diff --git a/boot.php b/boot.php index 116c712a83..b2d9f01927 100644 --- a/boot.php +++ b/boot.php @@ -1,6 +1,6 @@ - [code=php]function text_highlight($s,$lang)[/code] + [code=php]function text_highlight($s,$lang)[/code]1
  1.  function text_highlight($s,$lang)
@@ -208,6 +208,27 @@ code +1: Supported language parameter values for code highlighting: +- abap +- avrc +- cpp +- css +- diff +- dtd +- html +- java +- javascript +- js +- mysql +- perl +- php +- python +- ruby +- sh +- sql +- vbscript +- xml + ### Titles diff --git a/include/acl_selectors.php b/include/acl_selectors.php index 72f49a40ba..076bb2a498 100644 --- a/include/acl_selectors.php +++ b/include/acl_selectors.php @@ -2,12 +2,14 @@ /** * @file include/acl_selectors.php */ + use Friendica\App; use Friendica\Content\Feature; use Friendica\Content\Widget; use Friendica\Core\Addon; use Friendica\Core\Config; use Friendica\Core\L10n; +use Friendica\Core\Protocol; use Friendica\Database\DBM; use Friendica\Model\Contact; use Friendica\Model\GContact; @@ -248,7 +250,7 @@ function contact_select($selname, $selclass, $preselected = false, $size = 4, $p } if ($privmail) { - $trimmed = GetProfileUsername($rr['url'], $rr['name'], false); + $trimmed = Protocol::formatMention($rr['url'], $rr['name']); } else { $trimmed = mb_substr($rr['name'],0,20); } diff --git a/include/api.php b/include/api.php index 1e8ba13d41..41631567f3 100644 --- a/include/api.php +++ b/include/api.php @@ -2751,7 +2751,7 @@ function api_get_entitities(&$text, $bbcode) return []; } - $bbcode = bb_CleanPictureLinks($bbcode); + $bbcode = BBCode::cleanPictureLinks($bbcode); // Change pure links in text to bbcode uris $bbcode = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url=$2]$2[/url]', $bbcode); @@ -5171,7 +5171,7 @@ function api_clean_plain_items($Text) { $include_entities = strtolower(x($_REQUEST, 'include_entities') ? $_REQUEST['include_entities'] : "false"); - $Text = bb_CleanPictureLinks($Text); + $Text = BBCode::cleanPictureLinks($Text); $URLSearchString = "^\[\]"; $Text = preg_replace("/([!#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $Text); diff --git a/include/bb2diaspora.php b/include/bb2diaspora.php index 0cb50e95ee..c1c9fa4c3d 100644 --- a/include/bb2diaspora.php +++ b/include/bb2diaspora.php @@ -76,10 +76,10 @@ function diaspora2bb($s) { $s = preg_replace('/([^\]=]|^)(https?\:\/\/)([a-zA-Z0-9:\/\-?&;.=_~#%$!+,@]+(?' . Map::byCoordinates(str_replace('/', ' ', $match[1])) . '', $match[0]); -} -function bb_map_location($match) { - // the extra space in the following line is intentional - return str_replace($match[0], '
' . Map::byLocation($match[1]) . '
', $match[0]); -} - -/** - * Processes [attachment] tags - * - * Note: Can produce a [bookmark] tag in the returned string - * - * @brief Processes [attachment] tags - * @param string $return - * @param bool|int $simplehtml - * @param bool $tryoembed - * @return string - */ -function bb_attachment($return, $simplehtml = false, $tryoembed = true) -{ - $data = BBCode::getAttachmentData($return); - if (!$data) { - return $return; - } - - if (isset($data["title"])) { - $data["title"] = strip_tags($data["title"]); - $data["title"] = str_replace(["http://", "https://"], "", $data["title"]); - } - - if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) { - $data["preview"] = $data["image"]; - $data["image"] = ""; - } - - $return = ''; - if ($simplehtml == 7) { - $return = style_url_for_mastodon($data["url"]); - } elseif (($simplehtml != 4) && ($simplehtml != 0)) { - $return = sprintf('%s
', $data["url"], $data["title"]); - } else { - try { - if ($tryoembed && OEmbed::isAllowedURL($data['url'])) { - $return = OEmbed::getHTML($data['url'], $data['title']); - } else { - throw new Exception('OEmbed is disabled for this attachment.'); - } - } catch (Exception $e) { - if ($simplehtml != 4) { - $return = sprintf('
', $data["type"]); - } - - if ($data["image"] != "") { - $return .= sprintf('
', $data["url"], proxy_url($data["image"]), $data["title"]); - } elseif ($data["preview"] != "") { - $return .= sprintf('
', $data["url"], proxy_url($data["preview"]), $data["title"]); - } - - if (($data["type"] == "photo") && ($data["url"] != "") && ($data["image"] != "")) { - $return .= sprintf('', $data["url"], proxy_url($data["image"]), $data["title"]); - } else { - $return .= sprintf('

%s

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

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

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

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

$1

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

$1

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

$1

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

$1

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

$1

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


    ", "

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

    ', '', $Text); - //$Text = str_replace('
    ', '', $Text); - //$Text = str_replace('
  • ', '
  • ', $Text); - //$Text = str_replace('
    0) { - $abstract = fetch_abstract($b["body"]); + $abstract = self::getAbstract($b["body"]); } } } @@ -636,4 +646,1341 @@ class BBCode return $body; } } + + /** + * Processes [attachment] tags + * + * Note: Can produce a [bookmark] tag in the returned string + * + * @brief Processes [attachment] tags + * @param string $return + * @param bool|int $simplehtml + * @param bool $tryoembed + * @return string + */ + private static function convertAttachment($return, $simplehtml = false, $tryoembed = true) + { + $data = self::getAttachmentData($return); + if (!$data) { + return $return; + } + + if (isset($data["title"])) { + $data["title"] = strip_tags($data["title"]); + $data["title"] = str_replace(["http://", "https://"], "", $data["title"]); + } + + if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) { + $data["preview"] = $data["image"]; + $data["image"] = ""; + } + + $return = ''; + if ($simplehtml == 7) { + $return = self::convertUrlForMastodon($data["url"]); + } elseif (($simplehtml != 4) && ($simplehtml != 0)) { + $return = sprintf('%s
    ', $data["url"], $data["title"]); + } else { + try { + if ($tryoembed && OEmbed::isAllowedURL($data['url'])) { + $return = OEmbed::getHTML($data['url'], $data['title']); + } else { + throw new Exception('OEmbed is disabled for this attachment.'); + } + } catch (Exception $e) { + if ($simplehtml != 4) { + $return = sprintf('
    ', $data["type"]); + } + + if ($data["image"] != "") { + $return .= sprintf('
    ', $data["url"], proxy_url($data["image"]), $data["title"]); + } elseif ($data["preview"] != "") { + $return .= sprintf('
    ', $data["url"], proxy_url($data["preview"]), $data["title"]); + } + + if (($data["type"] == "photo") && ($data["url"] != "") && ($data["image"] != "")) { + $return .= sprintf('', $data["url"], proxy_url($data["image"]), $data["title"]); + } else { + $return .= sprintf('

    %s

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

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

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

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

    $1

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

    $1

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

    $1

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

    $1

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

    $1

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


    ", "

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

    ', '', $Text); + //$Text = str_replace('
    ', '', $Text); + //$Text = str_replace('
  • ', '
  • ', $Text); + //$Text = str_replace('
    $start_pos, 'end' => $end_pos]; + + return $res; + } } diff --git a/src/Core/Protocol.php b/src/Core/Protocol.php new file mode 100644 index 0000000000..9caa0ab152 --- /dev/null +++ b/src/Core/Protocol.php @@ -0,0 +1,124 @@ + + */ +class Protocol +{ + const DFRN = 'dfrn'; // Friendica, Mistpark, other DFRN implementations + const DIASPORA = 'dspr'; // Diaspora + const DIASPORA2 = 'dspc'; // Diaspora connector + const STATUSNET = 'stac'; // Statusnet connector + const OSTATUS = 'stat'; // GNU-social, Pleroma, Mastodon, other OStatus implementations + const FEED = 'feed'; // RSS/Atom feeds with no known "post/notify" protocol + const MAIL = 'mail'; // IMAP/POP + const XMPP = 'xmpp'; // XMPP - Currently unsupported + + const FACEBOOK = 'face'; // Facebook API + const LINKEDIN = 'lnkd'; // LinkedIn + const MYSPACE = 'mysp'; // MySpace - Currently unsupported + const GPLUS = 'goog'; // Google+ + const PUMPIO = 'pump'; // pump.io + const TWITTER = 'twit'; // Twitter + const APPNET = 'apdn'; // app.net - Dead protocol + + const NEWS = 'nntp'; // Network News Transfer Protocol - Currently unsupported + const ICALENDAR = 'ical'; // iCalendar - Currently unsupported + const PNUT = 'pnut'; // pnut.io - Currently unsupported + const ZOT = 'zot!'; // Zot! - Currently unsupported + + const PHANTOM = 'unkn'; // Place holder + + /** + * Returns the address string for the provided profile URL + * + * @param string $profile_url + * @return string + * @throws Exception + */ + public static function getAddrFromProfileUrl($profile_url) + { + $network = self::matchByProfileUrl($profile_url, $matches); + + if ($network === self::PHANTOM) { + throw new Exception('Unknown network for profile URL: ' . $profile_url); + } + + $addr = $matches[2] . '@' . $matches[1]; + + return $addr; + } + + /** + * Guesses the network from a profile URL + * + * @param string $profile_url + * @param array $matches preg_match return array: [0] => Full match [1] => hostname [2] => username + * @return type + */ + public static function matchByProfileUrl($profile_url, &$matches = []) + { + if (preg_match('=https?://(twitter\.com)/(.*)=ism', $profile_url, $matches)) { + return self::TWITTER; + } + + if (preg_match('=https?://(alpha\.app\.net)/(.*)=ism', $profile_url, $matches)) { + return self::APPNET; + } + + if (preg_match('=https?://(plus\.google\.com)/(.*)=ism', $profile_url, $matches)) { + return self::GPLUS; + } + + if (preg_match('=https?://(.*)/profile/(.*)=ism', $profile_url, $matches)) { + return self::DFRN; + } + + if (preg_match('=https?://(.*)/u/(.*)=ism', $profile_url, $matches)) { + return self::DIASPORA; + } + + if (preg_match('=https?://(.*)/channel/(.*)=ism', $profile_url, $matches)) { + // RedMatrix/Hubzilla is identified as Diaspora - friendica can't connect directly to it + return self::DIASPORA; + } + + if (preg_match('=https?://(.*)/user/(.*)=ism', $profile_url, $matches)) { + $statusnet_host = $matches[1]; + $statusnet_user = $matches[2]; + $UserData = Network::fetchUrl('http://' . $statusnet_host . '/api/users/show.json?user_id=' . $statusnet_user); + $user = json_decode($UserData); + if ($user) { + $matches[2] = $user->screen_name; + return self::STATUSNET; + } + } + + // pumpio (http://host.name/user) + if (preg_match('=https?://([\.\w]+)/([\.\w]+)$=ism', $profile_url, $matches)) { + return self::PUMPIO; + } + + return self::PHANTOM; + } + + /** + * Returns a formatted mention from a profile URL and a display name + * + * @param string $profile_url + * @param string $display_name + * @return string + */ + public static function formatMention($profile_url, $display_name) + { + return $display_name . '(' . self::getAddrFromProfileUrl($profile_url) . ')'; + } +} diff --git a/src/Model/Conversation.php b/src/Model/Conversation.php index e6888ca8f6..3f9b3c7e5e 100644 --- a/src/Model/Conversation.php +++ b/src/Model/Conversation.php @@ -11,6 +11,14 @@ require_once "include/dba.php"; class Conversation { + const PROTOCOL_UNKNOWN = 0; + const PROTOCOL_DFRN = 1; + const PROTOCOL_DIASPORA = 2; + const PROTOCOL_OSTATUS_SALMON = 3; + const PROTOCOL_OSTATUS_FEED = 4; // Deprecated + const PROTOCOL_GS_CONVERSATION = 5; // Deprecated + const PROTOCOL_SPLITTED_CONV = 6; + /** * @brief Store the conversation data * diff --git a/src/Protocol/DFRN.php b/src/Protocol/DFRN.php index 6ec83151f7..9f690847a7 100644 --- a/src/Protocol/DFRN.php +++ b/src/Protocol/DFRN.php @@ -10,6 +10,7 @@ namespace Friendica\Protocol; use Friendica\App; use Friendica\Content\OEmbed; +use Friendica\Content\Text\BBCode; use Friendica\Core\Addon; use Friendica\Core\Config; use Friendica\Core\L10n; @@ -29,8 +30,6 @@ use Friendica\Util\Crypto; use Friendica\Util\DateTimeFormat; use Friendica\Util\Network; use Friendica\Util\XML; -use Friendica\Content\Text\BBCode; - use dba; use DOMDocument; use DOMXPath; @@ -929,7 +928,7 @@ class DFRN } // Remove the abstract element. It is only locally important. - $body = remove_abstract($body); + $body = BBCode::stripAbstract($body); if ($type == 'html') { $htmlbody = $body; diff --git a/view/theme/frio/theme.php b/view/theme/frio/theme.php index 3c4d5120cf..3f401612e2 100644 --- a/view/theme/frio/theme.php +++ b/view/theme/frio/theme.php @@ -6,7 +6,9 @@ * Author: Rabuzarus * */ + use Friendica\App; +use Friendica\Content\Text\Plaintext; use Friendica\Content\Widget; use Friendica\Core\Addon; use Friendica\Core\Config; @@ -15,7 +17,6 @@ use Friendica\Core\PConfig; use Friendica\Core\System; use Friendica\Database\DBM; use Friendica\Model\Profile; -use Friendica\Object\Image; $frio = "view/theme/frio"; @@ -90,10 +91,8 @@ function frio_uninstall() */ function frio_item_photo_links(App $a, &$body_info) { - $phototypes = Image::supportedTypes(); - $occurence = 1; - $p = bb_find_open_close($body_info['html'], ""); - + $occurence = 0; + $p = Plaintext::getBoundariesPosition($body_info['html'], ""); while ($p !== false && ($occurence++ < 500)) { $link = substr($body_info['html'], $p['start'], $p['end'] - $p['start']); $matches = []; @@ -112,7 +111,7 @@ function frio_item_photo_links(App $a, &$body_info) $body_info['html'] = str_replace($link, $newlink, $body_info['html']); } - $p = bb_find_open_close($body_info['html'], "", $occurence); + $p = Plaintext::getBoundariesPosition($body_info['html'], "", $occurence); } } @@ -302,7 +301,7 @@ function frio_acl_lookup(App $a, &$results) $search_txt = dbesc(protect_sprintf(preg_quote($results["search"]))); $searching = true; } - + $sql_extra = ''; if ($searching) { $sql_extra .= " AND (`attag` LIKE '%%" . dbesc($search_txt) . "%%' OR `name` LIKE '%%" . dbesc($search_txt) . "%%' OR `nick` LIKE '%%" . dbesc($search_txt) . "%%') "; diff --git a/view/theme/frost/theme.php b/view/theme/frost/theme.php index 1692560341..dd1a0cebaf 100644 --- a/view/theme/frost/theme.php +++ b/view/theme/frost/theme.php @@ -10,9 +10,9 @@ */ use Friendica\App; +use Friendica\Content\Text\Plaintext; use Friendica\Core\Addon; use Friendica\Core\System; -use Friendica\Object\Image; function frost_init(App $a) { $a->videowidth = 400; @@ -49,10 +49,8 @@ function frost_uninstall() { function frost_item_photo_links(App $a, &$body_info) { - $phototypes = Image::supportedTypes(); - - $occurence = 1; - $p = bb_find_open_close($body_info['html'], ""); + $occurence = 0; + $p = Plaintext::getBoundariesPosition($body_info['html'], ''); while($p !== false && ($occurence++ < 500)) { $link = substr($body_info['html'], $p['start'], $p['end'] - $p['start']); @@ -73,7 +71,7 @@ function frost_item_photo_links(App $a, &$body_info) } - $p = bb_find_open_close($body_info['html'], "", $occurence); + $p = Plaintext::getBoundariesPosition($body_info['html'], '', $occurence); } }