<?php
/**
- * @file src/Content/Text/BBCode.php
+ * @copyright Copyright (C) 2020, Friendica
+ *
+ * @license GNU AGPL version 3 or any later version
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as
+ * published by the Free Software Foundation, either version 3 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program. If not, see <https://www.gnu.org/licenses/>.
+ *
*/
namespace Friendica\Content\Text;
use DOMDocument;
use DOMXPath;
use Exception;
-use Friendica\BaseObject;
use Friendica\Content\OEmbed;
use Friendica\Content\Smilies;
-use Friendica\Core\Cache;
-use Friendica\Core\Config;
use Friendica\Core\Hook;
-use Friendica\Core\L10n;
use Friendica\Core\Logger;
use Friendica\Core\Protocol;
use Friendica\Core\Renderer;
use Friendica\Core\System;
+use Friendica\DI;
use Friendica\Model\Contact;
use Friendica\Model\Event;
+use Friendica\Model\Photo;
use Friendica\Network\Probe;
use Friendica\Object\Image;
+use Friendica\Protocol\Activity;
+use Friendica\Util\Images;
use Friendica\Util\Map;
use Friendica\Util\Network;
use Friendica\Util\ParseUrl;
use Friendica\Util\Strings;
use Friendica\Util\XML;
-class BBCode extends BaseObject
+class BBCode
{
+ const INTERNAL = 0;
+ const API = 2;
+ const DIASPORA = 3;
+ const CONNECTORS = 4;
+ const OSTATUS = 7;
+ const TWITTER = 8;
+ const BACKLINK = 8;
+ const ACTIVITYPUB = 9;
+
/**
- * @brief Fetches attachment data that were generated the old way
+ * Fetches attachment data that were generated the old way
*
* @param string $body Message body
* @return array
- * 'type' -> Message type ("link", "video", "photo")
+ * 'type' -> Message type ('link', 'video', 'photo')
* 'text' -> Text before the shared message
* 'after' -> Text after the shared message
* 'image' -> Preview image of the message
if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
foreach ($attached as $data) {
- if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
+ if (!in_array($data[1], ['type-link', 'type-video', 'type-photo'])) {
continue;
}
- $post["type"] = substr($data[1], 5);
+ $post['type'] = substr($data[1], 5);
$pos = strpos($body, $data[0]);
if ($pos > 0) {
- $post["text"] = trim(substr($body, 0, $pos));
- $post["after"] = trim(substr($body, $pos + strlen($data[0])));
+ $post['text'] = trim(substr($body, 0, $pos));
+ $post['after'] = trim(substr($body, $pos + strlen($data[0])));
} else {
- $post["text"] = trim(str_replace($data[0], "", $body));
- $post["after"] = '';
+ $post['text'] = trim(str_replace($data[0], '', $body));
+ $post['after'] = '';
}
$attacheddata = $data[2];
- $URLSearchString = "^\[\]";
-
- if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
+ if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
- $picturedata = Image::getInfoFromURL($matches[1]);
+ $picturedata = Images::getInfoFromURLCached($matches[1]);
if ($picturedata) {
if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
- $post["image"] = $matches[1];
+ $post['image'] = $matches[1];
} else {
- $post["preview"] = $matches[1];
+ $post['preview'] = $matches[1];
}
}
}
- if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
- $post["url"] = $matches[1];
- $post["title"] = $matches[2];
+ if (preg_match("/\[bookmark\=(.*?)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
+ $post['url'] = $matches[1];
+ $post['title'] = $matches[2];
}
- if (!empty($post["url"]) && (in_array($post["type"], ["link", "video"]))
- && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
- $post["url"] = $matches[1];
+ if (!empty($post['url']) && (in_array($post['type'], ['link', 'video']))
+ && preg_match("/\[url\=(.*?)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
+ $post['url'] = $matches[1];
}
// Search for description
if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
- $post["description"] = $matches[1];
+ $post['description'] = $matches[1];
}
}
}
}
/**
- * @brief Fetches attachment data that were generated with the "attachment" element
+ * Fetches attachment data that were generated with the "attachment" element
*
* @param string $body Message body
* @return array
- * 'type' -> Message type ("link", "video", "photo")
+ * 'type' -> Message type ('link', 'video', 'photo')
* 'text' -> Text before the shared message
* 'after' -> Text after the shared message
* 'image' -> Preview image of the message
*/
public static function getAttachmentData($body)
{
- $data = [];
+ $data = [
+ 'type' => '',
+ 'text' => '',
+ 'after' => '',
+ 'image' => null,
+ 'url' => '',
+ 'title' => '',
+ 'description' => '',
+ ];
if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
return self::getOldAttachmentData($body);
$attributes = $match[2];
- $data["text"] = trim($match[1]);
+ $data['text'] = trim($match[1]);
- $type = "";
+ $type = '';
preg_match("/type='(.*?)'/ism", $attributes, $matches);
if (!empty($matches[1])) {
$type = strtolower($matches[1]);
$type = strtolower($matches[1]);
}
- if ($type == "") {
+ if ($type == '') {
return [];
}
- if (!in_array($type, ["link", "audio", "photo", "video"])) {
+ if (!in_array($type, ['link', 'audio', 'photo', 'video'])) {
return [];
}
- if ($type != "") {
- $data["type"] = $type;
+ if ($type != '') {
+ $data['type'] = $type;
}
- $url = "";
+ $url = '';
preg_match("/url='(.*?)'/ism", $attributes, $matches);
if (!empty($matches[1])) {
$url = $matches[1];
$url = $matches[1];
}
- if ($url != "") {
- $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
+ if ($url != '') {
+ $data['url'] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
}
- $title = "";
+ $title = '';
preg_match("/title='(.*?)'/ism", $attributes, $matches);
if (!empty($matches[1])) {
$title = $matches[1];
$title = $matches[1];
}
- if ($title != "") {
+ if ($title != '') {
$title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
$title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
- $title = str_replace(["[", "]"], ["[", "]"], $title);
- $data["title"] = $title;
+ $title = str_replace(['[', ']'], ['[', ']'], $title);
+ $data['title'] = $title;
}
- $image = "";
+ $image = '';
preg_match("/image='(.*?)'/ism", $attributes, $matches);
if (!empty($matches[1])) {
$image = $matches[1];
$image = $matches[1];
}
- if ($image != "") {
- $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
+ if ($image != '') {
+ $data['image'] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
}
- $preview = "";
+ $preview = '';
preg_match("/preview='(.*?)'/ism", $attributes, $matches);
if (!empty($matches[1])) {
$preview = $matches[1];
$preview = $matches[1];
}
- if ($preview != "") {
- $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
+ if ($preview != '') {
+ $data['preview'] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
}
- $data["description"] = trim($match[3]);
+ $data['description'] = trim($match[3]);
- $data["after"] = trim($match[4]);
+ $data['after'] = trim($match[4]);
return $data;
}
*/
$has_title = !empty($item['title']);
- $plink = defaults($item, 'plink', '');
+ $plink = $item['plink'] ?? '';
$post = self::getAttachmentData($body);
+ // Get all linked images with alternative image description
+ if (preg_match_all("/\[img=([^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
+ foreach ($pictures as $picture) {
+ if (Photo::isLocal($picture[1])) {
+ $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2]];
+ }
+ }
+ if (!empty($post['images']) && !empty($post['images'][0]['description'])) {
+ $post['image_description'] = $post['images'][0]['description'];
+ }
+ }
+
+ if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) {
+ foreach ($pictures as $picture) {
+ if (Photo::isLocal($picture[1])) {
+ $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => ''];
+ }
+ }
+ }
+
// if nothing is found, it maybe having an image.
- if (!isset($post["type"])) {
+ if (!isset($post['type'])) {
// Simplify image codes
$body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
+ $body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
+ $post['text'] = $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 (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $body, $pictures, PREG_SET_ORDER)) {
if ((count($pictures) == 1) && !$has_title) {
- // Checking, if the link goes to a picture
- $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
+ if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
+ // Replace the preview picture with the real picture
+ $url = str_replace('-1.', '-0.', $pictures[0][2]);
+ $data = ['url' => $url, 'type' => 'photo'];
+ } else {
+ // Checking, if the link goes to a picture
+ $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
+ }
// Workaround:
// Sometimes photo posts to the own album are not detected at the start.
// So we seem to cannot use the cache for these cases. That's strange.
- if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
+ if (($data['type'] != 'photo') && strstr($pictures[0][1], "/photos/")) {
$data = ParseUrl::getSiteinfo($pictures[0][1], true);
}
- if ($data["type"] == "photo") {
- $post["type"] = "photo";
- if (isset($data["images"][0])) {
- $post["image"] = $data["images"][0]["src"];
- $post["url"] = $data["url"];
+ if ($data['type'] == 'photo') {
+ $post['type'] = 'photo';
+ if (isset($data['images'][0])) {
+ $post['image'] = $data['images'][0]['src'];
+ $post['url'] = $data['url'];
} else {
- $post["image"] = $data["url"];
+ $post['image'] = $data['url'];
}
- $post["preview"] = $pictures[0][2];
- $post["text"] = str_replace($pictures[0][0], "", $body);
+ $post['preview'] = $pictures[0][2];
+ $post['text'] = trim(str_replace($pictures[0][0], '', $body));
} else {
- $imgdata = Image::getInfoFromURL($pictures[0][1]);
- if ($imgdata && substr($imgdata["mime"], 0, 6) == "image/") {
- $post["type"] = "photo";
- $post["image"] = $pictures[0][1];
- $post["preview"] = $pictures[0][2];
- $post["text"] = str_replace($pictures[0][0], "", $body);
+ $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
+ if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
+ $post['type'] = 'photo';
+ $post['image'] = $pictures[0][1];
+ $post['preview'] = $pictures[0][2];
+ $post['text'] = trim(str_replace($pictures[0][0], '', $body));
}
}
} elseif (count($pictures) > 0) {
- $post["type"] = "link";
- $post["url"] = $plink;
- $post["image"] = $pictures[0][2];
- $post["text"] = $body;
+ $post['type'] = 'link';
+ $post['url'] = $plink;
+ $post['image'] = $pictures[0][2];
+ $post['text'] = $body;
+
+ foreach ($pictures as $picture) {
+ $post['text'] = trim(str_replace($picture[0], '', $post['text']));
+ }
}
- } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
+ } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
if ((count($pictures) == 1) && !$has_title) {
- $post["type"] = "photo";
- $post["image"] = $pictures[0][1];
- $post["text"] = str_replace($pictures[0][0], "", $body);
+ $post['type'] = 'photo';
+ $post['image'] = $pictures[0][1];
+ $post['text'] = str_replace($pictures[0][0], '', $body);
} elseif (count($pictures) > 0) {
- $post["type"] = "link";
- $post["url"] = $plink;
- $post["image"] = $pictures[0][1];
- $post["text"] = $body;
+ $post['type'] = 'link';
+ $post['url'] = $plink;
+ $post['image'] = $pictures[0][1];
+ $post['text'] = $body;
+
+ foreach ($pictures as $picture) {
+ $post['text'] = trim(str_replace($picture[0], '', $post['text']));
+ }
}
}
// Test for the external links
- preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
- preg_match_all("(\[url\=([$URLSearchString]*)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
+ preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER);
+ preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $links2, PREG_SET_ORDER);
$links = array_merge($links1, $links2);
// If there is only a single one, then use it.
// This should cover link posts via API.
- if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
- $post["type"] = "link";
- $post["text"] = trim($body);
- $post["url"] = $links[0][1];
+ if ((count($links) == 1) && !isset($post['preview']) && !$has_title) {
+ $post['type'] = 'link';
+ $post['url'] = $links[0][1];
}
+ // Simplify "video" element
+ $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
+
// Now count the number of external media links
- preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
- preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
- preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
- preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
+ preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
+ preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
+ preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER);
+ preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $links4, PREG_SET_ORDER);
// Add them to the other external links
$links = array_merge($links, $links1, $links2, $links3, $links4);
// Are there more than one?
if (count($links) > 1) {
// The post will be the type "text", which means a blog post
- unset($post["type"]);
- $post["url"] = $plink;
+ unset($post['type']);
+ $post['url'] = $plink;
}
- if (!isset($post["type"])) {
- $post["type"] = "text";
- $post["text"] = trim($body);
+ if (!isset($post['type'])) {
+ $post['type'] = "text";
+ $post['text'] = trim($body);
}
- } elseif (isset($post["url"]) && ($post["type"] == "video")) {
- $data = ParseUrl::getSiteinfoCached($post["url"], true);
+ } elseif (isset($post['url']) && ($post['type'] == 'video')) {
+ $data = ParseUrl::getSiteinfoCached($post['url'], true);
- if (isset($data["images"][0])) {
- $post["image"] = $data["images"][0]["src"];
+ if (isset($data['images'][0])) {
+ $post['image'] = $data['images'][0]['src'];
}
}
}
/**
- * @brief Converts a BBCode text into plaintext
+ * Remove [attachment] BBCode and replaces it with a regular [url]
+ *
+ * @param string $body
+ * @param boolean $no_link_desc No link description
+ *
+ * @return string with replaced body
+ */
+ public static function removeAttachment($body, $no_link_desc = false)
+ {
+ return preg_replace_callback("/\s*\[attachment (.*)\](.*?)\[\/attachment\]\s*/ism",
+ function ($match) use ($no_link_desc) {
+ $attach_data = self::getAttachmentData($match[0]);
+ if (empty($attach_data['url'])) {
+ return $match[0];
+ } elseif (empty($attach_data['title']) || $no_link_desc) {
+ return "\n[url]" . $attach_data['url'] . "[/url]\n";
+ } else {
+ return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
+ }
+ }, $body);
+ }
+
+ /**
+ * Converts a BBCode text into plaintext
*
* @param $text
* @param bool $keep_urls Whether to keep URLs in the resulting plaintext
*/
public static function toPlaintext($text, $keep_urls = true)
{
- $naked_text = preg_replace('/\[(.+?)\]\s*/','', $text);
- if (!$keep_urls) {
- $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text);
- }
+ $naked_text = HTML::toPlaintext(self::convert($text, false, 0, true), 0, !$keep_urls);
return $naked_text;
}
- private static function proxyUrl($image, $simplehtml = false)
+ private static function proxyUrl($image, $simplehtml = self::INTERNAL)
{
// Only send proxied pictures to API and for internal display
- if (in_array($simplehtml, [false, 2])) {
+ if (in_array($simplehtml, [self::INTERNAL, self::API])) {
return ProxyUtils::proxifyUrl($image);
} else {
return $image;
}
}
- public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
+ /**
+ * This function changing the visual size (not the real size) of images.
+ * The function does not work for pictures with an alternate text description.
+ * This could only be changed by using some new "img" BBCode format.
+ *
+ * @param string $srctext The body with images
+ * @return string The body with possibly scaled images
+ */
+ public static function scaleExternalImages(string $srctext)
{
- // Suppress "view full size"
- if (intval(Config::get('system', 'no_view_full_size'))) {
- $include_link = false;
- }
+ $s = $srctext;
- // Picture addresses can contain special characters
- $s = htmlspecialchars_decode($srctext);
+ // Simplify image links
+ $s = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $s);
$matches = null;
$c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
foreach ($matches as $mtch) {
Logger::log('scale_external_image: ' . $mtch[1]);
- $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
+ $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
if (stristr($mtch[1], $hostname)) {
continue;
}
- // $scale_replace, if passed, is an array of two elements. The
- // first is the name of the full-size image. The second is the
- // name of a remote, scaled-down version of the full size image.
- // This allows Friendica to display the smaller remote image if
- // one exists, while still linking to the full-size image
- if ($scale_replace) {
- $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
- } else {
- $scaled = $mtch[1];
- }
- $i = Network::fetchUrl($scaled);
- if (!$i) {
- return $srctext;
+ $curlResult = Network::curl($mtch[1], true);
+ if (!$curlResult->isSuccess()) {
+ continue;
}
- // guess mimetype from headers or filename
- $type = Image::guessType($mtch[1], true);
+ $i = $curlResult->getBody();
+ $type = $curlResult->getContentType();
+ $type = Images::getMimeTypeByData($i, $mtch[1], $type);
if ($i) {
$Image = new Image($i, $type);
Logger::log('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], Logger::DEBUG);
$s = str_replace(
$mtch[0],
- '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
- . "\n" . (($include_link)
- ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
- : ''),
+ '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
+ . "\n",
$s
);
Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG);
}
}
- // replace the special char encoding
- $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
return $s;
}
/**
+ * Truncates imported message body string length to max_import_size
+ *
* The purpose of this function is to apply system message length limits to
* imported messages without including any embedded photos in the length
*
- * @brief Truncates imported message body string length to max_import_size
* @param string $body
* @return string
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
public static function limitBodySize($body)
{
- $maxlen = Config::get('config', 'max_import_size', 0);
+ $maxlen = DI::config()->get('config', 'max_import_size', 0);
// If the length of the body, including the embedded images, is smaller
// than the maximum, then don't waste time looking for the images
*
* Note: Can produce a [bookmark] tag in the returned string
*
- * @brief Processes [attachment] tags
- * @param string $return
- * @param bool|int $simplehtml
- * @param bool $tryoembed
+ * @param string $text
+ * @param integer $simplehtml
+ * @param bool $tryoembed
* @return string
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
+ private static function convertAttachment($text, $simplehtml = self::INTERNAL, $tryoembed = true)
{
- $data = self::getAttachmentData($return);
- if (empty($data) || empty($data["url"])) {
- return $return;
+ $data = self::getAttachmentData($text);
+ if (empty($data) || empty($data['url'])) {
+ return $text;
}
- if (isset($data["title"])) {
- $data["title"] = strip_tags($data["title"]);
- $data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
+ if (isset($data['title'])) {
+ $data['title'] = strip_tags($data['title']);
+ $data['title'] = str_replace(['http://', 'https://'], '', $data['title']);
} else {
- $data["title"] = null;
+ $data['title'] = null;
}
- if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data["image"])) {
- $data["preview"] = $data["image"];
- $data["image"] = "";
+ if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
+ $data['preview'] = $data['image'];
+ $data['image'] = '';
}
$return = '';
- if ($simplehtml == 7) {
- $return = self::convertUrlForOStatus($data["url"]);
- } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
- $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $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) {
- $data["title"] = defaults($data, 'title', $data['url']);
+ try {
+ if ($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) {
+ $data['title'] = ($data['title'] ?? '') ?: $data['url'];
- if ($simplehtml != 4) {
- $return = sprintf('<div class="type-%s">', $data["type"]);
- }
+ if ($simplehtml != self::CONNECTORS) {
+ $return = sprintf('<div class="type-%s">', $data['type']);
+ }
- if (!empty($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'], $data['title']);
+ if (!empty($data['title']) && !empty($data['url'])) {
+ if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
+ $return .= sprintf('<a href="%s" target="_blank" rel="noopener noreferrer"><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" rel="noopener noreferrer"><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" rel="noopener noreferrer"><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'], $data['title']);
}
+ }
- if (!empty($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 (!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 (!empty($data['url'])) {
- $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
- }
+ if (!empty($data['url'])) {
+ $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
+ }
- if ($simplehtml != 4) {
- $return .= '</div>';
- }
+ if ($simplehtml != self::CONNECTORS) {
+ $return .= '</div>';
}
}
- return trim(defaults($data, 'text', '') . ' ' . $return . ' ' . defaults($data, 'after', ''));
+ return trim(($data['text'] ?? '') . ' ' . $return . ' ' . ($data['after'] ?? ''));
}
public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
if (!$data) {
return $Text;
} elseif ($nolink) {
- return $data["text"] . defaults($data, 'after', '');
+ return $data['text'] . ($data['after'] ?? '');
}
- $title = htmlentities(defaults($data, 'title', ''), ENT_QUOTES, 'UTF-8', false);
- $text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
- if ($plaintext || (($title != "") && strstr($text, $title))) {
- $data["title"] = $data["url"];
- } elseif (($text != "") && strstr($title, $text)) {
- $data["text"] = $data["title"];
- $data["title"] = $data["url"];
+ $title = htmlentities($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'] = $data['url'];
}
- if (empty($data["text"]) && !empty($data["title"]) && empty($data["url"])) {
- return $data["title"] . $data["after"];
+ if (empty($data['text']) && !empty($data['title']) && empty($data['url'])) {
+ return $data['title'] . $data['after'];
}
// If the link already is included in the post, don't add it again
- if (!empty($data["url"]) && strpos($data["text"], $data["url"])) {
- return $data["text"] . $data["after"];
+ if (!empty($data['url']) && strpos($data['text'], $data['url'])) {
+ return $data['text'] . $data['after'];
}
- $text = $data["text"];
+ $text = $data['text'];
- if (!empty($data["url"]) && !empty($data["title"])) {
- $text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
- } elseif (!empty($data["url"])) {
- $text .= "\n[url]" . $data["url"] . "[/url]";
+ if (!empty($data['url']) && !empty($data['title'])) {
+ $text .= "\n[url=" . $data['url'] . ']' . $data['title'] . '[/url]';
+ } elseif (!empty($data['url'])) {
+ $text .= "\n[url]" . $data['url'] . '[/url]';
}
- return $text . "\n" . $data["after"];
+ return $text . "\n" . $data['after'];
}
/**
* 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 $match Array with the matching values
* @return string reformatted link including HTML codes
*/
- private static function convertUrlForOStatusCallback($match)
+ private static function convertUrlForActivityPubCallback($match)
{
$url = $match[1];
return $match[0];
}
- return self::convertUrlForOStatus($url);
+ return self::convertUrlForActivityPub($url);
}
/**
- * @brief Converts [url] BBCodes in a format that looks fine on OStatus systems.
+ * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
+ *
* @param string $url URL that is about to be reformatted
* @return string reformatted link including HTML codes
*/
- private static function convertUrlForOStatus($url)
+ private static function convertUrlForActivityPub($url)
+ {
+ $html = '<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>';
+ return sprintf($html, $url, self::getStyledURL($url));
+ }
+
+ /**
+ * Converts an URL in a nicer format (without the scheme and possibly shortened)
+ *
+ * @param string $url URL that is about to be reformatted
+ * @return string reformatted link
+ */
+ private static function getStyledURL($url)
{
$parts = parse_url($url);
$scheme = $parts['scheme'] . '://';
$styled_url = substr($styled_url, 0, 30) . "…";
}
- $html = '<a href="%s" target="_blank">%s</a>';
-
- return sprintf($html, $url, $styled_url);
+ return $styled_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="' . self::proxyUrl($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
+ '<img src="' . self::proxyUrl($image) . '" alt="' . DI::l10n()->t('Image/photo') . '" />', $newbody);
$cnt++;
}
public static function convertShare($text, callable $callback)
{
$return = preg_replace_callback(
- "/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
+ "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
function ($match) use ($callback) {
$attribute_string = $match[2];
-
$attributes = [];
- foreach(['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
+ foreach (['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
- $attributes[$field] = html_entity_decode(defaults($matches, 2, ''), ENT_QUOTES, 'UTF-8');
+ $attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
}
// We only call this so that a previously unknown contact can be added.
// This is important for the function "Model\Contact::getDetailsByURL()".
// This function then can fetch an entry from the contact table.
- Contact::getIdForURL($attributes['profile'], 0, true);
+ $default['url'] = $attributes['profile'];
+
+ if (!empty($attributes['author'])) {
+ $default['name'] = $attributes['author'];
+ }
+
+ if (!empty($attributes['avatar'])) {
+ $default['photo'] = $attributes['avatar'];
+ }
+
+ Contact::getIdForURL($attributes['profile'], 0, true, $default);
$author_contact = Contact::getDetailsByURL($attributes['profile']);
- $author_contact['addr'] = defaults($author_contact, 'addr' , Protocol::getAddrFromProfileUrl($attributes['profile']));
+ $author_contact['url'] = ($author_contact['url'] ?? $attributes['profile']);
+ $author_contact['addr'] = ($author_contact['addr'] ?? '') ?: Protocol::getAddrFromProfileUrl($attributes['profile']);
- $attributes['author'] = defaults($author_contact, 'name' , $attributes['author']);
- $attributes['avatar'] = defaults($author_contact, 'micro', $attributes['avatar']);
- $attributes['profile'] = defaults($author_contact, 'url' , $attributes['profile']);
+ $attributes['author'] = ($author_contact['name'] ?? '') ?: $attributes['author'];
+ $attributes['avatar'] = ($author_contact['micro'] ?? '') ?: $attributes['avatar'];
+ $attributes['profile'] = ($author_contact['url'] ?? '') ?: $attributes['profile'];
if ($attributes['avatar']) {
$attributes['avatar'] = ProxyUtils::proxifyUrl($attributes['avatar'], false, ProxyUtils::SIZE_THUMB);
$mention = Protocol::formatMention($attributes['profile'], $attributes['author']);
switch ($simplehtml) {
- case 1:
- $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' <a href="' . $attributes['profile'] . '">' . $mention . '</a>: </p>' . "\n" . '«' . $content . '»';
- break;
- case 2:
+ case self::API:
$text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
break;
- case 3: // Diaspora
+ case self::DIASPORA:
if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0) {
$text = ($is_quote_share? '<hr />' : '') . '<p><a href="' . $attributes['link'] . '">' . $attributes['link'] . '</a></p>' . "\n";
} else {
}
break;
- case 4:
+ case self::CONNECTORS:
$headline = '<p><b>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8');
- $headline .= L10n::t('<a href="%1$s" target="_blank">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
+ $headline .= DI::l10n()->t('<a href="%1$s" target="_blank" rel="noopener noreferrer">%2$s</a> %3$s', $attributes['link'], $mention, $attributes['posted']);
$headline .= ':</b></p>' . "\n";
$text = ($is_quote_share? '<hr />' : '') . $headline . '<blockquote class="shared_content">' . trim($content) . '</blockquote>' . "\n";
break;
- case 5:
- $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
- break;
- case 7: // statusnet/GNU Social
+ case self::OSTATUS:
$text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
break;
- case 9: // Google+
- $text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n";
- $text .= '<p>' . $content . '</p>' . "\n";
-
- if ($attributes['link'] != '') {
- $text .= '<p>' . $attributes['link'] . '</p>';
- }
+ case self::ACTIVITYPUB:
+ $author = '@<span class="vcard"><a href="' . $author_contact['url'] . '" class="url u-url mention" title="' . $author_contact['addr'] . '"><span class="fn nickname mention">' . $author_contact['addr'] . '</span></a>:</span>';
+ $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
break;
default:
- // Transforms quoted tweets in rich attachments to avoid nested tweets
- if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) {
- try {
- $text = ($is_quote_share? '<br />' : '') . OEmbed::getHTML($attributes['link']);
- } catch (Exception $e) {
- $text = ($is_quote_share? '<br />' : '') . sprintf('[bookmark=%s]%s[/bookmark]', $attributes['link'], $content);
- }
- } else {
- $text = ($is_quote_share? "\n" : '');
-
- $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
- $text .= Renderer::replaceMacros($tpl, [
- '$profile' => $attributes['profile'],
- '$avatar' => $attributes['avatar'],
- '$author' => $attributes['author'],
- '$link' => $attributes['link'],
- '$posted' => $attributes['posted'],
- '$content' => trim($content)
- ]);
- }
+ $text = ($is_quote_share? "\n" : '');
+
+ $tpl = Renderer::getMarkupTemplate('shared_content.tpl');
+ $text .= Renderer::replaceMacros($tpl, [
+ '$profile' => $attributes['profile'],
+ '$avatar' => $attributes['avatar'],
+ '$author' => $attributes['author'],
+ '$link' => $attributes['link'],
+ '$posted' => $attributes['posted'],
+ '$content' => trim($content)
+ ]);
break;
}
private static function removePictureLinksCallback($match)
{
- $text = Cache::get($match[1]);
+ $cache_key = 'remove:' . $match[1];
+ $text = DI::cache()->get($cache_key);
if (is_null($text)) {
- $a = self::getApp();
+ $a = DI::app();
$stamp1 = microtime(true);
@curl_exec($ch);
$curl_info = @curl_getinfo($ch);
- $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack());
+ DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
- if (substr($curl_info["content_type"], 0, 6) == "image/") {
- $text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
+ if (substr($curl_info['content_type'], 0, 6) == 'image/') {
+ $text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
} else {
- $text = "[url=" . $match[2] . "]" . $match[2] . "[/url]";
+ $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]);
}
}
- if (strtolower($attr["name"]) == "twitter:image") {
- $text = "[url=" . $attr["content"] . "]" . $attr["content"] . "[/url]";
+ if (strtolower($attr['name']) == 'twitter:image') {
+ $text = '[url=' . $attr['content'] . ']' . $attr['content'] . '[/url]';
}
}
}
- Cache::set($match[1], $text);
+ DI::cache()->set($cache_key, $text);
}
return $text;
private static function expandLinksCallback($match)
{
- if (($match[3] == "") || ($match[2] == $match[3]) || stristr($match[2], $match[3])) {
+ if (($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]);
+ $a = DI::app();
- if (is_null($text)) {
- $a = self::getApp();
+ // When the picture link is the own photo path then we can avoid fetching the link
+ $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
+ if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
+ if (!empty($match[3])) {
+ $text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
+ } else {
+ $text = '[img]' . str_replace('-1.', '-0.', $match[2]) . '[/img]';
+ }
+ return $text;
+ }
- $stamp1 = microtime(true);
+ $cache_key = 'clean:' . $match[1];
+ $text = DI::cache()->get($cache_key);
+ if (!is_null($text)) {
+ return $text;
+ }
- $ch = @curl_init($match[1]);
- @curl_setopt($ch, CURLOPT_NOBODY, true);
- @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
- @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
- @curl_exec($ch);
- $curl_info = @curl_getinfo($ch);
+ // Only fetch the header, not the content
+ $stamp1 = microtime(true);
- $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack());
+ $ch = @curl_init($match[1]);
+ @curl_setopt($ch, CURLOPT_NOBODY, true);
+ @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
+ @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
+ @curl_exec($ch);
+ $curl_info = @curl_getinfo($ch);
- // if its a link to a picture then embed this picture
- if (substr($curl_info["content_type"], 0, 6) == "image/") {
- $text = "[img]" . $match[1] . "[/img]";
+ DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
+
+ // if its a link to a picture then embed this picture
+ if (substr($curl_info['content_type'], 0, 6) == 'image/') {
+ $text = '[img]' . $match[1] . '[/img]';
+ } else {
+ if (!empty($match[3])) {
+ $text = '[img=' . $match[2] . ']' . $match[3] . '[/img]';
} else {
- $text = "[img]" . $match[2] . "[/img]";
+ $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]);
+ // 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;
- }
+ $doc = new DOMDocument();
+ @$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]" . $attr["content"] . "[/img]";
+ if (strtolower($attr['name']) == "twitter:image") {
+ if (!empty($match[3])) {
+ $text = "[img=" . $attr['content'] . "]" . $match[3] . "[/img]";
+ } else {
+ $text = "[img]" . $attr['content'] . "[/img]";
}
}
}
- Cache::set($match[1], $text);
}
+ DI::cache()->set($cache_key, $text);
return $text;
}
public static function cleanPictureLinks($text)
{
- $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
+ $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img=(.*)\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $text);
+ $return = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::cleanPictureLinksCallback', $return);
return $return;
}
/**
- * @brief Converts a BBCode message to HTML message
+ * Converts a BBCode message to HTML message
*
* BBcode 2 HTML was written by WAY2WEB.net
* extended to work with Mistpark/Friendica - Mike Macgirvin
* Simple HTML values meaning:
* - 0: Friendica display
* - 1: Unused
- * - 2: Used for Google+, Windows Phone push, Friendica API
+ * - 2: Used for 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: Unused
* - 7: Used for dfrn, OStatus
- * - 8: Used for WP backlink text setting
+ * - 8: Used for Twitter, WP backlink text setting
+ * - 9: ActivityPub
*
* @param string $text
* @param bool $try_oembed
* @return string
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false)
+ public static function convert($text, $try_oembed = true, $simple_html = self::INTERNAL, $for_plaintext = false)
{
- $a = self::getApp();
+ $a = DI::app();
+
+ $text = self::performWithEscapedTags($text, ['code'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
+ $text = self::performWithEscapedTags($text, ['noparse', 'nobb', 'pre'], function ($text) use ($try_oembed, $simple_html, $for_plaintext, $a) {
+ /*
+ * preg_match_callback function to replace potential Oembed tags with Oembed content
+ *
+ * $match[0] = [tag]$url[/tag] or [tag=$url]$title[/tag]
+ * $match[1] = $url
+ * $match[2] = $title or absent
+ */
+ $try_oembed_callback = function ($match)
+ {
+ $url = $match[1];
+ $title = $match[2] ?? null;
- /*
- * 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;
- };
-
- // Extracting multi-line code blocks before the whitespace processing
- $codeblocks = [];
+ try {
+ $return = OEmbed::getHTML($url, $title);
+ } catch (Exception $ex) {
+ $return = $match[0];
+ }
- $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
- function ($matches) use (&$codeblocks) {
- $return = $matches[0];
- if (strpos($matches[2], "\n") !== false) {
- $return = '#codeblock-' . count($codeblocks) . '#';
+ return $return;
+ };
- $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . trim($matches[2], "\n\r") . '</code></pre>';
- }
- return $return;
- },
- $text
- );
- // Hide all [noparse] contained bbtags by spacefying them
- // POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
- $text = preg_replace_callback("/\[noparse\](.*?)\[\/noparse\]/ism", 'self::escapeNoparseCallback', $text);
- $text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::escapeNoparseCallback', $text);
- $text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::escapeNoparseCallback', $text);
+ // Remove the abstract element. It is a non visible element.
+ $text = self::stripAbstract($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);
- // 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.
- // 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'];
- $extracted = self::extractImagesFromItemBody($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.
- // 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.
+ $ev = Event::fromBBCode($text);
- $ev = Event::fromBBCode($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
- // 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);
- $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);
- // 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]", $text);
+ // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
+ if (!$try_oembed) {
+ $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
+ }
- // when the content is meant exporting to other systems then remove the avatar picture since this doesn't really look good on these systems
- if (!$try_oembed) {
- $text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
- }
+ // Convert new line chars to html <br /> tags
+
+ // nlbr seems to be hopelessly messed up
+ // $Text = nl2br($Text);
+
+ // We'll emulate it.
+
+ $text = trim($text);
+ $text = str_replace("\r\n", "\n", $text);
+
+ // Remove linefeeds inside of the table elements. See issue #6799
+ $search = ["\n[th]", "[th]\n", " [th]", "\n[/th]", "[/th]\n", "[/th] ",
+ "\n[td]", "[td]\n", " [td]", "\n[/td]", "[/td]\n", "[/td] ",
+ "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
+ "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "];
+ $replace = ["[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
+ "[td]", "[td]", "[td]", "[/td]", "[/td]", "[/td]",
+ "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
+ "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"];
+ do {
+ $oldtext = $text;
+ $text = str_replace($search, $replace, $text);
+ } while ($oldtext != $text);
+
+ // Replace these here only once
+ $search = ["\n[table]", "[/table]\n"];
+ $replace = ["[table]", "[/table]"];
+ $text = str_replace($search, $replace, $text);
- // Convert new line chars to html <br /> tags
+ // removing multiplicated newlines
+ if (DI::config()->get('system', 'remove_multiplicated_lines')) {
+ $search = ["\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
+ "\n[h1]", "[/h1]\n", "\n[h2]", "[/h2]\n", "\n[h3]", "[/h3]\n", "\n[h4]", "[/h4]\n", "\n[h5]", "[/h5]\n", "\n[h6]", "[/h6]\n"];
+ $replace = ["\n\n", "\n", "\n", "[/quote]\n", "[/quote]", "[/li]", "[li]", "[ul]", "[/ul]", "\n[share ", "[/attachment]",
+ "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"];
+ do {
+ $oldtext = $text;
+ $text = str_replace($search, $replace, $text);
+ } while ($oldtext != $text);
+ }
- // nlbr seems to be hopelessly messed up
- // $Text = nl2br($Text);
+ /// @todo Have a closer look at the different html modes
+ // Handle attached links or videos
+ if ($simple_html == self::ACTIVITYPUB) {
+ $text = self::removeAttachment($text);
+ } elseif (!in_array($simple_html, [self::INTERNAL, self::CONNECTORS])) {
+ $text = self::removeAttachment($text, true);
+ } else {
+ $text = self::convertAttachment($text, $simple_html, $try_oembed);
+ }
- // We'll emulate it.
+ // leave open the posibility of [map=something]
+ // this is replaced in Item::prepareBody() which has knowledge of the item location
+ if (strpos($text, '[/map]') !== false) {
+ $text = preg_replace_callback(
+ "/\[map\](.*?)\[\/map\]/ism",
+ function ($match) use ($simple_html) {
+ return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
+ },
+ $text
+ );
+ }
- $text = trim($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 != $text);
- }
+ if (strpos($text, '[map=') !== false) {
+ $text = preg_replace_callback(
+ "/\[map=(.*?)\]/ism",
+ function ($match) use ($simple_html) {
+ return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
+ },
+ $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 https://daringfireball.net/2010/07/improved_regex_for_matching_urls)
- $autolink_regex = '@(?xi)
-(?<![=\'\]"/]) # Not preceded by =, \', ], ", /
-\b
-( # Capture 1: entire matched URL
- https?:// # http or https protocol
- (?:
- www\d{0,3}[.] # "www.", "www1.", "www2." … "www999."
- | # or
- [a-z0-9.\-]+[.][a-z]{2,4}/ # looks like domain name followed by a slash
- )
- (?: # One or more:
- [^\s()<>]+ # Run of non-space, non-()<>
- | # or
- \(([^\s()<>]+|(\([^\s()<>]+\)))*\) # balanced parens, up to 2 levels
- )+
- (?: # End with:
- \(([^\s()<>]+|(\([^\s()<>]+\)))*\) # balanced parens, up to 2 levels
- | # or
- [^\s`!()\[\]{};:\'".,<>?«»“”‘’] # not a space or one of these punct chars
- )
-)@';
- $text = preg_replace($autolink_regex, '[url]$1[/url]', $text);
- if ($simple_html == 7) {
- $text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
- $text = preg_replace_callback("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text);
- }
- } else {
- $text = preg_replace("(\[url\]([$URLSearchString]*)\[\/url\])ism", " $1 ", $text);
- $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
- }
+ if (strpos($text, '[map]') !== false) {
+ $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
+ }
+ // Check for headers
+ $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
+ $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
+ $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
+ $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
+ $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
+ $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
- // Handle attached links or videos
- $text = self::convertAttachment($text, $simple_html, $try_oembed);
-
- $text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
-
- // Remove all hashtag addresses
- if ((!$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",
- '@<a href="$2">$3</a>',
- $text);
- } elseif ($simple_html == 7) {
- $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
- '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
- $text);
- } elseif (!$simple_html) {
- $text = preg_replace("/([@!])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",
- '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
- $text);
- }
+ // Check for paragraph
+ $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $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);
+ // Check for bold text
+ $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
- if (in_array($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);
- }
+ // Check for Italics text
+ $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
- if ($simple_html == 5) {
- $text = preg_replace("/[^#@!]\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[url]$1[/url]', $text);
- }
+ // Check for Underline text
+ $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
- // Perform URL Search
- if ($try_oembed) {
- $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
- }
+ // Check for strike-through text
+ $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
- if ($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);
- }
+ // Check for over-line text
+ $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $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
- );
+ // Check for colored text
+ $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
- $text = preg_replace_callback(
- "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
- function ($match) {
- return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
- }, $text
- );
+ // Check for sized text
+ // [size=50] --> font-size: 50px (with the unit).
+ if ($simple_html != self::DIASPORA) {
+ $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
+ $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
+ } else {
+ // Issue 2199: Diaspora doesn't interpret the construct above, nor the <small> or <big> element
+ $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "$2", $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);
- /* Tag conversion
- * Supports:
- * - #[url=<anything>]<term>[/url]
- * - [url=<anything>]#<term>[/url]
- */
- $text = preg_replace_callback("/(?:#\[url\=[$URLSearchString]*\]|\[url\=[$URLSearchString]*\]#)(.*?)\[\/url\]/ism", function($matches) {
- return '#<a href="'
- . System::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
- . '" class="tag" title="' . XML::escape($matches[1]) . '">'
- . XML::escape($matches[1])
- . '</a>';
- }, $text);
+ // Check for centered text
+ $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
+
+ // Check for list text
+ $text = str_replace("[*]", "<li>", $text);
+
+ // Check for style sheet commands
+ $text = preg_replace_callback(
+ "(\[style=(.*?)\](.*?)\[\/style\])ism",
+ function ($match) {
+ return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
+ },
+ $text
+ );
+
+ // Check for CSS classes
+ $text = preg_replace_callback(
+ "(\[class=(.*?)\](.*?)\[\/class\])ism",
+ function ($match) {
+ return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
+ },
+ $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", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
+ $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
+ $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
+ $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
+ $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
+ $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
+ $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
+ $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
+ $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
+ $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
+ }
- // We need no target="_blank" for local links
- // convert links start with System::baseUrl() as local link without the target="_blank" attribute
- $escapedBaseUrl = preg_quote(System::baseUrl(), '/');
- $text = preg_replace("/\[url\](".$escapedBaseUrl."[$URLSearchString]*)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
- $text = preg_replace("/\[url\=(".$escapedBaseUrl."[$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
+ $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
+ $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
+ $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
+ $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
- $text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '<a href="$1" target="_blank">$1</a>', $text);
- $text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
+ $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
+ $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
- // Red compatibility, though the link can't be authenticated on Friendica
- $text = preg_replace("/\[zrl\=([$URLSearchString]*)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank">$2</a>', $text);
+ $text = str_replace('[hr]', '<hr />', $text);
+ if (!$for_plaintext) {
+ $escaped = [];
- // we may need to restrict this further if it picks up too many strays
- // link acct:user@host to a webfinger profile redirector
+ // Escaping BBCodes susceptible to contain rogue URL we don'' want the autolinker to catch
+ $text = preg_replace_callback('#\[(url|img|audio|video|youtube|vimeo|share|attachment|iframe|bookmark).+?\[/\1\]#ism',
+ function ($matches) use (&$escaped) {
+ $return = '{escaped-' . count($escaped) . '}';
+ $escaped[] = $matches[0];
- $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', '<a href="' . System::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
+ return $return;
+ },
+ $text
+ );
- // Perform MAIL Search
- $text = preg_replace("/\[mail\]([$MAILSearchString]*)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
- $text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
+ // Autolinker for isolated URLs
+ $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
- // leave open the posibility of [map=something]
- // this is replaced in Item::prepareBody() which has knowledge of the item location
+ // Restoring escaped blocks
+ $text = preg_replace_callback('/{escaped-([0-9]+)}/iU',
+ function ($matches) use ($escaped) {
+ return $escaped[intval($matches[1])] ?? $matches[0];
+ },
+ $text
+ );
+ }
- if (strpos($text, '[/map]') !== false) {
- $text = preg_replace_callback(
- "/\[map\](.*?)\[\/map\]/ism",
- function ($match) use ($simple_html) {
- return str_replace($match[0], '<p class="map">' . Map::byLocation($match[1], $simple_html) . '</p>', $match[0]);
- },
- $text
- );
- }
- if (strpos($text, '[map=') !== false) {
- $text = preg_replace_callback(
- "/\[map=(.*?)\]/ism",
- function ($match) use ($simple_html) {
- return str_replace($match[0], '<p class="map">' . Map::byCoordinates(str_replace('/', ' ', $match[1]), $simple_html) . '</p>', $match[0]);
- },
- $text
- );
- }
- if (strpos($text, '[map]') !== false) {
- $text = preg_replace("/\[map\]/", '<p class="map"></p>', $text);
- }
+ // This is actually executed in Item::prepareBody()
- // Check for headers
- $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
- $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
- $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
- $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
- $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
- $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
+ $nosmile = strpos($text, '[nosmile]') !== false;
+ $text = str_replace('[nosmile]', '', $text);
- // Check for paragraph
- $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
+ // Check for font change text
+ $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
- // Check for bold text
- $text = preg_replace("(\[b\](.*?)\[\/b\])ism", '<strong>$1</strong>', $text);
+ // Declare the format for [spoiler] layout
+ $SpoilerLayout = '<details class="spoiler"><summary>' . DI::l10n()->t('Click to open/close') . '</summary>$1</details>';
- // Check for Italics text
- $text = preg_replace("(\[i\](.*?)\[\/i\])ism", '<em>$1</em>', $text);
+ // Check for [spoiler] text
+ // handle nested quotes
+ $endlessloop = 0;
+ while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
+ $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", $SpoilerLayout, $text);
+ }
- // Check for Underline text
- $text = preg_replace("(\[u\](.*?)\[\/u\])ism", '<u>$1</u>', $text);
+ // Check for [spoiler=Title] text
- // Check for strike-through text
- $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '<s>$1</s>', $text);
+ // handle nested quotes
+ $endlessloop = 0;
+ while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
+ $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
+ '<details class="spoiler"><summary>$1</summary>$2</details>',
+ $text);
+ }
- // Check for over-line text
- $text = preg_replace("(\[o\](.*?)\[\/o\])ism", '<span class="overline">$1</span>', $text);
+ // Declare the format for [quote] layout
+ $QuoteLayout = '<blockquote>$1</blockquote>';
- // Check for colored text
- $text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])ism", "<span style=\"color: $1;\">$2</span>", $text);
+ // Check for [quote] text
+ // handle nested quotes
+ $endlessloop = 0;
+ while ((strpos($text, "[/quote]") !== false) && (strpos($text, "[quote]") !== false) && (++$endlessloop < 20)) {
+ $text = preg_replace("/\[quote\](.*?)\[\/quote\]/ism", "$QuoteLayout", $text);
+ }
- // Check for sized text
- // [size=50] --> font-size: 50px (with the unit).
- $text = preg_replace("(\[size=(\d*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1px; line-height: initial;\">$2</span>", $text);
- $text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])ism", "<span style=\"font-size: $1; line-height: initial;\">$2</span>", $text);
+ // Check for [quote=Author] text
- // Check for centered text
- $text = preg_replace("(\[center\](.*?)\[\/center\])ism", "<div style=\"text-align:center;\">$1</div>", $text);
+ $t_wrote = DI::l10n()->t('$1 wrote:');
- // Check for list text
- $text = str_replace("[*]", "<li>", $text);
+ // handle nested quotes
+ $endlessloop = 0;
+ while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
+ $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
+ "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
+ $text);
+ }
- // Check for style sheet commands
- $text = preg_replace_callback(
- "(\[style=(.*?)\](.*?)\[\/style\])ism",
- function ($match) {
- return "<span style=\"" . HTML::sanitizeCSS($match[1]) . ";\">" . $match[2] . "</span>";
- },
- $text
- );
- // Check for CSS classes
- $text = preg_replace_callback(
- "(\[class=(.*?)\](.*?)\[\/class\])ism",
- function ($match) {
- return "<span class=\"" . HTML::sanitizeCSS($match[1]) . "\">" . $match[2] . "</span>";
- },
- $text
- );
+ // [img=widthxheight]image source[/img]
+ $text = preg_replace_callback(
+ "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
+ function ($matches) use ($simple_html) {
+ if (strpos($matches[3], "data:image/") === 0) {
+ return $matches[0];
+ }
- // 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", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
- $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
- $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
- $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
- $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
- $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
- $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
- $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
- $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
- $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
- }
+ $matches[3] = self::proxyUrl($matches[3], $simple_html);
+ return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
+ },
+ $text
+ );
+
+ $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '<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("/\[img\=(.*?)\](.*?)\[\/img\]/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] . '" title="' . $matches[2] . '">';
+ },
+ $text);
+
+ // Images
+ // [img]pathtoimage[/img]
+ $text = preg_replace_callback(
+ "/\[img\](.*?)\[\/img\]/ism",
+ function ($matches) use ($simple_html) {
+ if (strpos($matches[1], "data:image/") === 0) {
+ return $matches[0];
+ }
- $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
- $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
- $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
- $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
+ $matches[1] = self::proxyUrl($matches[1], $simple_html);
+ return "[img]" . $matches[1] . "[/img]";
+ },
+ $text
+ );
- $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
- $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
+ $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
+ $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . DI::l10n()->t('Image/photo') . '" />', $text);
- $text = str_replace('[hr]', '<hr />', $text);
+ $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br/><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . DI::l10n()->t('Encrypted content') . '" title="' . DI::l10n()->t('Encrypted content') . '" /><br />', $text);
+ $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . DI::l10n()->t('Encrypted content') . '" title="' . '$1' . ' ' . DI::l10n()->t('Encrypted content') . '" /><br />', $text);
+ //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . DI::l10n()->t('Encrypted content') . '" title="' . '$1' . ' ' . DI::l10n()->t('Encrypted content') . '" /><br />', $Text);
- // This is actually executed in Item::prepareBody()
+ // Simplify "video" element
+ $text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
- $text = str_replace('[nosmile]', '', $text);
+ // Try to Oembed
+ if ($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("/\[audio\](.*?)\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
- // Check for font change text
- $text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
+ $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
+ $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
+ } else {
+ $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
+ '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
+ $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
+ '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
+ }
- // Declare the format for [code] layout
+ // html5 video and audio
- $CodeLayout = '<code>$1</code>';
- // Check for [code] text
- $text = preg_replace("/\[code\](.*?)\[\/code\]/ism", "$CodeLayout", $text);
- // Declare the format for [spoiler] layout
- $SpoilerLayout = '<blockquote class="spoiler">$1</blockquote>';
+ if ($try_oembed) {
+ $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
+ } else {
+ $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
+ }
- // Check for [spoiler] text
- // handle nested quotes
- $endlessloop = 0;
- while ((strpos($text, "[/spoiler]") !== false) && (strpos($text, "[spoiler]") !== false) && (++$endlessloop < 20)) {
- $text = preg_replace("/\[spoiler\](.*?)\[\/spoiler\]/ism", "$SpoilerLayout", $text);
- }
+ // Youtube extensions
+ if ($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);
+ }
- // Check for [spoiler=Author] 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);
- $t_wrote = L10n::t('$1 wrote:');
+ if ($try_oembed) {
+ $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '<iframe width="' . $a->videowidth . '" height="' . $a->videoheight . '" src="https://www.youtube.com/embed/$1" frameborder="0" ></iframe>', $text);
+ } else {
+ $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
+ '<a href="https://www.youtube.com/watch?v=$1" target="_blank" rel="noopener noreferrer">https://www.youtube.com/watch?v=$1</a>', $text);
+ }
- // handle nested quotes
- $endlessloop = 0;
- while ((strpos($text, "[/spoiler]")!== false) && (strpos($text, "[spoiler=") !== false) && (++$endlessloop < 20)) {
- $text = preg_replace("/\[spoiler=[\"\']*(.*?)[\"\']*\](.*?)\[\/spoiler\]/ism",
- "<br /><strong class=".'"spoiler"'.">" . $t_wrote . "</strong><blockquote class=".'"spoiler"'.">$2</blockquote>",
- $text);
- }
+ if ($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);
+ }
- // Declare the format for [quote] layout
- $QuoteLayout = '<blockquote>$1</blockquote>';
+ $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);
- // 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", $text);
- }
+ if ($try_oembed) {
+ $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '<iframe width="' . $a->videowidth . '" height="' . $a->videoheight . '" src="https://player.vimeo.com/video/$1" frameborder="0" ></iframe>', $text);
+ } else {
+ $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
+ '<a href="https://vimeo.com/$1" target="_blank" rel="noopener noreferrer">https://vimeo.com/$1</a>', $text);
+ }
- // Check for [quote=Author] text
+ // oembed tag
+ $text = OEmbed::BBCode2HTML($text);
- $t_wrote = L10n::t('$1 wrote:');
+ // Avoid triple linefeeds through oembed
+ $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
- // handle nested quotes
- $endlessloop = 0;
- while ((strpos($text, "[/quote]")!== false) && (strpos($text, "[quote=") !== false) && (++$endlessloop < 20)) {
- $text = preg_replace("/\[quote=[\"\']*(.*?)[\"\']*\](.*?)\[\/quote\]/ism",
- "<p><strong class=".'"author"'.">" . $t_wrote . "</strong></p><blockquote>$2</blockquote>",
- $text);
- }
+ // If we found an event earlier, strip out all the event code and replace with a reformatted version.
+ // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
+ // Summary (e.g. title) is required, earlier revisions only required description (in addition to
+ // start which is always required). Allow desc with a missing summary for compatibility.
+ if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
+ $sub = Event::getHTML($ev, $simple_html);
- // [img=widthxheight]image source[/img]
- $text = preg_replace_callback(
- "/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
- function ($matches) use ($simple_html) {
- if (strpos($matches[3], "data:image/") === 0) {
- return $matches[0];
+ $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/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);
}
- $matches[3] = self::proxyUrl($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 ($matches) use ($simple_html) {
- if (strpos($matches[1], "data:image/") === 0) {
- return $matches[0];
+ // Replace non graphical smilies for external posts
+ if (!$nosmile && !$for_plaintext) {
+ $text = Smilies::replace($text);
}
- $matches[1] = self::proxyUrl($matches[1], $simple_html);
- return "[img]" . $matches[1] . "[/img]";
- },
- $text
- );
-
- $text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
- $text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
-
- // Shared content
- $text = self::convertShare(
- $text,
- function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
- return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
- }
- );
+ if (!$for_plaintext && DI::config()->get('system', 'big_emojis') && ($simple_html != self::DIASPORA)) {
+ $conv = html_entity_decode(str_replace([' ', "\n", "\r"], '', $text));
+ // Emojis are always 4 byte Unicode characters
+ if (!empty($conv) && (strlen($conv) / mb_strlen($conv) == 4)) {
+ $text = '<span style="font-size: xx-large; line-height: initial;">' . $text . '</span>';
+ }
+ }
- $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . L10n::t('Encrypted content') . '" /><br />', $text);
- $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $text);
- //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $Text);
+ if (!$for_plaintext) {
+ if (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
+ $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
+ $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForActivityPubCallback', $text);
+ }
+ } else {
+ $text = preg_replace("(\[url\](.*?)\[\/url\])ism", " $1 ", $text);
+ $text = preg_replace_callback("&\[url=([^\[\]]*)\]\[img\](.*)\[\/img\]\[\/url\]&Usi", 'self::removePictureLinksCallback', $text);
+ }
- // Try to Oembed
- if ($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 = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
+
+ // Remove all hashtag addresses
+ if ($simple_html && !in_array($simple_html, [self::DIASPORA, self::OSTATUS, self::ACTIVITYPUB])) {
+ $text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
+ } elseif ($simple_html == self::DIASPORA) {
+ // The ! is converted to @ since Diaspora only understands the @
+ $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
+ '@<a href="$2">$3</a>',
+ $text);
+ } elseif (in_array($simple_html, [self::OSTATUS, self::ACTIVITYPUB])) {
+ $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
+ '$1<span class="vcard"><a href="$2" class="url u-url mention" title="$3"><span class="fn nickname mention">$3</span></a></span>',
+ $text);
+ } elseif (!$simple_html) {
+ $text = preg_replace("/([@!])\[url\=(.*?)\](.*?)\[\/url\]/ism",
+ '$1<a href="$2" class="userinfo mention" title="$3">$3</a>',
+ $text);
+ }
- $text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
- $text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
- } else {
- $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
- '<a href="$1" target="_blank">$1</a>', $text);
- $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
- '<a href="$1" target="_blank">$1</a>', $text);
- }
+ // Bookmarks in red - will be converted to bookmarks in friendica
+ $text = preg_replace("/#\^\[url\](.*?)\[\/url\]/ism", '[bookmark=$1]$1[/bookmark]', $text);
+ $text = preg_replace("/#\^\[url\=(.*?)\](.*?)\[\/url\]/ism", '[bookmark=$1]$2[/bookmark]', $text);
+ $text = preg_replace("/#\[url\=.*?\]\^\[\/url\]\[url\=(.*?)\](.*?)\[\/url\]/i",
+ "[bookmark=$1]$2[/bookmark]", $text);
- // html5 video and audio
+ if (in_array($simple_html, [self::API, self::OSTATUS, self::TWITTER])) {
+ $text = preg_replace_callback("/([^#@!])\[url\=([^\]]*)\](.*?)\[\/url\]/ism", "self::expandLinksCallback", $text);
+ //$Text = preg_replace("/[^#@!]\[url\=([^\]]*)\](.*?)\[\/url\]/ism", ' $2 [url]$1[/url]', $Text);
+ $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", ' $2 [url]$1[/url]',$text);
+ }
+ // Perform URL Search
+ if ($try_oembed) {
+ $text = preg_replace_callback("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $try_oembed_callback, $text);
+ }
- if ($try_oembed) {
- $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="' . $a->videowidth . '" height="' . $a->videoheight . '"><a href="$1">$1</a></iframe>', $text);
- } else {
- $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
- }
+ $text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", '[url=$1]$2[/url]', $text);
+
+ // Handle Diaspora posts
+ $text = preg_replace_callback(
+ "&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
+ function ($match) {
+ return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
+ }, $text
+ );
+
+ $text = preg_replace_callback(
+ "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
+ function ($match) {
+ return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
+ }, $text
+ );
+
+ // Server independent link to posts and comments
+ // See issue: https://github.com/diaspora/diaspora_federation/issues/75
+ $expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
+ $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
+
+ /* Tag conversion
+ * Supports:
+ * - #[url=<anything>]<term>[/url]
+ * - [url=<anything>]#<term>[/url]
+ */
+ $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) use ($simple_html) {
+ if ($simple_html == BBCode::ACTIVITYPUB) {
+ return '<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
+ . '" data-tag="' . XML::escape($matches[1]) . '" rel="tag ugc">#'
+ . XML::escape($matches[1]) . '</a>';
+ } else {
+ return '#<a href="' . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
+ . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
+ . XML::escape($matches[1]) . '</a>';
+ }
+ }, $text);
- // Youtube extensions
- if ($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);
- }
+ // We need no target="_blank" rel="noopener noreferrer" for local links
+ // convert links start with DI::baseUrl() as local link without the target="_blank" rel="noopener noreferrer" attribute
+ $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
+ $text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
+ $text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $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);
+ $text = preg_replace("/\[url\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$1</a>', $text);
+ $text = preg_replace("/\[url\=(.*?)\](.*?)\[\/url\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
- if ($try_oembed) {
- $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '<iframe width="' . $a->videowidth . '" height="' . $a->videoheight . '" src="https://www.youtube.com/embed/$1" frameborder="0" ></iframe>', $text);
- } else {
- $text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism",
- '<a href="https://www.youtube.com/watch?v=$1" target="_blank">https://www.youtube.com/watch?v=$1</a>', $text);
- }
+ // Red compatibility, though the link can't be authenticated on Friendica
+ $text = preg_replace("/\[zrl\=(.*?)\](.*?)\[\/zrl\]/ism", '<a href="$1" target="_blank" rel="noopener noreferrer">$2</a>', $text);
- if ($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]', $text);
+ // we may need to restrict this further if it picks up too many strays
+ // link acct:user@host to a webfinger profile redirector
- if ($try_oembed) {
- $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '<iframe width="' . $a->videowidth . '" height="' . $a->videoheight . '" src="https://player.vimeo.com/video/$1" frameborder="0" ></iframe>', $text);
- } else {
- $text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism",
- '<a href="https://vimeo.com/$1" target="_blank">https://vimeo.com/$1</a>', $text);
- }
+ $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', '<a href="' . DI::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
- // oembed tag
- $text = OEmbed::BBCode2HTML($text);
+ // Perform MAIL Search
+ $text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
+ $text = preg_replace("/\[mail\=(.*?)\](.*?)\[\/mail\]/", '<a href="mailto:$1">$2</a>', $text);
- // Avoid triple linefeeds through oembed
- $text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $text);
+ /// @todo What is the meaning of these lines?
+ $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
+ $text = preg_replace('/\&\#039\;/', '\'', $text);
- // If we found an event earlier, strip out all the event code and replace with a reformatted version.
- // Replace the event-start section with the entire formatted event. The other bbcode is stripped.
- // Summary (e.g. title) is required, earlier revisions only required description (in addition to
- // start which is always required). Allow desc with a missing summary for compatibility.
+ // Currently deactivated, it made problems with " inside of alt texts.
+ //$text = preg_replace('/\"\;/', '"', $text);
- if ((!empty($ev['desc']) || !empty($ev['summary'])) && !empty($ev['start'])) {
- $sub = Event::getHTML($ev, $simple_html);
+ // fix any escaped ampersands that may have been converted into links
+ $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
- $text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/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);
- }
+ // sanitizes src attributes (http and redir URLs for displaying in a web page, cid used for inline images in emails)
+ $allowed_src_protocols = ['//', 'http://', 'https://', 'redir/', 'cid:'];
- // Replace non graphical smilies for external posts
- if ($simple_html) {
- $text = Smilies::replace($text);
- }
+ array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
- // Unhide all [noparse] contained bbtags unspacefying them
- // and triming the [noparse] tag.
+ $text = preg_replace('#<([^>]*?)(src)="(?!' . implode('|', $allowed_src_protocols) . ')(.*?)"(.*?)>#ism',
+ '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $text);
- $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);
+ // sanitize href attributes (only allowlisted protocols URLs)
+ // default value for backward compatibility
+ $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
- /// @todo What is the meaning of these lines?
- $text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
- $text = preg_replace('/\&\#039\;/', '\'', $text);
+ // Always allowed protocol even if config isn't set or not including it
+ $allowed_link_protocols[] = '//';
+ $allowed_link_protocols[] = 'http://';
+ $allowed_link_protocols[] = 'https://';
+ $allowed_link_protocols[] = 'redir/';
- // Currently deactivated, it made problems with " inside of alt texts.
- //$text = preg_replace('/\"\;/', '"', $text);
+ array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
- // fix any escaped ampersands that may have been converted into links
- $text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
+ $regex = '#<([^>]*?)(href)="(?!' . implode('|', $allowed_link_protocols) . ')(.*?)"(.*?)>#ism';
+ $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
- // 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);
+ // Shared content
+ $text = self::convertShare(
+ $text,
+ function (array $attributes, array $author_contact, $content, $is_quote_share) use ($simple_html) {
+ return self::convertShareCallback($attributes, $author_contact, $content, $is_quote_share, $simple_html);
+ }
+ );
- // 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']);
+ $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
- // Always allowed protocol even if config isn't set or not including it
- $allowed_link_protocols[] = 'http';
- $allowed_link_protocols[] = 'redir/';
+ return $text;
+ }); // Escaped noparse, nobb, pre
- $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);
+ // Remove escaping tags
+ $text = preg_replace("/\[noparse\](.*?)\[\/noparse\]/ism", '\1', $text);
+ $text = preg_replace("/\[nobb\](.*?)\[\/nobb\]/ism", '\1', $text);
+ $text = preg_replace("/\[pre\](.*?)\[\/pre\]/ism", '\1', $text);
- if ($saved_image) {
- $text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
- }
+ return $text;
+ }); // Escaped code
- // Restore code blocks
- $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
- function ($matches) use ($codeblocks) {
- $return = $matches[0];
- if (isset($codeblocks[intval($matches[1])])) {
- $return = $codeblocks[$matches[1]];
+ $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#ism",
+ function ($matches) {
+ if (strpos($matches[2], "\n") !== false) {
+ $return = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
+ } else {
+ $return = '<code>' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
}
+
return $return;
},
$text
// Clean up the HTML by loading and saving the HTML with the DOM.
// Bad structured html can break a whole page.
// For performance reasons do it only with activated item cache or at export.
- if (!$try_oembed || (get_itemcachepath() != "")) {
+ if (!$try_oembed || (get_itemcachepath() != '')) {
$doc = new DOMDocument();
$doc->preserveWhiteSpace = false;
$doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
$encoding = '<?xml encoding="UTF-8">';
- @$doc->loadHTML($encoding.$doctype."<html><body>".$text."</body></html>");
+ @$doc->loadHTML($encoding . $doctype . '<html><body>' . $text . '</body></html>');
$doc->encoding = 'UTF-8';
$text = $doc->saveHTML();
- $text = str_replace(["<html><body>", "</body></html>", $doctype, $encoding], ["", "", "", ""], $text);
+ $text = str_replace(['<html><body>', '</body></html>', $doctype, $encoding], ['', '', '', ''], $text);
$text = str_replace('<br></li>', '</li>', $text);
}
/**
- * @brief Strips the "abstract" tag from the provided text
+ * Strips the "abstract" tag from the provided text
*
* @param string $text The text with BBCode
* @return string The same text - but without "abstract" element
}
/**
- * @brief Returns the value of the "abstract" element
+ * Returns the value of the "abstract" element
*
* @param string $text The text that maybe contains the element
* @param string $addon The addon for which the abstract is meant for
* @return string The abstract
*/
- public static function getAbstract($text, $addon = "")
+ public static function getAbstract($text, $addon = '')
{
- $abstract = "";
+ $abstract = '';
$abstracts = [];
$addon = strtolower($addon);
$abstract = $abstracts[$addon];
}
- if ($abstract == "" && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
+ if ($abstract == '' && preg_match("/\[abstract\](.*?)\[\/abstract\]/ism", $text, $result)) {
$abstract = $result[1];
}
}
/**
- * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
+ * Callback function to replace a Friendica style mention in a mention for Diaspora
*
* @param array $match Matching values for the callback
* [1] = Mention type (! or @)
}
/**
- * @brief Converts a BBCode text into Markdown
+ * Converts a BBCode text into Markdown
*
* This function converts a BBCode item body to be sent to Markdown-enabled
* systems like Diaspora and Libertree
*/
public static function toMarkdown($text, $for_diaspora = true)
{
- $a = self::getApp();
-
$original_text = $text;
// Since Diaspora is creating a summary for links, this function removes them before posting
// Convert it to HTML - don't try oembed
if ($for_diaspora) {
- $text = self::convert($text, false, 3);
+ $text = self::convert($text, false, self::DIASPORA);
// Add all tags that maybe were removed
if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
- $tagline = "";
+ $tagline = '';
foreach ($tags[2] as $tag) {
$tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
$text = $text . " " . $tagline;
}
} else {
- $text = self::convert($text, false, 4);
+ $text = self::convert($text, false, self::CONNECTORS);
}
- // mask some special HTML chars from conversation to markdown
- $text = str_replace(['<', '>', '&'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text);
-
// If a link is followed by a quote then there should be a newline before it
// Maybe we should make this newline at every time before a quote.
- $text = str_replace(["</a><blockquote>"], ["</a><br><blockquote>"], $text);
+ $text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
$stamp1 = microtime(true);
// Now convert HTML to Markdown
$text = HTML::toMarkdown($text);
- // unmask the special chars back to HTML
- $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
-
- $a->getProfiler()->saveTimestamp($stamp1, "parser", System::callstack());
+ DI::profiler()->saveTimestamp($stamp1, "parser", System::callstack());
// Libertree has a problem with escaped hashtags.
$text = str_replace(['\#'], ['#'], $text);
}
/**
- * @brief Pull out all #hashtags and @person tags from $string.
- *
- * We also get @person@domain.com - which would make
- * the regex quite complicated as tags can also
- * end a sentence. So we'll run through our results
- * and strip the period from any tags which end with one.
- * Returns array of tags found, or empty array.
- *
- * @param string $string Post content
- *
- * @return array List of tag and person names
- */
- public static function getTags($string)
- {
- $ret = [];
-
- // Convert hashtag links to hashtags
- $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2', $string);
-
- // ignore anything in a code block
- $string = preg_replace('/\[code.*?\].*?\[\/code\]/sm', '', $string);
-
- // Force line feeds at bbtags
- $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
-
- // ignore anything in a bbtag
- $string = preg_replace('/\[(.*?)\]/sm', '', $string);
-
- // Match full names against @tags including the space between first and last
- // We will look these up afterward to see if they are full names or not recognisable.
-
- if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
- foreach ($matches[1] as $match) {
- if (strstr($match, ']')) {
- // we might be inside a bbcode color tag - leave it alone
- continue;
- }
-
- if (substr($match, -1, 1) === '.') {
- $ret[] = substr($match, 0, -1);
- } else {
- $ret[] = $match;
- }
- }
- }
-
- // Otherwise pull out single word tags. These can be @nickname, @first_last
- // and #hash tags.
-
- if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
- foreach ($matches[1] as $match) {
- if (strstr($match, ']')) {
- // we might be inside a bbcode color tag - leave it alone
- continue;
- }
- if (substr($match, -1, 1) === '.') {
- $match = substr($match,0,-1);
- }
- // ignore strictly numeric tags like #1
- if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
- continue;
- }
- // try not to catch url fragments
- if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
- continue;
- }
- $ret[] = $match;
- }
- }
-
- return $ret;
- }
+ * Pull out all #hashtags and @person tags from $string.
+ *
+ * We also get @person@domain.com - which would make
+ * the regex quite complicated as tags can also
+ * end a sentence. So we'll run through our results
+ * and strip the period from any tags which end with one.
+ * Returns array of tags found, or empty array.
+ *
+ * @param string $string Post content
+ *
+ * @return array List of tag and person names
+ */
+ public static function getTags($string)
+ {
+ $ret = [];
+
+ // Convert hashtag links to hashtags
+ $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2 ', $string);
+
+ // ignore anything in a code block
+ $string = preg_replace('/\[code.*?\].*?\[\/code\]/sm', '', $string);
+
+ // Force line feeds at bbtags
+ $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
+
+ // ignore anything in a bbtag
+ $string = preg_replace('/\[(.*?)\]/sm', '', $string);
+
+ // Match full names against @tags including the space between first and last
+ // We will look these up afterward to see if they are full names or not recognisable.
+
+ if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
+ foreach ($matches[1] as $match) {
+ if (strstr($match, ']')) {
+ // we might be inside a bbcode color tag - leave it alone
+ continue;
+ }
+
+ if (substr($match, -1, 1) === '.') {
+ $ret[] = substr($match, 0, -1);
+ } else {
+ $ret[] = $match;
+ }
+ }
+ }
+
+ // Otherwise pull out single word tags. These can be @nickname, @first_last
+ // and #hash tags.
+
+ if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
+ foreach ($matches[1] as $match) {
+ if (strstr($match, ']')) {
+ // we might be inside a bbcode color tag - leave it alone
+ continue;
+ }
+
+ if (substr($match, -1, 1) === '.') {
+ $match = substr($match,0,-1);
+ }
+
+ // ignore strictly numeric tags like #1
+ if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
+ continue;
+ }
+
+ // try not to catch url fragments
+ if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
+ continue;
+ }
+ $ret[] = $match;
+ }
+ }
+
+ return $ret;
+ }
+
+ /**
+ * Perform a custom function on a text after having escaped blocks enclosed in the provided tag list.
+ *
+ * @param string $text
+ * @param array $tagList A list of tag names, e.g ['noparse', 'nobb', 'pre']
+ * @param callable $callback
+ * @return string
+ * @throws Exception
+ *@see Strings::performWithEscapedBlocks
+ *
+ */
+ public static function performWithEscapedTags(string $text, array $tagList, callable $callback)
+ {
+ $tagList = array_map('preg_quote', $tagList);
+
+ return Strings::performWithEscapedBlocks($text, '#\[(?:' . implode('|', $tagList) . ').*?\[/(?:' . implode('|', $tagList) . ')]#ism', $callback);
+ }
}