X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FContent%2FText%2FBBCode.php;h=66f4190b287eca731fe8b465dea18da5504b0feb;hb=5a02e39a65f8f685440228cc1d36738cbe15f32b;hp=13d4e1b055f66b71c1e369fa5e6151819aee47b8;hpb=29f7ebe307c22b275466390937b82ccb3820fb1c;p=friendica.git diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php index 13d4e1b055..66f4190b28 100644 --- a/src/Content/Text/BBCode.php +++ b/src/Content/Text/BBCode.php @@ -2,33 +2,31 @@ /** * @file src/Content/Text/BBCode.php */ + namespace Friendica\Content\Text; use DOMDocument; -use DomXPath; +use DOMXPath; use Exception; +use Friendica\BaseObject; 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\Protocol; -use Friendica\Core\PConfig; use Friendica\Core\System; use Friendica\Model\Contact; +use Friendica\Model\Event; +use Friendica\Network\Probe; use Friendica\Object\Image; use Friendica\Util\Map; use Friendica\Util\Network; use Friendica\Util\ParseUrl; +use Friendica\Util\Proxy as ProxyUtils; -require_once "include/event.php"; -require_once "include/html2plain.php"; -require_once "include/html2bbcode.php"; -require_once "mod/proxy.php"; - -class BBCode +class BBCode extends BaseObject { /** * @brief Fetches attachment data that were generated the old way @@ -74,10 +72,12 @@ class BBCode $picturedata = Image::getInfoFromURL($matches[1]); - if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) { - $post["image"] = $matches[1]; - } else { - $post["preview"] = $matches[1]; + if ($picturedata) { + if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) { + $post["image"] = $matches[1]; + } else { + $post["preview"] = $matches[1]; + } } } @@ -85,7 +85,7 @@ class BBCode $post["url"] = $matches[1]; $post["title"] = $matches[2]; } - if (($post["url"] == "") && (in_array($post["type"], ["link", "video"])) + if (!empty($post["url"]) && (in_array($post["type"], ["link", "video"])) && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) { $post["url"] = $matches[1]; } @@ -174,7 +174,7 @@ class BBCode } if ($title != "") { - $title = BBCode::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true); + $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true); $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8'); $title = str_replace(["[", "]"], ["[", "]"], $title); $data["title"] = $title; @@ -239,6 +239,9 @@ class BBCode $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body); $URLSearchString = "^\[\]"; + + $body = preg_replace("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body); + if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) { if ((count($pictures) == 1) && !$has_title) { // Checking, if the link goes to a picture @@ -264,7 +267,7 @@ class BBCode $post["text"] = str_replace($pictures[0][0], "", $body); } else { $imgdata = Image::getInfoFromURL($pictures[0][1]); - if (substr($imgdata["mime"], 0, 6) == "image/") { + if ($imgdata && substr($imgdata["mime"], 0, 6) == "image/") { $post["type"] = "photo"; $post["image"] = $pictures[0][1]; $post["preview"] = $pictures[0][2]; @@ -336,159 +339,30 @@ class BBCode } /** - * @brief Convert a message into plaintext for connectors to other networks + * @brief Converts a BBCode text into plaintext * - * @param array $b The message array that is about to be posted - * @param int $limit The maximum number of characters when posting to that network - * @param bool $includedlinks Has an attached link to be included into the message? - * @param int $htmlmode This triggers the behaviour of the bbcode conversion - * @param string $target_network Name of the network where the post should go to. + * @param bool $keep_urls Whether to keep URLs in the resulting plaintext * - * @return string The converted message + * @return string */ - public static function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "") + public static function toPlaintext($text, $keep_urls = true) { - // Remove the hash tags - $URLSearchString = "^\[\]"; - $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]); - - // Add an URL element if the text contains a raw link - $body = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2[/url]', $body); - - // Remove the abstract - $body = self::stripAbstract($body); - - // At first look at data that is attached via "type-..." stuff - // This will hopefully replaced with a dedicated bbcode later - //$post = self::getAttachedData($b["body"]); - $post = self::getAttachedData($body, $b); - - if (($b["title"] != "") && ($post["text"] != "")) { - $post["text"] = trim($b["title"]."\n\n".$post["text"]); - } elseif ($b["title"] != "") { - $post["text"] = trim($b["title"]); - } - - $abstract = ""; - - // Fetch the abstract from the given target network - if ($target_network != "") { - $default_abstract = self::getAbstract($b["body"]); - $abstract = self::getAbstract($b["body"], $target_network); - - // If we post to a network with no limit we only fetch - // an abstract exactly for this network - if (($limit == 0) && ($abstract == $default_abstract)) { - $abstract = ""; - } - } else {// Try to guess the correct target network - switch ($htmlmode) { - case 8: - $abstract = self::getAbstract($b["body"], NETWORK_TWITTER); - break; - case 7: - $abstract = self::getAbstract($b["body"], NETWORK_STATUSNET); - break; - case 6: - $abstract = self::getAbstract($b["body"], NETWORK_APPNET); - break; - default: // We don't know the exact target. - // We fetch an abstract since there is a posting limit. - if ($limit > 0) { - $abstract = self::getAbstract($b["body"]); - } - } - } - - if ($abstract != "") { - $post["text"] = $abstract; - - if ($post["type"] == "text") { - $post["type"] = "link"; - $post["url"] = $b["plink"]; - } - } - - $html = BBCode::convert($post["text"].$post["after"], false, $htmlmode); - $msg = html2plain($html, 0, true); - $msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8')); - - $link = ""; - if ($includedlinks) { - if ($post["type"] == "link") { - $link = $post["url"]; - } elseif ($post["type"] == "text") { - $link = $post["url"]; - } elseif ($post["type"] == "video") { - $link = $post["url"]; - } elseif ($post["type"] == "photo") { - $link = $post["image"]; - } - - if (($msg == "") && isset($post["title"])) { - $msg = trim($post["title"]); - } - - if (($msg == "") && isset($post["description"])) { - $msg = trim($post["description"]); - } - - // If the link is already contained in the post, then it neeedn't to be added again - // But: if the link is beyond the limit, then it has to be added. - if (($link != "") && strstr($msg, $link)) { - $pos = strpos($msg, $link); - - // Will the text be shortened in the link? - // Or is the link the last item in the post? - if (($limit > 0) && ($pos < $limit) && (($pos + 23 > $limit) || ($pos + strlen($link) == strlen($msg)))) { - $msg = trim(str_replace($link, "", $msg)); - } elseif (($limit == 0) || ($pos < $limit)) { - // The limit has to be increased since it will be shortened - but not now - // Only do it with Twitter (htmlmode = 8) - if (($limit > 0) && (strlen($link) > 23) && ($htmlmode == 8)) { - $limit = $limit - 23 + strlen($link); - } - - $link = ""; - - if ($post["type"] == "text") { - unset($post["url"]); - } - } - } + $naked_text = preg_replace('/\[(.+?)\]\s*/','', $text); + if (!$keep_urls) { + $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text); } - if ($limit > 0) { - // Reduce multiple spaces - // When posted to a network with limited space, we try to gain space where possible - while (strpos($msg, " ") !== false) { - $msg = str_replace(" ", " ", $msg); - } - - // Twitter is using its own limiter, so we always assume that shortened links will have this length - if (iconv_strlen($link, "UTF-8") > 0) { - $limit = $limit - 23; - } + return $naked_text; + } - if (iconv_strlen($msg, "UTF-8") > $limit) { - if (($post["type"] == "text") && isset($post["url"])) { - $post["url"] = $b["plink"]; - } elseif (!isset($post["url"])) { - $limit = $limit - 23; - $post["url"] = $b["plink"]; - // Which purpose has this line? It is now uncommented, but left as a reminder - //} elseif (strpos($b["body"], "[share") !== false) { - // $post["url"] = $b["plink"]; - } elseif (PConfig::get($b["uid"], "system", "no_intelligent_shortening")) { - $post["url"] = $b["plink"]; - } - $msg = Plaintext::shorten($msg, $limit); - } + private static function proxyUrl($image, $simplehtml = false) + { + // Only send proxied pictures to API and for internal display + if (in_array($simplehtml, [false, 2])) { + return ProxyUtils::proxifyUrl($image); + } else { + return $image; } - - $post["text"] = trim($msg); - - return($post); } public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false) @@ -571,7 +445,7 @@ class BBCode */ public static function limitBodySize($body) { - $maxlen = get_max_import_size(); + $maxlen = Config::get('config', 'max_import_size', 0); // If the length of the body, including the embedded images, is smaller // than the maximum, then don't waste time looking for the images @@ -662,16 +536,18 @@ class BBCode private static function convertAttachment($return, $simplehtml = false, $tryoembed = true) { $data = self::getAttachmentData($return); - if (!$data) { + if (empty($data) || empty($data["url"])) { return $return; } if (isset($data["title"])) { $data["title"] = strip_tags($data["title"]); $data["title"] = str_replace(["http://", "https://"], "", $data["title"]); + } else { + $data["title"] = null; } - if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) { + if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data["image"])) { $data["preview"] = $data["image"]; $data["image"] = ""; } @@ -689,28 +565,32 @@ class BBCode throw new Exception('OEmbed is disabled for this attachment.'); } } catch (Exception $e) { + $data["title"] = defaults($data, 'title', $data['url']); + if ($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'], $data['title']); + if (!empty($data['title']) && !empty($data['url'])) { + if (!empty($data["image"]) && empty($data["text"]) && ($data["type"] == "photo")) { + $return .= sprintf('', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]); + } else { + if (!empty($data["image"])) { + $return .= sprintf('
', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]); + } elseif (!empty($data["preview"])) { + $return .= sprintf('
', $data["url"], self::proxyUrl($data["preview"], $simplehtml), $data["title"]); + } + $return .= sprintf('

%s

', $data['url'], $data['title']); + } } - if ($data["description"] != "" && $data["description"] != $data["title"]) { + if (!empty($data["description"]) && $data["description"] != $data["title"]) { // Sanitize the HTML by converting it to BBCode - $bbcode = html2bbcode($data["description"]); + $bbcode = HTML::toBBCode($data["description"]); $return .= sprintf('
%s
', trim(self::convert($bbcode))); } - if ($data["type"] == "link") { + + if (!empty($data['url'])) { $return .= sprintf('%s', $data['url'], parse_url($data['url'], PHP_URL_HOST)); } @@ -720,7 +600,7 @@ class BBCode } } - return trim($data["text"] . ' ' . $return . ' ' . $data["after"]); + return trim(defaults($data, 'text', '') . ' ' . $return . ' ' . defaults($data, 'after', '')); } public static function removeShareInformation($Text, $plaintext = false, $nolink = false) @@ -730,10 +610,10 @@ class BBCode if (!$data) { return $Text; } elseif ($nolink) { - return $data["text"] . $data["after"]; + return $data["text"] . defaults($data, 'after', ''); } - $title = htmlentities($data["title"], ENT_QUOTES, 'UTF-8', false); + $title = htmlentities(defaults($data, 'title', ''), ENT_QUOTES, 'UTF-8', false); $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false); if ($plaintext || (($title != "") && strstr($text, $title))) { $data["title"] = $data["url"]; @@ -747,42 +627,21 @@ class BBCode } // If the link already is included in the post, don't add it again - if (($data["url"] != "") && strpos($data["text"], $data["url"])) { + if (!empty($data["url"]) && strpos($data["text"], $data["url"])) { return $data["text"] . $data["after"]; } $text = $data["text"]; - if (($data["url"] != "") && ($data["title"] != "")) { + if (!empty($data["url"]) && !empty($data["title"])) { $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]"; - } elseif (($data["url"] != "")) { + } elseif (!empty($data["url"])) { $text .= "\n[url]" . $data["url"] . "[/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) * @@ -992,7 +851,7 @@ class BBCode // 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); + '' . L10n::t('Image/photo') . '', $newbody); $cnt++; } @@ -1145,16 +1004,13 @@ class BBCode 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 + case 9: // Google+ $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ":
" . $share[3]; if ($link != "") { @@ -1174,7 +1030,7 @@ class BBCode } else { $text = trim($share[1]) . "\n"; - $avatar = proxy_url($avatar, false, PROXY_SIZE_THUMB); + $avatar = ProxyUtils::proxifyUrl($avatar, false, ProxyUtils::SIZE_THUMB); $tpl = get_markup_template('shared_content.tpl'); $text .= replace_macros($tpl, [ @@ -1197,18 +1053,18 @@ class BBCode $text = Cache::get($match[1]); if (is_null($text)) { - $a = get_app(); + $a = self::getApp(); $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, $a->get_useragent()); + @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent()); @curl_exec($ch); $curl_info = @curl_getinfo($ch); - $a->save_timestamp($stamp1, "network"); + $a->saveTimestamp($stamp1, "network"); if (substr($curl_info["content_type"], 0, 6) == "image/") { $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]"; @@ -1220,7 +1076,7 @@ class BBCode $doc = new DOMDocument(); @$doc->loadHTML($body); - $xpath = new DomXPath($doc); + $xpath = new DOMXPath($doc); $list = $xpath->query("//meta[@name]"); foreach ($list as $node) { $attr = []; @@ -1256,18 +1112,18 @@ class BBCode $text = Cache::get($match[1]); if (is_null($text)) { - $a = get_app(); + $a = self::getApp(); $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, $a->get_useragent()); + @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent()); @curl_exec($ch); $curl_info = @curl_getinfo($ch); - $a->save_timestamp($stamp1, "network"); + $a->saveTimestamp($stamp1, "network"); // if its a link to a picture then embed this picture if (substr($curl_info["content_type"], 0, 6) == "image/") { @@ -1280,7 +1136,7 @@ class BBCode $doc = new DOMDocument(); @$doc->loadHTML($body); - $xpath = new DomXPath($doc); + $xpath = new DOMXPath($doc); $list = $xpath->query("//meta[@name]"); foreach ($list as $node) { $attr = []; @@ -1307,17 +1163,6 @@ class BBCode 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 $match[0]; - } - /** * @brief Converts a BBCode message to HTML message * @@ -1327,11 +1172,11 @@ class BBCode * Simple HTML values meaning: * - 0: Friendica display * - 1: Unused - * - 2: Used for Facebook, Google+, Windows Phone push, Friendica API + * - 2: Used for Google+, Windows Phone push, Friendica API * - 3: Used before converting to Markdown in bb2diaspora.php * - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr * - 5: Unused - * - 6: Used for Appnet + * - 6: Unused * - 7: Used for dfrn, OStatus * - 8: Used for WP backlink text setting * @@ -1343,7 +1188,7 @@ class BBCode */ public static function convert($text, $try_oembed = true, $simple_html = false, $for_plaintext = false) { - $a = get_app(); + $a = self::getApp(); /* * preg_match_callback function to replace potential Oembed tags with Oembed content @@ -1366,6 +1211,22 @@ class BBCode return $return; }; + // Extracting multi-line code blocks before the whitespace processing + $codeblocks = []; + + $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#is", + function ($matches) use (&$codeblocks) { + $return = $matches[0]; + if (strpos($matches[2], "\n") !== false) { + $return = '#codeblock-' . count($codeblocks) . '#'; + + $codeblocks[] = '
' . trim($matches[2], "\n\r") . '
'; + } + return $return; + }, + $text + ); + // Hide all [noparse] contained bbtags by spacefying them // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image? @@ -1392,7 +1253,7 @@ class BBCode // After we're finished processing the bbcode we'll // replace all of the event code with a reformatted version. - $ev = bbtoevent($text); + $ev = Event::fromBBCode($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 @@ -1412,11 +1273,6 @@ class BBCode $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 @@ -1519,6 +1375,13 @@ class BBCode }, $text ); + $text = preg_replace_callback( + "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi", + function ($match) { + return "[url=" . System::baseUrl() . "/search?search=%40" . $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"; @@ -1593,7 +1456,7 @@ class BBCode $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '$1', $text); // Check for strike-through text - $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '$1', $text); + $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '$1', $text); // Check for over-line text $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '$1', $text); @@ -1616,7 +1479,7 @@ class BBCode $text = preg_replace_callback( "(\[style=(.*?)\](.*?)\[\/style\])ism", function ($match) { - return "" . $match[2] . ""; + return "" . $match[2] . ""; }, $text ); @@ -1625,7 +1488,7 @@ class BBCode $text = preg_replace_callback( "(\[class=(.*?)\](.*?)\[\/class\])ism", function ($match) { - return "" . $match[2] . ""; + return "" . $match[2] . ""; }, $text ); @@ -1720,12 +1583,12 @@ class BBCode // [img=widthxheight]image source[/img] $text = preg_replace_callback( "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", - function ($matches) { + function ($matches) use ($simple_html) { if (strpos($matches[3], "data:image/") === 0) { return $matches[0]; } - $matches[3] = proxy_url($matches[3]); + $matches[3] = self::proxyUrl($matches[3], $simple_html); return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]"; }, $text @@ -1734,16 +1597,24 @@ class BBCode $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '', $text); $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '', $text); + $text = preg_replace_callback("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism", + function ($matches) use ($simple_html) { + $matches[1] = self::proxyUrl($matches[1], $simple_html); + $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT); + return '' . $matches[2] . ''; + }, + $text); + // Images // [img]pathtoimage[/img] $text = preg_replace_callback( "/\[img\](.*?)\[\/img\]/ism", - function ($matches) { + function ($matches) use ($simple_html) { if (strpos($matches[1], "data:image/") === 0) { return $matches[0]; } - $matches[1] = proxy_url($matches[1]); + $matches[1] = self::proxyUrl($matches[1], $simple_html); return "[img]" . $matches[1] . "[/img]"; }, $text @@ -1764,15 +1635,15 @@ class BBCode // Try to Oembed if ($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("/\[video\](.*?\.(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("/\[video\](.*?)\[\/video\]/", + $text = preg_replace("/\[video\](.*?)\[\/video\]/ism", '$1', $text); - $text = preg_replace("/\[audio\](.*?)\[\/audio\]/", + $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", '$1', $text); } @@ -1830,7 +1701,7 @@ class BBCode // start which is always required). Allow desc with a missing summary for compatibility. if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) { - $sub = format_event_html($ev, $simple_html); + $sub = Event::getHTML($ev, $simple_html); $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text); $text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text); @@ -1865,10 +1736,12 @@ class BBCode $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text); $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text); - + /// @todo What is the meaning of these lines? $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text); $text = preg_replace('/\&\#039\;/', '\'', $text); - $text = preg_replace('/\"\;/', '"', $text); + + // Currently deactivated, it made problems with " inside of alt texts. + //$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); @@ -1893,6 +1766,18 @@ class BBCode $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image); } + // Restore code blocks + $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU', + function ($matches) use ($codeblocks) { + $return = $matches[0]; + if (isset($codeblocks[intval($matches[1])])) { + $return = $codeblocks[$matches[1]]; + } + return $return; + }, + $text + ); + // 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. @@ -1947,7 +1832,7 @@ class BBCode * @param string $addon The addon for which the abstract is meant for * @return string The abstract */ - private static function getAbstract($text, $addon = "") + public static function getAbstract($text, $addon = "") { $abstract = ""; $abstracts = []; @@ -1969,4 +1854,118 @@ class BBCode return $abstract; } + + /** + * @brief Callback function to replace a Friendica style mention in a mention for Diaspora + * + * @param array $match Matching values for the callback + * @return string Replaced mention + */ + private static function bbCodeMention2DiasporaCallback($match) + { + $contact = Contact::getDetailsByURL($match[3]); + + if (empty($contact['addr'])) { + $contact = Probe::uri($match[3]); + } + + if (empty($contact['addr'])) { + return $match[0]; + } + + $mention = '@{' . $match[2] . '; ' . $contact['addr'] . '}'; + return $mention; + } + + /** + * @brief Converts a BBCode text into Markdown + * + * This function converts a BBCode item body to be sent to Markdown-enabled + * systems like Diaspora and Libertree + * + * @param string $text + * @param bool $for_diaspora Diaspora requires more changes than Libertree + * @return string + */ + public static function toMarkdown($text, $for_diaspora = true) + { + $a = self::getApp(); + + $original_text = $text; + + // Since Diaspora is creating a summary for links, this function removes them before posting + if ($for_diaspora) { + $text = self::removeShareInformation($text); + } + + /** + * Transform #tags, strip off the [url] and replace spaces with underscore + */ + $url_search_string = "^\[\]"; + $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i", + function ($matches) { + return '#' . str_replace(' ', '_', $matches[2]); + }, + $text + ); + + // Converting images with size parameters to simple images. Markdown doesn't know it. + $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text); + + // Convert it to HTML - don't try oembed + if ($for_diaspora) { + $text = self::convert($text, false, 3); + + // Add all tags that maybe were removed + if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) { + $tagline = ""; + foreach ($tags[2] as $tag) { + $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8'); + if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) { + $tagline .= '#' . $tag . ' '; + } + } + $text = $text . " " . $tagline; + } + } else { + $text = self::convert($text, false, 4); + } + + // mask some special HTML chars from conversation to markdown + $text = str_replace(['<', '>', '&'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text); + + // If a link is followed by a quote then there should be a newline before it + // Maybe we should make this newline at every time before a quote. + $text = str_replace(["
"], ["
"], $text); + + $stamp1 = microtime(true); + + // Now convert HTML to Markdown + $text = HTML::toMarkdown($text); + + // unmask the special chars back to HTML + $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text); + + $a->saveTimestamp($stamp1, "parser"); + + // Libertree has a problem with escaped hashtags. + $text = str_replace(['\#'], ['#'], $text); + + // Remove any leading or trailing whitespace, as this will mess up + // the Diaspora signature verification and cause the item to disappear + $text = trim($text); + + if ($for_diaspora) { + $url_search_string = "^\[\]"; + $text = preg_replace_callback( + "/([@]\[(.*?)\])\(([$url_search_string]*?)\)/ism", + ['self', 'bbCodeMention2DiasporaCallback'], + $text + ); + } + + Addon::callHooks('bb2diaspora', $text); + + return $text; + } }