X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FContent%2FText%2FBBCode.php;h=a8b5ec2025546e7ffa0e694597d2f38524ccd8b1;hb=5e2cbce9b6efce2baaed69b6aad1537a51bdae9d;hp=ebaef2b29d1accf43625b1e56c9cdcafc7fb61e1;hpb=a0451e1c6285212ed406b343b97e4a53859a673d;p=friendica.git diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php index ebaef2b29d..a8b5ec2025 100644 --- a/src/Content/Text/BBCode.php +++ b/src/Content/Text/BBCode.php @@ -1,5 +1,4 @@ Message type ("link", "video", "photo") - * 'text' -> Text before the shared message - * 'after' -> Text after the shared message - * 'image' -> Preview image of the message - * 'url' -> Url to the attached message - * 'title' -> Title of the attachment - * 'description' -> Description of the attachment + * 'type' -> Message type ("link", "video", "photo") + * 'text' -> Text before the shared message + * 'after' -> Text after the shared message + * 'image' -> Preview image of the message + * 'url' -> Url to the attached message + * 'title' -> Title of the attachment + * 'description' -> Description of the attachment + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ private static function getOldAttachmentData($body) { @@ -76,10 +77,12 @@ class BBCode extends BaseObject $picturedata = Image::getInfoFromURL($matches[1]); - if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) { - $post["image"] = $matches[1]; - } else { - $post["preview"] = $matches[1]; + if ($picturedata) { + if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) { + $post["image"] = $matches[1]; + } else { + $post["preview"] = $matches[1]; + } } } @@ -87,7 +90,7 @@ class BBCode extends BaseObject $post["url"] = $matches[1]; $post["title"] = $matches[2]; } - if (($post["url"] == "") && (in_array($post["type"], ["link", "video"])) + if (!empty($post["url"]) && (in_array($post["type"], ["link", "video"])) && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) { $post["url"] = $matches[1]; } @@ -106,13 +109,14 @@ class BBCode extends BaseObject * * @param string $body Message body * @return array - * 'type' -> Message type ("link", "video", "photo") - * 'text' -> Text before the shared message - * 'after' -> Text after the shared message - * 'image' -> Preview image of the message - * 'url' -> Url to the attached message - * 'title' -> Title of the attachment - * 'description' -> Description of the attachment + * 'type' -> Message type ("link", "video", "photo") + * 'text' -> Text before the shared message + * 'after' -> Text after the shared message + * 'image' -> Preview image of the message + * 'url' -> Url to the attached message + * 'title' -> Title of the attachment + * 'description' -> Description of the attachment + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ public static function getAttachmentData($body) { @@ -128,12 +132,12 @@ class BBCode extends BaseObject $type = ""; preg_match("/type='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $type = strtolower($matches[1]); } preg_match('/type="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $type = strtolower($matches[1]); } @@ -151,12 +155,12 @@ class BBCode extends BaseObject $url = ""; preg_match("/url='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $url = $matches[1]; } preg_match('/url="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $url = $matches[1]; } @@ -166,12 +170,12 @@ class BBCode extends BaseObject $title = ""; preg_match("/title='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $title = $matches[1]; } preg_match('/title="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $title = $matches[1]; } @@ -184,12 +188,12 @@ class BBCode extends BaseObject $image = ""; preg_match("/image='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $image = $matches[1]; } preg_match('/image="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $image = $matches[1]; } @@ -199,12 +203,12 @@ class BBCode extends BaseObject $preview = ""; preg_match("/preview='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $preview = $matches[1]; } preg_match('/preview="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { + if (!empty($matches[1])) { $preview = $matches[1]; } @@ -232,7 +236,7 @@ class BBCode extends BaseObject */ $has_title = !empty($item['title']); - $plink = (!empty($item['plink']) ? $item['plink'] : ''); + $plink = defaults($item, 'plink', ''); $post = self::getAttachmentData($body); // if nothing is found, it maybe having an image. @@ -241,6 +245,9 @@ class BBCode extends BaseObject $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body); $URLSearchString = "^\[\]"; + + $body = preg_replace("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body); + if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) { if ((count($pictures) == 1) && !$has_title) { // Checking, if the link goes to a picture @@ -266,7 +273,7 @@ class BBCode extends BaseObject $post["text"] = str_replace($pictures[0][0], "", $body); } else { $imgdata = Image::getInfoFromURL($pictures[0][1]); - if (substr($imgdata["mime"], 0, 6) == "image/") { + if ($imgdata && substr($imgdata["mime"], 0, 6) == "image/") { $post["type"] = "photo"; $post["image"] = $pictures[0][1]; $post["preview"] = $pictures[0][2]; @@ -338,159 +345,31 @@ class BBCode extends BaseObject } /** - * @brief Convert a message into plaintext for connectors to other networks + * @brief Converts a BBCode text into plaintext * - * @param array $b The message array that is about to be posted - * @param int $limit The maximum number of characters when posting to that network - * @param bool $includedlinks Has an attached link to be included into the message? - * @param int $htmlmode This triggers the behaviour of the bbcode conversion - * @param string $target_network Name of the network where the post should go to. + * @param $text + * @param bool $keep_urls Whether to keep URLs in the resulting plaintext * - * @return string The converted message + * @return string */ - public static function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "") + public static function toPlaintext($text, $keep_urls = true) { - // Remove the hash tags - $URLSearchString = "^\[\]"; - $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]); - - // Add an URL element if the text contains a raw link - $body = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2[/url]', $body); - - // Remove the abstract - $body = self::stripAbstract($body); - - // At first look at data that is attached via "type-..." stuff - // This will hopefully replaced with a dedicated bbcode later - //$post = self::getAttachedData($b["body"]); - $post = self::getAttachedData($body, $b); - - if (($b["title"] != "") && ($post["text"] != "")) { - $post["text"] = trim($b["title"]."\n\n".$post["text"]); - } elseif ($b["title"] != "") { - $post["text"] = trim($b["title"]); + $naked_text = preg_replace('/\[(.+?)\]\s*/','', $text); + if (!$keep_urls) { + $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text); } - $abstract = ""; - - // Fetch the abstract from the given target network - if ($target_network != "") { - $default_abstract = self::getAbstract($b["body"]); - $abstract = self::getAbstract($b["body"], $target_network); - - // If we post to a network with no limit we only fetch - // an abstract exactly for this network - if (($limit == 0) && ($abstract == $default_abstract)) { - $abstract = ""; - } - } else {// Try to guess the correct target network - switch ($htmlmode) { - case 8: - $abstract = self::getAbstract($b["body"], NETWORK_TWITTER); - break; - case 7: - $abstract = self::getAbstract($b["body"], NETWORK_STATUSNET); - break; - case 6: - $abstract = self::getAbstract($b["body"], NETWORK_APPNET); - break; - default: // We don't know the exact target. - // We fetch an abstract since there is a posting limit. - if ($limit > 0) { - $abstract = self::getAbstract($b["body"]); - } - } - } - - if ($abstract != "") { - $post["text"] = $abstract; - - if ($post["type"] == "text") { - $post["type"] = "link"; - $post["url"] = $b["plink"]; - } - } - - $html = self::convert($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($post["url"]); - } - } - } - } - - if ($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) @@ -507,7 +386,7 @@ class BBCode extends BaseObject $c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER); if ($c) { foreach ($matches as $mtch) { - logger('scale_external_image: ' . $mtch[1]); + Logger::log('scale_external_image: ' . $mtch[1]); $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3)); if (stristr($mtch[1], $hostname)) { @@ -542,7 +421,7 @@ class BBCode extends BaseObject $Image->scaleDown(640); $new_width = $Image->getWidth(); $new_height = $Image->getHeight(); - logger('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], LOGGER_DEBUG); + 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]' @@ -551,7 +430,7 @@ class BBCode extends BaseObject : ''), $s ); - logger('scale_external_images: new string: ' . $s, LOGGER_DEBUG); + Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG); } } } @@ -570,16 +449,17 @@ class BBCode extends BaseObject * @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 = get_max_import_size(); + $maxlen = Config::get('config', 'max_import_size', 0); // If the length of the body, including the embedded images, is smaller // than the maximum, then don't waste time looking for the images if ($maxlen && (strlen($body) > $maxlen)) { - logger('the total body length exceeds the limit', LOGGER_DEBUG); + Logger::log('the total body length exceeds the limit', Logger::DEBUG); $orig_body = $body; $new_body = ''; @@ -599,7 +479,7 @@ class BBCode extends BaseObject if (($textlen + $img_start) > $maxlen) { if ($textlen < $maxlen) { - logger('the limit happens before an embedded image', LOGGER_DEBUG); + Logger::log('the limit happens before an embedded image', Logger::DEBUG); $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen); $textlen = $maxlen; } @@ -613,7 +493,7 @@ class BBCode extends BaseObject if (($textlen + $img_end) > $maxlen) { if ($textlen < $maxlen) { - logger('the limit happens before the end of a non-embedded image', LOGGER_DEBUG); + Logger::log('the limit happens before the end of a non-embedded image', Logger::DEBUG); $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen); $textlen = $maxlen; } @@ -636,11 +516,11 @@ class BBCode extends BaseObject if (($textlen + strlen($orig_body)) > $maxlen) { if ($textlen < $maxlen) { - logger('the limit happens after the end of the last image', LOGGER_DEBUG); + Logger::log('the limit happens after the end of the last image', Logger::DEBUG); $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen); } } else { - logger('the text size with embedded images extracted did not violate the limit', LOGGER_DEBUG); + Logger::log('the text size with embedded images extracted did not violate the limit', Logger::DEBUG); $new_body = $new_body . $orig_body; } @@ -656,24 +536,27 @@ class BBCode extends BaseObject * Note: Can produce a [bookmark] tag in the returned string * * @brief Processes [attachment] tags - * @param string $return + * @param string $return * @param bool|int $simplehtml - * @param bool $tryoembed + * @param bool $tryoembed * @return string + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ private static function convertAttachment($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"] = ""; } @@ -691,28 +574,32 @@ 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)); } @@ -722,7 +609,7 @@ 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, $nolink = false) @@ -732,10 +619,10 @@ class BBCode extends BaseObject if (!$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"]; @@ -744,20 +631,20 @@ class BBCode extends BaseObject $data["title"] = $data["url"]; } - if (($data["text"] == "") && ($data["title"] != "") && ($data["url"] == "")) { + if (empty($data["text"]) && !empty($data["title"]) && empty($data["url"])) { return $data["title"] . $data["after"]; } // If the link already is included in the post, don't add it again - if (($data["url"] != "") && strpos($data["text"], $data["url"])) { + if (!empty($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]"; } @@ -895,10 +782,10 @@ class BBCode extends BaseObject /** * Performs a preg_replace within the boundaries of all named BBCode tags in a text * - * @param type $pattern Preg pattern string - * @param type $replace Preg replace string - * @param type $name BBCode tag name - * @param type $text Text to search + * @param string $pattern Preg pattern string + * @param string $replace Preg replace string + * @param string $name BBCode tag name + * @param string $text Text to search * @return string */ public static function pregReplaceInTag($pattern, $replace, $name, $text) @@ -973,7 +860,7 @@ 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++; } @@ -981,190 +868,141 @@ class BBCode extends BaseObject } /** - * Processes [share] tags + * This function converts a [share] block to text according to a provided callback function whose signature is: * - * Note: Can produce a [bookmark] tag in the output + * function(array $attributes, array $author_contact, string $content, boolean $is_quote_share): string * - * @brief Processes [share] tags - * @param array $share preg_match_callback result array - * @param bool|int $simplehtml - * @return string + * Where: + * - $attributes is an array of attributes of the [share] block itself. Missing keys will be completed by the contact + * data lookup + * - $author_contact is a contact record array + * - $content is the inner content of the [share] block + * - $is_quote_share indicates whether there's any content before the [share] block + * - Return value is the string that should replace the [share] block in the provided text + * + * This function is intended to be used by addon connector to format a share block like the target network is expecting it. + * + * @param string $text A BBCode string + * @param callable $callback + * @return string The BBCode string with all [share] blocks replaced */ - private static function convertShare($share, $simplehtml) + public static function convertShare($text, callable $callback) { - $attributes = $share[2]; - - $author = ""; - preg_match("/author='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { - $author = html_entity_decode($matches[1], ENT_QUOTES, 'UTF-8'); - } - - preg_match('/author="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { - $author = $matches[1]; - } - - $profile = ""; - preg_match("/profile='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { - $profile = $matches[1]; - } - - preg_match('/profile="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { - $profile = $matches[1]; - } - - $avatar = ""; - preg_match("/avatar='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { - $avatar = $matches[1]; - } - - preg_match('/avatar="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { - $avatar = $matches[1]; - } - - $link = ""; - preg_match("/link='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { - $link = $matches[1]; - } - - preg_match('/link="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { - $link = $matches[1]; - } - - $posted = ""; - - preg_match("/posted='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { - $posted = $matches[1]; - } - - preg_match('/posted="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { - $posted = $matches[1]; - } + $return = preg_replace_callback( + "/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism", + function ($match) use ($callback) { + $attribute_string = $match[2]; + + $attributes = []; + foreach(['author', 'profile', 'avatar', 'link', 'posted'] as $field) { + preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches); + $attributes[$field] = html_entity_decode(defaults($matches, 2, ''), ENT_QUOTES, 'UTF-8'); + } - // We only call this so that a previously unknown contact can be added. - // This is important for the function "Model\Contact::getDetailsByURL()". - // This function then can fetch an entry from the contact table. - Contact::getIdForURL($profile, 0, true); + // We only call this so that a previously unknown contact can be added. + // This is important for the function "Model\Contact::getDetailsByURL()". + // This function then can fetch an entry from the contact table. + Contact::getIdForURL($attributes['profile'], 0, true); - $data = Contact::getDetailsByURL($profile); + $author_contact = Contact::getDetailsByURL($attributes['profile']); + $author_contact['addr'] = defaults($author_contact, 'addr' , Protocol::getAddrFromProfileUrl($attributes['profile'])); - if (x($data, "name") && x($data, "addr")) { - $userid_compact = $data["name"] . " (" . $data["addr"] . ")"; - } else { - $userid_compact = Protocol::getAddrFromProfileUrl($profile, $author); - } + $attributes['author'] = defaults($author_contact, 'name' , $attributes['author']); + $attributes['avatar'] = defaults($author_contact, 'micro', $attributes['avatar']); + $attributes['profile'] = defaults($author_contact, 'url' , $attributes['profile']); - if (x($data, "addr")) { - $userid = $data["addr"]; - } else { - $userid = Protocol::formatMention($profile, $author); - } + if ($attributes['avatar']) { + $attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB); + } - if (x($data, "name")) { - $author = $data["name"]; - } + return $match[1] . $callback($attributes, $author_contact, $match[3], trim($match[1]) != ''); + }, + $text + ); - if (x($data, "micro")) { - $avatar = $data["micro"]; - } + return $return; + } - $preshare = trim($share[1]); - if ($preshare != "") { - $preshare .= "
"; - } + /** + * Default [share] tag conversion callback + * + * Note: Can produce a [bookmark] tag in the output + * + * @see BBCode::convertShare() + * @param array $attributes [share] block attribute values + * @param array $author_contact Contact row of the shared author + * @param string $content Inner content of the [share] block + * @param boolean $is_quote_share Whether there is content before the [share] block + * @param integer $simplehtml Mysterious integer value depending on the target network/formatting style + * @return string + * @throws \Friendica\Network\HTTPException\InternalServerErrorException + */ + private static function convertShareCallback(array $attributes, array $author_contact, $content, $is_quote_share, $simplehtml) + { + $mention = Protocol::formatMention($attributes['profile'], $attributes['author']); switch ($simplehtml) { case 1: - $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid . ":
»" . $share[3] . "«"; + $text = ($is_quote_share? '
' : '') . '

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

' . "\n" . '«' . $content . '»'; break; case 2: - $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ":
" . $share[3]; + $text = ($is_quote_share? '
' : '') . '

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

' . "\n" . $content; break; case 3: // Diaspora - $headline = '' . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . $userid . ':
'; - - $text = trim($share[1]); + $headline = '

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

' . "\n"; - if ($text != "") { - $text .= "
"; - } - - if (stripos(normalise_link($link), 'http://twitter.com/') === 0) { - $text .= '
' . $link . ''; + if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) { + $text = ($is_quote_share? '
' : '') . '

' . $attributes['link'] . '

' . "\n"; } else { - $text .= $headline . '
' . trim($share[3]) . "

"; + $text = ($is_quote_share? '
' : '') . $headline . '
' . trim($content) . '
' . "\n"; - if ($link != "") { - $text .= '
[l]'; + if ($attributes['link'] != '') { + $text .= '

[l]

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

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

' . "\n"; - if ($text != "") { - $text .= "
"; - } - - $text .= $headline . '
' . trim($share[3]) . "

"; + $text = ($is_quote_share? '
' : '') . $headline . '
' . trim($content) . '
' . "\n"; break; case 5: - $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ":
" . $share[3]; - break; - case 6: // app.net - $text = $preshare . ">> @" . $userid_compact . ":
" . $share[3]; + $text = ($is_quote_share? '
' : '') . '

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

' . "\n" . $content; break; case 7: // statusnet/GNU Social - $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3]; + $text = ($is_quote_share? '
' : '') . '

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

' . "\n"; break; - case 8: // twitter - $text = $preshare . "RT @" . $userid_compact . ": " . $share[3]; - break; - case 9: // Google+/Facebook - $text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ":
" . $share[3]; + case 9: // Google+ + $text = ($is_quote_share? '
' : '') . '

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

' . "\n"; + $text .= '

' . $content . '

' . "\n"; - if ($link != "") { - $text .= "

" . $link; + if ($attributes['link'] != '') { + $text .= '

' . $attributes['link'] . '

'; } break; default: // Transforms quoted tweets in rich attachments to avoid nested tweets - if (stripos(normalise_link($link), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($link)) { + if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) { try { - $oembed = OEmbed::getHTML($link, $preshare); + $text = ($is_quote_share? '
' : '') . OEmbed::getHTML($attributes['link']); } catch (Exception $e) { - $oembed = sprintf('[bookmark=%s]%s[/bookmark]', $link, $preshare); + $text = ($is_quote_share? '
' : '') . sprintf('[bookmark=%s]%s[/bookmark]', $attributes['link'], $content); } - - $text = $preshare . $oembed; } else { - $text = trim($share[1]) . "\n"; - - $avatar = proxy_url($avatar, false, PROXY_SIZE_THUMB); - - $tpl = get_markup_template('shared_content.tpl'); - $text .= replace_macros($tpl, [ - '$profile' => $profile, - '$avatar' => $avatar, - '$author' => $author, - '$link' => $link, - '$posted' => $posted, - '$content' => trim($share[3]) + $text = ($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; @@ -1185,11 +1023,11 @@ 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]"; @@ -1244,11 +1082,11 @@ 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/") { @@ -1288,17 +1126,6 @@ class BBCode extends BaseObject return $return; } - private static function textHighlightCallback($match) - { - if (in_array(strtolower($match[1]), - ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby', - 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh']) - ) { - return text_highlight($match[2], strtolower($match[1])); - } - return $match[0]; - } - /** * @brief Converts a BBCode message to HTML message * @@ -1308,11 +1135,11 @@ 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 * @@ -1321,8 +1148,9 @@ class BBCode extends BaseObject * @param int $simple_html * @param bool $for_plaintext * @return string + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ - public static function convert($text, $try_oembed = true, $simple_html = false, $for_plaintext = false) + public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false) { $a = self::getApp(); @@ -1347,6 +1175,22 @@ class BBCode extends BaseObject return $return; }; + // Extracting multi-line code blocks before the whitespace processing + $codeblocks = []; + + $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism", + 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? @@ -1385,19 +1229,11 @@ class BBCode extends BaseObject $text = preg_replace("/\s?\[share(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "[share$1]$2[/share]", $text); $text = preg_replace("/\s?\[quote(.*?)\]\s?(.*?)\s?\[\/quote\]\s?/ism", "[quote$1]$2[/quote]", $text); - $text = preg_replace("/\n\[code\]/ism", "[code]", $text); - $text = preg_replace("/\[\/code\]\n/ism", "[/code]", $text); - // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems if (!$try_oembed) { $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text); } - // Check for [code] text here, before the linefeeds are messed with. - // The highlighter will unescape and re-escape the content. - if (strpos($text, '[code=') !== false) { - $text = preg_replace_callback("/\[code=(.*?)\](.*?)\[\/code\]/ism", 'self::textHighlightCallback', $text); - } // Convert new line chars to html
tags // nlbr seems to be hopelessly messed up @@ -1457,7 +1293,7 @@ class BBCode extends BaseObject $text); } elseif ($simple_html == 7) { $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", - '$1$3', + '$1$3', $text); } elseif (!$simple_html) { $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", @@ -1494,26 +1330,45 @@ class BBCode extends BaseObject // Handle Diaspora posts $text = preg_replace_callback( - "&\[url=/posts/([^\[\]]*)\](.*)\[\/url\]&Usi", + "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi", function ($match) { return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/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 + ); + // Server independent link to posts and comments // See issue: https://github.com/diaspora/diaspora_federation/issues/75 $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism"; $text = preg_replace($expression, System::baseUrl()."/display/$1", $text); - $text = preg_replace("/([#])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", - '$1$3', $text); - - $text = preg_replace("/\[url\=([$URLSearchString]*)\]#(.*?)\[\/url\]/ism", - '#$2', $text); + /* Tag conversion + * Supports: + * - #[url=][/url] + * - [url=]#[/url] + */ + $text = preg_replace_callback("/(?:#\[url\=[$URLSearchString]*\]|\[url\=[$URLSearchString]*\]#)(.*?)\[\/url\]/ism", function($matches) { + return '#' + . XML::escape($matches[1]) + . ''; + }, $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("/\[url\](".$escapedBaseUrl."[$URLSearchString]*)\[\/url\]/ism", '$1', $text); + $text = preg_replace("/\[url\=(".$escapedBaseUrl."[$URLSearchString]*)\](.*?)\[\/url\]/ism", '$2', $text); $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '$1', $text); $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$2', $text); - //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '$2', $Text); // Red compatibility, though the link can't be authenticated on Friendica $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '$2', $text); @@ -1529,7 +1384,7 @@ class BBCode extends BaseObject $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '$2', $text); // leave open the posibility of [map=something] - // this is replaced in prepare_body() which has knowledge of the item location + // this is replaced in Item::prepareBody() which has knowledge of the item location if (strpos($text, '[/map]') !== false) { $text = preg_replace_callback( @@ -1574,7 +1429,7 @@ 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); @@ -1640,7 +1495,7 @@ class BBCode extends BaseObject $text = str_replace('[hr]', '
', $text); - // This is actually executed in prepare_body() + // This is actually executed in Item::prepareBody() $text = str_replace('[nosmile]', '', $text); @@ -1701,12 +1556,12 @@ 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 @@ -1715,16 +1570,24 @@ 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 @@ -1734,10 +1597,12 @@ class BBCode extends BaseObject $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '' . L10n::t('Image/photo') . '', $text); // Shared content - $text = preg_replace_callback("/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism", - function ($match) use ($simple_html) { - return self::convertShare($match, $simple_html); - }, $text); + $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); + } + ); $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '
' . L10n::t('Encrypted content') . '
', $text); $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '
' . L10n::t('Encrypted content') . '
', $text); @@ -1745,15 +1610,15 @@ class BBCode extends BaseObject // Try to Oembed if ($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', $text); } @@ -1810,7 +1675,7 @@ class BBCode extends BaseObject // Summary (e.g. title) is required, earlier revisions only required description (in addition to // start which is always required). Allow desc with a missing summary for compatibility. - if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) { + if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) { $sub = Event::getHTML($ev, $simple_html); $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text); @@ -1824,21 +1689,9 @@ class BBCode extends BaseObject // Replace non graphical smilies for external posts if ($simple_html) { - $text = Smilies::replace($text, false, true); + $text = Smilies::replace($text); } - // Replace inline code blocks - $text = preg_replace_callback("|(?!]*>)([^<]*)(?!]*>)|ism", - function ($match) use ($simple_html) { - $return = '' . $match[1] . ''; - // Use for Diaspora inline code blocks - if ($simple_html === 3) { - $return = '' . $match[1] . ''; - } - return $return; - } - , $text); - // Unhide all [noparse] contained bbtags unspacefying them // and triming the [noparse] tag. @@ -1846,10 +1699,12 @@ 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); @@ -1874,9 +1729,21 @@ 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; + }, + $text + ); + // Clean up the HTML by loading and saving the HTML with the DOM. // Bad structured html can break a whole page. - // For performance reasons do it only with ativated item cache or at export. + // For performance reasons do it only with activated item cache or at export. if (!$try_oembed || (get_itemcachepath() != "")) { $doc = new DOMDocument(); $doc->preserveWhiteSpace = false; @@ -1902,7 +1769,7 @@ class BBCode extends BaseObject //$Text = str_replace('
  • ', '
  • ', $Text); //$Text = str_replace('
    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(['\#'], ['#'], $text); @@ -2072,26 +1927,88 @@ class BBCode extends BaseObject if ($for_diaspora) { $url_search_string = "^\[\]"; $text = preg_replace_callback( - "/([@]\[(.*?)\])\(([$url_search_string]*?)\)/ism", + "/([@!])\[(.*?)\]\(([$url_search_string]*?)\)/ism", ['self', 'bbCodeMention2DiasporaCallback'], $text ); } - // 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); + Hook::callAll('bb2diaspora', $text); return $text; } + + /** + * @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; + } }