text); + + $text = str_replace('
    ', '', $text); + + //$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES'); + } + + // Clean up some useless linebreaks in lists + //$Text = str_replace('

    ', '', $Text); + //$Text = str_replace('
    ', '', $Text); + //$Text = str_replace('
  • ', '
  • ', $Text); + //$Text = str_replace('
    $start_pos, 'end' => $end_pos]; + + return $res; + } } diff --git a/src/Core/Network.php b/src/Core/Network.php new file mode 100644 index 000000000..0d59c7e85 --- /dev/null +++ b/src/Core/Network.php @@ -0,0 +1,126 @@ + + */ +class Network +{ + const DFRN = 'dfrn'; // Friendica, Mistpark, other DFRN implementations + const DIASPORA = 'dspr'; // Diaspora + const DIASPORA2 = 'dspc'; // Diaspora connector + const STATUSNET = 'stac'; // Statusnet connector + const OSTATUS = 'stat'; // GNU-social, Pleroma, Mastodon, other OStatus implementations + const FEED = 'feed'; // RSS/Atom feeds with no known "post/notify" protocol + const MAIL = 'mail'; // IMAP/POP + const XMPP = 'xmpp'; // XMPP - Currently unsupported + + const FACEBOOK = 'face'; // Facebook API + const LINKEDIN = 'lnkd'; // LinkedIn + const MYSPACE = 'mysp'; // MySpace - Currently unsupported + const GPLUS = 'goog'; // Google+ + const PUMPIO = 'pump'; // pump.io + const TWITTER = 'twit'; // Twitter + const APPNET = 'apdn'; // app.net - Dead protocol + + const NEWS = 'nntp'; // Network News Transfer Protocol - Currently unsupported + const ICALENDAR = 'ical'; // iCalendar - Currently unsupported + const PNUT = 'pnut'; // pnut.io - Currently unsupported + const ZOT = 'zot!'; // Zot! - Currently unsupported + + const PHANTOM = 'unkn'; // Place holder + + /** + * Returns the address string for the provided profile URL + * + * @param string $profile_url + * @return string + * @throws Exception + */ + public static function getAddrFromProfileUrl($profile_url) + { + $network = self::matchByProfileUrl($profile_url, $matches); + + if ($network === self::PHANTOM) { + throw new Exception('Unknown network for profile URL: ' . $profile_url); + } + + $addr = $matches[2] . '@' . $matches[1]; + + return $addr; + } + + /** + * Guesses the network from a profile URL + * + * @param string $profile_url + * @param array $matches preg_match return array: [0] => Full match [1] => hostname [2] => username + * @return type + */ + public static function matchByProfileUrl($profile_url, &$matches = []) + { + if (preg_match('=https?://(twitter\.com)/(.*)=ism', $profile_url, $matches)) { + return self::TWITTER; + } + + if (preg_match('=https?://(alpha\.app\.net)/(.*)=ism', $profile_url, $matches)) { + return self::APPNET; + } + + if (preg_match('=https?://(plus\.google\.com)/(.*)=ism', $profile_url, $matches)) { + return self::GPLUS; + } + + if (preg_match('=https?://(.*)/profile/(.*)=ism', $profile_url, $matches)) { + return self::DFRN; + } + + if (preg_match('=https?://(.*)/u/(.*)=ism', $profile_url, $matches)) { + return self::DIASPORA; + } + + if (preg_match('=https?://(.*)/channel/(.*)=ism', $profile_url, $matches)) { + // RedMatrix/Hubzilla is identified as Diaspora - friendica can't connect directly to it + return self::DIASPORA; + } + + if (preg_match('=https?://(.*)/user/(.*)=ism', $profile_url, $matches)) { + $statusnet_host = $matches[1]; + $statusnet_user = $matches[2]; + $UserData = Util\Network::fetchUrl('http://' . $statusnet_host . '/api/users/show.json?user_id=' . $statusnet_user); + $user = json_decode($UserData); + if ($user) { + $matches[2] = $user->screen_name; + return self::STATUSNET; + } + } + + // pumpio (http://host.name/user) + if (preg_match('=https?://([\.\w]+)/([\.\w]+)$=ism', $profile_url, $matches)) { + return self::PUMPIO; + } + + return self::PHANTOM; + } + + /** + * Returns a formatted mention from a profile URL and a display name + * + * @param string $profile_url + * @param string $display_name + * @return string + */ + public static function formatMention($profile_url, $display_name) + { + return $display_name . '(' . self::getAddrFromProfileUrl($profile_url) . ')'; + } +}