<?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\Util\Strings;
use Friendica\Util\XML;
-class BBCode extends BaseObject
+class BBCode
{
/**
- * @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
}
/**
- * @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
$body = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
$post['text'] = $body;
- if (preg_match_all("(\[url=(.*?)\]\s*\[img\](.*?)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
+ if (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $body, $pictures, PREG_SET_ORDER)) {
if ((count($pictures) == 1) && !$has_title) {
if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) {
// Replace the preview picture with the real picture
}
/**
- * @brief Converts a BBCode text into plaintext
+ * Converts a BBCode text into plaintext
*
* @param $text
* @param bool $keep_urls Whether to keep URLs in the resulting plaintext
}
}
- public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
+ public static function scaleExternalImages($srctext)
{
- // Suppress "view full size"
- if (intval(Config::get('system', 'no_view_full_size'))) {
- $include_link = false;
- }
-
- // Picture addresses can contain special characters
- $s = htmlspecialchars_decode($srctext);
+ $s = $srctext;
$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);
+ $i = Network::fetchUrl($mtch[1]);
if (!$i) {
return $srctext;
}
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 $text
* @param bool|int $simplehtml
* @param bool $tryoembed
$data['title'] = null;
}
- if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data['image'])) {
+ if (((strpos($data['text'], "[img=") !== false) || (strpos($data['text'], "[img]") !== false) || DI::config()->get('system', 'always_show_preview')) && !empty($data['image'])) {
$data['preview'] = $data['image'];
$data['image'] = '';
}
/**
* 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
*/
}
/**
- * @brief Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
+ * Converts [url] BBCodes in a format that looks fine on ActivityPub systems.
+ *
* @param string $url URL that is about to be reformatted
* @return string reformatted link including HTML codes
*/
/**
* 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
*/
// 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++;
}
break;
case 4:
$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">%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";
private static function removePictureLinksCallback($match)
{
$cache_key = 'remove:' . $match[1];
- $text = Cache::get($cache_key);
+ $text = DI::cache()->get($cache_key);
if (is_null($text)) {
- $a = 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] . ']' . $match[1] . "[/url]";
}
}
}
- Cache::set($cache_key, $text);
+ DI::cache()->set($cache_key, $text);
}
return $text;
private static function cleanPictureLinksCallback($match)
{
- $a = self::getApp();
+ $a = DI::app();
// When the picture link is the own photo path then we can avoid fetching the link
- $own_photo_url = preg_quote(Strings::normaliseLink($a->getBaseURL()) . '/photos/');
+ $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
if (!empty($match[3])) {
$text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
}
$cache_key = 'clean:' . $match[1];
- $text = Cache::get($cache_key);
+ $text = DI::cache()->get($cache_key);
if (!is_null($text)) {
return $text;
}
@curl_exec($ch);
$curl_info = @curl_getinfo($ch);
- $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack());
+ 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/') {
}
}
}
- Cache::set($cache_key, $text);
+ DI::cache()->set($cache_key, $text);
return $text;
}
}
/**
- * @brief Converts a BBCode message to HTML message
+ * Converts a BBCode message to HTML message
*
* BBcode 2 HTML was written by WAY2WEB.net
* extended to work with Mistpark/Friendica - Mike Macgirvin
*/
public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false)
{
- $a = self::getApp();
+ $a = DI::app();
/*
* preg_match_callback function to replace potential Oembed tags with Oembed content
function ($matches) use (&$codeblocks) {
$return = '#codeblock-' . count($codeblocks) . '#';
if (strpos($matches[2], "\n") !== false) {
- $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . trim($matches[2], "\n\r") . '</code></pre>';
+ $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
} else {
- $codeblocks[] = '<code>' . $matches[2] . '</code>';
+ $codeblocks[] = '<code>' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
}
return $return;
$text = str_replace($search, $replace, $text);
// removing multiplicated newlines
- if (Config::get('system', 'remove_multiplicated_lines')) {
+ 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]",
$text = preg_replace("/\[font=(.*?)\](.*?)\[\/font\]/sm", "<span style=\"font-family: $1;\">$2</span>", $text);
// Declare the format for [spoiler] layout
- $SpoilerLayout = '<details class="spoiler"><summary>' . L10n::t('Click to open/close') . '</summary>$1</details>';
+ $SpoilerLayout = '<details class="spoiler"><summary>' . DI::l10n()->t('Click to open/close') . '</summary>$1</details>';
// Check for [spoiler] text
// handle nested quotes
// Check for [quote=Author] text
- $t_wrote = L10n::t('$1 wrote:');
+ $t_wrote = DI::l10n()->t('$1 wrote:');
// handle nested quotes
$endlessloop = 0;
$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);
+ $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 = 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);
+ $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);
// Simplify "video" element
$text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
$text = preg_replace_callback(
"&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
function ($match) {
- return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
+ return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
}, $text
);
$text = preg_replace_callback(
"&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
function ($match) {
- return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
+ 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, System::baseUrl()."/display/$1", $text);
+ $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
/* Tag conversion
* Supports:
*/
$text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) {
return '#<a href="'
- . System::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
+ . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
. '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
. XML::escape($matches[1])
. '</a>';
}, $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(), '/');
+ // convert links start with DI::baseUrl() as local link without the target="_blank" 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);
// we may need to restrict this further if it picks up too many strays
// link acct:user@host to a webfinger profile redirector
- $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', '<a href="' . System::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</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);
// Perform MAIL Search
$text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
array_walk($allowed_src_protocols, function(&$value) { $value = preg_quote($value, '#');});
$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);
+ '<$1$2=""$4 data-original-src="$3" class="invalid-src" title="' . DI::l10n()->t('Invalid source protocol') . '">', $text);
// sanitize href attributes (only whitelisted protocols URLs)
// default value for backward compatibility
- $allowed_link_protocols = Config::get('system', 'allowed_link_protocols', []);
+ $allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
// Always allowed protocol even if config isn't set or not including it
$allowed_link_protocols[] = '//';
array_walk($allowed_link_protocols, function(&$value) { $value = preg_quote($value, '#');});
$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);
+ $text = preg_replace($regex, '<$1$2="javascript:void(0)"$4 data-original-href="$3" class="invalid-href" title="' . DI::l10n()->t('Invalid link protocol') . '">', $text);
// Shared content
$text = self::convertShare(
}
/**
- * @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
}
/**
- * @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();
+ $a = DI::app();
$original_text = $text;
// Now convert HTML to Markdown
$text = HTML::toMarkdown($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;
+ }
}