X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FContent%2FText%2FBBCode.php;h=d7afa1cd2f617486c814c0111c1d53b8fa2f4fa2;hb=0bc7b89530e7f91f4876c6ada0a27d0d4288780c;hp=f7ca6efd9783946cfe282afaa59bad7a632770d0;hpb=181529f3df152dd0fdb6d19a35c64d0ff158ffd2;p=friendica.git diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php index f7ca6efd97..d7afa1cd2f 100644 --- a/src/Content/Text/BBCode.php +++ b/src/Content/Text/BBCode.php @@ -1,6 +1,22 @@ . + * */ namespace Friendica\Content\Text; @@ -10,10 +26,7 @@ use DOMXPath; use Exception; use Friendica\Content\OEmbed; use Friendica\Content\Smilies; -use Friendica\Core\Cache; -use Friendica\Core\Config; use Friendica\Core\Hook; -use Friendica\Core\L10n; use Friendica\Core\Logger; use Friendica\Core\Protocol; use Friendica\Core\Renderer; @@ -35,8 +48,17 @@ use Friendica\Util\XML; class BBCode { + const INTERNAL = 0; + const API = 2; + const DIASPORA = 3; + const CONNECTORS = 4; + const OSTATUS = 7; + const TWITTER = 8; + const BACKLINK = 8; + const ACTIVITYPUB = 9; + /** - * @brief Fetches attachment data that were generated the old way + * Fetches attachment data that were generated the old way * * @param string $body Message body * @return array @@ -107,7 +129,7 @@ class BBCode } /** - * @brief Fetches attachment data that were generated with the "attachment" element + * Fetches attachment data that were generated with the "attachment" element * * @param string $body Message body * @return array @@ -276,7 +298,7 @@ class BBCode $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body); $post['text'] = $body; - if (preg_match_all("(\[url=(.*?)\]\s*\[img\](.*?)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) { + if (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $body, $pictures, PREG_SET_ORDER)) { if ((count($pictures) == 1) && !$has_title) { if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) { // Replace the preview picture with the real picture @@ -412,7 +434,7 @@ class BBCode } /** - * @brief Converts a BBCode text into plaintext + * Converts a BBCode text into plaintext * * @param $text * @param bool $keep_urls Whether to keep URLs in the resulting plaintext @@ -421,30 +443,35 @@ class BBCode */ public static function toPlaintext($text, $keep_urls = true) { - $naked_text = HTML::toPlaintext(BBCode::convert($text, false, 0, true), 0, !$keep_urls); + $naked_text = HTML::toPlaintext(self::convert($text, false, 0, true), 0, !$keep_urls); return $naked_text; } - private static function proxyUrl($image, $simplehtml = false) + private static function proxyUrl($image, $simplehtml = self::INTERNAL) { // Only send proxied pictures to API and for internal display - if (in_array($simplehtml, [false, 2])) { + if (in_array($simplehtml, [self::INTERNAL, self::API])) { return ProxyUtils::proxifyUrl($image); } else { return $image; } } - public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false) + /** + * This function changing the visual size (not the real size) of images. + * The function does not work for pictures with an alternate text description. + * This could only be changed by using some new "img" BBCode format. + * + * @param string $srctext The body with images + * @return string The body with possibly scaled images + */ + public static function scaleExternalImages(string $srctext) { - // Suppress "view full size" - if (intval(Config::get('system', 'no_view_full_size'))) { - $include_link = false; - } + $s = $srctext; - // Picture addresses can contain special characters - $s = htmlspecialchars_decode($srctext); + // Simplify image links + $s = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $s); $matches = null; $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER); @@ -452,28 +479,19 @@ class BBCode foreach ($matches as $mtch) { Logger::log('scale_external_image: ' . $mtch[1]); - $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3)); + $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3)); if (stristr($mtch[1], $hostname)) { continue; } - // $scale_replace, if passed, is an array of two elements. The - // first is the name of the full-size image. The second is the - // name of a remote, scaled-down version of the full size image. - // This allows Friendica to display the smaller remote image if - // one exists, while still linking to the full-size image - if ($scale_replace) { - $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]); - } else { - $scaled = $mtch[1]; - } - $i = Network::fetchUrl($scaled); - if (!$i) { - return $srctext; + $curlResult = Network::curl($mtch[1], true); + if (!$curlResult->isSuccess()) { + continue; } - // guess mimetype from headers or filename - $type = Images::guessType($mtch[1], true); + $i = $curlResult->getBody(); + $type = $curlResult->getContentType(); + $type = Images::getMimeTypeByData($i, $mtch[1], $type); if ($i) { $Image = new Image($i, $type); @@ -488,10 +506,8 @@ class BBCode Logger::log('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], Logger::DEBUG); $s = str_replace( $mtch[0], - '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]' - . "\n" . (($include_link) - ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n" - : ''), + '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]' + . "\n", $s ); Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG); @@ -501,23 +517,22 @@ class BBCode } } - // replace the special char encoding - $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8'); return $s; } /** + * Truncates imported message body string length to max_import_size + * * The purpose of this function is to apply system message length limits to * imported messages without including any embedded photos in the length * - * @brief Truncates imported message body string length to max_import_size * @param string $body * @return string * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ public static function limitBodySize($body) { - $maxlen = Config::get('config', 'max_import_size', 0); + $maxlen = DI::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 @@ -599,14 +614,13 @@ class BBCode * * Note: Can produce a [bookmark] tag in the returned string * - * @brief Processes [attachment] tags - * @param string $text - * @param bool|int $simplehtml - * @param bool $tryoembed + * @param string $text + * @param integer $simplehtml + * @param bool $tryoembed * @return string * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ - private static function convertAttachment($text, $simplehtml = false, $tryoembed = true) + private static function convertAttachment($text, $simplehtml = self::INTERNAL, $tryoembed = true) { $data = self::getAttachmentData($text); if (empty($data) || empty($data['url'])) { @@ -620,7 +634,7 @@ class BBCode $data['title'] = null; } - if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data['image'])) { + if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) { $data['preview'] = $data['image']; $data['image'] = ''; } @@ -635,18 +649,18 @@ class BBCode } catch (Exception $e) { $data['title'] = ($data['title'] ?? '') ?: $data['url']; - if ($simplehtml != 4) { + if ($simplehtml != self::CONNECTORS) { $return = sprintf('
', $data['type']); } if (!empty($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']); + $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']); + $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('
', $data['url'], self::proxyUrl($data['preview'], $simplehtml), $data['title']); } $return .= sprintf('

%s

', $data['url'], $data['title']); } @@ -662,7 +676,7 @@ class BBCode $return .= sprintf('%s', $data['url'], parse_url($data['url'], PHP_URL_HOST)); } - if ($simplehtml != 4) { + if ($simplehtml != self::CONNECTORS) { $return .= '
'; } } @@ -712,7 +726,6 @@ class BBCode /** * 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 */ @@ -733,18 +746,20 @@ class BBCode } /** - * @brief Converts [url] BBCodes in a format that looks fine on ActivityPub systems. + * Converts [url] BBCodes in a format that looks fine on ActivityPub systems. + * * @param string $url URL that is about to be reformatted * @return string reformatted link including HTML codes */ private static function convertUrlForActivityPub($url) { - $html = '%s'; + $html = '%s'; return sprintf($html, $url, self::getStyledURL($url)); } /** * Converts an URL in a nicer format (without the scheme and possibly shortened) + * * @param string $url URL that is about to be reformatted * @return string reformatted link */ @@ -927,7 +942,7 @@ 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); + '' . DI::l10n()->t('Image/photo') . '', $newbody); $cnt++; } @@ -981,7 +996,8 @@ class BBCode Contact::getIdForURL($attributes['profile'], 0, true, $default); $author_contact = Contact::getDetailsByURL($attributes['profile']); - $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']); + $author_contact['url'] = ($author_contact['url'] ?? $attributes['profile']); + $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']); $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author']; $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar']; @@ -1018,13 +1034,10 @@ class BBCode $mention = Protocol::formatMention($attributes['profile'], $attributes['author']); switch ($simplehtml) { - case 1: - $text = ($is_quote_share? '
' : '') . '

' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $mention . ':

' . "\n" . '«' . $content . '»'; - break; - case 2: + case self::API: $text = ($is_quote_share? '
' : '') . '

' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ':

' . "\n" . $content; break; - case 3: // Diaspora + case self::DIASPORA: if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) { $text = ($is_quote_share? '
' : '') . '

' . $attributes['link'] . '

' . "\n"; } else { @@ -1042,45 +1055,33 @@ class BBCode } break; - case 4: + case self::CONNECTORS: $headline = '

' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8'); - $headline .= L10n::t('%2$s %3$s', $attributes['link'], $mention, $attributes['posted']); + $headline .= DI::l10n()->t('%2$s %3$s', $attributes['link'], $mention, $attributes['posted']); $headline .= ':

' . "\n"; $text = ($is_quote_share? '
' : '') . $headline . '
' . trim($content) . '
' . "\n"; break; - case 5: - $text = ($is_quote_share? '
' : '') . '

' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ':

' . "\n" . $content; - break; - case 7: // statusnet/GNU Social + case self::OSTATUS: $text = ($is_quote_share? '
' : '') . '

' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '

' . "\n"; break; - case 9: // ActivityPub + case self::ACTIVITYPUB: $author = '@' . $author_contact['addr'] . ':'; $text = '
' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . ' ' . $author . '
' . $content . '
' . "\n"; break; default: - // Transforms quoted tweets in rich attachments to avoid nested tweets - if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) { - try { - $text = ($is_quote_share? '
' : '') . OEmbed::getHTML($attributes['link']); - } catch (Exception $e) { - $text = ($is_quote_share? '
' : '') . sprintf('[bookmark=%s]%s[/bookmark]', $attributes['link'], $content); - } - } else { - $text = ($is_quote_share? "\n" : ''); - - $tpl = Renderer::getMarkupTemplate('shared_content.tpl'); - $text .= Renderer::replaceMacros($tpl, [ - '$profile' => $attributes['profile'], - '$avatar' => $attributes['avatar'], - '$author' => $attributes['author'], - '$link' => $attributes['link'], - '$posted' => $attributes['posted'], - '$content' => trim($content) - ]); - } + $text = ($is_quote_share? "\n" : ''); + + $tpl = Renderer::getMarkupTemplate('shared_content.tpl'); + $text .= Renderer::replaceMacros($tpl, [ + '$profile' => $attributes['profile'], + '$avatar' => $attributes['avatar'], + '$author' => $attributes['author'], + '$link' => $attributes['link'], + '$posted' => $attributes['posted'], + '$content' => trim($content) + ]); break; } @@ -1090,7 +1091,7 @@ class BBCode private static function removePictureLinksCallback($match) { $cache_key = 'remove:' . $match[1]; - $text = Cache::get($cache_key); + $text = DI::cache()->get($cache_key); if (is_null($text)) { $a = DI::app(); @@ -1132,7 +1133,7 @@ class BBCode } } } - Cache::set($cache_key, $text); + DI::cache()->set($cache_key, $text); } return $text; @@ -1152,7 +1153,7 @@ class BBCode $a = DI::app(); // When the picture link is the own photo path then we can avoid fetching the link - $own_photo_url = preg_quote(Strings::normaliseLink($a->getBaseURL()) . '/photos/'); + $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/'); if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) { if (!empty($match[3])) { $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]'; @@ -1163,7 +1164,7 @@ class BBCode } $cache_key = 'clean:' . $match[1]; - $text = Cache::get($cache_key); + $text = DI::cache()->get($cache_key); if (!is_null($text)) { return $text; } @@ -1214,7 +1215,7 @@ class BBCode } } } - Cache::set($cache_key, $text); + DI::cache()->set($cache_key, $text); return $text; } @@ -1227,7 +1228,7 @@ class BBCode } /** - * @brief Converts a BBCode message to HTML message + * Converts a BBCode message to HTML message * * BBcode 2 HTML was written by WAY2WEB.net * extended to work with Mistpark/Friendica - Mike Macgirvin @@ -1251,643 +1252,630 @@ class BBCode * @return string * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ - public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false) + public static function convert($text, $try_oembed = true, $simple_html = self::INTERNAL, $for_plaintext = false) { $a = DI::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 = $match[2] ?? null; - - try { - $return = OEmbed::getHTML($url, $title); - } catch (Exception $ex) { - $return = $match[0]; - } - - return $return; - }; + $text = self::performWithEscapedTags($text, ['code'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) { + $text = self::performWithEscapedTags($text, ['noparse', 'nobb', 'pre'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) { + /* + * 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 = $match[2] ?? null; - // Extracting code blocks before the whitespace processing and the autolinker - $codeblocks = []; - - $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism", - function ($matches) use (&$codeblocks) { - $return = '#codeblock-' . count($codeblocks) . '#'; - if (strpos($matches[2], "\n") !== false) { - $codeblocks[] = '
' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '
'; - } else { - $codeblocks[] = '' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . ''; - } - - return $return; - }, - $text - ); - - // 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); + try { + $return = OEmbed::getHTML($url, $title); + } catch (Exception $ex) { + $return = $match[0]; + } - // Remove the abstract element. It is a non visible element. - $text = self::stripAbstract($text); + return $return; + }; - // 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. - $extracted = self::extractImagesFromItemBody($text); - $text = $extracted['body']; - $saved_image = $extracted['images']; + // Remove the abstract element. It is a non visible element. + $text = self::stripAbstract($text); - // 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. + // 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); - $ev = Event::fromBBCode($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. - // 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 + $extracted = self::extractImagesFromItemBody($text); + $text = $extracted['body']; + $saved_image = $extracted['images']; - $text = str_replace("<", "<", $text); - $text = str_replace(">", ">", $text); + // 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. - // 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); + $ev = Event::fromBBCode($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); - } + // 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 - // Convert new line chars to html
tags + $text = str_replace("<", "<", $text); + $text = str_replace(">", ">", $text); - // nlbr seems to be hopelessly messed up - // $Text = nl2br($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]", $text); - // We'll emulate it. + // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems + if (!$try_oembed) { + $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text); + } - $text = trim($text); - $text = str_replace("\r\n", "\n", $text); - - // Remove linefeeds inside of the table elements. See issue #6799 - $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ", - "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ", - "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ", - "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "]; - $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]", - "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]", - "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]", - "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"]; - do { - $oldtext = $text; - $text = str_replace($search, $replace, $text); - } while ($oldtext != $text); - - // Replace these here only once - $search = ["\n[table]", "[/table]\n"]; - $replace = ["[table]", "[/table]"]; - $text = str_replace($search, $replace, $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; + // 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); + + // Remove linefeeds inside of the table elements. See issue #6799 + $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ", + "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ", + "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ", + "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "]; + $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]", + "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]", + "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]", + "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"]; + do { + $oldtext = $text; + $text = str_replace($search, $replace, $text); + } while ($oldtext != $text); + + // Replace these here only once + $search = ["\n[table]", "[/table]\n"]; + $replace = ["[table]", "[/table]"]; $text = str_replace($search, $replace, $text); - } while ($oldtext != $text); - } - - /// @todo Have a closer look at the different html modes - // Handle attached links or videos - if (in_array($simple_html, [9])) { - $text = self::removeAttachment($text); - } elseif (!in_array($simple_html, [0, 4])) { - $text = self::removeAttachment($text, true); - } else { - $text = self::convertAttachment($text, $simple_html, $try_oembed); - } - - // leave open the posibility of [map=something] - // this is replaced in Item::prepareBody() which has knowledge of the item location - if (strpos($text, '[/map]') !== false) { - $text = preg_replace_callback( - "/\[map\](.*?)\[\/map\]/ism", - function ($match) use ($simple_html) { - return str_replace($match[0], '

' . Map::byLocation($match[1], $simple_html) . '

', $match[0]); - }, - $text - ); - } - - if (strpos($text, '[map=') !== false) { - $text = preg_replace_callback( - "/\[map=(.*?)\]/ism", - function ($match) use ($simple_html) { - return str_replace($match[0], '

' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '

', $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); + // removing multiplicated newlines + if (DI::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); + } - // Check for strike-through text - $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '$1', $text); + /// @todo Have a closer look at the different html modes + // Handle attached links or videos + if ($simple_html == self::ACTIVITYPUB) { + $text = self::removeAttachment($text); + } elseif (!in_array($simple_html, [self::INTERNAL, self::CONNECTORS])) { + $text = self::removeAttachment($text, true); + } else { + $text = self::convertAttachment($text, $simple_html, $try_oembed); + } - // Check for over-line text - $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '$1', $text); + // leave open the posibility of [map=something] + // this is replaced in Item::prepareBody() which has knowledge of the item location + if (strpos($text, '[/map]') !== false) { + $text = preg_replace_callback( + "/\[map\](.*?)\[\/map\]/ism", + function ($match) use ($simple_html) { + return str_replace($match[0], '

' . Map::byLocation($match[1], $simple_html) . '

', $match[0]); + }, + $text + ); + } - // Check for colored text - $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "$2", $text); + if (strpos($text, '[map=') !== false) { + $text = preg_replace_callback( + "/\[map=(.*?)\]/ism", + function ($match) use ($simple_html) { + return str_replace($match[0], '

' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '

', $match[0]); + }, + $text + ); + } - // Check for sized text - // [size=50] --> font-size: 50px (with the unit). - if ($simple_html != 3) { - $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "$2", $text); - $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text); - } else { - // Issue 2199: Diaspora doesn't interpret the construct above, nor the or element - $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text); - } + if (strpos($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 centered text - $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "
$1
", $text); + // Check for paragraph + $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '

$1

', $text); - // Check for list text - $text = str_replace("[*]", "
  • ", $text); + // Check for bold text + $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '$1', $text); - // Check for style sheet commands - $text = preg_replace_callback( - "(\[style=(.*?)\](.*?)\[\/style\])ism", - function ($match) { - return "" . $match[2] . ""; - }, - $text - ); + // Check for Italics text + $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '$1', $text); - // Check for CSS classes - $text = preg_replace_callback( - "(\[class=(.*?)\](.*?)\[\/class\])ism", - function ($match) { - return "" . $match[2] . ""; - }, - $text - ); + // Check for Underline text + $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '$1', $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", '
      $1
    ', $text); - $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '
      $1
    ', $text); - $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '
      $1
    ', $text); - $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '
      $2
    ', $text); - $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '
      $2
    ', $text); - $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '
      $2
    ', $text); - $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '
      $2
    ', $text); - $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '
      $1
    ', $text); - $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '
      $1
    ', $text); - $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '
  • $1
  • ', $text); - } + // Check for strike-through text + $text = preg_replace("(\[s\](.*?)\[\/s\])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); + // Check for over-line text + $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '$1', $text); - $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '$1
    ', $text); - $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '$1
    ', $text); + // Check for colored text + $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "$2", $text); - $text = str_replace('[hr]', '
    ', $text); + // Check for sized text + // [size=50] --> font-size: 50px (with the unit). + if ($simple_html != self::DIASPORA) { + $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "$2", $text); + $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text); + } else { + // Issue 2199: Diaspora doesn't interpret the construct above, nor the or element + $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $text); + } - if (!$for_plaintext) { - $escaped = []; - // Escaping BBCodes susceptible to contain rogue URL we don'' want the autolinker to catch - $text = preg_replace_callback('#\[(url|img|audio|video|youtube|vimeo|share|attachment|iframe|bookmark).+?\[/\1\]#ism', - function ($matches) use (&$escaped) { - $return = '{escaped-' . count($escaped) . '}'; - $escaped[] = $matches[0]; + // 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", '
      $1
    ', $text); + $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '
      $1
    ', $text); + $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '
      $1
    ', $text); + $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '
      $2
    ', $text); + $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '
      $2
    ', $text); + $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '
      $2
    ', $text); + $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '
      $2
    ', $text); + $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '
      $1
    ', $text); + $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '
      $1
    ', $text); + $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '
  • $1
  • ', $text); + } - return $return; - }, - $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); - // Autolinker for isolated URLs - $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text); + $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '$1
    ', $text); + $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '$1
    ', $text); - // Restoring escaped blocks - $text = preg_replace_callback('/{escaped-([0-9]+)}/iU', - function ($matches) use ($escaped) { - return $escaped[intval($matches[1])] ?? $matches[0]; - }, - $text - ); - } + $text = str_replace('[hr]', '
    ', $text); - // This is actually executed in Item::prepareBody() + if (!$for_plaintext) { + $escaped = []; - $nosmile = strpos($text, '[nosmile]') !== false; - $text = str_replace('[nosmile]', '', $text); + // Escaping BBCodes susceptible to contain rogue URL we don'' want the autolinker to catch + $text = preg_replace_callback('#\[(url|img|audio|video|youtube|vimeo|share|attachment|iframe|bookmark).+?\[/\1\]#ism', + function ($matches) use (&$escaped) { + $return = '{escaped-' . count($escaped) . '}'; + $escaped[] = $matches[0]; - // Check for font change text - $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "$2", $text); + return $return; + }, + $text + ); - // Declare the format for [spoiler] layout - $SpoilerLayout = '
    ' . L10n::t('Click to open/close') . '$1
    '; + // Autolinker for isolated URLs + $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text); - // 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); - } + // Restoring escaped blocks + $text = preg_replace_callback('/{escaped-([0-9]+)}/iU', + function ($matches) use ($escaped) { + return $escaped[intval($matches[1])] ?? $matches[0]; + }, + $text + ); + } - // Check for [spoiler=Title] text + // This is actually executed in Item::prepareBody() - // handle nested quotes - $endlessloop = 0; - while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) { - $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism", - '
    $1$2
    ', - $text); - } + $nosmile = strpos($text, '[nosmile]') !== false; + $text = str_replace('[nosmile]', '', $text); - // Declare the format for [quote] layout - $QuoteLayout = '
    $1
    '; + // Check for font change text + $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "$2", $text); - // 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); - } + // Declare the format for [spoiler] layout + $SpoilerLayout = '
    ' . DI::l10n()->t('Click to open/close') . '$1
    '; - // Check for [quote=Author] text + // 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); + } - $t_wrote = L10n::t('$1 wrote:'); + // Check for [spoiler=Title] text - // 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); - } + // handle nested quotes + $endlessloop = 0; + while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) { + $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism", + '
    $1$2
    ', + $text); + } + // Declare the format for [quote] layout + $QuoteLayout = '
    $1
    '; - // [img=widthxheight]image source[/img] - $text = preg_replace_callback( - "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", - function ($matches) use ($simple_html) { - if (strpos($matches[3], "data:image/") === 0) { - return $matches[0]; + // 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); } - $matches[3] = self::proxyUrl($matches[3], $simple_html); - return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]"; - }, - $text - ); + // Check for [quote=Author] 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); + $t_wrote = DI::l10n()->t('$1 wrote:'); - $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/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 ($matches) use ($simple_html) { - if (strpos($matches[1], "data:image/") === 0) { - return $matches[0]; + // 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); } - $matches[1] = self::proxyUrl($matches[1], $simple_html); - 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); + // [img=widthxheight]image source[/img] + $text = preg_replace_callback( + "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", + function ($matches) use ($simple_html) { + if (strpos($matches[3], "data:image/") === 0) { + return $matches[0]; + } - $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); + $matches[3] = self::proxyUrl($matches[3], $simple_html); + 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); + + $text = preg_replace_callback("/\[img\=(.*?)\](.*?)\[\/img\]/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 ($matches) use ($simple_html) { + if (strpos($matches[1], "data:image/") === 0) { + return $matches[0]; + } - // Simplify "video" element - $text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text); + $matches[1] = self::proxyUrl($matches[1], $simple_html); + return "[img]" . $matches[1] . "[/img]"; + }, + $text + ); - // Try to Oembed - if ($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("/\[img\](.*?)\[\/img\]/ism", '' . DI::l10n()->t('Image/photo') . '', $text); + $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '' . DI::l10n()->t('Image/photo') . '', $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("/\[video\](.*?)\[\/video\]/ism", - '$1', $text); - $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", - '$1', $text); - } + $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '
    ' . DI::l10n()->t('Encrypted content') . '
    ', $text); + $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '
    ' . DI::l10n()->t('Encrypted content') . '
    ', $text); + //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '
    ' . DI::l10n()->t('Encrypted content') . '
    ', $Text); - // html5 video and audio + // Simplify "video" element + $text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text); + // Try to Oembed + if ($try_oembed) { + $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism", '', $text); + $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism", '', $text); - if ($try_oembed) { - $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '', $text); - } else { - $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '$1', $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("/\[video\](.*?)\[\/video\]/ism", + '$1', $text); + $text = preg_replace("/\[audio\](.*?)\[\/audio\]/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); - } + // html5 video and audio - $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("/\[iframe\](.*?)\[\/iframe\]/ism", '', $text); + } else { + $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '$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, $text); - } + // Youtube extensions + if ($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("/\[vimeo\]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); + $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("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '', $text); - } else { - $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", - 'https://vimeo.com/$1', $text); - } + if ($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); + } - // oembed tag - $text = OEmbed::BBCode2HTML($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); + } - // Avoid triple linefeeds through oembed - $text = str_replace("


    ", "

    ", $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]', $text); - // If we found an event earlier, strip out all the event code and replace with a reformatted version. - // Replace the event-start section with the entire formatted event. The other bbcode is stripped. - // Summary (e.g. title) is required, earlier revisions only required description (in addition to - // start which is always required). Allow desc with a missing summary for compatibility. + if ($try_oembed) { + $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '', $text); + } else { + $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", + 'https://vimeo.com/$1', $text); + } - if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) { - $sub = Event::getHTML($ev, $simple_html); + // oembed tag + $text = OEmbed::BBCode2HTML($text); - $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); - } + // Avoid triple linefeeds through oembed + $text = str_replace("


    ", "

    ", $text); - // Replace non graphical smilies for external posts - if (!$nosmile && !$for_plaintext) { - $text = Smilies::replace($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 (!$for_plaintext) { - if (in_array($simple_html, [7, 9])) { - $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text); - $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text); - } - } else { - $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text); - $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text); - } + if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) { + $sub = Event::getHTML($ev, $simple_html); - $text = str_replace(["\r","\n"], ['
    ', '
    '], $text); - - // Remove all hashtag addresses - if ($simple_html && !in_array($simple_html, [3, 7, 9])) { - $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text); - } elseif ($simple_html == 3) { - // The ! is converted to @ since Diaspora only understands the @ - $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", - '@$3', - $text); - } elseif (in_array($simple_html, [7, 9])) { - $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", - '$1$3', - $text); - } elseif (!$simple_html) { - $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", - '$1$3', - $text); - } + $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); + } - // Bookmarks in red - will be converted to bookmarks in friendica - $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text); - $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text); - $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i", - "[bookmark=$1]$2[/bookmark]", $text); + // Replace non graphical smilies for external posts + if (!$nosmile && !$for_plaintext) { + $text = Smilies::replace($text); + } - if (in_array($simple_html, [2, 6, 7, 8])) { - $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 (!$for_plaintext && DI::config()->get('system', 'big_emojis') && ($simple_html != self::DIASPORA)) { + $conv = html_entity_decode(str_replace([' ', "\n", "\r"], '', $text)); + // Emojis are always 4 byte Unicode characters + if (!empty($conv) && (strlen($conv) / mb_strlen($conv) == 4)) { + $text = '' . $text . ''; + } + } - if ($simple_html == 5) { - $text = preg_replace("/[^#@!]\[url\=(.*?)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text); - } + if (!$for_plaintext) { + if (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) { + $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text); + $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text); + } + } else { + $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text); + $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text); + } - // Perform URL Search - if ($try_oembed) { - $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text); - } + $text = str_replace(["\r","\n"], ['
    ', '
    '], $text); + + // Remove all hashtag addresses + if ($simple_html && !in_array($simple_html, [self::DIASPORA, self::OSTATUS, self::ACTIVITYPUB])) { + $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text); + } elseif ($simple_html == self::DIASPORA) { + // The ! is converted to @ since Diaspora only understands the @ + $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", + '@$3', + $text); + } elseif (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) { + $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", + '$1$3', + $text); + } elseif (!$simple_html) { + $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", + '$1$3', + $text); + } - if ($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); - } + // Bookmarks in red - will be converted to bookmarks in friendica + $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text); + $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text); + $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i", + "[bookmark=$1]$2[/bookmark]", $text); - // Handle Diaspora posts - $text = preg_replace_callback( - "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi", - function ($match) { - return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]"; - }, $text - ); + if (in_array($simple_html, [self::API, self::OSTATUS, self::TWITTER])) { + $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); + } - $text = preg_replace_callback( - "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi", - function ($match) { - return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]"; - }, $text - ); + // Perform URL Search + if ($try_oembed) { + $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $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("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text); + + // Handle Diaspora posts + $text = preg_replace_callback( + "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi", + function ($match) { + return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]"; + }, $text + ); + + $text = preg_replace_callback( + "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi", + function ($match) { + return "[url=" . DI::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"; + $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text); + + /* Tag conversion + * Supports: + * - #[url=][/url] + * - [url=]#[/url] + */ + $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) use ($simple_html) { + if ($simple_html == BBCode::ACTIVITYPUB) { + return '#' + . XML::escape($matches[1]) . ''; + } else { + return '#'; + } + }, $text); - /* Tag conversion - * Supports: - * - #[url=][/url] - * - [url=]#[/url] - */ - $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) { - return '#'; - }, $text); + // We need no target="_blank" rel="noopener noreferrer" for local links + // convert links start with DI::baseUrl() as local link without the target="_blank" rel="noopener noreferrer" attribute + $escapedBaseUrl = preg_quote(DI::baseUrl(), '/'); + $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '$1', $text); + $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '$2', $text); - // We need no target="_blank" for local links - // convert links start with System::baseUrl() as local link without the target="_blank" attribute - $escapedBaseUrl = preg_quote(System::baseUrl(), '/'); - $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '$1', $text); - $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '$2', $text); + $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '$1', $text); + $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '$2', $text); - $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '$1', $text); - $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '$2', $text); + // Red compatibility, though the link can't be authenticated on Friendica + $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '$2', $text); - // Red compatibility, though the link can't be authenticated on Friendica - $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/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 - // 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); - $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("/\[mail\](.*?)\[\/mail\]/", '$1', $text); + $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '$2', $text); - // Perform MAIL Search - $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '$1', $text); - $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '$2', $text); + /// @todo What is the meaning of these lines? + $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text); + $text = preg_replace('/\&\#039\;/', '\'', $text); - // Unhide all [noparse] contained bbtags unspacefying them - // and triming the [noparse] tag. + // Currently deactivated, it made problems with " inside of alt texts. + //$text = preg_replace('/\"\;/', '"', $text); - $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); + // fix any escaped ampersands that may have been converted into links + $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text); - /// @todo What is the meaning of these lines? - $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text); - $text = preg_replace('/\&\#039\;/', '\'', $text); + // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails) + $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:']; - // Currently deactivated, it made problems with " inside of alt texts. - //$text = preg_replace('/\"\;/', '"', $text); + array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');}); - // fix any escaped ampersands that may have been converted into links - $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text); + $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism', + '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $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://', 'https://', 'redir/', 'cid:']; + // sanitize href attributes (only allowlisted protocols URLs) + // default value for backward compatibility + $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []); - array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');}); + // Always allowed protocol even if config isn't set or not including it + $allowed_link_protocols[] = '//'; + $allowed_link_protocols[] = 'http://'; + $allowed_link_protocols[] = 'https://'; + $allowed_link_protocols[] = 'redir/'; - $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') . '">', $text); + array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');}); - // sanitize href attributes (only whitelisted protocols URLs) - // default value for backward compatibility - $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', []); + $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="' . DI::l10n()->t('Invalid link protocol') . '">', $text); - // Always allowed protocol even if config isn't set or not including it - $allowed_link_protocols[] = '//'; - $allowed_link_protocols[] = 'http://'; - $allowed_link_protocols[] = 'https://'; - $allowed_link_protocols[] = 'redir/'; + // Shared content + $text = self::convertShare( + $text, + function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) { + return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html); + } + ); - array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');}); + $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image); - $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); + return $text; + }); // Escaped noparse, nobb, pre - // Shared content - $text = self::convertShare( - $text, - function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) { - return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html); - } - ); + // Remove escaping tags + $text = preg_replace("/\[noparse\](.*?)\[\/noparse\]/ism", '\1', $text); + $text = preg_replace("/\[nobb\](.*?)\[\/nobb\]/ism", '\1', $text); + $text = preg_replace("/\[pre\](.*?)\[\/pre\]/ism", '\1', $text); - if ($saved_image) { - $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image); - } + return $text; + }); // Escaped code - // 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]]; + $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism", + function ($matches) { + if (strpos($matches[2], "\n") !== false) { + $return = '
    ' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '
    '; + } else { + $return = '' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . ''; } + return $return; }, $text @@ -1927,7 +1915,7 @@ class BBCode } /** - * @brief Strips the "abstract" tag from the provided text + * Strips the "abstract" tag from the provided text * * @param string $text The text with BBCode * @return string The same text - but without "abstract" element @@ -1941,7 +1929,7 @@ class BBCode } /** - * @brief Returns the value of the "abstract" element + * Returns the value of the "abstract" element * * @param string $text The text that maybe contains the element * @param string $addon The addon for which the abstract is meant for @@ -1971,7 +1959,7 @@ class BBCode } /** - * @brief Callback function to replace a Friendica style mention in a mention for Diaspora + * Callback function to replace a Friendica style mention in a mention for Diaspora * * @param array $match Matching values for the callback * [1] = Mention type (! or @) @@ -1998,7 +1986,7 @@ class BBCode } /** - * @brief Converts a BBCode text into Markdown + * Converts a BBCode text into Markdown * * This function converts a BBCode item body to be sent to Markdown-enabled * systems like Diaspora and Libertree @@ -2010,8 +1998,6 @@ class BBCode */ public static function toMarkdown($text, $for_diaspora = true) { - $a = DI::app(); - $original_text = $text; // Since Diaspora is creating a summary for links, this function removes them before posting @@ -2035,7 +2021,7 @@ class BBCode // Convert it to HTML - don't try oembed if ($for_diaspora) { - $text = self::convert($text, false, 3); + $text = self::convert($text, false, self::DIASPORA); // Add all tags that maybe were removed if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) { @@ -2049,7 +2035,7 @@ class BBCode $text = $text . " " . $tagline; } } else { - $text = self::convert($text, false, 4); + $text = self::convert($text, false, self::CONNECTORS); } // If a link is followed by a quote then there should be a newline before it @@ -2085,76 +2071,97 @@ class BBCode } /** - * @brief Pull out all #hashtags and @person tags from $string. - * - * We also get @person@domain.com - which would make - * the regex quite complicated as tags can also - * end a sentence. So we'll run through our results - * and strip the period from any tags which end with one. - * Returns array of tags found, or empty array. - * - * @param string $string Post content - * - * @return array List of tag and person names - */ - public static function getTags($string) - { - $ret = []; - - // Convert hashtag links to hashtags - $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2', $string); - - // ignore anything in a code block - $string = preg_replace('/\[code.*?\].*?\[\/code\]/sm', '', $string); - - // Force line feeds at bbtags - $string = str_replace(['[', ']'], ["\n[", "]\n"], $string); - - // ignore anything in a bbtag - $string = preg_replace('/\[(.*?)\]/sm', '', $string); - - // Match full names against @tags including the space between first and last - // We will look these up afterward to see if they are full names or not recognisable. - - if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) { - foreach ($matches[1] as $match) { - if (strstr($match, ']')) { - // we might be inside a bbcode color tag - leave it alone - continue; - } - - if (substr($match, -1, 1) === '.') { - $ret[] = substr($match, 0, -1); - } else { - $ret[] = $match; - } - } - } - - // Otherwise pull out single word tags. These can be @nickname, @first_last - // and #hash tags. - - if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) { - foreach ($matches[1] as $match) { - if (strstr($match, ']')) { - // we might be inside a bbcode color tag - leave it alone - continue; - } - if (substr($match, -1, 1) === '.') { - $match = substr($match,0,-1); - } - // ignore strictly numeric tags like #1 - if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) { - continue; - } - // try not to catch url fragments - if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) { - continue; - } - $ret[] = $match; - } - } - - return $ret; - } + * Pull out all #hashtags and @person tags from $string. + * + * We also get @person@domain.com - which would make + * the regex quite complicated as tags can also + * end a sentence. So we'll run through our results + * and strip the period from any tags which end with one. + * Returns array of tags found, or empty array. + * + * @param string $string Post content + * + * @return array List of tag and person names + */ + public static function getTags($string) + { + $ret = []; + + // Convert hashtag links to hashtags + $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2 ', $string); + + // ignore anything in a code block + $string = preg_replace('/\[code.*?\].*?\[\/code\]/sm', '', $string); + + // Force line feeds at bbtags + $string = str_replace(['[', ']'], ["\n[", "]\n"], $string); + + // ignore anything in a bbtag + $string = preg_replace('/\[(.*?)\]/sm', '', $string); + + // Match full names against @tags including the space between first and last + // We will look these up afterward to see if they are full names or not recognisable. + + if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) { + foreach ($matches[1] as $match) { + if (strstr($match, ']')) { + // we might be inside a bbcode color tag - leave it alone + continue; + } + + if (substr($match, -1, 1) === '.') { + $ret[] = substr($match, 0, -1); + } else { + $ret[] = $match; + } + } + } + + // Otherwise pull out single word tags. These can be @nickname, @first_last + // and #hash tags. + + if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) { + foreach ($matches[1] as $match) { + if (strstr($match, ']')) { + // we might be inside a bbcode color tag - leave it alone + continue; + } + + if (substr($match, -1, 1) === '.') { + $match = substr($match,0,-1); + } + + // ignore strictly numeric tags like #1 + if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) { + continue; + } + + // try not to catch url fragments + if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) { + continue; + } + $ret[] = $match; + } + } + + return $ret; + } + + /** + * Perform a custom function on a text after having escaped blocks enclosed in the provided tag list. + * + * @param string $text + * @param array $tagList A list of tag names, e.g ['noparse', 'nobb', 'pre'] + * @param callable $callback + * @return string + * @throws Exception + *@see Strings::performWithEscapedBlocks + * + */ + public static function performWithEscapedTags(string $text, array $tagList, callable $callback) + { + $tagList = array_map('preg_quote', $tagList); + + return Strings::performWithEscapedBlocks($text, '#\[(?:' . implode('|', $tagList) . ').*?\[/(?:' . implode('|', $tagList) . ')]#ism', $callback); + } }