X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FContent%2FText%2FBBCode.php;h=66f4190b287eca731fe8b465dea18da5504b0feb;hb=5a02e39a65f8f685440228cc1d36738cbe15f32b;hp=5b0aa2cff60d196a169dbd71acce2e9334f8f300;hpb=4997ccd3df3a5ad8fe7f3a15ced2691ee2ba670b;p=friendica.git diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php index 5b0aa2cff6..66f4190b28 100644 --- a/src/Content/Text/BBCode.php +++ b/src/Content/Text/BBCode.php @@ -1,5 +1,4 @@ 0) { - $abstract = self::getAbstract($b["body"]); - } - } - } - - if ($abstract != "") { - $post["text"] = $abstract; - - if ($post["type"] == "text") { - $post["type"] = "link"; - $post["url"] = $b["plink"]; - } - } - - $html = self::convert($post["text"].$post["after"], false, $htmlmode); - $msg = HTML::toPlaintext($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, $scale_replace = false) @@ -575,7 +445,7 @@ class BBCode extends BaseObject */ public static function limitBodySize($body) { - $maxlen = get_max_import_size(); + $maxlen = Config::get('config', 'max_import_size', 0); // If the length of the body, including the embedded images, is smaller // than the maximum, then don't waste time looking for the images @@ -666,16 +536,18 @@ class BBCode extends BaseObject private static function convertAttachment($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"] = ""; } @@ -693,28 +565,32 @@ class BBCode extends BaseObject 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 $bbcode = HTML::toBBCode($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)); } @@ -724,7 +600,7 @@ class BBCode extends BaseObject } } - return trim($data["text"] . ' ' . $return . ' ' . $data["after"]); + return trim(defaults($data, 'text', '') . ' ' . $return . ' ' . defaults($data, 'after', '')); } public static function removeShareInformation($Text, $plaintext = false, $nolink = false) @@ -734,10 +610,10 @@ class BBCode extends BaseObject if (!$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"] = $data["url"]; @@ -751,15 +627,15 @@ class BBCode extends BaseObject } // If the link already is included in the post, don't add it again - if (($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]"; } @@ -975,7 +851,7 @@ class BBCode extends BaseObject // 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++; } @@ -1128,16 +1004,13 @@ class BBCode extends BaseObject 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 != "") { @@ -1157,7 +1030,7 @@ class BBCode extends BaseObject } 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, [ @@ -1187,11 +1060,11 @@ class BBCode extends BaseObject $ch = @curl_init($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]"; @@ -1246,11 +1119,11 @@ class BBCode extends BaseObject $ch = @curl_init($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/") { @@ -1290,21 +1163,6 @@ class BBCode extends BaseObject return $return; } - private static function textHighlightCallback($match) - { - // Fallback in case the language doesn't exist - $return = '[code]' . $match[2] . '[/code]'; - - if (in_array(strtolower($match[1]), - ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby', - 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh', 'bash']) - ) { - $return = text_highlight($match[2], strtolower($match[1])); - } - - return $return; - } - /** * @brief Converts a BBCode message to HTML message * @@ -1314,11 +1172,11 @@ class BBCode extends BaseObject * 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 * @@ -1353,6 +1211,22 @@ class BBCode extends BaseObject 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? @@ -1399,11 +1273,6 @@ class BBCode extends BaseObject $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 @@ -1506,6 +1375,13 @@ class BBCode extends BaseObject }, $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"; @@ -1580,7 +1456,7 @@ class BBCode extends BaseObject $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); @@ -1707,12 +1583,12 @@ class BBCode extends BaseObject // [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 @@ -1721,16 +1597,24 @@ class BBCode extends BaseObject $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]"; }, $text @@ -1751,15 +1635,15 @@ class BBCode extends BaseObject // Try to Oembed if ($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); } @@ -1852,10 +1736,12 @@ class BBCode extends BaseObject $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); @@ -1880,6 +1766,18 @@ class BBCode extends BaseObject $text = self::interpolateSavedImagesIntoItemBody($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. @@ -1934,7 +1832,7 @@ class BBCode extends BaseObject * @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 = []; @@ -2014,23 +1912,6 @@ class BBCode extends BaseObject // 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); - // Extracting multi-line code blocks before the whitespace processing/code highlighter in self::convert() - $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) . '#'; - - $prefix = '````' . $matches[1] . PHP_EOL; - $codeblocks[] = $prefix . trim($matches[2]) . PHP_EOL . '````'; - } - return $return; - }, - $text - ); - // Convert it to HTML - don't try oembed if ($for_diaspora) { $text = self::convert($text, false, 3); @@ -2060,13 +1941,12 @@ class BBCode extends BaseObject $stamp1 = microtime(true); // Now convert HTML to Markdown - $converter = new HtmlConverter(); - $text = $converter->convert($text); + $text = HTML::toMarkdown($text); // unmask the special chars back to HTML $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text); - $a->save_timestamp($stamp1, "parser"); + $a->saveTimestamp($stamp1, "parser"); // Libertree has a problem with escaped hashtags. $text = str_replace(['\#'], ['#'], $text); @@ -2084,18 +1964,6 @@ class BBCode extends BaseObject ); } - // Restore code blocks - $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU', - function ($matches) use ($codeblocks) { - $return = ''; - if (isset($codeblocks[intval($matches[1])])) { - $return = $codeblocks[$matches[1]]; - } - return $return; - }, - $text - ); - Addon::callHooks('bb2diaspora', $text); return $text;