X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=include%2Fbb2diaspora.php;h=a8b39f741ac84a2d066066e129176cf0dff82b9a;hb=84238e1b89c0a40cfd2dfa72972fdfc2f65c4fab;hp=75ba57dff86d5b3a08a47802ed39a5d11f4ff9ec;hpb=1f968396749f8cac31aaeee770ae7fc5aff48f38;p=friendica.git
diff --git a/include/bb2diaspora.php b/include/bb2diaspora.php
index 75ba57dff8..a8b39f741a 100644
--- a/include/bb2diaspora.php
+++ b/include/bb2diaspora.php
@@ -5,7 +5,8 @@ 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/html-to-markdown/HTML_To_Markdown.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.
@@ -14,44 +15,47 @@ require_once("include/markdownify/markdownify.php");
function diaspora2bb($s) {
- // for testing purposes: Collect raw markdown articles
- // $file = tempnam("/tmp/friendica/", "markdown");
- // file_put_contents($file, $s);
-
$s = html_entity_decode($s,ENT_COMPAT,'UTF-8');
- // Too many new lines. So deactivated the following line
- // $s = str_replace("\r","\n",$s);
- // Simply remove cr.
+ // Remove CR to avoid problems with following code
$s = str_replace("\r","",$s);
- //
is invalid. Replace it with the valid expression
- $s = str_replace("
","
",$s);
+ $s = str_replace("\n"," \n",$s);
- $s = preg_replace('/\@\{(.+?)\; (.+?)\@(.+?)\}/','@[url=https://$3/u/$2]$1[/url]',$s);
+ // The parser cannot handle paragraphs correctly
+ $s = str_replace(array("
", "", '
'),array("
", "
", "
"),$s);
- // Escaping the hash tags - doesn't always seem to work
- // $s = preg_replace('/\#([^\s\#])/','\\#$1',$s);
- // This seems to work
+ // Escaping the hash tags
$s = preg_replace('/\#([^\s\#])/','#$1',$s);
$s = Markdown($s);
+ $s = preg_replace('/\@\{(.+?)\; (.+?)\@(.+?)\}/','@[url=https://$3/u/$2]$1[/url]',$s);
+
$s = str_replace('#','#',$s);
- $s = str_replace("\n",'
',$s);
+ $search = array(" \n", "\n ");
+ $replace = array("\n", "\n");
+ do {
+ $oldtext = $s;
+ $s = str_replace($search, $replace, $s);
+ } while ($oldtext != $s);
+
+ $s = str_replace("\n\n", "
", $s);
$s = html2bbcode($s);
-// $s = str_replace('*','*',$s);
+
+ // protect the recycle symbol from turning into a tag, but without unescaping angles and naked ampersands
+ $s = str_replace('♲',html_entity_decode('♲',ENT_QUOTES,'UTF-8'),$s);
// 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);
//$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 = 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 = 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]", $s);
@@ -61,247 +65,64 @@ function diaspora2bb($s) {
return $s;
}
+function bb2diaspora($Text,$preserve_nl = false, $fordiaspora = true) {
-function stripdcode_br_cb($s) {
- return '[code]' . str_replace('
', "\n\t", $s[1]) . '[/code]';
-}
-
-
-function diaspora_ul($s) {
- // Replace "[*]" followed by any number (including zero) of
- // spaces by "* " to match Diaspora's list format
- if( strpos($s[0], "[list]") === 0 )
- return '
' . preg_replace("/\[\*\]( *)/", "* ", $s[1]) . '
';
- elseif( strpos($s[0], "[ul]") === 0 )
- return '' . preg_replace("/\[\*\]( *)/", "* ", $s[1]) . '
';
- else
- return $s[0];
-}
-
-
-function diaspora_ol($s) {
- // A hack: Diaspora will create a properly-numbered ordered list even
- // if you use '1.' for each element of the list, like:
- // 1. First element
- // 1. Second element
- // 1. Third element
- if( strpos($s[0], "[list=1]") === 0 )
- return '' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '
';
- elseif( strpos($s[0], "[list=i]") === 0 )
- return '' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '
';
- elseif( strpos($s[0], "[list=I]") === 0 )
- return '' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '
';
- elseif( strpos($s[0], "[list=a]") === 0 )
- return '' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '
';
- elseif( strpos($s[0], "[list=A]") === 0 )
- return '' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '
';
- elseif( strpos($s[0], "[ol]") === 0 )
- return '' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '
';
- else
- return $s[0];
-}
-
-
-function bb2diaspora($Text,$preserve_nl = false) {
-
- // bbcode() will convert "[*]" into "" with no closing ""
- // Markdownify() is unable to handle these, as it makes each new
- // "" into a deeper nested element until it crashes. So pre-format
- // the lists as Diaspora lists before sending the $Text to bbcode()
- //
- // Note that regular expressions are really not suitable for parsing
- // text with opening and closing tags, so nested lists may make things
- // wonky
- $endlessloop = 0;
- while ((strpos($Text, "[/list]") !== false) && (strpos($Text, "[list") !== false) && (++$endlessloop < 20)) {
- $Text = preg_replace_callback("/\[list\](.*?)\[\/list\]/is", 'diaspora_ul', $Text);
- $Text = preg_replace_callback("/\[list=1\](.*?)\[\/list\]/is", 'diaspora_ol', $Text);
- $Text = preg_replace_callback("/\[list=i\](.*?)\[\/list\]/s",'diaspora_ol', $Text);
- $Text = preg_replace_callback("/\[list=I\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
- $Text = preg_replace_callback("/\[list=a\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
- $Text = preg_replace_callback("/\[list=A\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
- }
- $Text = preg_replace_callback("/\[ul\](.*?)\[\/ul\]/is", 'diaspora_ul', $Text);
- $Text = preg_replace_callback("/\[ol\](.*?)\[\/ol\]/is", 'diaspora_ol', $Text);
-
- // Convert it to HTML - don't try oembed
- $Text = bbcode($Text, $preserve_nl, false);
-
- // Now convert HTML to Markdown
- $md = new Markdownify(false, false, false);
- $Text = $md->parseString($Text);
-
- // Remove all unconverted tags
- $Text = strip_tags($Text);
-
-/*
- $ev = bbtoevent($Text);
-
- // 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
-
- $Text = str_replace("<", "<", $Text);
- $Text = str_replace(">", ">", $Text);
-
- // 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.
+ $a = get_app();
- if($preserve_nl)
- $Text = str_replace(array("\n","\r"), array('',''),$Text);
- else
- // Remove the "return" character, as Diaspora uses only the "newline"
- // character, so having the "return" character can cause signature
- // failures
- $Text = str_replace("\r", "", $Text);
+ $OriginalText = $Text;
+ // Since Diaspora is creating a summary for links, this function removes them before posting
+ if ($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_callback("/#\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/i", create_function('$match',
+ 'return \'#\'. str_replace(\' \', \'_\', $match[2]);'
+ ), $Text);
- $Text = preg_replace("/\[img\](.*?)\[\/img\]/", '![' . t('image/photo') . '](' . '$1' . ')', $Text);
- $Text = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/", '![' . t('image/photo') . '](' . '$2' . ')', $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);
- // 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);
-
- // Check for bold text
- $Text = preg_replace("(\[b\](.*?)\[\/b\])is",'**$1**',$Text);
-
- // Check for italics text
- $Text = preg_replace("(\[i\](.*?)\[\/i\])is",'_$1_',$Text);
-
- // Check for underline text
- // Replace with italics since Diaspora doesn't have underline
- $Text = preg_replace("(\[u\](.*?)\[\/u\])is",'_$1_',$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",'$1',$Text);
-
- // Check for colored text
- // Remove color since Diaspora doesn't support it
- $Text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])is","$2",$Text);
-
- // Check for sized text
- // Remove it since Diaspora doesn't support sizes very well
- $Text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])is","$2",$Text);
-
- // Check for list text
- $Text = preg_replace_callback("/\[list\](.*?)\[\/list\]/is", 'diaspora_ul', $Text);
- $Text = preg_replace_callback("/\[ul\](.*?)\[\/ul\]/is", 'diaspora_ul', $Text);
- $Text = preg_replace_callback("/\[list=1\](.*?)\[\/list\]/is", 'diaspora_ol', $Text);
- $Text = preg_replace_callback("/\[list=i\](.*?)\[\/list\]/s",'diaspora_ol', $Text);
- $Text = preg_replace_callback("/\[list=I\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
- $Text = preg_replace_callback("/\[list=a\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
- $Text = preg_replace_callback("/\[list=A\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
- $Text = preg_replace_callback("/\[ol\](.*?)\[\/ol\]/is", 'diaspora_ol', $Text);
-// $Text = preg_replace("/\[li\](.*?)\[\/li\]/s", '$1' ,$Text);
-
- // Just get rid of table tags since Diaspora doesn't support tables
- $Text = preg_replace("/\[th\](.*?)\[\/th\]/s", '$1' ,$Text);
- $Text = preg_replace("/\[td\](.*?)\[\/td\]/s", '$1' ,$Text);
- $Text = preg_replace("/\[tr\](.*?)\[\/tr\]/s", '$1' ,$Text);
- $Text = preg_replace("/\[table\](.*?)\[\/table\]/s", '$1' ,$Text);
-
- $Text = preg_replace("/\[table border=(.*?)\](.*?)\[\/table\]/s", '$2' ,$Text);
-// $Text = preg_replace("/\[table border=0\](.*?)\[\/table\]/s", '' ,$Text);
-
-
-// $Text = str_replace("[*]", "", $Text);
-
- // Check for font change text
-// $Text = preg_replace("(\[font=(.*?)\](.*?)\[\/font\])","$2",$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 = '$1
';
- // Check for [quote] text
- $Text = preg_replace("/\[quote\](.*?)\[\/quote\]/is",">$1\n\n", $Text);
- $Text = preg_replace("/\[quote=(.*?)\](.*?)\[\/quote\]/is",">$2\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("/\[iframe\](.*?)\[\/iframe\]/", '', $Text);
-
- // [img=widthxheight]image source[/img]
-// $Text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/", '
', $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 = 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);
+ // Convert it to HTML - don't try oembed
+ if ($fordiaspora) {
+ $Text = bbcode($Text, $preserve_nl, false, 3);
+ // Add all tags that maybe were removed
+ if (preg_match_all("/#\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism",$OriginalText, $tags)) {
+ $tagline = "";
+ foreach($tags[2] as $tag) {
+ $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
+ if (!strpos(html_entity_decode($Text, ENT_QUOTES, 'UTF-8'), "#".$tag))
+ $tagline .= "#".$tag." ";
+ }
+ $Text = $Text." ".$tagline;
+ }
- $Text = str_replace('[nosmile]','',$Text);
+ } else
+ $Text = bbcode($Text, $preserve_nl, false, 4);
- // oembed tag
- // $Text = oembed_bbcode2html($Text);
+ // mask some special HTML chars from conversation to markdown
+ $Text = str_replace(array('<','>','&'),array('&_lt_;','&_gt_;','&_amp_;'),$Text);
- // If we found an event earlier, strip out all the event code and replace with a reformatted version.
+ // If a link is followed by a quote then there should be a newline before it
+ // Maybe we should make this newline at every time before a quote.
+ $Text = str_replace(array(""), array("
"), $Text);
- if(x($ev,'desc') && x($ev,'start')) {
+ $stamp1 = microtime(true);
- $sub = format_event_diaspora($ev);
+ // Now convert HTML to Markdown
+ $Text = new HTML_To_Markdown($Text);
- $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);
- }
+ // unmask the special chars back to HTML
+ $Text = str_replace(array('&_lt_;','&_gt_;','&_amp_;'),array('<','>','&'),$Text);
- $Text = preg_replace("/\<(.*?)(src|href)=(.*?)\&\;(.*?)\>/ism",'<$1$2=$3&$4>',$Text);
+ $a->save_timestamp($stamp1, "parser");
- $Text = preg_replace_callback('/\[(.*?)\]\((.*?)\)/ism','unescape_underscores_in_links',$Text);
-*/
+ // Libertree has a problem with escaped hashtags.
+ $Text = str_replace(array('\#'), array('#'), $Text);
// Remove any leading or trailing whitespace, as this will mess up
// the Diaspora signature verification and cause the item to disappear
@@ -328,25 +149,25 @@ function format_event_diaspora($ev) {
$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',
+ . (($ev['adjust']) ? day_translate(datetime_convert('UTC', 'UTC',
$ev['start'] , $bd_format ))
- : day_translate(datetime_convert('UTC', 'UTC',
+ : day_translate(datetime_convert('UTC', 'UTC',
$ev['start'] , $bd_format)))
. '](' . $a->get_baseurl() . '/localtime/?f=&time=' . urlencode(datetime_convert('UTC','UTC',$ev['start'])) . ")\n";
if(! $ev['nofinish'])
- $o .= t('Finishes:') . ' ' . '['
- . (($ev['adjust']) ? day_translate(datetime_convert('UTC', 'UTC',
+ $o .= t('Finishes:') . ' ' . '['
+ . (($ev['adjust']) ? day_translate(datetime_convert('UTC', 'UTC',
$ev['finish'] , $bd_format ))
- : day_translate(datetime_convert('UTC', 'UTC',
+ : day_translate(datetime_convert('UTC', 'UTC',
$ev['finish'] , $bd_format )))
. '](' . $a->get_baseurl() . '/localtime/?f=&time=' . urlencode(datetime_convert('UTC','UTC',$ev['finish'])) . ")\n";
if(strlen($ev['location']))
- $o .= t('Location:') . bb2diaspora($ev['location'])
+ $o .= t('Location:') . bb2diaspora($ev['location'])
. "\n";
$o .= "\n";