<?php
-
/**
* @file src/Content/Text/BBCode.php
*/
use Friendica\Core\Cache;
use Friendica\Core\Config;
use Friendica\Core\L10n;
-use Friendica\Core\PConfig;
use Friendica\Core\Protocol;
use Friendica\Core\System;
use Friendica\Model\Contact;
use Friendica\Util\Map;
use Friendica\Util\Network;
use Friendica\Util\ParseUrl;
+use Friendica\Util\Proxy as ProxyUtils;
use League\HTMLToMarkdown\HtmlConverter;
-require_once "mod/proxy.php";
-
class BBCode extends BaseObject
{
/**
$post["url"] = $matches[1];
$post["title"] = $matches[2];
}
- if (($post["url"] == "") && (in_array($post["type"], ["link", "video"]))
+ if (!empty($post["url"]) && (in_array($post["type"], ["link", "video"]))
&& preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
$post["url"] = $matches[1];
}
$body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
$URLSearchString = "^\[\]";
+
+ $body = preg_replace("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
+
if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
if ((count($pictures) == 1) && !$has_title) {
// Checking, if the link goes to a picture
}
/**
- * @brief Convert a message into plaintext for connectors to other networks
+ * @brief Converts a BBCode text into plaintext
*
- * @param array $b The message array that is about to be posted
- * @param int $limit The maximum number of characters when posting to that network
- * @param bool $includedlinks Has an attached link to be included into the message?
- * @param int $htmlmode This triggers the behaviour of the bbcode conversion
- * @param string $target_network Name of the network where the post should go to.
+ * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
*
- * @return string The converted message
+ * @return string
*/
- public static function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "")
+ public static function toPlaintext($text, $keep_urls = true)
{
- // Remove the hash tags
- $URLSearchString = "^\[\]";
- $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]);
-
- // Add an URL element if the text contains a raw link
- $body = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2[/url]', $body);
-
- // Remove the abstract
- $body = self::stripAbstract($body);
-
- // At first look at data that is attached via "type-..." stuff
- // This will hopefully replaced with a dedicated bbcode later
- //$post = self::getAttachedData($b["body"]);
- $post = self::getAttachedData($body, $b);
-
- if (($b["title"] != "") && ($post["text"] != "")) {
- $post["text"] = trim($b["title"]."\n\n".$post["text"]);
- } elseif ($b["title"] != "") {
- $post["text"] = trim($b["title"]);
- }
-
- $abstract = "";
-
- // Fetch the abstract from the given target network
- if ($target_network != "") {
- $default_abstract = self::getAbstract($b["body"]);
- $abstract = self::getAbstract($b["body"], $target_network);
-
- // If we post to a network with no limit we only fetch
- // an abstract exactly for this network
- if (($limit == 0) && ($abstract == $default_abstract)) {
- $abstract = "";
- }
- } else {// Try to guess the correct target network
- switch ($htmlmode) {
- case 8:
- $abstract = self::getAbstract($b["body"], NETWORK_TWITTER);
- break;
- case 7:
- $abstract = self::getAbstract($b["body"], NETWORK_STATUSNET);
- break;
- case 6:
- $abstract = self::getAbstract($b["body"], NETWORK_APPNET);
- break;
- default: // We don't know the exact target.
- // We fetch an abstract since there is a posting limit.
- if ($limit > 0) {
- $abstract = self::getAbstract($b["body"]);
- }
- }
- }
-
- if ($abstract != "") {
- $post["text"] = $abstract;
-
- if ($post["type"] == "text") {
- $post["type"] = "link";
- $post["url"] = $b["plink"];
- }
- }
-
- $html = self::convert($post["text"].$post["after"], false, $htmlmode);
- $msg = HTML::toPlaintext($html, 0, true);
- $msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8'));
-
- $link = "";
- if ($includedlinks) {
- if ($post["type"] == "link") {
- $link = $post["url"];
- } elseif ($post["type"] == "text") {
- $link = $post["url"];
- } elseif ($post["type"] == "video") {
- $link = $post["url"];
- } elseif ($post["type"] == "photo") {
- $link = $post["image"];
- }
-
- if (($msg == "") && isset($post["title"])) {
- $msg = trim($post["title"]);
- }
-
- if (($msg == "") && isset($post["description"])) {
- $msg = trim($post["description"]);
- }
-
- // If the link is already contained in the post, then it neeedn't to be added again
- // But: if the link is beyond the limit, then it has to be added.
- if (($link != "") && strstr($msg, $link)) {
- $pos = strpos($msg, $link);
-
- // Will the text be shortened in the link?
- // Or is the link the last item in the post?
- if (($limit > 0) && ($pos < $limit) && (($pos + 23 > $limit) || ($pos + strlen($link) == strlen($msg)))) {
- $msg = trim(str_replace($link, "", $msg));
- } elseif (($limit == 0) || ($pos < $limit)) {
- // The limit has to be increased since it will be shortened - but not now
- // Only do it with Twitter (htmlmode = 8)
- if (($limit > 0) && (strlen($link) > 23) && ($htmlmode == 8)) {
- $limit = $limit - 23 + strlen($link);
- }
-
- $link = "";
-
- if ($post["type"] == "text") {
- unset($post["url"]);
- }
- }
- }
+ $naked_text = preg_replace('/\[(.+?)\]/','', $text);
+ if (!$keep_urls) {
+ $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text);
}
- if ($limit > 0) {
- // Reduce multiple spaces
- // When posted to a network with limited space, we try to gain space where possible
- while (strpos($msg, " ") !== false) {
- $msg = str_replace(" ", " ", $msg);
- }
-
- // Twitter is using its own limiter, so we always assume that shortened links will have this length
- if (iconv_strlen($link, "UTF-8") > 0) {
- $limit = $limit - 23;
- }
+ return $naked_text;
+ }
- if (iconv_strlen($msg, "UTF-8") > $limit) {
- if (($post["type"] == "text") && isset($post["url"])) {
- $post["url"] = $b["plink"];
- } elseif (!isset($post["url"])) {
- $limit = $limit - 23;
- $post["url"] = $b["plink"];
- // Which purpose has this line? It is now uncommented, but left as a reminder
- //} elseif (strpos($b["body"], "[share") !== false) {
- // $post["url"] = $b["plink"];
- } elseif (PConfig::get($b["uid"], "system", "no_intelligent_shortening")) {
- $post["url"] = $b["plink"];
- }
- $msg = Plaintext::shorten($msg, $limit);
- }
+ private static function proxyUrl($image, $simplehtml = false)
+ {
+ // Only send proxied pictures to API and for internal display
+ if (in_array($simplehtml, [false, 2])) {
+ return ProxyUtils::proxifyUrl($image);
+ } else {
+ return $image;
}
-
- $post["text"] = trim($msg);
-
- return($post);
}
public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
*/
public static function limitBodySize($body)
{
- $maxlen = get_max_import_size();
+ $maxlen = Config::get('config', 'max_import_size', 0);
// If the length of the body, including the embedded images, is smaller
// than the maximum, then don't waste time looking for the images
private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
{
$data = self::getAttachmentData($return);
- if (!$data) {
+ if (empty($data) || empty($data["url"])) {
return $return;
}
if (isset($data["title"])) {
$data["title"] = strip_tags($data["title"]);
$data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
+ } else {
+ $data["title"] = null;
}
- if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) {
+ if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data["image"])) {
$data["preview"] = $data["image"];
$data["image"] = "";
}
throw new Exception('OEmbed is disabled for this attachment.');
}
} catch (Exception $e) {
+ $data["title"] = defaults($data, 'title', $data['url']);
+
if ($simplehtml != 4) {
$return = sprintf('<div class="type-%s">', $data["type"]);
}
- if ($data["image"] != "") {
- $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], proxy_url($data["image"]), $data["title"]);
- } elseif ($data["preview"] != "") {
- $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], proxy_url($data["preview"]), $data["title"]);
- }
-
- if (($data["type"] == "photo") && ($data["url"] != "") && ($data["image"] != "")) {
- $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], proxy_url($data["image"]), $data["title"]);
- } else {
- $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
+ if (!empty($data['title']) && !empty($data['url'])) {
+ if (!empty($data["image"]) && empty($data["text"]) && ($data["type"] == "photo")) {
+ $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
+ } else {
+ if (!empty($data["image"])) {
+ $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], self::proxyUrl($data["image"], $simplehtml), $data["title"]);
+ } elseif (!empty($data["preview"])) {
+ $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], self::proxyUrl($data["preview"], $simplehtml), $data["title"]);
+ }
+ $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
+ }
}
- if ($data["description"] != "" && $data["description"] != $data["title"]) {
+ if (!empty($data["description"]) && $data["description"] != $data["title"]) {
// Sanitize the HTML by converting it to BBCode
$bbcode = HTML::toBBCode($data["description"]);
$return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
}
- if ($data["type"] == "link") {
+
+ if (!empty($data['url'])) {
$return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
}
}
}
- return trim($data["text"] . ' ' . $return . ' ' . $data["after"]);
+ return trim(defaults($data, 'text', '') . ' ' . $return . ' ' . defaults($data, 'after', ''));
}
public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
if (!$data) {
return $Text;
} elseif ($nolink) {
- return $data["text"] . $data["after"];
+ return $data["text"] . defaults($data, 'after', '');
}
- $title = htmlentities($data["title"], ENT_QUOTES, 'UTF-8', false);
+ $title = htmlentities(defaults($data, 'title', ''), ENT_QUOTES, 'UTF-8', false);
$text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
if ($plaintext || (($title != "") && strstr($text, $title))) {
$data["title"] = $data["url"];
}
// If the link already is included in the post, don't add it again
- if (($data["url"] != "") && strpos($data["text"], $data["url"])) {
+ if (!empty($data["url"]) && strpos($data["text"], $data["url"])) {
return $data["text"] . $data["after"];
}
$text = $data["text"];
- if (($data["url"] != "") && ($data["title"] != "")) {
+ if (!empty($data["url"]) && !empty($data["title"])) {
$text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
- } elseif (($data["url"] != "")) {
+ } elseif (!empty($data["url"])) {
$text .= "\n[url]" . $data["url"] . "[/url]";
}
// it loops over the array starting from the first element and going sequentially
// to the last element
$newbody = str_replace('[$#saved_image' . $cnt . '#$]',
- '<img src="' . proxy_url($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
+ '<img src="' . self::proxyUrl($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
$cnt++;
}
case 5:
$text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
break;
- case 6: // app.net
- $text = $preshare . ">> @" . $userid_compact . ": <br />" . $share[3];
- break;
case 7: // statusnet/GNU Social
$text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3];
break;
case 8: // twitter
$text = $preshare . "RT @" . $userid_compact . ": " . $share[3];
break;
- case 9: // Google+/Facebook
+ case 9: // Google+
$text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
if ($link != "") {
} else {
$text = trim($share[1]) . "\n";
- $avatar = proxy_url($avatar, false, PROXY_SIZE_THUMB);
+ $avatar = ProxyUtils::proxifyUrl($avatar, false, ProxyUtils::SIZE_THUMB);
$tpl = get_markup_template('shared_content.tpl');
$text .= replace_macros($tpl, [
* Simple HTML values meaning:
* - 0: Friendica display
* - 1: Unused
- * - 2: Used for Facebook, Google+, Windows Phone push, Friendica API
+ * - 2: Used for Google+, Windows Phone push, Friendica API
* - 3: Used before converting to Markdown in bb2diaspora.php
* - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
* - 5: Unused
- * - 6: Used for Appnet
+ * - 6: Unused
* - 7: Used for dfrn, OStatus
* - 8: Used for WP backlink text setting
*
}, $text
);
+ $text = preg_replace_callback(
+ "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
+ function ($match) {
+ return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
+ }, $text
+ );
+
// Server independent link to posts and comments
// See issue: https://github.com/diaspora/diaspora_federation/issues/75
$expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
$text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
// Check for strike-through text
- $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<strike>$1</strike>', $text);
+ $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
// Check for over-line text
$text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
// [img=widthxheight]image source[/img]
$text = preg_replace_callback(
"/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
- function ($matches) {
+ function ($matches) use ($simple_html) {
if (strpos($matches[3], "data:image/") === 0) {
return $matches[0];
}
- $matches[3] = proxy_url($matches[3]);
+ $matches[3] = self::proxyUrl($matches[3], $simple_html);
return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
},
$text
$text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<img src="$3" style="width: $1px;" >', $text);
$text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
+ $text = preg_replace_callback("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism",
+ function ($matches) use ($simple_html) {
+ $matches[1] = self::proxyUrl($matches[1], $simple_html);
+ $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
+ return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '">';
+ },
+ $text);
+
// Images
// [img]pathtoimage[/img]
$text = preg_replace_callback(
"/\[img\](.*?)\[\/img\]/ism",
- function ($matches) {
+ function ($matches) use ($simple_html) {
if (strpos($matches[1], "data:image/") === 0) {
return $matches[0];
}
- $matches[1] = proxy_url($matches[1]);
+ $matches[1] = self::proxyUrl($matches[1], $simple_html);
return "[img]" . $matches[1] . "[/img]";
},
$text
// Try to Oembed
if ($try_oembed) {
- $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4))\[\/video\]/ism", '<video src="$1" controls="controls" width="' . $a->videowidth . '" height="' . $a->videoheight . '" loop="true"><a href="$1">$1</a></video>', $text);
- $text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3))\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
+ $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4)
.*?)\[\/video\]/ism", '<video src="$1" controls="controls" width="' . $a->videowidth . '" height="' . $a->videoheight . '" loop="true"><a href="$1">$1</a></video>', $text);
+ $text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3)
.*?)\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
$text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
$text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
} else {
- $text = preg_replace("/\[video\](.*?)\[\/video\]/",
+ $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
'<a href="$1" target="_blank">$1</a>', $text);
- $text = preg_replace("/\[audio\](.*?)\[\/audio\]/",
+ $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
'<a href="$1" target="_blank">$1</a>', $text);
}
$text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
$text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
-
+ /// @todo What is the meaning of these lines?
$text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
$text = preg_replace('/\&\#039\;/', '\'', $text);
- $text = preg_replace('/\"\;/', '"', $text);
+
+ // Currently deactivated, it made problems with " inside of alt texts.
+ //$text = preg_replace('/\"\;/', '"', $text);
// fix any escaped ampersands that may have been converted into links
$text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
* @param string $addon The addon for which the abstract is meant for
* @return string The abstract
*/
- private static function getAbstract($text, $addon = "")
+ public static function getAbstract($text, $addon = "")
{
$abstract = "";
$abstracts = [];
https://reisub.nsupdate.info/git / friendica.git / .git / blobdiff