]> git.mxchange.org Git - friendica.git/blobdiff - include/bb2diaspora.php
Just removing unused codelines
[friendica.git] / include / bb2diaspora.php
old mode 100755 (executable)
new mode 100644 (file)
index 8f17f2e..7107c49
 <?php
 
 require_once("include/oembed.php");
-require_once('include/event.php');
+require_once("include/event.php");
+require_once("library/markdown.php");
+require_once("include/html2bbcode.php");
+require_once("include/bbcode.php");
+require_once("include/markdownify/markdownify.php");
 
-require_once('library/markdown.php');
-require_once('include/html2bbcode.php');
 
 // we don't want to support a bbcode specific markdown interpreter
 // and the markdown library we have is pretty good, but provides HTML output.
-// So we'll use that to convert to HTML, then convert the HTML back to bbcode, 
+// So we'll use that to convert to HTML, then convert the HTML back to bbcode,
 // and then clean up a few Diaspora specific constructs.
 
 function diaspora2bb($s) {
 
        $s = html_entity_decode($s,ENT_COMPAT,'UTF-8');
 
-       $s = preg_replace('/\@\{(.+?)\; (.+?)\@(.+?)\}/','@[url=https://$3/u/$2]$1[/url]',$s);
-       $s = Markdown($s);
-       $s = html2bbcode($s);
-//     $s = str_replace('&#42;','*',$s);
+       // Simply remove cr.
+       $s = str_replace("\r","",$s);
 
-       // we seem to get a lot of text smushed together with links from Diaspora.
-       // if it's a url that we haven't already parsed into a bbcode structure, put a space before it.
-       $s = preg_replace("/([^=\"\]])(https?:\/\/)/ism",'$1 $2',$s);
+       // <br/> is invalid. Replace it with the valid expression
+       $s = str_replace(array("<br/>", "</p>", "<p>", '<p dir="ltr">'),array("<br />", "<br />", "<br />", "<br />"),$s);
 
-    $s = preg_replace("/\[url\=?(.*?)\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/url\]/ism",'[youtube]$2[/youtube]',$s); 
-    $s = preg_replace("/\[url\=https?:\/\/www.youtube.com\/watch\?v\=(.*?)\].*?\[\/url\]/ism",'[youtube]$1[/youtube]',$s); 
-       $s = preg_replace("/\[url\=?(.*?)\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/url\]/ism",'[vimeo]$2[/vimeo]',$s); 
-       $s = preg_replace("/\[url\=https?:\/\/vimeo.com\/([0-9]+)\](.*?)\[\/url\]/ism",'[vimeo]$1[/vimeo]',$s); 
-       $s = preg_replace("/([^\]\=]|^)(https?\:\/\/)(vimeo|youtu|www\.youtube|soundcloud)([a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2$3$4[/url]',$s);
-       // remove duplicate adjacent code tags
-       $s = preg_replace("/(\[code\])+(.*?)(\[\/code\])+/ism","[code]$2[/code]", $s);
-       $s = scale_diaspora_images($s);
-       return $s;
-}
+       $s = preg_replace('/\@\{(.+?)\; (.+?)\@(.+?)\}/','@[url=https://$3/u/$2]$1[/url]',$s);
 
+       // Escaping the hash tags
+       $s = preg_replace('/\#([^\s\#])/','&#35;$1',$s);
 
-function scale_diaspora_images($s,$include_link = true) {
-
-       $matches = null;
-       $c = preg_match_all('/\[img\](.*?)\[\/img\]/ism',$s,$matches,PREG_SET_ORDER);
-       if($c) {
-               require_once('include/Photo.php');
-               foreach($matches as $mtch) {
-                       logger('scale_diaspora_image: ' . $mtch[1]);
-                       $i = fetch_url($mtch[1]);
-                       if($i) {
-                               $ph = new Photo($i);
-                               if($ph->is_valid()) {
-                                       if($ph->getWidth() > 600 || $ph->getHeight() > 600) {
-                                               $ph->scaleImage(600);
-                                               $new_width = $ph->getWidth();
-                                               $new_height = $ph->getHeight();
-                                               logger('scale_diaspora_image: ' . $new_width . 'w ' . $new_height . 'h' . 'match: ' . $mtch[0], LOGGER_DEBUG);
-                                               $s = str_replace($mtch[0],'[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
-                                                       . "\n" . (($include_link) 
-                                                               ? '[url=' . $mtch[1] . ']' . t('view full size') . '[/url]' . "\n"
-                                                               : ''),$s);
-                                               logger('scale_diaspora_image: new string: ' . $s, LOGGER_DEBUG);
-                                       }
-                               }
-                       }
-               }
-       }
-       return $s;
-}
+       $s = Markdown($s);
 
-function stripdcode_br_cb($s) {
-       return '[code]' . str_replace('<br />', "\n\t", $s[1]) . '[/code]';
-}
+       $s = str_replace('&#35;','#',$s);
 
+       $s = html2bbcode($s);
 
+       // protect the recycle symbol from turning into a tag, but without unescaping angles and naked ampersands
+       $s = str_replace('&#x2672;',html_entity_decode('&#x2672;',ENT_QUOTES,'UTF-8'),$s);
 
-function bb2diaspora($Text,$preserve_nl = false) {
+       // Convert everything that looks like a link to a link
+       $s = preg_replace("/([^\]\=]|^)(https?\:\/\/)([a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url=$2$3]$2$3[/url]',$s);
 
-       $ev = bbtoevent($Text);
+       //$s = preg_replace("/([^\]\=]|^)(https?\:\/\/)(vimeo|youtu|www\.youtube|soundcloud)([a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url=$2$3$4]$2$3$4[/url]',$s);
+       $s = bb_tag_preg_replace("/\[url\=?(.*?)\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/url\]/ism",'[youtube]$2[/youtube]','url',$s);
+       $s = bb_tag_preg_replace("/\[url\=https?:\/\/www.youtube.com\/watch\?v\=(.*?)\].*?\[\/url\]/ism",'[youtube]$1[/youtube]','url',$s);
+       $s = bb_tag_preg_replace("/\[url\=?(.*?)\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/url\]/ism",'[vimeo]$2[/vimeo]','url',$s);
+       $s = bb_tag_preg_replace("/\[url\=https?:\/\/vimeo.com\/([0-9]+)\](.*?)\[\/url\]/ism",'[vimeo]$1[/vimeo]','url',$s);
+       // remove duplicate adjacent code tags
+       $s = preg_replace("/(\[code\])+(.*?)(\[\/code\])+/ism","[code]$2[/code]", $s);
 
-       // Replace any html brackets with HTML Entities to prevent executing HTML or script
-       // Don't use strip_tags here because it breaks [url] search by replacing & with amp
+       // Don't show link to full picture (until it is fixed)
+       $s = scale_external_images($s, false);
 
-       $Text = str_replace("<", "&lt;", $Text);
-       $Text = str_replace(">", "&gt;", $Text);
+       return $s;
+}
 
-       // If we find any event code, turn it into an event.
-       // After we're finished processing the bbcode we'll 
-       // replace all of the event code with a reformatted version.
+function bb2diaspora($Text,$preserve_nl = false, $fordiaspora = true) {
 
+       $OriginalText = $Text;
 
-       if($preserve_nl)
-               $Text = str_replace(array("\n","\r"), array('',''),$Text);
+       // Since Diaspora is creating a summary for links, this function removes them before posting
+       if ($fordiaspora)
+               $Text = bb_remove_share_information($Text);
 
-       // Set up the parameters for a URL search string
+       /**
+        * Transform #tags, strip off the [url] and replace spaces with underscore
+        */
        $URLSearchString = "^\[\]";
-       // Set up the parameters for a MAIL search string
-       $MAILSearchString = $URLSearchString;
-
-       // Perform URL Search
-
-       // [img]pathtoimage[/img]
-
-       // the following was added on 10-January-2012 due to an inability of Diaspora's
-       // new javascript markdown processor to handle links with images as the link "text"
-       // It is not optimal and may be removed if this ability is restored in the future
-
-       $Text = preg_replace("/\[url\=([$URLSearchString]*)\]\[img\](.*?)\[\/img\]\[\/url\]/ism", 
-               '![' . t('image/photo') . '](' . '$2' . ')' . "\n" . '[' . t('link') . '](' . '$1' . ')', $Text);
-
-       $Text = preg_replace("/\[bookmark\]([$URLSearchString]*)\[\/bookmark\]/ism", '[$1]($1)', $Text);
-       $Text = preg_replace("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", '[$2]($1)', $Text);
-
-       $Text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '[$1]($1)', $Text);
-       $Text = preg_replace("/\#\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[#$2]($1)', $Text);
-       $Text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[$2]($1)', $Text);
-
-
-       $Text = preg_replace("/\[img\](.*?)\[\/img\]/", '![' . t('image/photo') . '](' . '$1' . ')', $Text);
+       $Text = preg_replace_callback("/#\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i", create_function('$match',
+               'return \'#\'. str_replace(\' \', \'_\', $match[2]);'
+       ), $Text);
 
-       // Perform MAIL Search
-       $Text = preg_replace("(\[mail\]([$MAILSearchString]*)\[/mail\])", '[$1](mailto:$1)', $Text);
-       $Text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '[$2](mailto:$1)', $Text);
-         
-       $Text = str_replace('*', '\\*', $Text);
-       $Text = str_replace('_', '\\_', $Text);
 
-       $Text = str_replace('`','\\`', $Text);
+       // Converting images with size parameters to simple images. Markdown doesn't know it.
+       $Text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $Text);
 
-       // Check for bold text
-       $Text = preg_replace("(\[b\](.*?)\[\/b\])is",'**$1**',$Text);
+       // Convert it to HTML - don't try oembed
+       if ($fordiaspora) {
+               $Text = bbcode($Text, $preserve_nl, false, 3);
 
-       // Check for Italics text
-       $Text = preg_replace("(\[i\](.*?)\[\/i\])is",'_$1_',$Text);
+               // Add all tags that maybe were removed
+               if (preg_match_all("/#\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",$OriginalText, $tags)) {
+                       $tagline = "";
+                       foreach($tags[2] as $tag)
+                               if (!strpos($Text, "#".$tag))
+                                       $tagline .= "#".$tag." ";
 
-       // Check for Underline text
-//     $Text = preg_replace("(\[u\](.*?)\[\/u\])is",'<u>$1</u>',$Text);
-
-       // Check for strike-through text
-//     $Text = preg_replace("(\[s\](.*?)\[\/s\])is",'<strike>$1</strike>',$Text);
-
-       // Check for over-line text
-//     $Text = preg_replace("(\[o\](.*?)\[\/o\])is",'<span class="overline">$1</span>',$Text);
-
-       // Check for colored text
-//     $Text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])is","<span style=\"color: $1;\">$2</span>",$Text);
-
-       // Check for sized text
-//     $Text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])is","<span style=\"font-size: $1;\">$2</span>",$Text);
-
-       // Check for list text
-//     $Text = preg_replace("/\[list\](.*?)\[\/list\]/is", '<ul class="listbullet">$1</ul>' ,$Text);
-//     $Text = preg_replace("/\[list=1\](.*?)\[\/list\]/is", '<ul class="listdecimal">$1</ul>' ,$Text);
-//     $Text = preg_replace("/\[list=i\](.*?)\[\/list\]/s",'<ul class="listlowerroman">$1</ul>' ,$Text);
-//     $Text = preg_replace("/\[list=I\](.*?)\[\/list\]/s", '<ul class="listupperroman">$1</ul>' ,$Text);
-//     $Text = preg_replace("/\[list=a\](.*?)\[\/list\]/s", '<ul class="listloweralpha">$1</ul>' ,$Text);
-//     $Text = preg_replace("/\[list=A\](.*?)\[\/list\]/s", '<ul class="listupperalpha">$1</ul>' ,$Text);
-//     $Text = preg_replace("/\[li\](.*?)\[\/li\]/s", '<li>$1</li>' ,$Text);
-
-//     $Text = preg_replace("/\[td\](.*?)\[\/td\]/s", '<td>$1</td>' ,$Text);
-//     $Text = preg_replace("/\[tr\](.*?)\[\/tr\]/s", '<tr>$1</tr>' ,$Text);
-//     $Text = preg_replace("/\[table\](.*?)\[\/table\]/s", '<table>$1</table>' ,$Text);
-
-//     $Text = preg_replace("/\[table border=1\](.*?)\[\/table\]/s", '<table border="1" >$1</table>' ,$Text);
-//     $Text = preg_replace("/\[table border=0\](.*?)\[\/table\]/s", '<table border="0" >$1</table>' ,$Text);
-
-       
-//     $Text = str_replace("[*]", "<li>", $Text);
-
-       // Check for font change text
-//     $Text = preg_replace("(\[font=(.*?)\](.*?)\[\/font\])","<span style=\"font-family: $1;\">$2</span>",$Text);
-
-
-    $Text = preg_replace_callback("/\[code\](.*?)\[\/code\]/is",'stripdcode_br_cb',$Text);
-
-       // Check for [code] text
-       $Text = preg_replace("/(\[code\])+(.*?)(\[\/code\])+/is","\t$2\n", $Text);
-
-
-
-
-       // Declare the format for [quote] layout
-       //      $QuoteLayout = '<blockquote>$1</blockquote>';                     
-       // Check for [quote] text
-       $Text = preg_replace("/\[quote\](.*?)\[\/quote\]/is",">$1\n\n", $Text);
-         
-       // Images
-
-       // html5 video and audio
-
-       $Text = preg_replace("/\[video\](.*?)\[\/video\]/", '$1', $Text);
-
-       $Text = preg_replace("/\[audio\](.*?)\[\/audio\]/", '$1', $Text);
-
-//     $Text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/", '<iframe src="$1" width="425" height="350"><a href="$1">$1</a></iframe>', $Text);
-         
-       // [img=widthxheight]image source[/img]
-//     $Text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/", '<img src="$3" style="height:{$2}px; width:{$1}px;" >', $Text);
-
-    $Text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism",'http://www.youtube.com/watch?v=$1',$Text); 
-    $Text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism",'http://www.youtube.com/watch?v=$1',$Text); 
-    $Text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism",'http://www.youtube.com/watch?v=$1',$Text); 
-       $Text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", 'http://www.youtube.com/watch?v=$1', $Text);
+                       $Text = $Text."<br />".$tagline;
+               }
 
-       $Text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism",'http://vimeo.com/$1',$Text); 
-       $Text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism",'http://vimeo.com/$1',$Text); 
-       $Text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", 'http://vimeo.com/$1',$Text);
+       } else {
+               $Text = bbcode($Text, $preserve_nl, false, 4);
 
+               // Libertree doesn't convert a harizontal rule if there isn't a linefeed
+               $Text = str_replace(array("<hr />", "<hr>"), array("<br /><hr />", "<br><hr>"), $Text);
+       }
 
+       // Now convert HTML to Markdown
+       $md = new Markdownify(false, false, false);
+       $Text = $md->parseString($Text);
 
+       // The Markdownify converter converts underscores '_' in URLs to '\_', which
+       // messes up the URL. Manually fix these
+       $count = 1;
+       $pos = bb_find_open_close($Text, '[', ']', $count);
+       while($pos !== false) {
+               $start = substr($Text, 0, $pos['start']);
+               $subject = substr($Text, $pos['start'], $pos['end'] - $pos['start'] + 1);
+               $end = substr($Text, $pos['end'] + 1);
 
-       // oembed tag
-       //      $Text = oembed_bbcode2html($Text);
+               $subject = str_replace('\_', '_', $subject);
+               $Text = $start . $subject . $end;
 
-       // If we found an event earlier, strip out all the event code and replace with a reformatted version.
+               $count++;
+               $pos = bb_find_open_close($Text, '[', ']', $count);
+       }
 
-       if(x($ev,'desc') && x($ev,'start')) {
+       // If the text going into bbcode() has a plain URL in it, i.e.
+       // with no [url] tags around it, it will come out of parseString()
+       // looking like: <http://url.com>, which gets removed by strip_tags().
+       // So take off the angle brackets of any such URL
+       $Text = preg_replace("/<http(.*?)>/is", "http$1", $Text);
 
-               $sub = format_event_diaspora($ev);
-       
-               $Text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/is",$sub,$Text);
-               $Text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/is",'',$Text);
-               $Text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/is",'',$Text);
-               $Text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/is",'',$Text);
-               $Text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/is",'',$Text);
-       }
+       // Remove all unconverted tags
+       $Text = strip_tags($Text);
 
-       $Text = preg_replace("/\<(.*?)(src|href)=(.*?)\&amp\;(.*?)\>/ism",'<$1$2=$3&$4>',$Text);
+       // Remove any leading or trailing whitespace, as this will mess up
+       // the Diaspora signature verification and cause the item to disappear
+       $Text = trim($Text);
 
-       $Text = preg_replace('/\[(.*?)\]\((.*?)\\\\_(.*?)\)/ism','[$1]($2_$3)',$Text);
-       
        call_hooks('bb2diaspora',$Text);
 
        return $Text;
 }
 
+function unescape_underscores_in_links($m) {
+       $y = str_replace('\\_','_', $m[2]);
+       return('[' . $m[1] . '](' . $y . ')');
+}
+
 function format_event_diaspora($ev) {
 
        $a = get_app();
@@ -241,9 +147,9 @@ function format_event_diaspora($ev) {
 
        $bd_format = t('l F d, Y \@ g:i A') ; // Friday January 18, 2011 @ 8 AM
 
-       $o = 'Friendika event notification:' . "\n";
+       $o = 'Friendica event notification:' . "\n";
 
-       $o .= '**' . bb2diaspora($ev['desc']) .  '**' . "\n";
+       $o .= '**' . (($ev['summary']) ? bb2diaspora($ev['summary']) : bb2diaspora($ev['desc'])) .  '**' . "\n";
 
        $o .= t('Starts:') . ' ' . '['
                . (($ev['adjust']) ? day_translate(datetime_convert('UTC', 'UTC',