<?php
/**
- * @copyright Copyright (C) 2010-2023, the Friendica project
+ * @copyright Copyright (C) 2010-2024, the Friendica project
*
* @license GNU AGPL version 3 or any later version
*
{
DI::profiler()->startRecording('rendering');
// Remove pictures in advance to avoid unneeded proxy calls
+ $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", ' ', $text);
$text = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $2 ', $text);
$text = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $text);
// Remove attachment
$text = self::replaceAttachment($text);
- $naked_text = HTML::toPlaintext(self::convert($text, false, BBCode::EXTERNAL, true), 0, !$keep_urls);
+ $naked_text = HTML::toPlaintext(self::convert($text, false, self::EXTERNAL, true), 0, !$keep_urls);
DI::profiler()->stopRecording();
return $naked_text;
}
+ /**
+ * Converts text into a format that can be used for the channel search and the language detection.
+ *
+ * @param string $text
+ * @param integer $uri_id
+ * @return string
+ */
+ public static function toSearchText(string $text, int $uri_id): string
+ {
+ // Removes attachments
+ $text = self::removeAttachment($text);
+
+ // Add images because of possible alt texts
+ if (!empty($uri_id)) {
+ $text = Post\Media::addAttachmentsToBody($uri_id, $text, [Post\Media::IMAGE]);
+ }
+
+ if (empty($text)) {
+ return '';
+ }
+
+ // Remove links without a link description
+ $text = preg_replace("~\[url\=.*\]https?:.*\[\/url\]~", ' ', $text);
+
+ // Remove pictures
+ $text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", ' ', $text);
+
+ // Replace picture with the alt description
+ $text = preg_replace("/\[img\=.*?\](.*?)\[\/img\]/ism", ' $1 ', $text);
+
+ // Remove the other pictures
+ $text = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $text);
+
+ // Removes mentions, remove links from hashtags
+ $text = preg_replace('/[@!]\[url\=.*?\].*?\[\/url\]/ism', ' ', $text);
+ $text = preg_replace('/[#]\[url\=.*?\](.*?)\[\/url\]/ism', ' #$1 ', $text);
+ $text = preg_replace('/[@!#]?\[url.*?\[\/url\]/ism', ' ', $text);
+ $text = preg_replace("/\[url=[^\[\]]*\](.*)\[\/url\]/Usi", ' $1 ', $text);
+
+ // Convert it to plain text
+ $text = self::toPlaintext($text, false);
+
+ // Remove possibly remaining links
+ $text = preg_replace(Strings::autoLinkRegEx(), '', $text);
+
+ // Remove all unneeded white space
+ do {
+ $oldtext = $text;
+ $text = str_replace([' ', "\n", "\r", '"'], ' ', $text);
+ } while ($oldtext != $text);
+
+ return trim($text);
+ }
+
private static function proxyUrl(string $image, int $simplehtml = self::INTERNAL, int $uriid = 0, string $size = ''): string
{
// Only send proxied pictures to API and for internal display
$network = $contact['network'] ?? Protocol::PHANTOM;
$tpl = Renderer::getMarkupTemplate('shared_content.tpl');
- $text .= BBCode::SHARED_ANCHOR . Renderer::replaceMacros($tpl, [
+ $text .= self::SHARED_ANCHOR . Renderer::replaceMacros($tpl, [
'$profile' => $attributes['profile'],
'$avatar' => $attributes['avatar'],
'$author' => $attributes['author'],
public static function removeLinks(string $bbcode): string
{
DI::profiler()->startRecording('rendering');
+ $bbcode = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", ' ', $bbcode);
$bbcode = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", ' $1 ', $bbcode);
$bbcode = preg_replace("/\[img.*?\[\/img\]/ism", ' ', $bbcode);
}
/**
- * Expand Youtube and Vimeo links to
+ * Expand Youtube and Vimeo links to
*
* @param string $text
* @return string
"\n[hr]", "[hr]\n", " [hr]", "[hr] ",
"\n[attachment ", " [attachment ", "\n[/attachment]", "[/attachment]\n", " [/attachment]", "[/attachment] ",
"[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] ",
- " \n", "\t\n", "[/li]\n", "\n[li]", "\n[*]",
+ " \n", "\t\n", "[/li]\n", "\n[li]", "\n[*]",
];
$replace = [
"[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
if ($simple_html == self::INTERNAL) {
//Ensure to always start with <h4> if possible
$heading_count = 0;
- for ($level = 6; $level > 0; $level--) {
+ for ($level = 6; $level > 0; $level--) {
if (preg_match("(\[h$level\].*?\[\/h$level\])ism", $text)) {
$heading_count++;
}
}
if ($heading_count > 0) {
$heading = min($heading_count + 3, 6);
- for ($level = 6; $level > 0; $level--) {
+ for ($level = 6; $level > 0; $level--) {
if (preg_match("(\[h$level\].*?\[\/h$level\])ism", $text)) {
$text = preg_replace("(\[h$level\](.*?)\[\/h$level\])ism", "</p><h$heading>$1</h$heading><p>", $text);
$heading--;
$text = preg_replace("(\[style=(.*?)\](.*?)\[\/style\])ism", '<span style="$1">$2</span>', $text);
// Mastodon Emoji (internal tag, do not document for users)
- $text = preg_replace("(\[emoji=(.*?)](.*?)\[/emoji])ism", '<span class="mastodon emoji"><img src="$1" alt="$2" title="$2"/></span>', $text);
+ if ($simple_html == self::MASTODON_API) {
+ $text = preg_replace("(\[emoji=(.*?)](.*?)\[/emoji])ism", '$2', $text);
+ } else {
+ $text = preg_replace("(\[emoji=(.*?)](.*?)\[/emoji])ism", '<span class="mastodon emoji"><img src="$1" alt="$2" title="$2"/></span>', $text);
+ }
// Check for CSS classes
// @deprecated since 2021.12, left for backward-compatibility reasons
$text
);
- // sanitize href attributes (only allowlisted protocols URLs)
+ // sanitize href attributes (only whitelisted protocols URLs)
// default value for backward compatibility
$allowed_link_protocols = DI::config()->get('system', 'allowed_link_protocols', []);
{
DI::profiler()->startRecording('rendering');
- $text = BBCode::performWithEscapedTags($text, ['code', 'noparse', 'nobb', 'pre'], function ($text) {
+ $text = self::performWithEscapedTags($text, ['code', 'noparse', 'nobb', 'pre'], function ($text) {
$text = preg_replace("/[\s|\n]*\[abstract\].*?\[\/abstract\][\s|\n]*/ism", ' ', $text);
$text = preg_replace("/[\s|\n]*\[abstract=.*?\].*?\[\/abstract][\s|\n]*/ism", ' ', $text);
return $text;
DI::profiler()->startRecording('rendering');
$addon = strtolower($addon);
- $abstract = BBCode::performWithEscapedTags($text, ['code', 'noparse', 'nobb', 'pre'], function ($text) use ($addon) {
+ $abstract = self::performWithEscapedTags($text, ['code', 'noparse', 'nobb', 'pre'], function ($text) use ($addon) {
if ($addon && preg_match('#\[abstract=' . preg_quote($addon, '#') . '](.*?)\[/abstract]#ism', $text, $matches)) {
return $matches[1];
}
// Maybe we should make this newline at every time before a quote.
$text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
+ // The converter doesn't convert these elements
+ $text = str_replace(['<div>', '</div>'], ['<p>', '</p>'], $text);
+
// Now convert HTML to Markdown
$text = HTML::toMarkdown($text);