3 require_once("include/oembed.php");
4 require_once("include/event.php");
5 require_once("library/markdown.php");
6 require_once("include/html2bbcode.php");
7 require_once("include/bbcode.php");
8 require_once("include/markdownify/markdownify.php");
10 // we don't want to support a bbcode specific markdown interpreter
11 // and the markdown library we have is pretty good, but provides HTML output.
12 // So we'll use that to convert to HTML, then convert the HTML back to bbcode,
13 // and then clean up a few Diaspora specific constructs.
15 function diaspora2bb($s) {
17 // for testing purposes: Collect raw markdown articles
18 // $file = tempnam("/tmp/friendica/", "markdown");
19 // file_put_contents($file, $s);
21 $s = html_entity_decode($s,ENT_COMPAT,'UTF-8');
23 // Too many new lines. So deactivated the following line
24 // $s = str_replace("\r","\n",$s);
26 $s = str_replace("\r","",$s);
28 // <br/> is invalid. Replace it with the valid expression
29 $s = str_replace("<br/>","<br />",$s);
31 $s = preg_replace('/\@\{(.+?)\; (.+?)\@(.+?)\}/','@[url=https://$3/u/$2]$1[/url]',$s);
33 // Escaping the hash tags - doesn't always seem to work
34 // $s = preg_replace('/\#([^\s\#])/','\\#$1',$s);
36 $s = preg_replace('/\#([^\s\#])/','#$1',$s);
40 $s = str_replace('#','#',$s);
41 // we seem to have double linebreaks
42 // $s = str_replace("\n",'<br />',$s);
45 // $s = str_replace('*','*',$s);
47 // protect the recycle symbol from turning into a tag, but without unescaping angles and naked ampersands
48 $s = str_replace('♲',html_entity_decode('♲',ENT_QUOTES,'UTF-8'),$s);
50 // Convert everything that looks like a link to a link
51 $s = preg_replace("/([^\]\=]|^)(https?\:\/\/)([a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url=$2$3]$2$3[/url]',$s);
53 //$s = preg_replace("/([^\]\=]|^)(https?\:\/\/)(vimeo|youtu|www\.youtube|soundcloud)([a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url=$2$3$4]$2$3$4[/url]',$s);
54 $s = preg_replace("/\[url\=?(.*?)\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/url\]/ism",'[youtube]$2[/youtube]',$s);
55 $s = preg_replace("/\[url\=https?:\/\/www.youtube.com\/watch\?v\=(.*?)\].*?\[\/url\]/ism",'[youtube]$1[/youtube]',$s);
56 $s = preg_replace("/\[url\=?(.*?)\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/url\]/ism",'[vimeo]$2[/vimeo]',$s);
57 $s = preg_replace("/\[url\=https?:\/\/vimeo.com\/([0-9]+)\](.*?)\[\/url\]/ism",'[vimeo]$1[/vimeo]',$s);
58 // remove duplicate adjacent code tags
59 $s = preg_replace("/(\[code\])+(.*?)(\[\/code\])+/ism","[code]$2[/code]", $s);
61 // Don't show link to full picture (until it is fixed)
62 $s = scale_external_images($s, false);
68 function stripdcode_br_cb($s) {
69 return '[code]' . str_replace('<br />', "\n\t", $s[1]) . '[/code]';
73 //////////////////////
74 // The following "diaspora_ul" and "diaspora_ol" are only appropriate for the
75 // pre-Markdownify conversion. If Markdownify isn't used, use the non-Markdownify
77 //////////////////////
79 function diaspora_ul($s) {
80 // Replace "[*]" followed by any number (including zero) of
81 // spaces by "* " to match Diaspora's list format
82 if( strpos($s[0], "[list]") === 0 )
83 return '<ul class="listbullet" style="list-style-type: circle;">' . preg_replace("/\[\*\]( *)/", "* ", $s[1]) . '</ul>';
84 elseif( strpos($s[0], "[ul]") === 0 )
85 return '<ul class="listbullet" style="list-style-type: circle;">' . preg_replace("/\[\*\]( *)/", "* ", $s[1]) . '</ul>';
91 function diaspora_ol($s) {
92 // A hack: Diaspora will create a properly-numbered ordered list even
93 // if you use '1.' for each element of the list, like:
97 if( strpos($s[0], "[list=1]") === 0 )
98 return '<ul class="listdecimal" style="list-style-type: decimal;">' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '</ul>';
99 elseif( strpos($s[0], "[list=i]") === 0 )
100 return '<ul class="listlowerroman" style="list-style-type: lower-roman;">' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '</ul>';
101 elseif( strpos($s[0], "[list=I]") === 0 )
102 return '<ul class="listupperroman" style="list-style-type: upper-roman;">' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '</ul>';
103 elseif( strpos($s[0], "[list=a]") === 0 )
104 return '<ul class="listloweralpha" style="list-style-type: lower-alpha;">' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '</ul>';
105 elseif( strpos($s[0], "[list=A]") === 0 )
106 return '<ul class="listupperalpha" style="list-style-type: upper-alpha;">' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '</ul>';
107 elseif( strpos($s[0], "[ol]") === 0 )
108 return '<ul class="listdecimal" style="list-style-type: decimal;">' . preg_replace("/\[\*\]( *)/", "1. ", $s[1]) . '</ul>';
114 //////////////////////
115 // Non-Markdownify versions of "diaspora_ol" and "diaspora_ul"
116 //////////////////////
117 function diaspora_ul($s) {
118 // Replace "[\\*]" followed by any number (including zero) of
119 // spaces by "* " to match Diaspora's list format
120 return preg_replace("/\[\\\\\*\]( *)/", "* ", $s[1]);
123 function diaspora_ol($s) {
124 // A hack: Diaspora will create a properly-numbered ordered list even
125 // if you use '1.' for each element of the list, like:
129 return preg_replace("/\[\\\\\*\]( *)/", "1. ", $s[1]);
133 function bb2diaspora($Text,$preserve_nl = false) {
135 //////////////////////
136 // An attempt was made to convert bbcode to html and then to markdown
137 // consisting of the following lines.
138 // I'm undoing this as we have a lot of bbcode constructs which
139 // were simply getting lost, for instance bookmark, vimeo, video, youtube, events, etc.
140 // We can try this again, but need a very good test sequence to verify
141 // all the major bbcode constructs that we use are getting through.
142 //////////////////////
144 // bbcode() will convert "[*]" into "<li>" with no closing "</li>"
145 // Markdownify() is unable to handle these, as it makes each new
146 // "<li>" into a deeper nested element until it crashes. So pre-format
147 // the lists as Diaspora lists before sending the $Text to bbcode()
149 // Note that to get nested lists to work for Diaspora, we would need
150 // to define the closing tag for the list elements. So nested lists
151 // are going to be flattened out in Diaspora for now
154 while ((((strpos($Text, "[/list]") !== false) && (strpos($Text, "[list") !== false)) ||
155 ((strpos($Text, "[/ol]") !== false) && (strpos($Text, "[ol]") !== false)) ||
156 ((strpos($Text, "[/ul]") !== false) && (strpos($Text, "[ul]") !== false))) && (++$endlessloop < 20)) {
157 $Text = preg_replace_callback("/\[list\](.*?)\[\/list\]/is", 'diaspora_ul', $Text);
158 $Text = preg_replace_callback("/\[list=1\](.*?)\[\/list\]/is", 'diaspora_ol', $Text);
159 $Text = preg_replace_callback("/\[list=i\](.*?)\[\/list\]/s",'diaspora_ol', $Text);
160 $Text = preg_replace_callback("/\[list=I\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
161 $Text = preg_replace_callback("/\[list=a\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
162 $Text = preg_replace_callback("/\[list=A\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
163 $Text = preg_replace_callback("/\[ul\](.*?)\[\/ul\]/is", 'diaspora_ul', $Text);
164 $Text = preg_replace_callback("/\[ol\](.*?)\[\/ol\]/is", 'diaspora_ol', $Text);
169 // Convert it to HTML - don't try oembed
170 // $Text = bbcode($Text, $preserve_nl, false);
172 // Now convert HTML to Markdown
173 // $md = new Markdownify(false, false, false);
174 // $Text = $md->parseString($Text);
176 // If the text going into bbcode() has a plain URL in it, i.e.
177 // with no [url] tags around it, it will come out of parseString()
178 // looking like: <http://url.com>, which gets removed by strip_tags().
179 // So take off the angle brackets of any such URL
180 // $Text = preg_replace("/<http(.*?)>/is", "http$1", $Text);
182 // Remove all unconverted tags
183 // $Text = strip_tags($Text);
186 // end of bb->html->md conversion attempt
191 $ev = bbtoevent($Text);
193 // Replace any html brackets with HTML Entities to prevent executing HTML or script
194 // Don't use strip_tags here because it breaks [url] search by replacing & with amp
196 $Text = str_replace("<", "<", $Text);
197 $Text = str_replace(">", ">", $Text);
199 // If we find any event code, turn it into an event.
200 // After we're finished processing the bbcode we'll
201 // replace all of the event code with a reformatted version.
204 $Text = str_replace(array("\n","\r"), array('',''),$Text);
206 // Remove the "return" character, as Diaspora uses only the "newline"
207 // character, so having the "return" character can cause signature
209 $Text = str_replace("\r", "", $Text);
212 // Set up the parameters for a URL search string
213 $URLSearchString = "^\[\]";
214 // Set up the parameters for a MAIL search string
215 $MAILSearchString = $URLSearchString;
217 // Perform URL Search
219 // [img]pathtoimage[/img]
221 // the following was added on 10-January-2012 due to an inability of Diaspora's
222 // new javascript markdown processor to handle links with images as the link "text"
223 // It is not optimal and may be removed if this ability is restored in the future
225 $Text = preg_replace("/\[url\=([$URLSearchString]*)\]\[img\](.*?)\[\/img\]\[\/url\]/ism",
226 '![' . t('image/photo') . '](' . '$2' . ')' . "\n" . '[' . t('link') . '](' . '$1' . ')', $Text);
228 $Text = preg_replace("/\[bookmark\]([$URLSearchString]*)\[\/bookmark\]/ism", '[$1]($1)', $Text);
229 $Text = preg_replace("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", '[$2]($1)', $Text);
231 $Text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '[$1]($1)', $Text);
232 $Text = preg_replace("/\#\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[#$2]($1)', $Text);
233 $Text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '[$2]($1)', $Text);
236 $Text = preg_replace("/\[img\](.*?)\[\/img\]/", '![' . t('image/photo') . '](' . '$1' . ')', $Text);
237 $Text = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/", '![' . t('image/photo') . '](' . '$2' . ')', $Text);
239 // Perform MAIL Search
240 $Text = preg_replace("(\[mail\]([$MAILSearchString]*)\[/mail\])", '[$1](mailto:$1)', $Text);
241 $Text = preg_replace("/\[mail\=([$MAILSearchString]*)\](.*?)\[\/mail\]/", '[$2](mailto:$1)', $Text);
243 $Text = str_replace('*', '\\*', $Text);
244 $Text = str_replace('_', '\\_', $Text);
246 $Text = str_replace('`','\\`', $Text);
248 // Check for bold text
249 $Text = preg_replace("(\[b\](.*?)\[\/b\])is",'**$1**',$Text);
251 // Check for italics text
252 $Text = preg_replace("(\[i\](.*?)\[\/i\])is",'_$1_',$Text);
254 // Check for underline text
255 // Replace with italics since Diaspora doesn't have underline
256 $Text = preg_replace("(\[u\](.*?)\[\/u\])is",'_$1_',$Text);
258 // Check for strike-through text
259 $Text = preg_replace("(\[s\](.*?)\[\/s\])is",'**[strike]**$1**[/strike]**',$Text);
261 // Check for over-line text
262 // $Text = preg_replace("(\[o\](.*?)\[\/o\])is",'<span class="overline">$1</span>',$Text);
264 // Check for colored text
265 // Remove color since Diaspora doesn't support it
266 $Text = preg_replace("(\[color=(.*?)\](.*?)\[\/color\])is","$2",$Text);
268 // Check for sized text
269 // Remove it since Diaspora doesn't support sizes very well
270 $Text = preg_replace("(\[size=(.*?)\](.*?)\[\/size\])is","$2",$Text);
272 // Check for list text
274 while ((((strpos($Text, "[/list]") !== false) && (strpos($Text, "[list") !== false)) ||
275 ((strpos($Text, "[/ol]") !== false) && (strpos($Text, "[ol]") !== false)) ||
276 ((strpos($Text, "[/ul]") !== false) && (strpos($Text, "[ul]") !== false)) ||
277 ((strpos($Text, "[/li]") !== false) && (strpos($Text, "[li]") !== false))) && (++$endlessloop < 20)) {
278 $Text = preg_replace_callback("/\[list\](.*?)\[\/list\]/is", 'diaspora_ul', $Text);
279 $Text = preg_replace_callback("/\[list=1\](.*?)\[\/list\]/is", 'diaspora_ol', $Text);
280 $Text = preg_replace_callback("/\[list=i\](.*?)\[\/list\]/s",'diaspora_ol', $Text);
281 $Text = preg_replace_callback("/\[list=I\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
282 $Text = preg_replace_callback("/\[list=a\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
283 $Text = preg_replace_callback("/\[list=A\](.*?)\[\/list\]/s", 'diaspora_ol', $Text);
284 $Text = preg_replace_callback("/\[ul\](.*?)\[\/ul\]/is", 'diaspora_ul', $Text);
285 $Text = preg_replace_callback("/\[ol\](.*?)\[\/ol\]/is", 'diaspora_ol', $Text);
286 $Text = preg_replace("/\[li\]( *)(.*?)\[\/li\]/s", '* $2' ,$Text);
289 // Just get rid of table tags since Diaspora doesn't support tables
290 $Text = preg_replace("/\[th\](.*?)\[\/th\]/s", '$1' ,$Text);
291 $Text = preg_replace("/\[td\](.*?)\[\/td\]/s", '$1' ,$Text);
292 $Text = preg_replace("/\[tr\](.*?)\[\/tr\]/s", '$1' ,$Text);
293 $Text = preg_replace("/\[table\](.*?)\[\/table\]/s", '$1' ,$Text);
295 $Text = preg_replace("/\[table border=(.*?)\](.*?)\[\/table\]/s", '$2' ,$Text);
296 // $Text = preg_replace("/\[table border=0\](.*?)\[\/table\]/s", '<table border="0" >$1</table>' ,$Text);
299 // $Text = str_replace("[*]", "<li>", $Text);
301 // Check for font change text
302 // $Text = preg_replace("(\[font=(.*?)\](.*?)\[\/font\])","<span style=\"font-family: $1;\">$2</span>",$Text);
305 $Text = preg_replace_callback("/\[code\](.*?)\[\/code\]/is",'stripdcode_br_cb',$Text);
307 // Check for [code] text
308 $Text = preg_replace("/(\[code\])+(.*?)(\[\/code\])+/is","\t$2\n", $Text);
313 // Declare the format for [quote] layout
314 // $QuoteLayout = '<blockquote>$1</blockquote>';
315 // Check for [quote] text
316 $Text = preg_replace("/\[quote\](.*?)\[\/quote\]/is",">$1\n\n", $Text);
317 $Text = preg_replace("/\[quote=(.*?)\](.*?)\[\/quote\]/is",">$2\n\n", $Text);
321 // html5 video and audio
323 $Text = preg_replace("/\[video\](.*?)\[\/video\]/", '$1', $Text);
325 $Text = preg_replace("/\[audio\](.*?)\[\/audio\]/", '$1', $Text);
327 // $Text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/", '<iframe src="$1" width="425" height="350"><a href="$1">$1</a></iframe>', $Text);
329 // [img=widthxheight]image source[/img]
330 // $Text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/", '<img src="$3" style="height:{$2}px; width:{$1}px;" >', $Text);
332 $Text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/watch\?v\=(.*?)\[\/youtube\]/ism",'http://www.youtube.com/watch?v=$1',$Text);
333 $Text = preg_replace("/\[youtube\]https?:\/\/www.youtube.com\/embed\/(.*?)\[\/youtube\]/ism",'http://www.youtube.com/watch?v=$1',$Text);
334 $Text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism",'http://www.youtube.com/watch?v=$1',$Text);
335 $Text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", 'http://www.youtube.com/watch?v=$1', $Text);
337 $Text = preg_replace("/\[vimeo\]https?:\/\/player.vimeo.com\/video\/([0-9]+)(.*?)\[\/vimeo\]/ism",'http://vimeo.com/$1',$Text);
338 $Text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism",'http://vimeo.com/$1',$Text);
339 $Text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", 'http://vimeo.com/$1',$Text);
342 $Text = str_replace('[nosmile]','',$Text);
345 // $Text = oembed_bbcode2html($Text);
347 // If we found an event earlier, strip out all the event code and replace with a reformatted version.
351 $sub = format_event_diaspora($ev);
353 $Text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/is",'',$Text);
354 $Text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/is",'',$Text);
355 $Text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/is",$sub,$Text);
356 $Text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/is",'',$Text);
357 $Text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/is",'',$Text);
358 $Text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/is",'',$Text);
361 $Text = preg_replace("/\<(.*?)(src|href)=(.*?)\&\;(.*?)\>/ism",'<$1$2=$3&$4>',$Text);
363 $Text = preg_replace_callback('/\[(.*?)\]\((.*?)\)/ism','unescape_underscores_in_links',$Text);
366 // Remove any leading or trailing whitespace, as this will mess up
367 // the Diaspora signature verification and cause the item to disappear
370 call_hooks('bb2diaspora',$Text);
375 function unescape_underscores_in_links($m) {
376 $y = str_replace('\\_','_', $m[2]);
377 return('[' . $m[1] . '](' . $y . ')');
380 function format_event_diaspora($ev) {
384 if(! ((is_array($ev)) && count($ev)))
387 $bd_format = t('l F d, Y \@ g:i A') ; // Friday January 18, 2011 @ 8 AM
389 $o = 'Friendica event notification:' . "\n";
391 $o .= '**' . (($ev['summary']) ? bb2diaspora($ev['summary']) : bb2diaspora($ev['desc'])) . '**' . "\n";
393 $o .= t('Starts:') . ' ' . '['
394 . (($ev['adjust']) ? day_translate(datetime_convert('UTC', 'UTC',
395 $ev['start'] , $bd_format ))
396 : day_translate(datetime_convert('UTC', 'UTC',
397 $ev['start'] , $bd_format)))
398 . '](' . $a->get_baseurl() . '/localtime/?f=&time=' . urlencode(datetime_convert('UTC','UTC',$ev['start'])) . ")\n";
400 if(! $ev['nofinish'])
401 $o .= t('Finishes:') . ' ' . '['
402 . (($ev['adjust']) ? day_translate(datetime_convert('UTC', 'UTC',
403 $ev['finish'] , $bd_format ))
404 : day_translate(datetime_convert('UTC', 'UTC',
405 $ev['finish'] , $bd_format )))
406 . '](' . $a->get_baseurl() . '/localtime/?f=&time=' . urlencode(datetime_convert('UTC','UTC',$ev['finish'])) . ")\n";
408 if(strlen($ev['location']))
409 $o .= t('Location:') . bb2diaspora($ev['location'])