<?php
/**
- * @file src/Content/Text/HTML.php
+ * @copyright Copyright (C) 2010-2023, the Friendica project
+ *
+ * @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 Friendica\Protocol\HTTP\MediaType;
use Friendica\Content\Widget\ContactBlock;
use Friendica\Core\Hook;
-use Friendica\Core\L10n;
-use Friendica\Core\Config;
use Friendica\Core\Renderer;
+use Friendica\Core\Search;
+use Friendica\DI;
use Friendica\Model\Contact;
use Friendica\Util\Network;
-use Friendica\Util\Proxy as ProxyUtils;
use Friendica\Util\Strings;
use Friendica\Util\XML;
use League\HTMLToMarkdown\HtmlConverter;
+use Psr\Http\Message\UriInterface;
class HTML
{
- public static function sanitizeCSS($input)
- {
- $cleaned = "";
-
- $input = strtolower($input);
-
- for ($i = 0; $i < strlen($input); $i++) {
- $char = substr($input, $i, 1);
-
- if (($char >= "a") && ($char <= "z")) {
- $cleaned .= $char;
- }
-
- if (!(strpos(" #;:0123456789-_.%", $char) === false)) {
- $cleaned .= $char;
- }
- }
-
- return $cleaned;
- }
-
- private static function tagToBBCode(DOMDocument $doc, $tag, $attributes, $startbb, $endbb)
+ /**
+ * Search all instances of a specific HTML tag node in the provided DOM document and replaces them with BBCode text nodes.
+ *
+ * @see HTML::tagToBBCodeSub()
+ */
+ private static function tagToBBCode(DOMDocument $doc, string $tag, array $attributes, string $startbb, string $endbb, bool $ignoreChildren = false)
{
do {
- $done = self::tagToBBCodeSub($doc, $tag, $attributes, $startbb, $endbb);
+ $done = self::tagToBBCodeSub($doc, $tag, $attributes, $startbb, $endbb, $ignoreChildren);
} while ($done);
}
- private static function tagToBBCodeSub(DOMDocument $doc, $tag, $attributes, $startbb, $endbb)
+ /**
+ * Search the first specific HTML tag node in the provided DOM document and replaces it with BBCode text nodes.
+ *
+ * @param DOMDocument $doc
+ * @param string $tag HTML tag name
+ * @param array $attributes Array of attributes to match and optionally use the value from
+ * @param string $startbb BBCode tag opening
+ * @param string $endbb BBCode tag closing
+ * @param bool $ignoreChildren If set to false, the HTML tag children will be appended as text inside the BBCode tag
+ * Otherwise, they will be entirely ignored. Useful for simple BBCode that draw their
+ * inner value from an attribute value and disregard the tag children.
+ * @return bool Whether a replacement was done
+ */
+ private static function tagToBBCodeSub(DOMDocument $doc, string $tag, array $attributes, string $startbb, string $endbb, bool $ignoreChildren = false): bool
{
$savestart = str_replace('$', '\x01', $startbb);
$replace = false;
$node->parentNode->insertBefore($StartCode, $node);
- if ($node->hasChildNodes()) {
+ if (!$ignoreChildren && $node->hasChildNodes()) {
/** @var \DOMNode $child */
foreach ($node->childNodes as $key => $child) {
/* Remove empty text nodes at the start or at the end of the children list */
- if ($key > 0 && $key < $node->childNodes->length - 1 || trim($child->nodeValue)) {
+ if ($key > 0 && $key < $node->childNodes->length - 1 || $child->nodeName != '#text' || trim($child->nodeValue) !== '') {
$newNode = $child->cloneNode(true);
$node->parentNode->insertBefore($newNode, $node);
}
}
/**
+ * Converter for HTML to BBCode
+ *
* Made by: ike@piratenpartei.de
* Originally made for the syncom project: http://wiki.piratenpartei.de/Syncom
* https://github.com/annando/Syncom
*
- * @brief Converter for HTML to BBCode
* @param string $message
* @param string $basepath
* @return string
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function toBBCode($message, $basepath = '')
+ public static function toBBCode(string $message, string $basepath = ''): string
{
+ /*
+ * Check if message is empty to prevent a lot of code below from being executed
+ * for just an empty message.
+ */
+ if ($message === '') {
+ return '';
+ }
+
+ DI::profiler()->startRecording('rendering');
$message = str_replace("\r", "", $message);
- // Removing code blocks before the whitespace removal processing below
- $codeblocks = [];
- $message = preg_replace_callback(
- '#<pre><code(?: class="language-([^"]*)")?>(.*)</code></pre>#iUs',
- function ($matches) use (&$codeblocks) {
- $return = '[codeblock-' . count($codeblocks) . ']';
+ $message = Strings::performWithEscapedBlocks($message, '#<pre><code.*</code></pre>#iUs', function ($message) {
+ $message = str_replace(
+ [
+ "<li><p>",
+ "</p></li>",
+ ],
+ [
+ "<li>",
+ "</li>",
+ ],
+ $message
+ );
- $prefix = '[code]';
- if ($matches[1] != '') {
- $prefix = '[code=' . $matches[1] . ']';
- }
+ // remove namespaces
+ $message = preg_replace('=<(\w+):(.+?)>=', '<removeme>', $message);
+ $message = preg_replace('=</(\w+):(.+?)>=', '</removeme>', $message);
- $codeblocks[] = $prefix . PHP_EOL . trim($matches[2]) . PHP_EOL . '[/code]';
- return $return;
- },
- $message
- );
+ $doc = new DOMDocument();
+ $doc->preserveWhiteSpace = false;
- $message = str_replace(
- [
- "<li><p>",
- "</p></li>",
- ],
- [
- "<li>",
- "</li>",
- ],
- $message
- );
+ $message = mb_convert_encoding($message, 'HTML-ENTITIES', "UTF-8");
- // remove namespaces
- $message = preg_replace('=<(\w+):(.+?)>=', '<removeme>', $message);
- $message = preg_replace('=</(\w+):(.+?)>=', '</removeme>', $message);
+ if (empty($message)) {
+ return '';
+ }
- $doc = new DOMDocument();
- $doc->preserveWhiteSpace = false;
+ @$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
+
+ XML::deleteNode($doc, 'style');
+ XML::deleteNode($doc, 'head');
+ XML::deleteNode($doc, 'title');
+ XML::deleteNode($doc, 'meta');
+ XML::deleteNode($doc, 'xml');
+ XML::deleteNode($doc, 'removeme');
+
+ $xpath = new DomXPath($doc);
+ $list = $xpath->query("//pre");
+ foreach ($list as $node) {
+ // Ensure to escape unescaped & - they will otherwise raise a warning
+ $safe_value = preg_replace('/&(?!\w+;)/', '&', $node->nodeValue);
+ $node->nodeValue = str_replace("\n", "\r", $safe_value);
+ }
- $message = mb_convert_encoding($message, 'HTML-ENTITIES', "UTF-8");
+ $message = $doc->saveHTML();
+ $message = str_replace(["\n<", ">\n", "\r", "\n", "\xC3\x82\xC2\xA0"], ["<", ">", "<br />", " ", ""], $message);
+ $message = preg_replace('= [\s]*=i', " ", $message);
- @$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
+ if (empty($message)) {
+ return '';
+ }
- XML::deleteNode($doc, 'style');
- XML::deleteNode($doc, 'head');
- XML::deleteNode($doc, 'title');
- XML::deleteNode($doc, 'meta');
- XML::deleteNode($doc, 'xml');
- XML::deleteNode($doc, 'removeme');
+ @$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
- $xpath = new DomXPath($doc);
- $list = $xpath->query("//pre");
- foreach ($list as $node) {
- // Ensure to escape unescaped & - they will otherwise raise a warning
- $safe_value = preg_replace('/&(?!\w+;)/', '&', $node->nodeValue);
- $node->nodeValue = str_replace("\n", "\r", $safe_value);
- }
+ self::tagToBBCode($doc, 'html', [], "", "");
+ self::tagToBBCode($doc, 'body', [], "", "");
- $message = $doc->saveHTML();
- $message = str_replace(["\n<", ">\n", "\r", "\n", "\xC3\x82\xC2\xA0"], ["<", ">", "<br />", " ", ""], $message);
- $message = preg_replace('= [\s]*=i', " ", $message);
+ // Outlook-Quote - Variant 1
+ self::tagToBBCode($doc, 'p', ['class' => 'MsoNormal', 'style' => 'margin-left:35.4pt'], '[quote]', '[/quote]');
- @$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
+ // Outlook-Quote - Variant 2
+ self::tagToBBCode(
+ $doc,
+ 'div',
+ ['style' => 'border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt'],
+ '[quote]',
+ '[/quote]'
+ );
- self::tagToBBCode($doc, 'html', [], "", "");
- self::tagToBBCode($doc, 'body', [], "", "");
+ // MyBB-Stuff
+ self::tagToBBCode($doc, 'span', ['style' => 'text-decoration: underline;'], '[u]', '[/u]');
+ self::tagToBBCode($doc, 'span', ['style' => 'font-style: italic;'], '[i]', '[/i]');
+ self::tagToBBCode($doc, 'span', ['style' => 'font-weight: bold;'], '[b]', '[/b]');
- // Outlook-Quote - Variant 1
- self::tagToBBCode($doc, 'p', ['class' => 'MsoNormal', 'style' => 'margin-left:35.4pt'], '[quote]', '[/quote]');
+ /* self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/', 'size'=>'/(\d+)/', 'color'=>'/(.+)/'), '[font=$1][size=$2][color=$3]', '[/color][/size][/font]');
+ self::node2BBCode($doc, 'font', array('size'=>'/(\d+)/', 'color'=>'/(.+)/'), '[size=$1][color=$2]', '[/color][/size]');
+ self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/', 'size'=>'/(.+)/'), '[font=$1][size=$2]', '[/size][/font]');
+ self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/', 'color'=>'/(.+)/'), '[font=$1][color=$3]', '[/color][/font]');
+ self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/'), '[font=$1]', '[/font]');
+ self::node2BBCode($doc, 'font', array('size'=>'/(\d+)/'), '[size=$1]', '[/size]');
+ self::node2BBCode($doc, 'font', array('color'=>'/(.+)/'), '[color=$1]', '[/color]');
+ */
+ // Untested
+ //self::node2BBCode($doc, 'span', array('style'=>'/.*font-size:\s*(.+?)[,;].*font-family:\s*(.+?)[,;].*color:\s*(.+?)[,;].*/'), '[size=$1][font=$2][color=$3]', '[/color][/font][/size]');
+ //self::node2BBCode($doc, 'span', array('style'=>'/.*font-size:\s*(\d+)[,;].*/'), '[size=$1]', '[/size]');
+ //self::node2BBCode($doc, 'span', array('style'=>'/.*font-size:\s*(.+?)[,;].*/'), '[size=$1]', '[/size]');
- // Outlook-Quote - Variant 2
- self::tagToBBCode(
- $doc,
- 'div',
- ['style' => 'border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt'],
- '[quote]',
- '[/quote]'
- );
+ self::tagToBBCode($doc, 'span', ['style' => '/.*color:\s*(.+?)[,;].*/'], '[color="$1"]', '[/color]');
- // MyBB-Stuff
- self::tagToBBCode($doc, 'span', ['style' => 'text-decoration: underline;'], '[u]', '[/u]');
- self::tagToBBCode($doc, 'span', ['style' => 'font-style: italic;'], '[i]', '[/i]');
- self::tagToBBCode($doc, 'span', ['style' => 'font-weight: bold;'], '[b]', '[/b]');
-
- /* self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/', 'size'=>'/(\d+)/', 'color'=>'/(.+)/'), '[font=$1][size=$2][color=$3]', '[/color][/size][/font]');
- self::node2BBCode($doc, 'font', array('size'=>'/(\d+)/', 'color'=>'/(.+)/'), '[size=$1][color=$2]', '[/color][/size]');
- self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/', 'size'=>'/(.+)/'), '[font=$1][size=$2]', '[/size][/font]');
- self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/', 'color'=>'/(.+)/'), '[font=$1][color=$3]', '[/color][/font]');
- self::node2BBCode($doc, 'font', array('face'=>'/([\w ]+)/'), '[font=$1]', '[/font]');
- self::node2BBCode($doc, 'font', array('size'=>'/(\d+)/'), '[size=$1]', '[/size]');
- self::node2BBCode($doc, 'font', array('color'=>'/(.+)/'), '[color=$1]', '[/color]');
- */
- // Untested
- //self::node2BBCode($doc, 'span', array('style'=>'/.*font-size:\s*(.+?)[,;].*font-family:\s*(.+?)[,;].*color:\s*(.+?)[,;].*/'), '[size=$1][font=$2][color=$3]', '[/color][/font][/size]');
- //self::node2BBCode($doc, 'span', array('style'=>'/.*font-size:\s*(\d+)[,;].*/'), '[size=$1]', '[/size]');
- //self::node2BBCode($doc, 'span', array('style'=>'/.*font-size:\s*(.+?)[,;].*/'), '[size=$1]', '[/size]');
-
- self::tagToBBCode($doc, 'span', ['style' => '/.*color:\s*(.+?)[,;].*/'], '[color="$1"]', '[/color]');
-
- //self::node2BBCode($doc, 'span', array('style'=>'/.*font-family:\s*(.+?)[,;].*/'), '[font=$1]', '[/font]');
- //self::node2BBCode($doc, 'div', array('style'=>'/.*font-family:\s*(.+?)[,;].*font-size:\s*(\d+?)pt.*/'), '[font=$1][size=$2]', '[/size][/font]');
- //self::node2BBCode($doc, 'div', array('style'=>'/.*font-family:\s*(.+?)[,;].*font-size:\s*(\d+?)px.*/'), '[font=$1][size=$2]', '[/size][/font]');
- //self::node2BBCode($doc, 'div', array('style'=>'/.*font-family:\s*(.+?)[,;].*/'), '[font=$1]', '[/font]');
- // Importing the classes - interesting for importing of posts from third party networks that were exported from friendica
- // Test
- //self::node2BBCode($doc, 'span', array('class'=>'/([\w ]+)/'), '[class=$1]', '[/class]');
- self::tagToBBCode($doc, 'span', ['class' => 'type-link'], '[class=type-link]', '[/class]');
- self::tagToBBCode($doc, 'span', ['class' => 'type-video'], '[class=type-video]', '[/class]');
-
- self::tagToBBCode($doc, 'strong', [], '[b]', '[/b]');
- self::tagToBBCode($doc, 'em', [], '[i]', '[/i]');
- self::tagToBBCode($doc, 'b', [], '[b]', '[/b]');
- self::tagToBBCode($doc, 'i', [], '[i]', '[/i]');
- self::tagToBBCode($doc, 'u', [], '[u]', '[/u]');
- self::tagToBBCode($doc, 's', [], '[s]', '[/s]');
- self::tagToBBCode($doc, 'del', [], '[s]', '[/s]');
- self::tagToBBCode($doc, 'strike', [], '[s]', '[/s]');
-
- self::tagToBBCode($doc, 'big', [], "[size=large]", "[/size]");
- self::tagToBBCode($doc, 'small', [], "[size=small]", "[/size]");
-
- self::tagToBBCode($doc, 'blockquote', [], '[quote]', '[/quote]');
+ //self::node2BBCode($doc, 'span', array('style'=>'/.*font-family:\s*(.+?)[,;].*/'), '[font=$1]', '[/font]');
+ //self::node2BBCode($doc, 'div', array('style'=>'/.*font-family:\s*(.+?)[,;].*font-size:\s*(\d+?)pt.*/'), '[font=$1][size=$2]', '[/size][/font]');
+ //self::node2BBCode($doc, 'div', array('style'=>'/.*font-family:\s*(.+?)[,;].*font-size:\s*(\d+?)px.*/'), '[font=$1][size=$2]', '[/size][/font]');
+ //self::node2BBCode($doc, 'div', array('style'=>'/.*font-family:\s*(.+?)[,;].*/'), '[font=$1]', '[/font]');
+ // Importing the classes - interesting for importing of posts from third party networks that were exported from friendica
+ // Test
+ //self::node2BBCode($doc, 'span', array('class'=>'/([\w ]+)/'), '[class=$1]', '[/class]');
+ self::tagToBBCode($doc, 'span', ['class' => 'type-link'], '[class=type-link]', '[/class]');
+ self::tagToBBCode($doc, 'span', ['class' => 'type-video'], '[class=type-video]', '[/class]');
- self::tagToBBCode($doc, 'br', [], "\n", '');
+ self::tagToBBCode($doc, 'strong', [], '[b]', '[/b]');
+ self::tagToBBCode($doc, 'em', [], '[i]', '[/i]');
+ self::tagToBBCode($doc, 'b', [], '[b]', '[/b]');
+ self::tagToBBCode($doc, 'i', [], '[i]', '[/i]');
+ self::tagToBBCode($doc, 'u', [], '[u]', '[/u]');
+ self::tagToBBCode($doc, 's', [], '[s]', '[/s]');
+ self::tagToBBCode($doc, 'del', [], '[s]', '[/s]');
+ self::tagToBBCode($doc, 'strike', [], '[s]', '[/s]');
- self::tagToBBCode($doc, 'p', ['class' => 'MsoNormal'], "\n", "");
- self::tagToBBCode($doc, 'div', ['class' => 'MsoNormal'], "\r", "");
+ self::tagToBBCode($doc, 'big', [], "[size=large]", "[/size]");
+ self::tagToBBCode($doc, 'small', [], "[size=small]", "[/size]");
- self::tagToBBCode($doc, 'span', [], "", "");
+ self::tagToBBCode($doc, 'blockquote', [], '[quote]', '[/quote]');
- self::tagToBBCode($doc, 'span', [], "", "");
- self::tagToBBCode($doc, 'pre', [], "", "");
+ self::tagToBBCode($doc, 'br', [], "\n", '');
- self::tagToBBCode($doc, 'div', [], "\r", "\r");
- self::tagToBBCode($doc, 'p', [], "\n", "\n");
+ self::tagToBBCode($doc, 'p', ['class' => 'MsoNormal'], "\n", "");
+ self::tagToBBCode($doc, 'div', ['class' => 'MsoNormal'], "\r", "");
- self::tagToBBCode($doc, 'ul', [], "[list]", "[/list]");
- self::tagToBBCode($doc, 'ol', [], "[list=1]", "[/list]");
- self::tagToBBCode($doc, 'li', [], "[*]", "");
+ self::tagToBBCode($doc, 'span', [], "", "");
- self::tagToBBCode($doc, 'hr', [], "[hr]", "");
+ self::tagToBBCode($doc, 'span', [], "", "");
+ self::tagToBBCode($doc, 'pre', [], "", "");
- self::tagToBBCode($doc, 'table', [], "", "");
- self::tagToBBCode($doc, 'tr', [], "\n", "");
- self::tagToBBCode($doc, 'td', [], "\t", "");
- //self::node2BBCode($doc, 'table', array(), "[table]", "[/table]");
- //self::node2BBCode($doc, 'th', array(), "[th]", "[/th]");
- //self::node2BBCode($doc, 'tr', array(), "[tr]", "[/tr]");
- //self::node2BBCode($doc, 'td', array(), "[td]", "[/td]");
- //self::node2BBCode($doc, 'h1', array(), "\n\n[size=xx-large][b]", "[/b][/size]\n");
- //self::node2BBCode($doc, 'h2', array(), "\n\n[size=x-large][b]", "[/b][/size]\n");
- //self::node2BBCode($doc, 'h3', array(), "\n\n[size=large][b]", "[/b][/size]\n");
- //self::node2BBCode($doc, 'h4', array(), "\n\n[size=medium][b]", "[/b][/size]\n");
- //self::node2BBCode($doc, 'h5', array(), "\n\n[size=small][b]", "[/b][/size]\n");
- //self::node2BBCode($doc, 'h6', array(), "\n\n[size=x-small][b]", "[/b][/size]\n");
-
- self::tagToBBCode($doc, 'h1', [], "[h1]", "[/h1]");
- self::tagToBBCode($doc, 'h2', [], "[h2]", "[/h2]");
- self::tagToBBCode($doc, 'h3', [], "[h3]", "[/h3]");
- self::tagToBBCode($doc, 'h4', [], "[h4]", "[/h4]");
- self::tagToBBCode($doc, 'h5', [], "[h5]", "[/h5]");
- self::tagToBBCode($doc, 'h6', [], "[h6]", "[/h6]");
-
- self::tagToBBCode($doc, 'a', ['href' => '/mailto:(.+)/'], '[mail=$1]', '[/mail]');
- self::tagToBBCode($doc, 'a', ['href' => '/(.+)/'], '[url=$1]', '[/url]');
-
- self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'alt' => '/(.+)/'], '[img=$1]$2', '[/img]');
- self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'width' => '/(\d+)/', 'height' => '/(\d+)/'], '[img=$2x$3]$1', '[/img]');
- self::tagToBBCode($doc, 'img', ['src' => '/(.+)/'], '[img]$1', '[/img]');
-
-
- self::tagToBBCode($doc, 'video', ['src' => '/(.+)/'], '[video]$1', '[/video]');
- self::tagToBBCode($doc, 'audio', ['src' => '/(.+)/'], '[audio]$1', '[/audio]');
- self::tagToBBCode($doc, 'iframe', ['src' => '/(.+)/'], '[iframe]$1', '[/iframe]');
-
- self::tagToBBCode($doc, 'key', [], '[code]', '[/code]');
- self::tagToBBCode($doc, 'code', [], '[code]', '[/code]');
+ self::tagToBBCode($doc, 'div', [], "\r", "\r");
+ self::tagToBBCode($doc, 'p', [], "\n", "\n");
- $message = $doc->saveHTML();
+ self::tagToBBCode($doc, 'ul', [], "[ul]", "\n[/ul]");
+ self::tagToBBCode($doc, 'ol', [], "[ol]", "\n[/ol]");
+ self::tagToBBCode($doc, 'li', [], "\n[li]", "[/li]");
- // I'm removing something really disturbing
- // Don't know exactly what it is
- $message = str_replace(chr(194) . chr(160), ' ', $message);
+ self::tagToBBCode($doc, 'hr', [], "[hr]", "");
- $message = str_replace(" ", " ", $message);
+ self::tagToBBCode($doc, 'table', [], "[table]", "[/table]");
+ self::tagToBBCode($doc, 'th', [], "[th]", "[/th]");
+ self::tagToBBCode($doc, 'tr', [], "[tr]", "[/tr]");
+ self::tagToBBCode($doc, 'td', [], "[td]", "[/td]");
- // removing multiple DIVs
- $message = preg_replace('=\r *\r=i', "\n", $message);
- $message = str_replace("\r", "\n", $message);
+ self::tagToBBCode($doc, 'h1', [], "[h1]", "[/h1]");
+ self::tagToBBCode($doc, 'h2', [], "[h2]", "[/h2]");
+ self::tagToBBCode($doc, 'h3', [], "[h3]", "[/h3]");
+ self::tagToBBCode($doc, 'h4', [], "[h4]", "[/h4]");
+ self::tagToBBCode($doc, 'h5', [], "[h5]", "[/h5]");
+ self::tagToBBCode($doc, 'h6', [], "[h6]", "[/h6]");
- Hook::callAll('html2bbcode', $message);
+ self::tagToBBCode($doc, 'a', ['href' => '/mailto:(.+)/'], '[mail=$1]', '[/mail]');
+ self::tagToBBCode($doc, 'a', ['href' => '/(.+)/'], '[url=$1]', '[/url]');
- $message = strip_tags($message);
+ self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'alt' => '/(.+)/'], '[img=$1]$2', '[/img]', true);
+ self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'width' => '/(\d+)/', 'height' => '/(\d+)/'], '[img=$2x$3]$1', '[/img]', true);
+ self::tagToBBCode($doc, 'img', ['src' => '/(.+)/'], '[img]$1', '[/img]', true);
- $message = html_entity_decode($message, ENT_QUOTES, 'UTF-8');
- $message = str_replace(["<"], ["<"], $message);
+ self::tagToBBCode($doc, 'video', ['src' => '/(.+)/'], '[video]$1', '[/video]', true);
+ self::tagToBBCode($doc, 'audio', ['src' => '/(.+)/'], '[audio]$1', '[/audio]', true);
+ // Backward compatibility, [iframe] support has been removed in version 2020.12
+ self::tagToBBCode($doc, 'iframe', ['src' => '/(.+)/'], '[url]$1', '[/url]', true);
- // remove quotes if they don't make sense
- $message = preg_replace('=\[/quote\][\s]*\[quote\]=i', "\n", $message);
+ self::tagToBBCode($doc, 'key', [], '[code]', '[/code]');
+ self::tagToBBCode($doc, 'code', [], '[code]', '[/code]');
- $message = preg_replace('=\[quote\]\s*=i', "[quote]", $message);
- $message = preg_replace('=\s*\[/quote\]=i', "[/quote]", $message);
+ $message = $doc->saveHTML();
- do {
- $oldmessage = $message;
- $message = str_replace("\n \n", "\n\n", $message);
- } while ($oldmessage != $message);
+ // I'm removing something really disturbing
+ // Don't know exactly what it is
+ $message = str_replace(chr(194) . chr(160), ' ', $message);
- do {
- $oldmessage = $message;
- $message = str_replace("\n\n\n", "\n\n", $message);
- } while ($oldmessage != $message);
+ $message = str_replace(" ", " ", $message);
+
+ // removing multiple DIVs
+ $message = preg_replace('=\r *\r=i', "\n", $message);
+ $message = str_replace("\r", "\n", $message);
+
+ Hook::callAll('html2bbcode', $message);
+
+ $message = strip_tags($message);
+
+ $message = html_entity_decode($message, ENT_QUOTES, 'UTF-8');
+
+ // remove quotes if they don't make sense
+ $message = preg_replace('=\[/quote\][\s]*\[quote\]=i', "\n", $message);
+
+ $message = preg_replace('=\[quote\]\s*=i', "[quote]", $message);
+ $message = preg_replace('=\s*\[/quote\]=i', "[/quote]", $message);
+
+ do {
+ $oldmessage = $message;
+ $message = str_replace("\n \n", "\n\n", $message);
+ } while ($oldmessage != $message);
+
+ do {
+ $oldmessage = $message;
+ $message = str_replace("\n\n\n", "\n\n", $message);
+ } while ($oldmessage != $message);
- do {
- $oldmessage = $message;
$message = str_replace(
- [
- "[/size]\n\n",
- "\n[hr]",
- "[hr]\n",
- "\n[list",
- "[/list]\n",
- "\n[/",
- "[list]\n",
- "[list=1]\n",
- "\n[*]"],
- [
- "[/size]\n",
- "[hr]",
- "[hr]",
- "[list",
- "[/list]",
- "[/",
- "[list]",
- "[list=1]",
- "[*]"],
+ ['[b][b]', '[/b][/b]', '[i][i]', '[/i][/i]'],
+ ['[b]', '[/b]', '[i]', '[/i]'],
$message
);
- } while ($message != $oldmessage);
- $message = str_replace(
- ['[b][b]', '[/b][/b]', '[i][i]', '[/i][/i]'],
- ['[b]', '[/b]', '[i]', '[/i]'],
- $message
- );
+ // Handling Yahoo style of mails
+ $message = str_replace('[hr][b]From:[/b]', '[quote][b]From:[/b]', $message);
- // Handling Yahoo style of mails
- $message = str_replace('[hr][b]From:[/b]', '[quote][b]From:[/b]', $message);
+ return $message;
+ });
- // Restore code blocks
$message = preg_replace_callback(
- '#\[codeblock-([0-9]+)\]#iU',
- function ($matches) use ($codeblocks) {
- $return = '';
- if (isset($codeblocks[intval($matches[1])])) {
- $return = $codeblocks[$matches[1]];
+ '#<pre><code(?: class="language-([^"]*)")?>(.*)</code></pre>#iUs',
+ function ($matches) {
+ $prefix = '[code]';
+ if ($matches[1] != '') {
+ $prefix = '[code=' . $matches[1] . ']';
}
- return $return;
+
+ return $prefix . "\n" . html_entity_decode($matches[2]) . "\n" . '[/code]';
},
$message
);
$message = self::qualifyURLs($message, $basepath);
}
+ DI::profiler()->stopRecording();
return $message;
}
/**
- * @brief Sub function to complete incomplete URL
+ * Sub function to complete incomplete URL
*
* @param array $matches Result of preg_replace_callback
* @param string $basepath Basepath that is used to complete the URL
*
* @return string The expanded URL
*/
- private static function qualifyURLsSub($matches, $basepath)
+ private static function qualifyURLsSub(array $matches, string $basepath): string
{
$base = parse_url($basepath);
unset($base['query']);
$link = $matches[0];
$url = $matches[1];
+ if (empty($url) || empty(parse_url($url))) {
+ return $matches[0];
+ }
+
$parts = array_merge($base, parse_url($url));
$url2 = Network::unparseURL($parts);
}
/**
- * @brief Complete incomplete URLs in BBCode
+ * Complete incomplete URLs in BBCode
*
* @param string $body Body with URLs
* @param string $basepath Base path that is used to complete the URL
*
* @return string Body with expanded URLs
*/
- private static function qualifyURLs($body, $basepath)
+ private static function qualifyURLs(string $body, string $basepath): string
{
$URLSearchString = "^\[\]";
return $body;
}
- private static function breakLines($line, $level, $wraplength = 75)
+ private static function breakLines(string $line, int $level, int $wraplength = 75): string
{
if ($wraplength == 0) {
$wraplength = 2000000;
$newlines[] = $line;
- return implode($newlines, "\n");
+ return implode("\n", $newlines);
}
- private static function quoteLevel($message, $wraplength = 75)
+ private static function quoteLevel(string $message, int $wraplength = 75): string
{
$lines = explode("\n", $message);
}
}
- return implode($newlines, "\n");
+ return implode("\n", $newlines);
}
- private static function collectURLs($message)
+ private static function collectURLs(string $message): array
{
$pattern = '/<a.*?href="(.*?)".*?>(.*?)<\/a>/is';
preg_match_all($pattern, $message, $result, PREG_SET_ORDER);
* @param bool $compact True: Completely strips image tags; False: Keeps image URLs
* @return string
*/
- public static function toPlaintext(string $html, $wraplength = 75, $compact = false)
+ public static function toPlaintext(string $html, int $wraplength = 75, bool $compact = false): string
{
+ DI::profiler()->startRecording('rendering');
$message = str_replace("\r", "", $html);
$doc = new DOMDocument();
$message = mb_convert_encoding($message, 'HTML-ENTITIES', "UTF-8");
+ if (empty($message)) {
+ DI::profiler()->stopRecording();
+ return '';
+ }
+
@$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
$message = $doc->saveHTML();
// Collecting all links
$urls = self::collectURLs($message);
+ if (empty($message)) {
+ DI::profiler()->stopRecording();
+ return '';
+ }
+
@$doc->loadHTML($message, LIBXML_HTML_NODEFDTD);
self::tagToBBCode($doc, 'html', [], '', '');
self::tagToBBCode($doc, 'img', ['src' => '/(.+)/'], ' ', ' ');
}
+ // Backward compatibility, [iframe] support has been removed in version 2020.12
self::tagToBBCode($doc, 'iframe', ['src' => '/(.+)/'], ' $1 ', '');
$message = $doc->saveHTML();
$message = self::quoteLevel(trim($message), $wraplength);
+ DI::profiler()->stopRecording();
return trim($message);
}
* @param string $html
* @return string
*/
- public static function toMarkdown($html)
+ public static function toMarkdown(string $html): string
{
+ DI::profiler()->startRecording('rendering');
$converter = new HtmlConverter(['hard_break' => true]);
$markdown = $converter->convert($html);
+ DI::profiler()->stopRecording();
return $markdown;
}
/**
- * @brief Convert video HTML to BBCode tags
+ * Convert video HTML to BBCode tags
*
* @param string $s
* @return string
*/
- public static function toBBCodeVideo($s)
+ public static function toBBCodeVideo(string $s): string
{
$s = preg_replace(
'#<object[^>]+>(.*?)https?://www.youtube.com/((?:v|cp)/[A-Za-z0-9\-_=]+)(.*?)</object>#ism',
'[youtube]$2[/youtube]',
$s
);
-
+
$s = preg_replace(
'#<iframe[^>](.*?)https?://www.youtube.com/embed/([A-Za-z0-9\-_=]+)(.*?)</iframe>#ism',
'[youtube]$2[/youtube]',
$s
);
-
+
$s = preg_replace(
'#<iframe[^>](.*?)https?://player.vimeo.com/video/([0-9]+)(.*?)</iframe>#ism',
'[vimeo]$2[/vimeo]',
$s
);
-
+
return $s;
}
-
+
/**
* transform link href and img src from relative to absolute
*
* @param string $base base url
* @return string
*/
- public static function relToAbs($text, $base)
+ public static function relToAbs(string $text, string $base): string
{
if (empty($base)) {
return $text;
}
-
+
$base = rtrim($base, '/');
-
+
$base2 = $base . "/";
-
+
// Replace links
$pattern = "/<a([^>]*) href=\"(?!http|https|\/)([^\"]*)\"/";
$replace = "<a\${1} href=\"" . $base2 . "\${2}\"";
$text = preg_replace($pattern, $replace, $text);
-
+
$pattern = "/<a([^>]*) href=\"(?!http|https)([^\"]*)\"/";
$replace = "<a\${1} href=\"" . $base . "\${2}\"";
$text = preg_replace($pattern, $replace, $text);
-
+
// Replace images
$pattern = "/<img([^>]*) src=\"(?!http|https|\/)([^\"]*)\"/";
$replace = "<img\${1} src=\"" . $base2 . "\${2}\"";
$text = preg_replace($pattern, $replace, $text);
-
+
$pattern = "/<img([^>]*) src=\"(?!http|https)([^\"]*)\"/";
$replace = "<img\${1} src=\"" . $base . "\${2}\"";
$text = preg_replace($pattern, $replace, $text);
-
-
+
+
// Done
return $text;
}
- /**
- * return div element with class 'clear'
- * @return string
- * @deprecated
- */
- public static function clearDiv()
- {
- return '<div class="clear"></div>';
- }
-
/**
* Loader for infinite scrolling
*
* @return string html for loader
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function scrollLoader()
+ public static function scrollLoader(): string
{
$tpl = Renderer::getMarkupTemplate("scroll_loader.tpl");
return Renderer::replaceMacros($tpl, [
- 'wait' => L10n::t('Loading more entries...'),
- 'end' => L10n::t('The end')
+ 'wait' => DI::l10n()->t('Loading more entries...'),
+ 'end' => DI::l10n()->t('The end')
]);
}
/**
- * Get html for contact block.
- *
- * @deprecated since version 2019.03
- * @see ContactBlock::getHTML()
- * @return string
- * @throws \Friendica\Network\HTTPException\InternalServerErrorException
- * @throws \ImagickException
- */
- public static function contactBlock()
- {
- $a = \get_app();
-
- return ContactBlock::getHTML($a->profile);
- }
-
- /**
- * @brief Format contacts as picture links or as text links
+ * Format contacts as picture links or as text links
*
* @param array $contact Array with contacts which contains an array with
* int 'id' => The ID of the contact
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
* @throws \ImagickException
*/
- public static function micropro($contact, $redirect = false, $class = '', $textmode = false)
+ public static function micropro(array $contact, bool $redirect = false, string $class = '', bool $textmode = false): string
{
// Use the contact URL if no address is available
if (empty($contact['addr'])) {
$redir = false;
if ($redirect) {
- $url = Contact::magicLink($contact['url']);
- if (strpos($url, 'redir/') === 0) {
+ $url = Contact::magicLinkByContact($contact);
+ if (strpos($url, 'contact/redir/') === 0) {
$sparkle = ' sparkle';
}
}
$url = '';
}
- return Renderer::replaceMacros(Renderer::getMarkupTemplate(($textmode)?'micropro_txt.tpl':'micropro_img.tpl'), [
- '$click' => defaults($contact, 'click', ''),
+ return Renderer::replaceMacros(Renderer::getMarkupTemplate($textmode ? 'micropro_txt.tpl' : 'micropro_img.tpl'), [
+ '$click' => $contact['click'] ?? '',
'$class' => $class,
'$url' => $url,
- '$photo' => ProxyUtils::proxifyUrl($contact['thumb'], false, ProxyUtils::SIZE_THUMB),
+ '$photo' => Contact::getThumb($contact),
'$name' => $contact['name'],
'title' => $contact['name'] . ' [' . $contact['addr'] . ']',
'$parkle' => $sparkle,
*
* @param string $s Search query.
* @param string $id HTML id
- * @param string $url Search url.
* @param bool $aside Display the search widgit aside.
*
* @return string Formatted HTML.
- * @throws \Friendica\Network\HTTPException\InternalServerErrorException
+ * @throws \Exception
*/
- public static function search($s, $id = 'search-box', $url = 'search', $aside = true)
+ public static function search(string $s, string $id = 'search-box', bool $aside = true): string
{
$mode = 'text';
if (strpos($s, '#') === 0) {
$mode = 'tag';
}
- $save_label = $mode === 'text' ? L10n::t('Save') : L10n::t('Follow');
+ $save_label = $mode === 'text' ? DI::l10n()->t('Save') : DI::l10n()->t('Follow');
$values = [
- '$s' => $s,
- '$id' => $id,
- '$action_url' => $url,
- '$search_label' => L10n::t('Search'),
- '$save_label' => $save_label,
- '$savedsearch' => 'savedsearch',
- '$search_hint' => L10n::t('@name, !forum, #tags, content'),
- '$mode' => $mode
- ];
+ '$s' => $s,
+ '$q' => urlencode($s),
+ '$id' => $id,
+ '$search_label' => DI::l10n()->t('Search'),
+ '$save_label' => $save_label,
+ '$search_hint' => DI::l10n()->t('@name, !forum, #tags, content'),
+ '$mode' => $mode,
+ '$return_url' => urlencode(Search::getSearchPath($s)),
+ ];
if (!$aside) {
- $values['$searchoption'] = [
- L10n::t("Full Text"),
- L10n::t("Tags"),
- L10n::t("Contacts")];
+ $values['$search_options'] = [
+ 'fulltext' => DI::l10n()->t('Full Text'),
+ 'tags' => DI::l10n()->t('Tags'),
+ 'contacts' => DI::l10n()->t('Contacts')
+ ];
- if (Config::get('system', 'poco_local_search')) {
- $values['$searchoption'][] = L10n::t("Forums");
+ if (DI::config()->get('system', 'poco_local_search')) {
+ $values['$searchoption']['forums'] = DI::l10n()->t('Forums');
}
}
return Renderer::replaceMacros(Renderer::getMarkupTemplate('searchbox.tpl'), $values);
}
- /**
- * Replace naked text hyperlink with HTML formatted hyperlink
- *
- * @param string $s
- * @return string
- */
- public static function toLink($s)
- {
- $s = preg_replace("/(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\'\%\$\!\+]*)/", ' <a href="$1" target="_blank">$1</a>', $s);
- $s = preg_replace("/\<(.*?)(src|href)=(.*?)\&\;(.*?)\>/ism", '<$1$2=$3&$4>', $s);
- return $s;
- }
-
/**
* Given a HTML text and a set of filtering reasons, adds a content hiding header with the provided reasons
*
* @return string
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function applyContentFilter($html, array $reasons)
+ public static function applyContentFilter(string $html, array $reasons): string
{
if (count($reasons)) {
$tpl = Renderer::getMarkupTemplate('wall/content_filter.tpl');
$html = Renderer::replaceMacros($tpl, [
'$reasons' => $reasons,
'$rnd' => Strings::getRandomHex(8),
- '$openclose' => L10n::t('Click to open/close'),
+ '$openclose' => DI::l10n()->t('Click to open/close'),
'$html' => $html
]);
}
* @param string $s
* @return string
*/
- public static function unamp($s)
+ public static function unamp(string $s): string
{
return str_replace('&', '&', $s);
}
+
+ /**
+ * Clean an HTML text for potentially harmful code
+ *
+ * @param string $text
+ * @param array $allowedIframeDomains List of allowed iframe source domains without the scheme
+ * @return string
+ */
+ public static function purify(string $text, array $allowedIframeDomains = []): string
+ {
+ // Allows cid: URL scheme
+ \HTMLPurifier_URISchemeRegistry::instance()->register('cid', new HTMLPurifier_URIScheme_cid());
+
+ $config = \HTMLPurifier_HTML5Config::createDefault();
+ $config->set('HTML.Doctype', 'HTML5');
+
+ // Used to remove iframe with src attribute filtered out
+ $config->set('AutoFormat.RemoveEmpty', true);
+
+ $config->set('HTML.SafeIframe', true);
+
+ array_walk($allowedIframeDomains, function (&$domain) {
+ // Allow the domain and all its eventual sub-domains
+ $domain = '(?:(?!-)[A-Za-z0-9-]{1,63}(?<!-)\.)*' . preg_quote(trim($domain, '/'), '%');
+ });
+
+ $config->set('URI.SafeIframeRegexp',
+ '%^https://(?:
+ ' . implode('|', $allowedIframeDomains) . '
+ )
+ (?:/|$) # Prevents bogus domains like youtube.com.fake.tld
+ %xi'
+ );
+
+ $config->set('Attr.AllowedRel', [
+ 'noreferrer' => true,
+ 'noopener' => true,
+ 'tag' => true,
+ ]);
+ $config->set('Attr.AllowedFrameTargets', [
+ '_blank' => true,
+ ]);
+
+ $config->set('AutoFormat.RemoveEmpty.Predicate', [
+ 'colgroup' => [], // |
+ 'th' => [], // |
+ 'td' => [], // |
+ 'iframe' => ['src'], // ↳ Default HTMLPurify values
+ 'i' => ['class'], // Allows forkawesome icons
+ ]);
+
+ // Uncomment to debug HTMLPurifier behavior
+ //$config->set('Core.CollectErrors', true);
+ //$config->set('Core.MaintainLineNumbers', true);
+
+ $HTMLPurifier = new \HTMLPurifier($config);
+
+ $text = $HTMLPurifier->purify($text);
+
+ /** @var \HTMLPurifier_ErrorCollector $errorCollector */
+ // Uncomment to debug HTML Purifier behavior
+ //$errorCollector = $HTMLPurifier->context->get('ErrorCollector');
+ //var_dump($errorCollector->getRaw());
+
+ return $text;
+ }
+
+ /**
+ * XPath arbitrary string quoting
+ *
+ * @see https://stackoverflow.com/a/45228168
+ * @param string $value
+ * @return string
+ */
+ public static function xpathQuote(string $value): string
+ {
+ if (false === strpos($value, '"')) {
+ return '"' . $value . '"';
+ }
+
+ if (false === strpos($value, "'")) {
+ return "'" . $value . "'";
+ }
+
+ // if the value contains both single and double quotes, construct an
+ // expression that concatenates all non-double-quote substrings with
+ // the quotes, e.g.:
+ //
+ // concat("'foo'", '"', "bar")
+ return 'concat(' . implode(', \'"\', ', array_map([self::class, 'xpathQuote'], explode('"', $value))) . ')';
+ }
+
+ /**
+ * Checks if the provided URL is present in the DOM document in an element with the rel="me" attribute
+ *
+ * XHTML Friends Network http://gmpg.org/xfn/
+ *
+ * @param DOMDocument $doc
+ * @param UriInterface $meUrl
+ * @return bool
+ */
+ public static function checkRelMeLink(DOMDocument $doc, UriInterface $meUrl): bool
+ {
+ $xpath = new \DOMXpath($doc);
+
+ // This expression checks that "me" is among the space-delimited values of the "rel" attribute.
+ // And that the href attribute contains exactly the provided URL
+ $expression = "//*[contains(concat(' ', normalize-space(@rel), ' '), ' me ')][@href = " . self::xpathQuote($meUrl) . "]";
+
+ $result = $xpath->query($expression);
+
+ return $result !== false && $result->length > 0;
+ }
+
+ /**
+ * @param DOMDocument $doc
+ * @return string|null Lowercase charset
+ */
+ public static function extractCharset(DOMDocument $doc): ?string
+ {
+ $xpath = new DOMXPath($doc);
+
+ $expression = "string(//meta[@charset]/@charset)";
+ if ($charset = $xpath->evaluate($expression)) {
+ return strtolower($charset);
+ }
+
+ try {
+ // This expression looks for a meta tag with the http-equiv attribute set to "content-type" ignoring case
+ // whose content attribute contains a "charset" string and returns its value
+ $expression = "string(//meta[@http-equiv][translate(@http-equiv, 'ABCDEFGHIJKLMNOPQRSTUVWXYZ', 'abcdefghijklmnopqrstuvwxyz') = 'content-type'][contains(translate(@content, 'ABCDEFGHIJKLMNOPQRSTUVWXYZ', 'abcdefghijklmnopqrstuvwxyz'), 'charset')]/@content)";
+ $mediaType = MediaType::fromContentType($xpath->evaluate($expression));
+ if (isset($mediaType->parameters['charset'])) {
+ return strtolower($mediaType->parameters['charset']);
+ }
+ } catch(\InvalidArgumentException $e) {}
+
+ return null;
+ }
}