]> git.mxchange.org Git - friendica.git/blobdiff - src/Content/Text/BBCode.php
Function moved to the "Strings" class
[friendica.git] / src / Content / Text / BBCode.php
index c9bbc02c934f414d68140c01c7ec8452c497bf2e..0e5ee9d4e2e03600e470e8ad9a5f5fcb91a0f79e 100644 (file)
@@ -36,14 +36,10 @@ use Friendica\Core\Renderer;
 use Friendica\DI;
 use Friendica\Model\Contact;
 use Friendica\Model\Event;
-use Friendica\Model\Photo;
 use Friendica\Model\Post;
 use Friendica\Model\Tag;
 use Friendica\Network\HTTPClient\Client\HttpClientAccept;
 use Friendica\Network\HTTPClient\Client\HttpClientOptions;
-use Friendica\Object\Image;
-use Friendica\Protocol\Activity;
-use Friendica\Util\Images;
 use Friendica\Util\Map;
 use Friendica\Util\ParseUrl;
 use Friendica\Util\Proxy;
@@ -61,6 +57,7 @@ class BBCode
        const DIASPORA     = 3;
        const CONNECTORS   = 4;
        const TWITTER_API  = 5;
+       const NPF          = 6;
        const OSTATUS      = 7;
        const TWITTER      = 8;
        const BACKLINK     = 8;
@@ -511,26 +508,7 @@ class BBCode
         */
        private static function convertUrlForActivityPub(string $url): string
        {
-               return sprintf('<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>', $url, self::getStyledURL($url));
-       }
-
-       /**
-        * Converts an URL in a nicer format (without the scheme and possibly shortened)
-        *
-        * @param string $url URL that is about to be reformatted
-        * @return string reformatted link
-        */
-       private static function getStyledURL(string $url): string
-       {
-               $parts = parse_url($url);
-               $scheme = $parts['scheme'] . '://';
-               $styled_url = str_replace($scheme, '', $url);
-
-               if (strlen($styled_url) > 30) {
-                       $styled_url = substr($styled_url, 0, 30) . "…";
-               }
-
-               return $styled_url;
+               return sprintf('<a href="%s" target="_blank" rel="noopener noreferrer">%s</a>', $url, Strings::getStyledURL($url));
        }
 
        /*
@@ -552,7 +530,7 @@ class BBCode
 
        /*
         * The previously spacefied [noparse][ i ]italic[ /i ][/noparse],
-        * now turns back and the [noparse] tags are trimed
+        * now turns back and the [noparse] tags are trimmed
         * returning [i]italic[/i]
         *
         * @param array $match
@@ -1316,7 +1294,8 @@ class BBCode
                                        "\n[tr]", "[tr]\n", " [tr]", "[tr] ", "\n[/tr]", "[/tr]\n", " [/tr]", "[/tr] ",
                                        "\n[hr]", "[hr]\n", " [hr]", "[hr] ",
                                        "\n[attachment ", " [attachment ", "\n[/attachment]", "[/attachment]\n", " [/attachment]", "[/attachment] ",
-                                       "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] "
+                                       "[table]\n", "[table] ", " [table]", "\n[/table]", " [/table]", "[/table] ",
+                                       " \n", "\t\n", "[/li]\n", "\n[li]", "\n[*]", 
                                ];
                                $replace = [
                                        "[th]", "[th]", "[th]", "[/th]", "[/th]", "[/th]",
@@ -1324,7 +1303,8 @@ class BBCode
                                        "[tr]", "[tr]", "[tr]", "[tr]", "[/tr]", "[/tr]", "[/tr]", "[/tr]",
                                        "[hr]", "[hr]", "[hr]", "[hr]",
                                        "[attachment ", "[attachment ", "[/attachment]", "[/attachment]", "[/attachment]", "[/attachment]",
-                                       "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]"
+                                       "[table]", "[table]", "[table]", "[/table]", "[/table]", "[/table]",
+                                       "\n", "\n", "[/li]", "[li]", "[*]",
                                ];
                                do {
                                        $oldtext = $text;
@@ -1342,11 +1322,11 @@ class BBCode
                                // removing multiplicated newlines
                                if (DI::config()->get('system', 'remove_multiplicated_lines')) {
                                        $search = [
-                                               "\n\n\n", "\n ", " \n", "[/quote]\n\n", "\n[/quote]", "[/li]\n", "\n[li]", "\n[*]", "\n[ul]", "[/ul]\n", "\n\n[share ", "[/attachment]\n",
+                                               "\n\n\n", "[/quote]\n\n", "\n[/quote]", "\n[ul]", "[/ul]\n", "\n[ol]", "[/ol]\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]",
+                                               "\n\n", "[/quote]\n", "[/quote]", "[ul]", "[/ul]", "[ol]", "[/ol]", "\n[share ", "[/attachment]",
                                                "[h1]", "[/h1]", "[h2]", "[/h2]", "[h3]", "[/h3]", "[h4]", "[/h4]", "[h5]", "[/h5]", "[h6]", "[/h6]"
                                        ];
                                        do {
@@ -1357,7 +1337,9 @@ class BBCode
 
                                /// @todo Have a closer look at the different html modes
                                // Handle attached links or videos
-                               if (in_array($simple_html, [self::MASTODON_API, self::TWITTER_API, self::ACTIVITYPUB])) {
+                               if ($simple_html == self::NPF) {
+                                       $text = self::removeAttachment($text);
+                               } elseif (in_array($simple_html, [self::MASTODON_API, self::TWITTER_API, self::ACTIVITYPUB])) {
                                        $text = self::replaceAttachment($text);
                                } elseif (!in_array($simple_html, [self::INTERNAL, self::EXTERNAL, self::CONNECTORS])) {
                                        $text = self::replaceAttachment($text, true);
@@ -1375,7 +1357,7 @@ class BBCode
                                        });
                                }
 
-                               // leave open the posibility of [map=something]
+                               // leave open the possibility of [map=something]
                                // this is replaced in Item::prepareBody() which has knowledge of the item location
                                if (strpos($text, '[/map]') !== false) {
                                        $text = preg_replace_callback(
@@ -1402,12 +1384,32 @@ class BBCode
                                }
 
                                // Check for headers
-                               $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '<h1>$1</h1>', $text);
-                               $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '<h2>$1</h2>', $text);
-                               $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '<h3>$1</h3>', $text);
-                               $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '<h4>$1</h4>', $text);
-                               $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '<h5>$1</h5>', $text);
-                               $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '<h6>$1</h6>', $text);
+
+                               if ($simple_html == self::INTERNAL) {
+                                       //Ensure to always start with <h4> if possible
+                                       $heading_count = 0;
+                                       for ($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--) { 
+                                                       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--;
+                                                       }
+                                               }
+                                       }
+                               } else {
+                                       $text = preg_replace("(\[h1\](.*?)\[\/h1\])ism", '</p><h1>$1</h1><p>', $text);
+                                       $text = preg_replace("(\[h2\](.*?)\[\/h2\])ism", '</p><h2>$1</h2><p>', $text);
+                                       $text = preg_replace("(\[h3\](.*?)\[\/h3\])ism", '</p><h3>$1</h3><p>', $text);
+                                       $text = preg_replace("(\[h4\](.*?)\[\/h4\])ism", '</p><h4>$1</h4><p>', $text);
+                                       $text = preg_replace("(\[h5\](.*?)\[\/h5\])ism", '</p><h5>$1</h5><p>', $text);
+                                       $text = preg_replace("(\[h6\](.*?)\[\/h6\])ism", '</p><h6>$1</h6><p>', $text);
+                               }
 
                                // Check for paragraph
                                $text = preg_replace("(\[p\](.*?)\[\/p\])ism", '<p>$1</p>', $text);
@@ -1470,27 +1472,27 @@ class BBCode
                                        ((strpos($text, "[/ol]") !== false) && (strpos($text, "[ol]") !== false)) ||
                                        ((strpos($text, "[/ul]") !== false) && (strpos($text, "[ul]") !== false)) ||
                                        ((strpos($text, "[/li]") !== false) && (strpos($text, "[li]") !== false))) && (++$endlessloop < 20)) {
-                                       $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
-                                       $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '<ul class="listnone" style="list-style-type: none;">$1</ul>', $text);
-                                       $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
-                                       $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '<ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul>', $text);
-                                       $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '<ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul>', $text);
-                                       $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '<ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul>', $text);
-                                       $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '<ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul>', $text);
-                                       $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '<ul class="listbullet" style="list-style-type: circle;">$1</ul>', $text);
-                                       $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '<ul class="listdecimal" style="list-style-type: decimal;">$1</ul>', $text);
+                                       $text = preg_replace("/\[list\](.*?)\[\/list\]/ism", '</p><ul class="listbullet" style="list-style-type: circle;">$1</ul><p>', $text);
+                                       $text = preg_replace("/\[list=\](.*?)\[\/list\]/ism", '</p><ul class="listnone" style="list-style-type: none;">$1</ul><p>', $text);
+                                       $text = preg_replace("/\[list=1\](.*?)\[\/list\]/ism", '</p><ul class="listdecimal" style="list-style-type: decimal;">$1</ul><p>', $text);
+                                       $text = preg_replace("/\[list=((?-i)i)\](.*?)\[\/list\]/ism", '</p><ul class="listlowerroman" style="list-style-type: lower-roman;">$2</ul><p>', $text);
+                                       $text = preg_replace("/\[list=((?-i)I)\](.*?)\[\/list\]/ism", '</p><ul class="listupperroman" style="list-style-type: upper-roman;">$2</ul><p>', $text);
+                                       $text = preg_replace("/\[list=((?-i)a)\](.*?)\[\/list\]/ism", '</p><ul class="listloweralpha" style="list-style-type: lower-alpha;">$2</ul><p>', $text);
+                                       $text = preg_replace("/\[list=((?-i)A)\](.*?)\[\/list\]/ism", '</p><ul class="listupperalpha" style="list-style-type: upper-alpha;">$2</ul><p>', $text);
+                                       $text = preg_replace("/\[ul\](.*?)\[\/ul\]/ism", '</p><ul>$1</ul><p>', $text);
+                                       $text = preg_replace("/\[ol\](.*?)\[\/ol\]/ism", '</p><ol>$1</ol><p>', $text);
                                        $text = preg_replace("/\[li\](.*?)\[\/li\]/ism", '<li>$1</li>', $text);
                                }
 
                                $text = preg_replace("/\[th\](.*?)\[\/th\]/sm", '<th>$1</th>', $text);
                                $text = preg_replace("/\[td\](.*?)\[\/td\]/sm", '<td>$1</td>', $text);
                                $text = preg_replace("/\[tr\](.*?)\[\/tr\]/sm", '<tr>$1</tr>', $text);
-                               $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '<table>$1</table>', $text);
+                               $text = preg_replace("/\[table\](.*?)\[\/table\]/sm", '</p><table>$1</table><p>', $text);
 
-                               $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '<table border="1" >$1</table>', $text);
-                               $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '<table border="0" >$1</table>', $text);
+                               $text = preg_replace("/\[table border=1\](.*?)\[\/table\]/sm", '</p><table border="1" >$1</table><p>', $text);
+                               $text = preg_replace("/\[table border=0\](.*?)\[\/table\]/sm", '</p><table border="0" >$1</table><p>', $text);
 
-                               $text = str_replace('[hr]', '<hr />', $text);
+                               $text = str_replace('[hr]', '</p><hr /><p>', $text);
 
                                if (!$for_plaintext) {
                                        $text = self::performWithEscapedTags($text, ['url', 'img', 'audio', 'video', 'youtube', 'vimeo', 'share', 'attachment', 'iframe', 'bookmark'], function ($text) {
@@ -1524,7 +1526,7 @@ class BBCode
                                }
 
                                // Declare the format for [quote] layout
-                               $QuoteLayout = '<blockquote>$1</blockquote>';
+                               $QuoteLayout = '</p><blockquote>$1</blockquote><p>';
 
                                // Check for [quote] text
                                // handle nested quotes
@@ -1566,11 +1568,16 @@ class BBCode
                                $text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
 
                                $text = preg_replace_callback(
-                                       "/\[img\=(.*?)\](.*?)\[\/img\]/ism",
+                                       "/\[[iz]mg\=(.*?)\](.*?)\[\/[iz]mg\]/ism",
                                        function ($matches) use ($simple_html, $uriid) {
                                                $matches[1] = self::proxyUrl($matches[1], $simple_html, $uriid);
-                                               $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
-                                               return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '" title="' . $matches[2] . '">';
+                                               $alt = htmlspecialchars($matches[2], ENT_COMPAT);
+                                               // Fix for Markdown problems with Diaspora, see issue #12701
+                                               if (($simple_html != self::DIASPORA) || strpos($matches[2], '"') === false) {
+                                                       return '<img src="' . $matches[1] . '" alt="' . $alt . '" title="' . $alt . '">';
+                                               } else {
+                                                       return '<img src="' . $matches[1] . '" alt="' . $alt . '">';
+                                               }
                                        },
                                        $text
                                );
@@ -1578,7 +1585,7 @@ class BBCode
                                // Images
                                // [img]pathtoimage[/img]
                                $text = preg_replace_callback(
-                                       "/\[img\](.*?)\[\/img\]/ism",
+                                       "/\[[iz]mg\](.*?)\[\/[iz]mg\]/ism",
                                        function ($matches) use ($simple_html, $uriid) {
                                                if (strpos($matches[1], "data:image/") === 0) {
                                                        return $matches[0];
@@ -1602,7 +1609,18 @@ class BBCode
                                // Simplify "video" element
                                $text = preg_replace('(\[video[^\]]*?\ssrc\s?=\s?([^\s\]]+)[^\]]*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
 
-                               if ($try_oembed) {
+                               if ($simple_html == self::NPF) {
+                                       $text = preg_replace(
+                                               "/\[video\](.*?)\[\/video\]/ism",
+                                               '</p><video src="$1" controls width="100%" height="auto">$1</video><p>',
+                                               $text
+                                       );
+                                       $text = preg_replace(
+                                               "/\[audio\](.*?)\[\/audio\]/ism",
+                                               '</p><audio src="$1" controls>$1">$1</audio><p>',
+                                               $text
+                                       );
+                               } elseif ($try_oembed) {
                                        // html5 video and audio
                                        $text = preg_replace(
                                                "/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism",
@@ -1636,13 +1654,6 @@ class BBCode
                                $text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $text);
 
                                // Youtube extensions
-                               if ($try_oembed) {
-                                       $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
-                                       $text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
-                                       $text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/shorts\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
-                                       $text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism", $try_oembed_callback, $text);
-                               }
-
                                $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
                                $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
                                $text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/shorts\/(.*?)\[\/youtube\]/ism", '[youtube]$1[/youtube]', $text);
@@ -1658,11 +1669,6 @@ class BBCode
                                        );
                                }
 
-                               if ($try_oembed) {
-                                       $text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
-                                       $text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism", $try_oembed_callback, $text);
-                               }
-
                                $text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
                                $text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism", '[vimeo]$1[/vimeo]', $text);
 
@@ -1939,7 +1945,11 @@ class BBCode
                                : []
                );
 
-               $text = HTML::purify('<p>' . $text . '</p>', $allowedIframeDomains);
+               if (strpos($text, '<p>') !== false || strpos($text, '</p>') !== false) {
+                       $text = '<p>' . $text . '</p>';
+               }
+
+               $text = HTML::purify($text, $allowedIframeDomains);
                DI::profiler()->stopRecording();
 
                return trim($text);