3 * @copyright Copyright (C) 2010-2022, the Friendica project
5 * @license GNU AGPL version 3 or any later version
7 * This program is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU Affero General Public License as
9 * published by the Free Software Foundation, either version 3 of the
10 * License, or (at your option) any later version.
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU Affero General Public License for more details.
17 * You should have received a copy of the GNU Affero General Public License
18 * along with this program. If not, see <https://www.gnu.org/licenses/>.
22 namespace Friendica\Content\Text;
26 use Friendica\Content\Widget\ContactBlock;
27 use Friendica\Core\Hook;
28 use Friendica\Core\Renderer;
29 use Friendica\Core\Search;
31 use Friendica\Model\Contact;
32 use Friendica\Util\Network;
33 use Friendica\Util\Strings;
34 use Friendica\Util\XML;
35 use League\HTMLToMarkdown\HtmlConverter;
36 use Psr\Http\Message\UriInterface;
41 * Search all instances of a specific HTML tag node in the provided DOM document and replaces them with BBCode text nodes.
43 * @see HTML::tagToBBCodeSub()
45 private static function tagToBBCode(DOMDocument $doc, string $tag, array $attributes, string $startbb, string $endbb, bool $ignoreChildren = false)
48 $done = self::tagToBBCodeSub($doc, $tag, $attributes, $startbb, $endbb, $ignoreChildren);
53 * Search the first specific HTML tag node in the provided DOM document and replaces it with BBCode text nodes.
55 * @param DOMDocument $doc
56 * @param string $tag HTML tag name
57 * @param array $attributes Array of attributes to match and optionally use the value from
58 * @param string $startbb BBCode tag opening
59 * @param string $endbb BBCode tag closing
60 * @param bool $ignoreChildren If set to false, the HTML tag children will be appended as text inside the BBCode tag
61 * Otherwise, they will be entirely ignored. Useful for simple BBCode that draw their
62 * inner value from an attribute value and disregard the tag children.
63 * @return bool Whether a replacement was done
65 private static function tagToBBCodeSub(DOMDocument $doc, string $tag, array $attributes, string $startbb, string $endbb, bool $ignoreChildren = false): bool
67 $savestart = str_replace('$', '\x01', $startbb);
70 $xpath = new DOMXPath($doc);
72 /** @var \DOMNode[] $list */
73 $list = $xpath->query("//" . $tag);
74 foreach ($list as $node) {
76 if ($node->attributes->length) {
77 foreach ($node->attributes as $attribute) {
78 $attr[$attribute->name] = $attribute->value;
84 $startbb = $savestart;
88 foreach ($attributes as $attribute => $value) {
89 $startbb = str_replace('\x01' . ++$i, '$1', $startbb);
90 if (strpos('*' . $startbb, '$1') > 0) {
91 if ($replace && (@$attr[$attribute] != '')) {
92 $startbb = preg_replace($value, $startbb, $attr[$attribute], -1, $count);
94 // If nothing could be changed
102 if (@$attr[$attribute] != $value) {
109 $StartCode = $doc->createTextNode($startbb);
110 $EndCode = $doc->createTextNode($endbb);
112 $node->parentNode->insertBefore($StartCode, $node);
114 if (!$ignoreChildren && $node->hasChildNodes()) {
115 /** @var \DOMNode $child */
116 foreach ($node->childNodes as $key => $child) {
117 /* Remove empty text nodes at the start or at the end of the children list */
118 if ($key > 0 && $key < $node->childNodes->length - 1 || $child->nodeName != '#text' || trim($child->nodeValue) !== '') {
119 $newNode = $child->cloneNode(true);
120 $node->parentNode->insertBefore($newNode, $node);
125 $node->parentNode->insertBefore($EndCode, $node);
126 $node->parentNode->removeChild($node);
134 * Converter for HTML to BBCode
136 * Made by: ike@piratenpartei.de
137 * Originally made for the syncom project: http://wiki.piratenpartei.de/Syncom
138 * https://github.com/annando/Syncom
140 * @param string $message
141 * @param string $basepath
143 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
145 public static function toBBCode(string $message, string $basepath = ''): string
148 * Check if message is empty to prevent a lot of code below from being executed
149 * for just an empty message.
151 if ($message === '') {
155 DI::profiler()->startRecording('rendering');
156 $message = str_replace("\r", "", $message);
158 $message = Strings::performWithEscapedBlocks($message, '#<pre><code.*</code></pre>#iUs', function ($message) {
159 $message = str_replace(
172 $message = preg_replace('=<(\w+):(.+?)>=', '<removeme>', $message);
173 $message = preg_replace('=</(\w+):(.+?)>=', '</removeme>', $message);
175 $doc = new DOMDocument();
176 $doc->preserveWhiteSpace = false;
178 $message = mb_convert_encoding($message, 'HTML-ENTITIES', "UTF-8");
180 if (empty($message)) {
184 @$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
186 XML::deleteNode($doc, 'style');
187 XML::deleteNode($doc, 'head');
188 XML::deleteNode($doc, 'title');
189 XML::deleteNode($doc, 'meta');
190 XML::deleteNode($doc, 'xml');
191 XML::deleteNode($doc, 'removeme');
193 $xpath = new DomXPath($doc);
194 $list = $xpath->query("//pre");
195 foreach ($list as $node) {
196 // Ensure to escape unescaped & - they will otherwise raise a warning
197 $safe_value = preg_replace('/&(?!\w+;)/', '&', $node->nodeValue);
198 $node->nodeValue = str_replace("\n", "\r", $safe_value);
201 $message = $doc->saveHTML();
202 $message = str_replace(["\n<", ">\n", "\r", "\n", "\xC3\x82\xC2\xA0"], ["<", ">", "<br />", " ", ""], $message);
203 $message = preg_replace('= [\s]*=i', " ", $message);
205 if (empty($message)) {
209 @$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
211 self::tagToBBCode($doc, 'html', [], "", "");
212 self::tagToBBCode($doc, 'body', [], "", "");
214 // Outlook-Quote - Variant 1
215 self::tagToBBCode($doc, 'p', ['class' => 'MsoNormal', 'style' => 'margin-left:35.4pt'], '[quote]', '[/quote]');
217 // Outlook-Quote - Variant 2
221 ['style' => 'border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt'],
227 self::tagToBBCode($doc, 'span', ['style' => 'text-decoration: underline;'], '[u]', '[/u]');
228 self::tagToBBCode($doc, 'span', ['style' => 'font-style: italic;'], '[i]', '[/i]');
229 self::tagToBBCode($doc, 'span', ['style' => 'font-weight: bold;'], '[b]', '[/b]');
231 /* self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/', 'size'=>'/(\d+)/', 'color'=>'/(.+)/'), '[font=$1][size=$2][color=$3]', '[/color][/size][/font]');
232 self::node2BBCode($doc, 'font', array('size'=>'/(\d+)/', 'color'=>'/(.+)/'), '[size=$1][color=$2]', '[/color][/size]');
233 self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/', 'size'=>'/(.+)/'), '[font=$1][size=$2]', '[/size][/font]');
234 self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/', 'color'=>'/(.+)/'), '[font=$1][color=$3]', '[/color][/font]');
235 self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/'), '[font=$1]', '[/font]');
236 self::node2BBCode($doc, 'font', array('size'=>'/(\d+)/'), '[size=$1]', '[/size]');
237 self::node2BBCode($doc, 'font', array('color'=>'/(.+)/'), '[color=$1]', '[/color]');
240 //self::node2BBCode($doc, 'span', array('style'=>'/.*font-size:\s*(.+?)[,;].*font-family:\s*(.+?)[,;].*color:\s*(.+?)[,;].*/'), '[size=$1][font=$2][color=$3]', '[/color][/font][/size]');
241 //self::node2BBCode($doc, 'span', array('style'=>'/.*font-size:\s*(\d+)[,;].*/'), '[size=$1]', '[/size]');
242 //self::node2BBCode($doc, 'span', array('style'=>'/.*font-size:\s*(.+?)[,;].*/'), '[size=$1]', '[/size]');
244 self::tagToBBCode($doc, 'span', ['style' => '/.*color:\s*(.+?)[,;].*/'], '[color="$1"]', '[/color]');
246 //self::node2BBCode($doc, 'span', array('style'=>'/.*font-family:\s*(.+?)[,;].*/'), '[font=$1]', '[/font]');
247 //self::node2BBCode($doc, 'div', array('style'=>'/.*font-family:\s*(.+?)[,;].*font-size:\s*(\d+?)pt.*/'), '[font=$1][size=$2]', '[/size][/font]');
248 //self::node2BBCode($doc, 'div', array('style'=>'/.*font-family:\s*(.+?)[,;].*font-size:\s*(\d+?)px.*/'), '[font=$1][size=$2]', '[/size][/font]');
249 //self::node2BBCode($doc, 'div', array('style'=>'/.*font-family:\s*(.+?)[,;].*/'), '[font=$1]', '[/font]');
250 // Importing the classes - interesting for importing of posts from third party networks that were exported from friendica
252 //self::node2BBCode($doc, 'span', array('class'=>'/([\w ]+)/'), '[class=$1]', '[/class]');
253 self::tagToBBCode($doc, 'span', ['class' => 'type-link'], '[class=type-link]', '[/class]');
254 self::tagToBBCode($doc, 'span', ['class' => 'type-video'], '[class=type-video]', '[/class]');
256 self::tagToBBCode($doc, 'strong', [], '[b]', '[/b]');
257 self::tagToBBCode($doc, 'em', [], '[i]', '[/i]');
258 self::tagToBBCode($doc, 'b', [], '[b]', '[/b]');
259 self::tagToBBCode($doc, 'i', [], '[i]', '[/i]');
260 self::tagToBBCode($doc, 'u', [], '[u]', '[/u]');
261 self::tagToBBCode($doc, 's', [], '[s]', '[/s]');
262 self::tagToBBCode($doc, 'del', [], '[s]', '[/s]');
263 self::tagToBBCode($doc, 'strike', [], '[s]', '[/s]');
265 self::tagToBBCode($doc, 'big', [], "[size=large]", "[/size]");
266 self::tagToBBCode($doc, 'small', [], "[size=small]", "[/size]");
268 self::tagToBBCode($doc, 'blockquote', [], '[quote]', '[/quote]');
270 self::tagToBBCode($doc, 'br', [], "\n", '');
272 self::tagToBBCode($doc, 'p', ['class' => 'MsoNormal'], "\n", "");
273 self::tagToBBCode($doc, 'div', ['class' => 'MsoNormal'], "\r", "");
275 self::tagToBBCode($doc, 'span', [], "", "");
277 self::tagToBBCode($doc, 'span', [], "", "");
278 self::tagToBBCode($doc, 'pre', [], "", "");
280 self::tagToBBCode($doc, 'div', [], "\r", "\r");
281 self::tagToBBCode($doc, 'p', [], "\n", "\n");
283 self::tagToBBCode($doc, 'ul', [], "[list]", "[/list]");
284 self::tagToBBCode($doc, 'ol', [], "[list=1]", "[/list]");
285 self::tagToBBCode($doc, 'li', [], "[*]", "");
287 self::tagToBBCode($doc, 'hr', [], "[hr]", "");
289 self::tagToBBCode($doc, 'table', [], "[table]", "[/table]");
290 self::tagToBBCode($doc, 'th', [], "[th]", "[/th]");
291 self::tagToBBCode($doc, 'tr', [], "[tr]", "[/tr]");
292 self::tagToBBCode($doc, 'td', [], "[td]", "[/td]");
294 self::tagToBBCode($doc, 'h1', [], "[h1]", "[/h1]");
295 self::tagToBBCode($doc, 'h2', [], "[h2]", "[/h2]");
296 self::tagToBBCode($doc, 'h3', [], "[h3]", "[/h3]");
297 self::tagToBBCode($doc, 'h4', [], "[h4]", "[/h4]");
298 self::tagToBBCode($doc, 'h5', [], "[h5]", "[/h5]");
299 self::tagToBBCode($doc, 'h6', [], "[h6]", "[/h6]");
301 self::tagToBBCode($doc, 'a', ['href' => '/mailto:(.+)/'], '[mail=$1]', '[/mail]');
302 self::tagToBBCode($doc, 'a', ['href' => '/(.+)/'], '[url=$1]', '[/url]');
304 self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'alt' => '/(.+)/'], '[img=$1]$2', '[/img]', true);
305 self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'width' => '/(\d+)/', 'height' => '/(\d+)/'], '[img=$2x$3]$1', '[/img]', true);
306 self::tagToBBCode($doc, 'img', ['src' => '/(.+)/'], '[img]$1', '[/img]', true);
309 self::tagToBBCode($doc, 'video', ['src' => '/(.+)/'], '[video]$1', '[/video]', true);
310 self::tagToBBCode($doc, 'audio', ['src' => '/(.+)/'], '[audio]$1', '[/audio]', true);
311 // Backward compatibility, [iframe] support has been removed in version 2020.12
312 self::tagToBBCode($doc, 'iframe', ['src' => '/(.+)/'], '[url]$1', '[/url]', true);
314 self::tagToBBCode($doc, 'key', [], '[code]', '[/code]');
315 self::tagToBBCode($doc, 'code', [], '[code]', '[/code]');
317 $message = $doc->saveHTML();
319 // I'm removing something really disturbing
320 // Don't know exactly what it is
321 $message = str_replace(chr(194) . chr(160), ' ', $message);
323 $message = str_replace(" ", " ", $message);
325 // removing multiple DIVs
326 $message = preg_replace('=\r *\r=i', "\n", $message);
327 $message = str_replace("\r", "\n", $message);
329 Hook::callAll('html2bbcode', $message);
331 $message = strip_tags($message);
333 $message = html_entity_decode($message, ENT_QUOTES, 'UTF-8');
335 // remove quotes if they don't make sense
336 $message = preg_replace('=\[/quote\][\s]*\[quote\]=i', "\n", $message);
338 $message = preg_replace('=\[quote\]\s*=i', "[quote]", $message);
339 $message = preg_replace('=\s*\[/quote\]=i', "[/quote]", $message);
342 $oldmessage = $message;
343 $message = str_replace("\n \n", "\n\n", $message);
344 } while ($oldmessage != $message);
347 $oldmessage = $message;
348 $message = str_replace("\n\n\n", "\n\n", $message);
349 } while ($oldmessage != $message);
352 $oldmessage = $message;
353 $message = str_replace(
376 } while ($message != $oldmessage);
378 $message = str_replace(
379 ['[b][b]', '[/b][/b]', '[i][i]', '[/i][/i]'],
380 ['[b]', '[/b]', '[i]', '[/i]'],
384 // Handling Yahoo style of mails
385 $message = str_replace('[hr][b]From:[/b]', '[quote][b]From:[/b]', $message);
390 $message = preg_replace_callback(
391 '#<pre><code(?: class="language-([^"]*)")?>(.*)</code></pre>#iUs',
392 function ($matches) {
394 if ($matches[1] != '') {
395 $prefix = '[code=' . $matches[1] . ']';
398 return $prefix . "\n" . html_entity_decode($matches[2]) . "\n" . '[/code]';
403 $message = trim($message);
405 if ($basepath != '') {
406 $message = self::qualifyURLs($message, $basepath);
409 DI::profiler()->stopRecording();
414 * Sub function to complete incomplete URL
416 * @param array $matches Result of preg_replace_callback
417 * @param string $basepath Basepath that is used to complete the URL
419 * @return string The expanded URL
421 private static function qualifyURLsSub(array $matches, string $basepath): string
423 $base = parse_url($basepath);
424 unset($base['query']);
425 unset($base['fragment']);
430 if (empty($url) || empty(parse_url($url))) {
434 $parts = array_merge($base, parse_url($url));
435 $url2 = Network::unparseURL($parts);
437 return str_replace($url, $url2, $link);
441 * Complete incomplete URLs in BBCode
443 * @param string $body Body with URLs
444 * @param string $basepath Base path that is used to complete the URL
446 * @return string Body with expanded URLs
448 private static function qualifyURLs(string $body, string $basepath): string
450 $URLSearchString = "^\[\]";
452 $matches = ["/\[url\=([$URLSearchString]*)\].*?\[\/url\]/ism",
453 "/\[url\]([$URLSearchString]*)\[\/url\]/ism",
454 "/\[img\=[0-9]*x[0-9]*\](.*?)\[\/img\]/ism",
455 "/\[img\](.*?)\[\/img\]/ism",
456 "/\[zmg\=[0-9]*x[0-9]*\](.*?)\[\/img\]/ism",
457 "/\[zmg\](.*?)\[\/zmg\]/ism",
458 "/\[video\](.*?)\[\/video\]/ism",
459 "/\[audio\](.*?)\[\/audio\]/ism",
462 foreach ($matches as $match) {
463 $body = preg_replace_callback(
465 function ($match) use ($basepath) {
466 return self::qualifyURLsSub($match, $basepath);
474 private static function breakLines(string $line, int $level, int $wraplength = 75): string
476 if ($wraplength == 0) {
477 $wraplength = 2000000;
480 $wraplen = $wraplength - $level;
487 $subline = substr($line, 0, $wraplen);
489 $pos = strrpos($subline, ' ');
492 $pos = strpos($line, ' ');
495 if (($pos > 0) && strlen($line) > $wraplen) {
496 $newline = trim(substr($line, 0, $pos));
498 $newline = str_repeat(">", $level) . ' ' . $newline;
501 $newlines[] = $newline . " ";
502 $line = substr($line, $pos + 1);
504 } while ((strlen($line) > $wraplen) && !($oldline == $line));
507 $line = str_repeat(">", $level) . ' ' . $line;
512 return implode("\n", $newlines);
515 private static function quoteLevel(string $message, int $wraplength = 75): string
517 $lines = explode("\n", $message);
521 foreach ($lines as $line) {
524 while (strpos("*" . $line, '[quote]') > 0) {
526 $pos = strpos($line, '[quote]');
527 $line = substr($line, 0, $pos) . substr($line, $pos + 7);
533 while (strpos("*" . $line, '[/quote]') > 0) {
539 $pos = strpos($line, '[/quote]');
540 $line = substr($line, 0, $pos) . substr($line, $pos + 8);
543 if (!$startquote || ($line != '')) {
544 $newlines[] = self::breakLines($line, $currlevel, $wraplength);
548 return implode("\n", $newlines);
551 private static function collectURLs(string $message): array
553 $pattern = '/<a.*?href="(.*?)".*?>(.*?)<\/a>/is';
554 preg_match_all($pattern, $message, $result, PREG_SET_ORDER);
557 foreach ($result as $treffer) {
560 // A list of some links that should be ignored
561 $list = ["/user/", "/tag/", "/group/", "/profile/", "/search?search=", "/search?tag=", "mailto:", "/u/", "/node/",
562 "//plus.google.com/", "//twitter.com/"];
563 foreach ($list as $listitem) {
564 if (strpos($treffer[1], $listitem) !== false) {
569 if ((strpos($treffer[1], "//twitter.com/") !== false) && (strpos($treffer[1], "/status/") !== false)) {
573 if ((strpos($treffer[1], "//plus.google.com/") !== false) && (strpos($treffer[1], "/posts") !== false)) {
577 if ((strpos($treffer[1], "//plus.google.com/") !== false) && (strpos($treffer[1], "/photos") !== false)) {
581 $ignore = $ignore || strpos($treffer[1], '#') === 0;
584 $urls[$treffer[1]] = $treffer[1];
592 * @param string $html
593 * @param int $wraplength Ensures individual lines aren't longer than this many characters. Doesn't break words.
594 * @param bool $compact True: Completely strips image tags; False: Keeps image URLs
597 public static function toPlaintext(string $html, int $wraplength = 75, bool $compact = false): string
599 DI::profiler()->startRecording('rendering');
600 $message = str_replace("\r", "", $html);
602 $doc = new DOMDocument();
603 $doc->preserveWhiteSpace = false;
605 $message = mb_convert_encoding($message, 'HTML-ENTITIES', "UTF-8");
607 if (empty($message)) {
608 DI::profiler()->stopRecording();
612 @$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
614 $message = $doc->saveHTML();
615 // Remove eventual UTF-8 BOM
616 $message = str_replace("\xC3\x82\xC2\xA0", "", $message);
618 // Collecting all links
619 $urls = self::collectURLs($message);
621 if (empty($message)) {
622 DI::profiler()->stopRecording();
626 @$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
628 self::tagToBBCode($doc, 'html', [], '', '');
629 self::tagToBBCode($doc, 'body', [], '', '');
632 self::tagToBBCode($doc, 'blockquote', [], "»", "«");
634 self::tagToBBCode($doc, 'blockquote', [], '[quote]', "[/quote]\n");
637 self::tagToBBCode($doc, 'br', [], "\n", '');
639 self::tagToBBCode($doc, 'span', [], "", "");
640 self::tagToBBCode($doc, 'pre', [], "", "");
641 self::tagToBBCode($doc, 'div', [], "\r", "\r");
642 self::tagToBBCode($doc, 'p', [], "\n", "\n");
644 self::tagToBBCode($doc, 'li', [], "\n* ", "\n");
646 self::tagToBBCode($doc, 'hr', [], "\n" . str_repeat("-", 70) . "\n", "");
648 self::tagToBBCode($doc, 'tr', [], "\n", "");
649 self::tagToBBCode($doc, 'td', [], "\t", "");
651 self::tagToBBCode($doc, 'h1', [], "\n\n*", "*\n");
652 self::tagToBBCode($doc, 'h2', [], "\n\n*", "*\n");
653 self::tagToBBCode($doc, 'h3', [], "\n\n*", "*\n");
654 self::tagToBBCode($doc, 'h4', [], "\n\n*", "*\n");
655 self::tagToBBCode($doc, 'h5', [], "\n\n*", "*\n");
656 self::tagToBBCode($doc, 'h6', [], "\n\n*", "*\n");
659 self::tagToBBCode($doc, 'img', ['src' => '/(.+)/'], ' [img]$1', '[/img] ');
661 self::tagToBBCode($doc, 'img', ['src' => '/(.+)/'], ' ', ' ');
664 // Backward compatibility, [iframe] support has been removed in version 2020.12
665 self::tagToBBCode($doc, 'iframe', ['src' => '/(.+)/'], ' $1 ', '');
667 $message = $doc->saveHTML();
670 $message = str_replace("[img]", "", $message);
671 $message = str_replace("[/img]", "", $message);
674 // was ersetze ich da?
675 // Irgendein stoerrisches UTF-Zeug
676 $message = str_replace(chr(194) . chr(160), ' ', $message);
678 $message = str_replace(" ", " ", $message);
680 // Aufeinanderfolgende DIVs
681 $message = preg_replace('=\r *\r=i', "\n", $message);
682 $message = str_replace("\r", "\n", $message);
684 $message = strip_tags($message);
686 $message = html_entity_decode($message, ENT_QUOTES, 'UTF-8');
688 if (!$compact && ($message != '')) {
689 foreach ($urls as $id => $url) {
690 if ($url != '' && strpos($message, $url) === false) {
691 $message .= "\n" . $url . ' ';
696 $message = str_replace("\n«", "«\n", $message);
697 $message = str_replace("»\n", "\n»", $message);
700 $oldmessage = $message;
701 $message = str_replace("\n\n\n", "\n\n", $message);
702 } while ($oldmessage != $message);
704 $message = self::quoteLevel(trim($message), $wraplength);
706 DI::profiler()->stopRecording();
707 return trim($message);
711 * Converts provided HTML code to Markdown. The hardwrap parameter maximizes
712 * compatibility with Diaspora in spite of the Markdown standards.
714 * @param string $html
717 public static function toMarkdown(string $html): string
719 DI::profiler()->startRecording('rendering');
720 $converter = new HtmlConverter(['hard_break' => true]);
721 $markdown = $converter->convert($html);
723 DI::profiler()->stopRecording();
728 * Convert video HTML to BBCode tags
733 public static function toBBCodeVideo(string $s): string
736 '#<object[^>]+>(.*?)https?://www.youtube.com/((?:v|cp)/[A-Za-z0-9\-_=]+)(.*?)</object>#ism',
737 '[youtube]$2[/youtube]',
742 '#<iframe[^>](.*?)https?://www.youtube.com/embed/([A-Za-z0-9\-_=]+)(.*?)</iframe>#ism',
743 '[youtube]$2[/youtube]',
748 '#<iframe[^>](.*?)https?://player.vimeo.com/video/([0-9]+)(.*?)</iframe>#ism',
757 * transform link href and img src from relative to absolute
759 * @param string $text
760 * @param string $base base url
763 public static function relToAbs(string $text, string $base): string
769 $base = rtrim($base, '/');
771 $base2 = $base . "/";
774 $pattern = "/<a([^>]*) href=\"(?!http|https|\/)([^\"]*)\"/";
775 $replace = "<a\${1} href=\"" . $base2 . "\${2}\"";
776 $text = preg_replace($pattern, $replace, $text);
778 $pattern = "/<a([^>]*) href=\"(?!http|https)([^\"]*)\"/";
779 $replace = "<a\${1} href=\"" . $base . "\${2}\"";
780 $text = preg_replace($pattern, $replace, $text);
783 $pattern = "/<img([^>]*) src=\"(?!http|https|\/)([^\"]*)\"/";
784 $replace = "<img\${1} src=\"" . $base2 . "\${2}\"";
785 $text = preg_replace($pattern, $replace, $text);
787 $pattern = "/<img([^>]*) src=\"(?!http|https)([^\"]*)\"/";
788 $replace = "<img\${1} src=\"" . $base . "\${2}\"";
789 $text = preg_replace($pattern, $replace, $text);
797 * Loader for infinite scrolling
799 * @return string html for loader
800 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
802 public static function scrollLoader(): string
804 $tpl = Renderer::getMarkupTemplate("scroll_loader.tpl");
805 return Renderer::replaceMacros($tpl, [
806 'wait' => DI::l10n()->t('Loading more entries...'),
807 'end' => DI::l10n()->t('The end')
812 * Format contacts as picture links or as text links
814 * @param array $contact Array with contacts which contains an array with
815 * int 'id' => The ID of the contact
816 * int 'uid' => The user ID of the user who owns this data
817 * string 'name' => The name of the contact
818 * string 'url' => The url to the profile page of the contact
819 * string 'addr' => The webbie of the contact (e.g.) username@friendica.com
820 * string 'network' => The network to which the contact belongs to
821 * string 'thumb' => The contact picture
822 * string 'click' => js code which is performed when clicking on the contact
823 * @param boolean $redirect If true try to use the redir url if it's possible
824 * @param string $class CSS class for the
825 * @param boolean $textmode If true display the contacts as text links
826 * if false display the contacts as picture links
827 * @return string Formatted html
828 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
829 * @throws \ImagickException
831 public static function micropro(array $contact, bool $redirect = false, string $class = '', bool $textmode = false): string
833 // Use the contact URL if no address is available
834 if (empty($contact['addr'])) {
835 $contact["addr"] = $contact["url"];
838 $url = $contact['url'];
843 $url = Contact::magicLinkByContact($contact);
844 if (strpos($url, 'contact/redir/') === 0) {
845 $sparkle = ' sparkle';
849 // If there is some js available we don't need the url
850 if (!empty($contact['click'])) {
854 return Renderer::replaceMacros(Renderer::getMarkupTemplate($textmode ? 'micropro_txt.tpl' : 'micropro_img.tpl'), [
855 '$click' => $contact['click'] ?? '',
858 '$photo' => Contact::getThumb($contact),
859 '$name' => $contact['name'],
860 'title' => $contact['name'] . ' [' . $contact['addr'] . ']',
861 '$parkle' => $sparkle,
869 * @param string $s Search query.
870 * @param string $id HTML id
871 * @param bool $aside Display the search widgit aside.
873 * @return string Formatted HTML.
876 public static function search(string $s, string $id = 'search-box', bool $aside = true): string
880 if (strpos($s, '#') === 0) {
883 $save_label = $mode === 'text' ? DI::l10n()->t('Save') : DI::l10n()->t('Follow');
887 '$q' => urlencode($s),
889 '$search_label' => DI::l10n()->t('Search'),
890 '$save_label' => $save_label,
891 '$search_hint' => DI::l10n()->t('@name, !forum, #tags, content'),
893 '$return_url' => urlencode(Search::getSearchPath($s)),
897 $values['$search_options'] = [
898 'fulltext' => DI::l10n()->t('Full Text'),
899 'tags' => DI::l10n()->t('Tags'),
900 'contacts' => DI::l10n()->t('Contacts')
903 if (DI::config()->get('system', 'poco_local_search')) {
904 $values['$searchoption']['forums'] = DI::l10n()->t('Forums');
908 return Renderer::replaceMacros(Renderer::getMarkupTemplate('searchbox.tpl'), $values);
912 * Given a HTML text and a set of filtering reasons, adds a content hiding header with the provided reasons
914 * Reasons are expected to have been translated already.
916 * @param string $html
917 * @param array $reasons
919 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
921 public static function applyContentFilter(string $html, array $reasons): string
923 if (count($reasons)) {
924 $tpl = Renderer::getMarkupTemplate('wall/content_filter.tpl');
925 $html = Renderer::replaceMacros($tpl, [
926 '$reasons' => $reasons,
927 '$rnd' => Strings::getRandomHex(8),
928 '$openclose' => DI::l10n()->t('Click to open/close'),
937 * replace html amp entity with amp char
941 public static function unamp(string $s): string
943 return str_replace('&', '&', $s);
947 * Clean an HTML text for potentially harmful code
949 * @param string $text
950 * @param array $allowedIframeDomains List of allowed iframe source domains without the scheme
953 public static function purify(string $text, array $allowedIframeDomains = []): string
955 // Allows cid: URL scheme
956 \HTMLPurifier_URISchemeRegistry::instance()->register('cid', new HTMLPurifier_URIScheme_cid());
958 $config = \HTMLPurifier_HTML5Config::createDefault();
959 $config->set('HTML.Doctype', 'HTML5');
961 // Used to remove iframe with src attribute filtered out
962 $config->set('AutoFormat.RemoveEmpty', true);
964 $config->set('HTML.SafeIframe', true);
966 array_walk($allowedIframeDomains, function (&$domain) {
967 // Allow the domain and all its eventual sub-domains
968 $domain = '(?:(?!-)[A-Za-z0-9-]{1,63}(?<!-)\.)*' . preg_quote(trim($domain, '/'), '%');
971 $config->set('URI.SafeIframeRegexp',
973 ' . implode('|', $allowedIframeDomains) . '
975 (?:/|$) # Prevents bogus domains like youtube.com.fake.tld
979 $config->set('Attr.AllowedRel', [
980 'noreferrer' => true,
984 $config->set('Attr.AllowedFrameTargets', [
988 $config->set('AutoFormat.RemoveEmpty.Predicate', [
989 'colgroup' => [], // |
992 'iframe' => ['src'], // ↳ Default HTMLPurify values
993 'i' => ['class'], // Allows forkawesome icons
996 // Uncomment to debug HTMLPurifier behavior
997 //$config->set('Core.CollectErrors', true);
998 //$config->set('Core.MaintainLineNumbers', true);
1000 $HTMLPurifier = new \HTMLPurifier($config);
1002 $text = $HTMLPurifier->purify($text);
1004 /** @var \HTMLPurifier_ErrorCollector $errorCollector */
1005 // Uncomment to debug HTML Purifier behavior
1006 //$errorCollector = $HTMLPurifier->context->get('ErrorCollector');
1007 //var_dump($errorCollector->getRaw());
1013 * XPath arbitrary string quoting
1015 * @see https://stackoverflow.com/a/45228168
1016 * @param string $value
1019 public static function xpathQuote(string $value): string
1021 if (false === strpos($value, '"')) {
1022 return '"' . $value . '"';
1025 if (false === strpos($value, "'")) {
1026 return "'" . $value . "'";
1029 // if the value contains both single and double quotes, construct an
1030 // expression that concatenates all non-double-quote substrings with
1031 // the quotes, e.g.:
1033 // concat("'foo'", '"', "bar")
1034 return 'concat(' . implode(', \'"\', ', array_map(['self', 'xpathQuote'], explode('"', $value))) . ')';
1038 * Checks if the provided URL is present in the DOM document in an element with the rel="me" attribute
1040 * XHTML Friends Network http://gmpg.org/xfn/
1042 * @param DOMDocument $doc
1043 * @param UriInterface $meUrl
1046 public static function checkRelMeLink(DOMDocument $doc, UriInterface $meUrl): bool
1048 $xpath = new \DOMXpath($doc);
1050 // This expression checks that "me" is among the space-delimited values of the "rel" attribute.
1051 // And that the href attribute contains exactly the provided URL
1052 $expression = "//*[contains(concat(' ', normalize-space(@rel), ' '), ' me ')][@href = " . self::xpathQuote($meUrl) . "]";
1054 $result = $xpath->query($expression);
1056 return $result !== false && $result->length > 0;