- $1
- $1
- $1
- $1
- $1
- $1
$1'; + // $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", $Text); + // Images // html5 video and audio @@ -187,13 +327,13 @@ function bb2diaspora($Text,$preserve_nl = false) { $Text = preg_replace("/\[audio\](.*?)\[\/audio\]/", '$1', $Text); // $Text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/", '', $Text); - + // [img=widthxheight]image source[/img] // $Text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/", '
tags - $Text = nl2br($Text); + // nlbr seems to be hopelessly messed up + // $Text = nl2br($Text); + + // We'll emulate it. + + $Text = str_replace("\r\n","\n", $Text); + $Text = str_replace(array("\r","\n"), array('
','
'), $Text); + if($preserve_nl) $Text = str_replace(array("\n","\r"), array('',''),$Text); + // Set up the parameters for a URL search string $URLSearchString = "^\[\]"; // Set up the parameters for a MAIL search string @@ -108,10 +256,14 @@ function bbcode($Text,$preserve_nl = false) { $Text = preg_replace("/([^\]\=]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1$2', $Text); - $Text = preg_replace_callback("/\[bookmark\=([^\]]*)\].*?\[\/bookmark\]/ism",'tryoembed',$Text); + if ($tryoembed) + $Text = preg_replace_callback("/\[bookmark\=([^\]]*)\].*?\[\/bookmark\]/ism",'tryoembed',$Text); + $Text = preg_replace("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism",'[url=$1]$2[/url]',$Text); - $Text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism",'tryoembed',$Text); + if ($tryoembed) + $Text = preg_replace_callback("/\[url\]([$URLSearchString]*)\[\/url\]/ism",'tryoembed',$Text); + $Text = preg_replace("/\[url\]([$URLSearchString]*)\[\/url\]/ism", '$1', $Text); $Text = preg_replace("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$2', $Text); //$Text = preg_replace("/\[url\=([$URLSearchString]*)\]([$URLSearchString]*)\[\/url\]/ism", '$2', $Text); @@ -154,11 +306,20 @@ function bbcode($Text,$preserve_nl = false) { // Check for list text $Text = str_replace("[*]", "
- $1
- $1
- $1
- $2
- $2
- $2
- $1
- $1
- $1
- $1
$1';
// Check for [code] text
@@ -245,40 +406,60 @@ function bbcode($Text,$preserve_nl = false) {
$Text = preg_replace("/\[img\](.*?)\[\/img\]/ism", ' . '/images/lock_icon.gif "' . t('Encrypted content') . '"){kind=icon}
', $Text); + $Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism",'
', $Text); + // Try to Oembed - $Text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", 'tryoembed', $Text); - $Text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", 'tryoembed', $Text); + if ($tryoembed) { + $Text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4))\[\/video\]/ism", '', $Text); + $Text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3))\[\/audio\]/ism", '', $Text); + $Text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", 'tryoembed', $Text); + $Text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", 'tryoembed', $Text); + } else { + $Text = preg_replace("/\[video\](.*?)\[\/video\]/", '$1', $Text); + $Text = preg_replace("/\[audio\](.*?)\[\/audio\]/", '$1', $Text); + } // html5 video and audio - $Text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '', $Text); - + if ($tryoembed) + $Text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '', $Text); + else + $Text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '$1', $Text); // Youtube extensions - $Text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", 'tryoembed', $Text); - $Text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", 'tryoembed', $Text); - $Text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism",'tryoembed',$Text); - + if ($tryoembed) { + $Text = preg_replace_callback("/\[youtube\](https?:\/\/www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", 'tryoembed', $Text); + $Text = preg_replace_callback("/\[youtube\](www.youtube.com\/watch\?v\=.*?)\[\/youtube\]/ism", 'tryoembed', $Text); + $Text = preg_replace_callback("/\[youtube\](https?:\/\/youtu.be\/.*?)\[\/youtube\]/ism",'tryoembed',$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?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism",'[youtube]$1[/youtube]',$Text); - - - $Text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '', $Text); + $Text = preg_replace("/\[youtube\]https?:\/\/youtu.be\/(.*?)\[\/youtube\]/ism",'[youtube]$1[/youtube]',$Text); + + if ($tryoembed) + $Text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '', $Text); + else + $Text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", "http://www.youtube.com/watch?v=$1", $Text); - $Text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism",'tryoembed',$Text); - $Text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism",'tryoembed',$Text); + if ($tryoembed) { + $Text = preg_replace_callback("/\[vimeo\](https?:\/\/player.vimeo.com\/video\/[0-9]+).*?\[\/vimeo\]/ism",'tryoembed',$Text); + $Text = preg_replace_callback("/\[vimeo\](https?:\/\/vimeo.com\/[0-9]+).*?\[\/vimeo\]/ism",'tryoembed',$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); - $Text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '', $Text); + $Text = preg_replace("/\[vimeo\]https?:\/\/vimeo.com\/([0-9]+)(.*?)\[\/vimeo\]/ism",'[vimeo]$1[/vimeo]',$Text); + + if ($tryoembed) + $Text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", '', $Text); + else + $Text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", "http://vimeo.com/$1", $Text); // $Text = preg_replace("/\[youtube\](.*?)\[\/youtube\]/", '', $Text); @@ -286,13 +467,20 @@ function bbcode($Text,$preserve_nl = false) { // oembed tag $Text = oembed_bbcode2html($Text); - // If we found an event earlier, strip out all the event code and replace with a reformatted version. + // Avoid triple linefeeds through oembed + $Text = str_replace("
", "
", $Text); - if(x($ev,'desc') && x($ev,'start')) { + // If we found an event earlier, strip out all the event code and replace with a reformatted version. + // Replace the event-start section with the entire formatted event. The other bbcode is stripped. + // Summary (e.g. title) is required, earlier revisions only required description (in addition to + // start which is always required). Allow desc with a missing summary for compatibility. + + if((x($ev,'desc') || x($ev,'summary')) && x($ev,'start')) { $sub = format_event_html($ev); - $Text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism",$sub,$Text); - $Text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism",'',$Text); + $Text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism",'',$Text); + $Text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism",'',$Text); + $Text = preg_replace("/\[event\-start\](.*?)\[\/event\-start\]/ism",$sub,$Text); $Text = preg_replace("/\[event\-finish\](.*?)\[\/event\-finish\]/ism",'',$Text); $Text = preg_replace("/\[event\-location\](.*?)\[\/event\-location\]/ism",'',$Text); $Text = preg_replace("/\[event\-adjust\](.*?)\[\/event\-adjust\]/ism",'',$Text); @@ -307,11 +495,34 @@ function bbcode($Text,$preserve_nl = false) { $Text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/','&$1;',$Text); + $Text = preg_replace('/\&\#039\;/','\'',$Text); + $Text = preg_replace('/\"\;/','"',$Text); // fix any escaped ampersands that may have been converted into links $Text = preg_replace("/\<(.*?)(src|href)=(.*?)\&\;(.*?)\>/ism",'<$1$2=$3&$4>',$Text); - if(strlen($saved_image)) - $Text = str_replace('[$#saved_image#$]','
','', $Text); + + $Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES'); + } call_hooks('bbcode',$Text); diff --git a/include/config.php b/include/config.php index 1f2a70e5a1..44606e329a 100644 --- a/include/config.php +++ b/include/config.php @@ -68,7 +68,7 @@ function get_config($family, $key, $instore = false) { ); if(count($ret)) { // manage array value - $val = (preg_match("|^a:[0-9]+:{.*}$|", $ret[0]['v'])?unserialize( $ret[0]['v']):$ret[0]['v']); + $val = (preg_match("|^a:[0-9]+:{.*}$|s", $ret[0]['v'])?unserialize( $ret[0]['v']):$ret[0]['v']); $a->config[$family][$key] = $val; return $val; } @@ -162,7 +162,7 @@ function get_pconfig($uid,$family, $key, $instore = false) { ); if(count($ret)) { - $val = (preg_match("|^a:[0-9]+:{.*}$|", $ret[0]['v'])?unserialize( $ret[0]['v']):$ret[0]['v']); + $val = (preg_match("|^a:[0-9]+:{.*}$|s", $ret[0]['v'])?unserialize( $ret[0]['v']):$ret[0]['v']); $a->config[$uid][$family][$key] = $val; return $val; } diff --git a/include/contact_widgets.php b/include/contact_widgets.php index ce1cdbad55..ea71b3b707 100644 --- a/include/contact_widgets.php +++ b/include/contact_widgets.php @@ -142,9 +142,16 @@ function common_friends_visitor_widget($profile_uid) { $cid = $zcid = 0; - if(can_write_wall($a,$profile_uid)) - $cid = remote_user(); - else { + if(is_array($_SESSION['remote'])) { + foreach($_SESSION['remote'] as $visitor) { + if($visitor['uid'] == $profile_uid) { + $cid = $visitor['cid']; + break; + } + } + } + + if(! $cid) { if(get_my_url()) { $r = q("select id from contact where nurl = '%s' and uid = %d limit 1", dbesc(normalise_link(get_my_url())), diff --git a/include/conversation.php b/include/conversation.php index 1d5a92284a..1d927d2016 100644 --- a/include/conversation.php +++ b/include/conversation.php @@ -1,43 +1,115 @@ $new_body, 'images' => $saved_image); +}} + +if(! function_exists('item_redir_and_replace_images')) { +function item_redir_and_replace_images($body, $images, $cid) { + + $origbody = $body; + $newbody = ''; + + $cnt = 1; + $pos = get_bb_tag_pos($origbody, 'url', 1); + while($pos !== false && $cnt < 1000) { + + $search = '/\[url\=(.*?)\]\[!#saved_image([0-9]*)#!\]\[\/url\]' . '/is'; + $replace = '[url=' . z_path() . '/redir/' . $cid + . '?f=1&url=' . '$1' . '][!#saved_image' . '$2' .'#!][/url]'; + + $newbody .= substr($origbody, 0, $pos['start']['open']); + $subject = substr($origbody, $pos['start']['open'], $pos['end']['close'] - $pos['start']['open']); + $origbody = substr($origbody, $pos['end']['close']); + if($origbody === false) + $origbody = ''; + + $subject = preg_replace($search, $replace, $subject); + $newbody .= $subject; + + $cnt++; + $pos = get_bb_tag_pos($origbody, 'url', 1); + } + $newbody .= $origbody; + + $cnt = 0; + foreach($images as $image) { + // We're depending on the property of 'foreach' (specified on the PHP website) that + // it loops over the array starting from the first element and going sequentially + // to the last element + $newbody = str_replace('[!#saved_image' . $cnt . '#!]', '[img]' . $image . '[/img]', $newbody); + $cnt++; + } + return $newbody; +}} + + + /** * Render actions localized */ function localize_item(&$item){ - $Text = $item['body']; - $saved_image = ''; - $img_start = strpos($Text,'[img]data:'); - $img_end = strpos($Text,'[/img]'); - - if($img_start !== false && $img_end !== false && $img_end > $img_start) { - $start_fragment = substr($Text,0,$img_start); - $img_start += strlen('[img]'); - $saved_image = substr($Text,$img_start,$img_end - $img_start); - $end_fragment = substr($Text,$img_end + strlen('[/img]')); - $Text = $start_fragment . '[!#saved_image#!]' . $end_fragment; - $search = '/\[url\=(.*?)\]\[!#saved_image#!\]\[\/url\]' . '/is'; - $replace = '[url=' . z_path() . '/redir/' . $item['contact-id'] - . '?f=1&url=' . '$1' . '][!#saved_image#!][/url]' ; - - $Text = preg_replace($search,$replace,$Text); - - if(strlen($saved_image)) - $item['body'] = str_replace('[!#saved_image#!]', '[img]' . $saved_image . '[/img]',$Text); - } + $extracted = item_extract_images($item['body']); + if($extracted['images']) + $item['body'] = item_redir_and_replace_images($extracted['body'], $extracted['images'], $item['contact-id']); $xmlhead="<"."?xml version='1.0' encoding='UTF-8' ?".">"; - if ($item['verb']=== ACTIVITY_LIKE || $item['verb']=== ACTIVITY_DISLIKE){ + if (activity_match($item['verb'],ACTIVITY_LIKE) || activity_match($item['verb'],ACTIVITY_DISLIKE)){ - $r = q("SELECT * from `item`,`contact` WHERE + $r = q("SELECT * from `item`,`contact` WHERE `item`.`contact-id`=`contact`.`id` AND `item`.`uri`='%s';", dbesc($item['parent-uri'])); if(count($r)==0) return; $obj=$r[0]; - - $author = '[url=' . $item['author-link'] . ']' . $item['author-name'] . '[/url]'; + + $author = '[url=' . $item['author-link'] . ']' . $item['author-name'] . '[/url]'; $objauthor = '[url=' . $obj['author-link'] . ']' . $obj['author-name'] . '[/url]'; - + switch($obj['verb']){ case ACTIVITY_POST: switch ($obj['object-type']){ @@ -51,38 +123,36 @@ function localize_item(&$item){ default: if($obj['resource-id']){ $post_type = t('photo'); - $m=array(); preg_match("/\[url=([^]]*)\]/", $obj['body'], $m); + $m=array(); preg_match("/\[url=([^]]*)\]/", $obj['body'], $m); $rr['plink'] = $m[1]; } else { $post_type = t('status'); } } - + $plink = '[url=' . $obj['plink'] . ']' . $post_type . '[/url]'; - - switch($item['verb']){ - case ACTIVITY_LIKE : - $bodyverb = t('%1$s likes %2$s\'s %3$s'); - break; - case ACTIVITY_DISLIKE: - $bodyverb = t('%1$s doesn\'t like %2$s\'s %3$s'); - break; + + if(activity_match($item['verb'],ACTIVITY_LIKE)) { + $bodyverb = t('%1$s likes %2$s\'s %3$s'); + } + elseif(activity_match($item['verb'],ACTIVITY_DISLIKE)) { + $bodyverb = t('%1$s doesn\'t like %2$s\'s %3$s'); } $item['body'] = sprintf($bodyverb, $author, $objauthor, $plink); - + } - if ($item['verb']=== ACTIVITY_FRIEND){ + if (activity_match($item['verb'],ACTIVITY_FRIEND)) { if ($item['object-type']=="" || $item['object-type']!== ACTIVITY_OBJ_PERSON) return; $Aname = $item['author-name']; $Alink = $item['author-link']; - + $xmlhead="<"."?xml version='1.0' encoding='UTF-8' ?".">"; - + $obj = parse_xml_string($xmlhead.$item['object']); $links = parse_xml_string($xmlhead."
' . bbcode($ev['summary']) . '
' . "\r\n"; + $o .= '' . bbcode($ev['desc']) . '
' . "\r\n"; $o .= '' . t('Starts:') . ' nodeValue = str_replace("\n", "\r", $node->nodeValue);
$message = $doc->saveHTML();
- $message = str_replace(array("\n<", ">\n", "\r", "\n", "\xC3\x82\xC2\xA0"), array("<", ">", " tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_p() {
+ if (!$this->parser->isStartTag) {
+ $this->setLineBreaks(2);
+ }
+ }
+ /**
+ * handle tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_a() {
+ if ($this->parser->isStartTag) {
+ $this->buffer();
+ if (isset($this->parser->tagAttributes['title'])) {
+ $this->parser->tagAttributes['title'] = $this->decode($this->parser->tagAttributes['title']);
+ } else {
+ $this->parser->tagAttributes['title'] = null;
+ }
+ $this->parser->tagAttributes['href'] = $this->decode(trim($this->parser->tagAttributes['href']));
+ $this->stack();
+ } else {
+ $tag = $this->unstack();
+ $buffer = $this->unbuffer();
+
+ if (empty($tag['href']) && empty($tag['title'])) {
+ # empty links... testcase mania, who would possibly do anything like that?!
+ $this->out('['.$buffer.']()', true);
+ return;
+ }
+
+ if ($buffer == $tag['href'] && empty($tag['title'])) {
+ # ') {
+ # next comes a paragraph, so we'll need an extra line
+ $this->out("\n".$this->indent);
+ } elseif (substr($this->output, -2) == "\n\n") {
+ $this->output = substr($this->output, 0, -1);
+ }
+ $this->out(': ');
+ $this->indent(' ', false);
+ } else {
+ # lookahead for next dt
+ if (substr(ltrim($this->parser->html), 0, 4) == ' \s*
", " ", ""), $message);
+ $message = str_replace(array("\n<", ">\n", "\r", "\n", "\xC3\x82\xC2\xA0"), array("<", ">", "
", " ", ""), $message);
$message = preg_replace('= [\s]*=i', " ", $message);
@$doc->loadHTML($message);
diff --git a/include/items.php b/include/items.php
index 05134ef8f5..f70e96fcbe 100755
--- a/include/items.php
+++ b/include/items.php
@@ -4,6 +4,8 @@ require_once('include/bbcode.php');
require_once('include/oembed.php');
require_once('include/salmon.php');
require_once('include/crypto.php');
+require_once('include/Photo.php');
+
function get_feed_for(&$a, $dfrn_id, $owner_nick, $last_update, $direction = 0) {
@@ -278,8 +280,123 @@ function construct_activity_target($item) {
}
return '';
-}
+}
+/* limit_body_size()
+ *
+ * The purpose of this function is to apply system message length limits to
+ * imported messages without including any embedded photos in the length
+ */
+if(! function_exists('limit_body_size')) {
+function limit_body_size($body) {
+
+ logger('limit_body_size: start', LOGGER_DEBUG);
+
+ $maxlen = get_max_import_size();
+
+ // If the length of the body, including the embedded images, is smaller
+ // than the maximum, then don't waste time looking for the images
+ if($maxlen && (strlen($body) > $maxlen)) {
+
+ logger('limit_body_size: the total body length exceeds the limit', LOGGER_DEBUG);
+
+ $orig_body = $body;
+ $new_body = '';
+ $textlen = 0;
+ $max_found = false;
+
+ $img_start = strpos($orig_body, '[img');
+ $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
+ $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
+ while(($img_st_close !== false) && ($img_end !== false)) {
+
+ $img_st_close++; // make it point to AFTER the closing bracket
+ $img_end += $img_start;
+ $img_end += strlen('[/img]');
+
+ if(! strcmp(substr($orig_body, $img_start + $img_st_close, 5), 'data:')) {
+ // This is an embedded image
+
+ if( ($textlen + $img_start) > $maxlen ) {
+ if($textlen < $maxlen) {
+ logger('limit_body_size: the limit happens before an embedded image', LOGGER_DEBUG);
+ $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
+ $textlen = $maxlen;
+ }
+ }
+ else {
+ $new_body = $new_body . substr($orig_body, 0, $img_start);
+ $textlen += $img_start;
+ }
+
+ $new_body = $new_body . substr($orig_body, $img_start, $img_end - $img_start);
+ }
+ else {
+
+ if( ($textlen + $img_end) > $maxlen ) {
+ if($textlen < $maxlen) {
+ logger('limit_body_size: the limit happens before the end of a non-embedded image', LOGGER_DEBUG);
+ $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
+ $textlen = $maxlen;
+ }
+ }
+ else {
+ $new_body = $new_body . substr($orig_body, 0, $img_end);
+ $textlen += $img_end;
+ }
+ }
+ $orig_body = substr($orig_body, $img_end);
+
+ if($orig_body === false) // in case the body ends on a closing image tag
+ $orig_body = '';
+
+ $img_start = strpos($orig_body, '[img');
+ $img_st_close = ($img_start !== false ? strpos(substr($orig_body, $img_start), ']') : false);
+ $img_end = ($img_start !== false ? strpos(substr($orig_body, $img_start), '[/img]') : false);
+ }
+
+ if( ($textlen + strlen($orig_body)) > $maxlen) {
+ if($textlen < $maxlen) {
+ logger('limit_body_size: the limit happens after the end of the last image', LOGGER_DEBUG);
+ $new_body = $new_body . substr($orig_body, 0, $maxlen - $textlen);
+ $textlen = $maxlen;
+ }
+ }
+ else {
+ logger('limit_body_size: the text size with embedded images extracted did not violate the limit', LOGGER_DEBUG);
+ $new_body = $new_body . $orig_body;
+ $textlen += strlen($orig_body);
+ }
+
+ return $new_body;
+ }
+ else
+ return $body;
+}}
+
+function title_is_body($title, $body) {
+
+ $title = strip_tags($title);
+ $title = trim($title);
+ $title = str_replace(array("\n", "\r", "\t", " "), array("","","",""), $title);
+
+ $body = strip_tags($body);
+ $body = trim($body);
+ $body = str_replace(array("\n", "\r", "\t", " "), array("","","",""), $body);
+
+ if (strlen($title) < strlen($body))
+ $body = substr($body, 0, strlen($title));
+
+ if (($title != $body) and (substr($title, -3) == "...")) {
+ $pos = strrpos($title, "...");
+ if ($pos > 0) {
+ $title = substr($title, 0, $pos);
+ $body = substr($body, 0, $pos);
+ }
+ }
+
+ return($title == $body);
+}
@@ -306,6 +423,11 @@ function get_atom_elements($feed,$item) {
$res['body'] = unxmlify($item->get_content());
$res['plink'] = unxmlify($item->get_link(0));
+ // removing the content of the title if its identically to the body
+ // This helps with auto generated titles e.g. from tumblr
+ if (title_is_body($res["title"], $res["body"]))
+ $res['title'] = "";
+
if($res['plink'])
$base_url = implode('/', array_slice(explode('/',$res['plink']),0,3));
else
@@ -324,7 +446,7 @@ function get_atom_elements($feed,$item) {
$res['author-avatar'] = unxmlify($link['attribs']['']['href']);
}
}
- }
+ }
$rawactor = $item->get_item_tags(NAMESPACE_ACTIVITY, 'actor');
@@ -356,7 +478,7 @@ function get_atom_elements($feed,$item) {
$res['author-avatar'] = unxmlify($link['attribs']['']['href']);
}
}
- }
+ }
$rawactor = $feed->get_feed_tags(NAMESPACE_ACTIVITY, 'subject');
@@ -381,7 +503,7 @@ function get_atom_elements($feed,$item) {
$res['app'] = strip_tags(unxmlify($apps[0]['attribs']['']['source']));
if($res['app'] === 'web')
$res['app'] = 'OStatus';
- }
+ }
// base64 encoded json structure representing Diaspora signature
@@ -414,9 +536,8 @@ function get_atom_elements($feed,$item) {
$res['body'] = notags(base64url_decode($res['body']));
}
- $maxlen = get_max_import_size();
- if($maxlen && (strlen($res['body']) > $maxlen))
- $res['body'] = substr($res['body'],0, $maxlen);
+
+ $res['body'] = limit_body_size($res['body']);
// It isn't certain at this point whether our content is plaintext or html and we'd be foolish to trust
// the content type. Our own network only emits text normally, though it might have been converted to
@@ -446,6 +567,8 @@ function get_atom_elements($feed,$item) {
$res['body'] = $purifier->purify($res['body']);
$res['body'] = @html2bbcode($res['body']);
+
+
}
elseif(! $have_real_body) {
@@ -455,6 +578,7 @@ function get_atom_elements($feed,$item) {
$res['body'] = escape_tags($res['body']);
}
+
// this tag is obsolete but we keep it for really old sites
$allow = $item->get_item_tags(NAMESPACE_DFRN,'comment-allow');
@@ -464,8 +588,8 @@ function get_atom_elements($feed,$item) {
$res['last-child'] = 0;
$private = $item->get_item_tags(NAMESPACE_DFRN,'private');
- if($private && $private[0]['data'] == 1)
- $res['private'] = 1;
+ if($private && intval($private[0]['data']) > 0)
+ $res['private'] = intval($private[0]['data']);
else
$res['private'] = 0;
@@ -523,7 +647,7 @@ function get_atom_elements($feed,$item) {
foreach($base as $link) {
if(!x($res, 'owner-avatar') || !$res['owner-avatar']) {
- if($link['attribs']['']['rel'] === 'photo' || $link['attribs']['']['rel'] === 'avatar')
+ if($link['attribs']['']['rel'] === 'photo' || $link['attribs']['']['rel'] === 'avatar')
$res['owner-avatar'] = unxmlify($link['attribs']['']['href']);
}
}
@@ -663,10 +787,41 @@ function get_atom_elements($feed,$item) {
$res['target'] .= '' . "\n";
}
+ // This is some experimental stuff. By now retweets are shown with "RT:"
+ // But: There is data so that the message could be shown similar to native retweets
+ // There is some better way to parse this array - but it didn't worked for me.
+ $child = $item->feed->data["child"][SIMPLEPIE_NAMESPACE_ATOM_10]["feed"][0]["child"][SIMPLEPIE_NAMESPACE_ATOM_10]["entry"][0]["child"]["http://activitystrea.ms/spec/1.0/"][object][0]["child"];
+ if (is_array($child)) {
+ $message = $child["http://activitystrea.ms/spec/1.0/"]["object"][0]["child"][SIMPLEPIE_NAMESPACE_ATOM_10]["content"][0]["data"];
+ $author = $child[SIMPLEPIE_NAMESPACE_ATOM_10]["author"][0]["child"][SIMPLEPIE_NAMESPACE_ATOM_10];
+ $uri = $author["uri"][0]["data"];
+ $name = $author["name"][0]["data"];
+ $avatar = @array_shift($author["link"][2]["attribs"]);
+ $avatar = $avatar["href"];
+
+ if (($name != "") and ($uri != "") and ($avatar != "") and ($message != "")) {
+ $res["owner-name"] = $res["author-name"];
+ $res["owner-link"] = $res["author-link"];
+ $res["owner-avatar"] = $res["author-avatar"];
+
+ $res["author-name"] = $name;
+ $res["author-link"] = $uri;
+ $res["author-avatar"] = $avatar;
+
+ $res["body"] = html2bbcode($message);
+ }
+ }
+
$arr = array('feed' => $feed, 'item' => $item, 'result' => $res);
call_hooks('parse_atom', $arr);
+ //if (($res["title"] != "") or (strpos($res["body"], "RT @") > 0)) {
+ //if (strpos($res["body"], "RT @") !== false) {
+ // $debugfile = tempnam("/home/ike/log", "item-res2-");
+ // file_put_contents($debugfile, serialize($arr));
+ //}
+
return $res;
}
@@ -722,6 +877,14 @@ function item_store($arr,$force_parent = false) {
$arr['body'] = strip_tags($arr['body']);
+ if (version_compare(PHP_VERSION, '5.3.0', '>=')) {
+ require_once('Text/LanguageDetect.php');
+ $naked_body = preg_replace('/\[(.+?)\]/','',$arr['body']);
+ $l = new Text_LanguageDetect;
+ $lng = $l->detectConfidence($naked_body);
+ $arr['postopts'] = (($lng['language']) ? 'lang=' . $lng['language'] . ';' . $lng['confidence'] : '');
+ }
+
$arr['wall'] = ((x($arr,'wall')) ? intval($arr['wall']) : 0);
$arr['uri'] = ((x($arr,'uri')) ? notags(trim($arr['uri'])) : random_string());
$arr['extid'] = ((x($arr,'extid')) ? notags(trim($arr['extid'])) : '');
@@ -762,6 +925,8 @@ function item_store($arr,$force_parent = false) {
$arr['origin'] = ((x($arr,'origin')) ? intval($arr['origin']) : 0 );
$arr['guid'] = ((x($arr,'guid')) ? notags(trim($arr['guid'])) : get_guid());
+
+ $arr['thr-parent'] = $arr['parent-uri'];
if($arr['parent-uri'] === $arr['uri']) {
$parent_id = 0;
$parent_deleted = 0;
@@ -787,9 +952,8 @@ function item_store($arr,$force_parent = false) {
// and re-attach to the conversation parent.
if($r[0]['uri'] != $r[0]['parent-uri']) {
- $arr['thr-parent'] = $arr['parent-uri'];
$arr['parent-uri'] = $r[0]['parent-uri'];
- $z = q("SELECT `id` FROM `item` WHERE `uri` = '%s' AND `parent-uri` = '%s' AND `uid` = %d
+ $z = q("SELECT * FROM `item` WHERE `uri` = '%s' AND `parent-uri` = '%s' AND `uid` = %d
ORDER BY `id` ASC LIMIT 1",
dbesc($r[0]['parent-uri']),
dbesc($r[0]['parent-uri']),
@@ -812,7 +976,7 @@ function item_store($arr,$force_parent = false) {
// email correspondents to be private even if the overall thread is not.
if($r[0]['private'])
- $arr['private'] = 1;
+ $arr['private'] = $r[0]['private'];
// Edge case. We host a public forum that was originally posted to privately.
// The original author commented, but as this is a comment, the permissions
@@ -829,7 +993,6 @@ function item_store($arr,$force_parent = false) {
if($force_parent) {
logger('item_store: $force_parent=true, reply converted to top-level post.');
$parent_id = 0;
- $arr['thr-parent'] = $arr['parent-uri'];
$arr['parent-uri'] = $arr['uri'];
$arr['gravity'] = 0;
}
@@ -1021,6 +1184,15 @@ function tag_deliver($uid,$item_id) {
// send a notification
+ // use a local photo if we have one
+
+ $r = q("select thumb from contact where uid = %d and nurl = '%s' limit 1",
+ intval($u[0]['uid']),
+ dbesc(normalise_link($item['author-link']))
+ );
+ $photo = (($r && count($r)) ? $r[0]['thumb'] : $item['author-avatar']);
+
+
require_once('include/enotify.php');
notification(array(
'type' => NOTIFY_TAGSELF,
@@ -1033,7 +1205,7 @@ function tag_deliver($uid,$item_id) {
'link' => $a->get_baseurl() . '/display/' . $u[0]['nickname'] . '/' . $item['id'],
'source_name' => $item['author-name'],
'source_link' => $item['author-link'],
- 'source_photo' => $item['author-avatar'],
+ 'source_photo' => $photo,
'verb' => ACTIVITY_TAG,
'otype' => 'item'
));
@@ -1084,6 +1256,59 @@ function tag_deliver($uid,$item_id) {
+function tgroup_check($uid,$item) {
+
+ $a = get_app();
+
+ $mention = false;
+
+ // check that the message originated elsewhere and is a top-level post
+
+ if(($item['wall']) || ($item['origin']) || ($item['uri'] != $item['parent-uri']))
+ return false;
+
+
+ $u = q("select * from user where uid = %d limit 1",
+ intval($uid)
+ );
+ if(! count($u))
+ return false;
+
+ $community_page = (($u[0]['page-flags'] == PAGE_COMMUNITY) ? true : false);
+ $prvgroup = (($u[0]['page-flags'] == PAGE_PRVGROUP) ? true : false);
+
+
+ $link = normalise_link($a->get_baseurl() . '/profile/' . $u[0]['nickname']);
+
+ // Diaspora uses their own hardwired link URL in @-tags
+ // instead of the one we supply with webfinger
+
+ $dlink = normalise_link($a->get_baseurl() . '/u/' . $u[0]['nickname']);
+
+ $cnt = preg_match_all('/[\@\!]\[url\=(.*?)\](.*?)\[\/url\]/ism',$item['body'],$matches,PREG_SET_ORDER);
+ if($cnt) {
+ foreach($matches as $mtch) {
+ if(link_compare($link,$mtch[1]) || link_compare($dlink,$mtch[1])) {
+ $mention = true;
+ logger('tgroup_check: mention found: ' . $mtch[2]);
+ }
+ }
+ }
+
+ if(! $mention)
+ return false;
+
+ if((! $community_page) && (! $prvgroup))
+ return false;
+
+
+
+ return true;
+
+}
+
+
+
@@ -1264,6 +1489,12 @@ function dfrn_deliver($owner,$contact,$atom, $dissolve = false) {
return 3;
}
+ if($contact['term-date'] != '0000-00-00 00:00:00') {
+ logger("dfrn_deliver: $url back from the dead - removing mark for death");
+ require_once('include/Contact.php');
+ unmark_for_death($contact);
+ }
+
$res = parse_xml_string($xml);
return $res->status;
@@ -1449,11 +1680,12 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
*
*/
- $bdtext = t('Birthday:') . ' [url=' . $contact['url'] . ']' . $contact['name'] . '[/url]' ;
+ $bdtext = sprintf( t('%s\'s birthday'), $contact['name']);
+ $bdtext2 = sprintf( t('Happy Birthday %s'), ' [url=' . $contact['url'] . ']' . $contact['name'] . '[/url]' ) ;
- $r = q("INSERT INTO `event` (`uid`,`cid`,`created`,`edited`,`start`,`finish`,`desc`,`type`)
- VALUES ( %d, %d, '%s', '%s', '%s', '%s', '%s', '%s' ) ",
+ $r = q("INSERT INTO `event` (`uid`,`cid`,`created`,`edited`,`start`,`finish`,`summary`,`desc`,`type`)
+ VALUES ( %d, %d, '%s', '%s', '%s', '%s', '%s', '%s', '%s' ) ",
intval($contact['uid']),
intval($contact['id']),
dbesc(datetime_convert()),
@@ -1461,6 +1693,7 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
dbesc(datetime_convert('UTC','UTC', $birthday)),
dbesc(datetime_convert('UTC','UTC', $birthday . ' + 1 day ')),
dbesc($bdtext),
+ dbesc($bdtext2),
dbesc('birthday')
);
@@ -1606,7 +1839,7 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
// Now process the feed
- if($feed->get_item_quantity()) {
+ if($feed->get_item_quantity()) {
logger('consume_feed: feed item count = ' . $feed->get_item_quantity());
@@ -1619,7 +1852,7 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
foreach($items as $item) {
- $is_reply = false;
+ $is_reply = false;
$item_id = $item->get_id();
$rawthread = $item->get_item_tags( NAMESPACE_THREAD,'in-reply-to');
if(isset($rawthread[0]['attribs']['']['ref'])) {
@@ -1632,12 +1865,17 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
if($pass == 1)
continue;
+ // not allowed to post
+
+ if($contact['rel'] == CONTACT_IS_FOLLOWER)
+ continue;
+
+
// Have we seen it? If not, import it.
-
+
$item_id = $item->get_id();
$datarray = get_atom_elements($feed,$item);
-
if((! x($datarray,'author-name')) && ($contact['network'] != NETWORK_DFRN))
$datarray['author-name'] = $contact['name'];
if((! x($datarray,'author-link')) && ($contact['network'] != NETWORK_DFRN))
@@ -1650,6 +1888,21 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
continue;
}
+ $force_parent = false;
+ if($contact['network'] === NETWORK_OSTATUS || stristr($contact['url'],'twitter.com')) {
+ if($contact['network'] === NETWORK_OSTATUS)
+ $force_parent = true;
+ if(strlen($datarray['title']))
+ unset($datarray['title']);
+ $r = q("UPDATE `item` SET `last-child` = 0, `changed` = '%s' WHERE `parent-uri` = '%s' AND `uid` = %d",
+ dbesc(datetime_convert()),
+ dbesc($parent_uri),
+ intval($importer['uid'])
+ );
+ $datarray['last-child'] = 1;
+ }
+
+
$r = q("SELECT `uid`, `last-child`, `edited`, `body` FROM `item` WHERE `uri` = '%s' AND `uid` = %d LIMIT 1",
dbesc($item_id),
intval($importer['uid'])
@@ -1693,19 +1946,6 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
continue;
}
- $force_parent = false;
- if($contact['network'] === NETWORK_OSTATUS || stristr($contact['url'],'twitter.com')) {
- if($contact['network'] === NETWORK_OSTATUS)
- $force_parent = true;
- if(strlen($datarray['title']))
- unset($datarray['title']);
- $r = q("UPDATE `item` SET `last-child` = 0, `changed` = '%s' WHERE `parent-uri` = '%s' AND `uid` = %d",
- dbesc(datetime_convert()),
- dbesc($parent_uri),
- intval($importer['uid'])
- );
- $datarray['last-child'] = 1;
- }
if(($contact['network'] === NETWORK_FEED) || (! strlen($contact['notify']))) {
// one way feed - no remote comment ability
@@ -1718,10 +1958,12 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
$datarray['type'] = 'activity';
$datarray['gravity'] = GRAVITY_LIKE;
// only one like or dislike per person
- $r = q("select id from item where uid = %d and `contact-id` = %d and verb ='%s' and deleted = 0 limit 1",
+ $r = q("select id from item where uid = %d and `contact-id` = %d and verb ='%s' and deleted = 0 and (`parent-uri` = '%s' OR `thr-parent` = '%s') limit 1",
intval($datarray['uid']),
intval($datarray['contact-id']),
- dbesc($datarray['verb'])
+ dbesc($datarray['verb']),
+ dbesc($parent_uri),
+ dbesc($parent_uri)
);
if($r && count($r))
continue;
@@ -1801,6 +2043,13 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
}
}
+ if($contact['network'] === NETWORK_OSTATUS || stristr($contact['url'],'twitter.com')) {
+ if(strlen($datarray['title']))
+ unset($datarray['title']);
+ $datarray['last-child'] = 1;
+ }
+
+
$r = q("SELECT `uid`, `last-child`, `edited`, `body` FROM `item` WHERE `uri` = '%s' AND `uid` = %d LIMIT 1",
dbesc($item_id),
intval($importer['uid'])
@@ -1863,24 +2112,23 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
if(! is_array($contact))
return;
- if($contact['network'] === NETWORK_OSTATUS || stristr($contact['url'],'twitter.com')) {
- if(strlen($datarray['title']))
- unset($datarray['title']);
- $datarray['last-child'] = 1;
- }
if(($contact['network'] === NETWORK_FEED) || (! strlen($contact['notify']))) {
// one way feed - no remote comment ability
$datarray['last-child'] = 0;
}
if($contact['network'] === NETWORK_FEED)
- $datarray['private'] = 1;
+ $datarray['private'] = 2;
// This is my contact on another system, but it's really me.
// Turn this into a wall post.
- if($contact['remote_self'])
+ if($contact['remote_self']) {
$datarray['wall'] = 1;
+ if($contact['network'] === NETWORK_FEED) {
+ $datarray['private'] = 0;
+ }
+ }
$datarray['parent-uri'] = $item_id;
$datarray['uid'] = $importer['uid'];
@@ -1897,6 +2145,14 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
$datarray['owner-avatar'] = $contact['thumb'];
}
+ // We've allowed "followers" to reach this point so we can decide if they are
+ // posting an @-tag delivery, which followers are allowed to do for certain
+ // page types. Now that we've parsed the post, let's check if it is legit. Otherwise ignore it.
+
+ if(($contact['rel'] == CONTACT_IS_FOLLOWER) && (! tgroup_check($importer['uid'],$datarray)))
+ continue;
+
+
$r = item_store($datarray);
continue;
@@ -1928,6 +2184,121 @@ function local_delivery($importer,$data) {
$feed->enable_order_by_date(false);
$feed->init();
+
+ if($feed->error())
+ logger('local_delivery: Error parsing XML: ' . $feed->error());
+
+
+ // Check at the feed level for updated contact name and/or photo
+
+ $name_updated = '';
+ $new_name = '';
+ $photo_timestamp = '';
+ $photo_url = '';
+
+
+ $rawtags = $feed->get_feed_tags( NAMESPACE_DFRN, 'owner');
+
+// Fallback should not be needed here. If it isn't DFRN it won't have DFRN updated tags
+// if(! $rawtags)
+// $rawtags = $feed->get_feed_tags( SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
+
+ if($rawtags) {
+ $elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10];
+ if($elems['name'][0]['attribs'][NAMESPACE_DFRN]['updated']) {
+ $name_updated = $elems['name'][0]['attribs'][NAMESPACE_DFRN]['updated'];
+ $new_name = $elems['name'][0]['data'];
+ }
+ if((x($elems,'link')) && ($elems['link'][0]['attribs']['']['rel'] === 'photo') && ($elems['link'][0]['attribs'][NAMESPACE_DFRN]['updated'])) {
+ $photo_timestamp = datetime_convert('UTC','UTC',$elems['link'][0]['attribs'][NAMESPACE_DFRN]['updated']);
+ $photo_url = $elems['link'][0]['attribs']['']['href'];
+ }
+ }
+
+ if(($photo_timestamp) && (strlen($photo_url)) && ($photo_timestamp > $importer['avatar-date'])) {
+ logger('local_delivery: Updating photo for ' . $importer['name']);
+ require_once("Photo.php");
+ $photo_failure = false;
+ $have_photo = false;
+
+ $r = q("SELECT `resource-id` FROM `photo` WHERE `contact-id` = %d AND `uid` = %d LIMIT 1",
+ intval($importer['id']),
+ intval($importer['importer_uid'])
+ );
+ if(count($r)) {
+ $resource_id = $r[0]['resource-id'];
+ $have_photo = true;
+ }
+ else {
+ $resource_id = photo_new_resource();
+ }
+
+ $img_str = fetch_url($photo_url,true);
+ // guess mimetype from headers or filename
+ $type = guess_image_type($photo_url,true);
+
+
+ $img = new Photo($img_str, $type);
+ if($img->is_valid()) {
+ if($have_photo) {
+ q("DELETE FROM `photo` WHERE `resource-id` = '%s' AND `contact-id` = %d AND `uid` = %d",
+ dbesc($resource_id),
+ intval($importer['id']),
+ intval($importer['importer_uid'])
+ );
+ }
+
+ $img->scaleImageSquare(175);
+
+ $hash = $resource_id;
+ $r = $img->store($importer['importer_uid'], $importer['id'], $hash, basename($photo_url), 'Contact Photos', 4);
+
+ $img->scaleImage(80);
+ $r = $img->store($importer['importer_uid'], $importer['id'], $hash, basename($photo_url), 'Contact Photos', 5);
+
+ $img->scaleImage(48);
+ $r = $img->store($importer['importer_uid'], $importer['id'], $hash, basename($photo_url), 'Contact Photos', 6);
+
+ $a = get_app();
+
+ q("UPDATE `contact` SET `avatar-date` = '%s', `photo` = '%s', `thumb` = '%s', `micro` = '%s'
+ WHERE `uid` = %d AND `id` = %d LIMIT 1",
+ dbesc(datetime_convert()),
+ dbesc($a->get_baseurl() . '/photo/' . $hash . '-4.'.$img->getExt()),
+ dbesc($a->get_baseurl() . '/photo/' . $hash . '-5.'.$img->getExt()),
+ dbesc($a->get_baseurl() . '/photo/' . $hash . '-6.'.$img->getExt()),
+ intval($importer['importer_uid']),
+ intval($importer['id'])
+ );
+ }
+ }
+
+ if(($name_updated) && (strlen($new_name)) && ($name_updated > $importer['name-date'])) {
+ $r = q("select * from contact where uid = %d and id = %d limit 1",
+ intval($importer['importer_uid']),
+ intval($importer['id'])
+ );
+
+ $x = q("UPDATE `contact` SET `name` = '%s', `name-date` = '%s' WHERE `uid` = %d AND `id` = %d LIMIT 1",
+ dbesc(notags(trim($new_name))),
+ dbesc(datetime_convert()),
+ intval($importer['importer_uid']),
+ intval($importer['id'])
+ );
+
+ // do our best to update the name on content items
+
+ if(count($r)) {
+ q("update item set `author-name` = '%s' where `author-name` = '%s' and `author-link` = '%s' and uid = %d",
+ dbesc(notags(trim($new_name))),
+ dbesc($r[0]['name']),
+ dbesc($r[0]['url']),
+ intval($importer['importer_uid'])
+ );
+ }
+ }
+
+
/*
// Currently unsupported - needs a lot of work
$reloc = $feed->get_feed_tags( NAMESPACE_DFRN, 'relocate' );
@@ -2138,6 +2509,68 @@ function local_delivery($importer,$data) {
}
if($deleted) {
+ // check for relayed deletes to our conversation
+
+ $is_reply = false;
+ $r = q("select * from item where uri = '%s' and uid = %d limit 1",
+ dbesc($uri),
+ intval($importer['importer_uid'])
+ );
+ if(count($r)) {
+ $parent_uri = $r[0]['parent-uri'];
+ if($r[0]['id'] != $r[0]['parent'])
+ $is_reply = true;
+ }
+
+ if($is_reply) {
+ $community = false;
+
+ if($importer['page-flags'] == PAGE_COMMUNITY || $importer['page-flags'] == PAGE_PRVGROUP ) {
+ $sql_extra = '';
+ $community = true;
+ logger('local_delivery: possible community delete');
+ }
+ else
+ $sql_extra = " and contact.self = 1 and item.wall = 1 ";
+
+ // was the top-level post for this reply written by somebody on this site?
+ // Specifically, the recipient?
+
+ $is_a_remote_delete = false;
+
+ // POSSIBLE CLEANUP --> Why select so many fields when only forum_mode and wall are used?
+ $r = q("select `item`.`id`, `item`.`uri`, `item`.`tag`, `item`.`forum_mode`,`item`.`origin`,`item`.`wall`,
+ `contact`.`name`, `contact`.`url`, `contact`.`thumb` from `item`
+ LEFT JOIN `contact` ON `contact`.`id` = `item`.`contact-id`
+ WHERE `item`.`uri` = '%s' AND (`item`.`parent-uri` = '%s' or `item`.`thr-parent` = '%s')
+ AND `item`.`uid` = %d
+ $sql_extra
+ LIMIT 1",
+ dbesc($parent_uri),
+ dbesc($parent_uri),
+ dbesc($parent_uri),
+ intval($importer['importer_uid'])
+ );
+ if($r && count($r))
+ $is_a_remote_delete = true;
+
+ // Does this have the characteristics of a community or private group comment?
+ // If it's a reply to a wall post on a community/prvgroup page it's a
+ // valid community comment. Also forum_mode makes it valid for sure.
+ // If neither, it's not.
+
+ if($is_a_remote_delete && $community) {
+ if((! $r[0]['forum_mode']) && (! $r[0]['wall'])) {
+ $is_a_remote_delete = false;
+ logger('local_delivery: not a community delete');
+ }
+ }
+
+ if($is_a_remote_delete) {
+ logger('local_delivery: received remote delete');
+ }
+ }
+
$r = q("SELECT `item`.*, `contact`.`self` FROM `item` left join contact on `item`.`contact-id` = `contact`.`id`
WHERE `uri` = '%s' AND `item`.`uid` = %d AND `contact-id` = %d AND NOT `item`.`file` LIKE '%%[%%' LIMIT 1",
dbesc($uri),
@@ -2190,7 +2623,8 @@ function local_delivery($importer,$data) {
}
if($item['uri'] == $item['parent-uri']) {
- $r = q("UPDATE `item` SET `deleted` = 1, `edited` = '%s', `changed` = '%s'
+ $r = q("UPDATE `item` SET `deleted` = 1, `edited` = '%s', `changed` = '%s',
+ `body` = '', `title` = ''
WHERE `parent-uri` = '%s' AND `uid` = %d",
dbesc($when),
dbesc(datetime_convert()),
@@ -2199,7 +2633,8 @@ function local_delivery($importer,$data) {
);
}
else {
- $r = q("UPDATE `item` SET `deleted` = 1, `edited` = '%s', `changed` = '%s'
+ $r = q("UPDATE `item` SET `deleted` = 1, `edited` = '%s', `changed` = '%s',
+ `body` = '', `title` = ''
WHERE `uri` = '%s' AND `uid` = %d LIMIT 1",
dbesc($when),
dbesc(datetime_convert()),
@@ -2225,7 +2660,11 @@ function local_delivery($importer,$data) {
);
}
}
- }
+ // if this is a relayed delete, propagate it to other recipients
+
+ if($is_a_remote_delete)
+ proc_run('php',"include/notifier.php","drop",$item['id']);
+ }
}
}
}
@@ -2257,20 +2696,32 @@ function local_delivery($importer,$data) {
// Specifically, the recipient?
$is_a_remote_comment = false;
-
- $r = q("select `item`.`id`, `item`.`uri`, `item`.`tag`, `item`.`forum_mode`,`item`.`origin`,`item`.`wall`,
- `contact`.`name`, `contact`.`url`, `contact`.`thumb` from `item`
- LEFT JOIN `contact` ON `contact`.`id` = `item`.`contact-id`
- WHERE `item`.`uri` = '%s' AND `item`.`parent-uri` = '%s'
- AND `item`.`uid` = %d
- $sql_extra
+ $top_uri = $parent_uri;
+
+ $r = q("select `item`.`parent-uri` from `item`
+ WHERE `item`.`uri` = '%s'
LIMIT 1",
- dbesc($parent_uri),
- dbesc($parent_uri),
- intval($importer['importer_uid'])
+ dbesc($parent_uri)
);
- if($r && count($r))
- $is_a_remote_comment = true;
+ if($r && count($r)) {
+ $top_uri = $r[0]['parent-uri'];
+
+ // POSSIBLE CLEANUP --> Why select so many fields when only forum_mode and wall are used?
+ $r = q("select `item`.`id`, `item`.`uri`, `item`.`tag`, `item`.`forum_mode`,`item`.`origin`,`item`.`wall`,
+ `contact`.`name`, `contact`.`url`, `contact`.`thumb` from `item`
+ LEFT JOIN `contact` ON `contact`.`id` = `item`.`contact-id`
+ WHERE `item`.`uri` = '%s' AND (`item`.`parent-uri` = '%s' or `item`.`thr-parent` = '%s')
+ AND `item`.`uid` = %d
+ $sql_extra
+ LIMIT 1",
+ dbesc($top_uri),
+ dbesc($top_uri),
+ dbesc($top_uri),
+ intval($importer['importer_uid'])
+ );
+ if($r && count($r))
+ $is_a_remote_comment = true;
+ }
// Does this have the characteristics of a community or private group comment?
// If it's a reply to a wall post on a community/prvgroup page it's a
@@ -2324,15 +2775,6 @@ function local_delivery($importer,$data) {
}
- // TODO: make this next part work against both delivery threads of a community post
-
-// if((! link_compare($datarray['author-link'],$importer['url'])) && (! $community)) {
-// logger('local_delivery: received relay claiming to be from ' . $importer['url'] . ' however comment author url is ' . $datarray['author-link'] );
- // they won't know what to do so don't report an error. Just quietly die.
-// return 0;
-// }
-
- // our user with $importer['importer_uid'] is the owner
$own = q("select name,url,thumb from contact where uid = %d and self = 1 limit 1",
intval($importer['importer_uid'])
@@ -2354,10 +2796,13 @@ function local_delivery($importer,$data) {
$datarray['gravity'] = GRAVITY_LIKE;
$datarray['last-child'] = 0;
// only one like or dislike per person
- $r = q("select id from item where uid = %d and `contact-id` = %d and verb ='%s' and deleted = 0 limit 1",
+ $r = q("select id from item where uid = %d and `contact-id` = %d and verb = '%s' and (`thr-parent` = '%s' or `parent-uri` = '%s') and deleted = 0 limit 1",
intval($datarray['uid']),
intval($datarray['contact-id']),
- dbesc($datarray['verb'])
+ dbesc($datarray['verb']),
+ dbesc($datarray['parent-uri']),
+ dbesc($datarray['parent-uri'])
+
);
if($r && count($r))
continue;
@@ -2399,26 +2844,19 @@ function local_delivery($importer,$data) {
}
}
-// if($community) {
-// $newtag = '@[url=' . $a->get_baseurl() . '/profile/' . $importer['nickname'] . ']' . $importer['username'] . '[/url]';
-// if(! stristr($datarray['tag'],$newtag)) {
-// if(strlen($datarray['tag']))
-// $datarray['tag'] .= ',';
-// $datarray['tag'] .= $newtag;
-// }
-// }
-
$posted_id = item_store($datarray);
$parent = 0;
if($posted_id) {
- $r = q("SELECT `parent` FROM `item` WHERE `id` = %d AND `uid` = %d LIMIT 1",
+ $r = q("SELECT `parent`, `parent-uri` FROM `item` WHERE `id` = %d AND `uid` = %d LIMIT 1",
intval($posted_id),
intval($importer['importer_uid'])
);
- if(count($r))
+ if(count($r)) {
$parent = $r[0]['parent'];
+ $parent_uri = $r[0]['parent-uri'];
+ }
if(! $is_like) {
$r1 = q("UPDATE `item` SET `last-child` = 0, `changed` = '%s' WHERE `uid` = %d AND `parent` = %d",
@@ -2435,7 +2873,7 @@ function local_delivery($importer,$data) {
}
if($posted_id && $parent) {
-
+
proc_run('php',"include/notifier.php","comment-import","$posted_id");
if((! $is_like) && (! $importer['self'])) {
@@ -2458,7 +2896,7 @@ function local_delivery($importer,$data) {
'verb' => ACTIVITY_POST,
'otype' => 'item',
'parent' => $parent,
-
+ 'parent_uri' => $parent_uri,
));
}
@@ -2475,6 +2913,9 @@ function local_delivery($importer,$data) {
$item_id = $item->get_id();
$datarray = get_atom_elements($feed,$item);
+ if($importer['rel'] == CONTACT_IS_FOLLOWER)
+ continue;
+
$r = q("SELECT `uid`, `last-child`, `edited`, `body` FROM `item` WHERE `uri` = '%s' AND `uid` = %d LIMIT 1",
dbesc($item_id),
intval($importer['importer_uid'])
@@ -2525,10 +2966,12 @@ function local_delivery($importer,$data) {
$datarray['type'] = 'activity';
$datarray['gravity'] = GRAVITY_LIKE;
// only one like or dislike per person
- $r = q("select id from item where uid = %d and `contact-id` = %d and verb ='%s' and deleted = 0 limit 1",
+ $r = q("select id from item where uid = %d and `contact-id` = %d and verb ='%s' and deleted = 0 and (`parent-uri` = '%s' OR `thr-parent` = '%s') limit 1",
intval($datarray['uid']),
intval($datarray['contact-id']),
- dbesc($datarray['verb'])
+ dbesc($datarray['verb']),
+ dbesc($parent_uri),
+ dbesc($parent_uri)
);
if($r && count($r))
continue;
@@ -2567,7 +3010,7 @@ function local_delivery($importer,$data) {
if(!x($datarray['type']) || $datarray['type'] != 'activity') {
$myconv = q("SELECT `author-link`, `author-avatar`, `parent` FROM `item` WHERE `parent-uri` = '%s' AND `uid` = %d AND `parent` != 0 AND `deleted` = 0",
- dbesc($parent_uri),
+ dbesc($top_uri),
intval($importer['importer_uid'])
);
@@ -2605,6 +3048,7 @@ function local_delivery($importer,$data) {
'verb' => ACTIVITY_POST,
'otype' => 'item',
'parent' => $conv_parent,
+ 'parent_uri' => $parent_uri
));
@@ -2694,7 +3138,8 @@ function local_delivery($importer,$data) {
$datarray['uid'] = $importer['importer_uid'];
$datarray['contact-id'] = $importer['id'];
- if(! link_compare($datarray['owner-link'],$contact['url'])) {
+
+ if(! link_compare($datarray['owner-link'],$importer['url'])) {
// The item owner info is not our contact. It's OK and is to be expected if this is a tgroup delivery,
// but otherwise there's a possible data mixup on the sender's system.
// the tgroup delivery code called from item_store will correct it if it's a forum,
@@ -2705,7 +3150,60 @@ function local_delivery($importer,$data) {
$datarray['owner-avatar'] = $importer['thumb'];
}
- $r = item_store($datarray);
+ if(($importer['rel'] == CONTACT_IS_FOLLOWER) && (! tgroup_check($importer['importer_uid'],$datarray)))
+ continue;
+
+ $posted_id = item_store($datarray);
+
+ if(stristr($datarray['verb'],ACTIVITY_POKE)) {
+ $verb = urldecode(substr($datarray['verb'],strpos($datarray['verb'],'#')+1));
+ if(! $verb)
+ continue;
+ $xo = parse_xml_string($datarray['object'],false);
+
+ if(($xo->type == ACTIVITY_OBJ_PERSON) && ($xo->id)) {
+
+ // somebody was poked/prodded. Was it me?
+
+ $links = parse_xml_string("
`)
+* organize methods better (i.e. flushlinebreaks & setlinebreaks close to each other)
+* take a look at function names etc.
+* is the new (in rev. 93) lastclosedtag property needed?
+* word wrapping (some work is done but it's still very buggy)
+
+
+Markdownify Extra
+=================
+
+* handle table alignment with KEEP_HTML=false
+* handle tables without headings when KEEP_HTML=false is set
+* handle Markdown inside non-markdownable tags
+
+
+Implementation Thoughts
+=======================
+* non-markdownifiable lists and markdown inside non-markdownable tags as well as the current
+ table implementation could be rewritten by using a rollback mechanism.
+
+ example:
+
+
+
+ we come to ``, know that this might fail and create a snapshot of our current parser
+ we keep on parsing and when we reach `
BACK
+
+
+
+
\ No newline at end of file
diff --git a/include/markdownify/markdownify.php b/include/markdownify/markdownify.php
new file mode 100644
index 0000000000..0d4429a013
--- /dev/null
+++ b/include/markdownify/markdownify.php
@@ -0,0 +1,1197 @@
+, tags
+ $this->out($this->parser->node);
+ static $indent;
+ $indent = $this->indent;
+ $this->indent = '';
+ }
+ } else {
+ if (!$this->parser->keepWhitespace) {
+ $this->output = rtrim($this->output);
+ }
+ if ($this->parser->tagName != 'pre') {
+ $this->indent(' ');
+ $this->out("\n".$this->indent.$this->parser->node);
+ } else {
+ # reset indentation
+ $this->out($this->parser->node);
+ static $indent;
+ $this->indent = $indent;
+ }
+
+ if (in_array($this->parent(), array('ins', 'del'))) {
+ # ins or del was block element
+ $this->out("\n");
+ $this->indent(' ');
+ }
+ if ($this->parser->tagName == 'li') {
+ $this->setLineBreaks(1);
+ } else {
+ $this->setLineBreaks(2);
+ }
+ }
+ } else {
+ $this->out($this->parser->node);
+ }
+ if (in_array($this->parser->tagName, array('code', 'pre'))) {
+ if ($this->parser->isStartTag) {
+ $this->buffer();
+ } else {
+ # add stuff so cleanup just reverses this
+ $this->out(str_replace('<', '<', str_replace('>', '>', $this->unbuffer())));
+ }
+ }
+ }
+ }
+ /**
+ * handle plain text
+ *
+ * @param void
+ * @return void
+ */
+ function handleText() {
+ if ($this->hasParent('pre') && strpos($this->parser->node, "\n") !== false) {
+ $this->parser->node = str_replace("\n", "\n".$this->indent, $this->parser->node);
+ }
+ if (!$this->hasParent('code') && !$this->hasParent('pre')) {
+ # entity decode
+ $this->parser->node = $this->decode($this->parser->node);
+ if (!$this->skipConversion) {
+ # escape some chars in normal Text
+ $this->parser->node = preg_replace($this->escapeInText['search'], $this->escapeInText['replace'], $this->parser->node);
+ }
+ } else {
+ $this->parser->node = str_replace(array('"', '&apos'), array('"', '\''), $this->parser->node);
+ }
+ $this->out($this->parser->node);
+ $this->lastClosedTag = '';
+ }
+ /**
+ * handle and tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_em() {
+ $this->out('*', true);
+ }
+ function handleTag_i() {
+ $this->handleTag_em();
+ }
+ /**
+ * handle and tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_strong() {
+ $this->out('**', true);
+ }
+ function handleTag_b() {
+ $this->handleTag_strong();
+ }
+ /**
+ * handle tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_h1() {
+ $this->handleHeader(1);
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_h2() {
+ $this->handleHeader(2);
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_h3() {
+ $this->handleHeader(3);
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_h4() {
+ $this->handleHeader(4);
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_h5() {
+ $this->handleHeader(5);
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_h6() {
+ $this->handleHeader(6);
+ }
+ /**
+ * number of line breaks before next inline output
+ */
+ var $lineBreaks = 0;
+ /**
+ * handle header tags (
-
)
+ *
+ * @param int $level 1-6
+ * @return void
+ */
+ function handleHeader($level) {
+ if ($this->parser->isStartTag) {
+ $this->out(str_repeat('#', $level).' ', true);
+ } else {
+ $this->setLineBreaks(2);
+ }
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_img() {
+ if (!$this->parser->isStartTag) {
+ return; # just to be sure this is really an empty tag...
+ }
+
+ if (isset($this->parser->tagAttributes['title'])) {
+ $this->parser->tagAttributes['title'] = $this->decode($this->parser->tagAttributes['title']);
+ } else {
+ $this->parser->tagAttributes['title'] = null;
+ }
+ if (isset($this->parser->tagAttributes['alt'])) {
+ $this->parser->tagAttributes['alt'] = $this->decode($this->parser->tagAttributes['alt']);
+ } else {
+ $this->parser->tagAttributes['alt'] = null;
+ }
+
+ if (empty($this->parser->tagAttributes['src'])) {
+ # support for "empty" images... dunno if this is really needed
+ # but there are some testcases which do that...
+ if (!empty($this->parser->tagAttributes['title'])) {
+ $this->parser->tagAttributes['title'] = ' '.$this->parser->tagAttributes['title'].' ';
+ }
+ $this->out('!['.$this->parser->tagAttributes['alt'].']('.$this->parser->tagAttributes['title'].')', true);
+ return;
+ } else {
+ $this->parser->tagAttributes['src'] = $this->decode($this->parser->tagAttributes['src']);
+ }
+
+// 
+ if ($this->parser->tagAttributes['title'] != "")
+ $this->out('!['.$this->parser->tagAttributes['alt'].']('.$this->parser->tagAttributes['src'].' "'.$this->parser->tagAttributes['title'].'")', true);
+ else
+ $this->out('!['.$this->parser->tagAttributes['alt'].']('.$this->parser->tagAttributes['src'].')', true);
+
+/*
+ # [This link][id]
+ $link_id = false;
+ if (!empty($this->stack['a'])) {
+ foreach ($this->stack['a'] as $tag) {
+ if ($tag['href'] == $this->parser->tagAttributes['src']
+ && $tag['title'] === $this->parser->tagAttributes['title']) {
+ $link_id = $tag['linkID'];
+ break;
+ }
+ }
+ } else {
+ $this->stack['a'] = array();
+ }
+ if (!$link_id) {
+ $link_id = count($this->stack['a']) + 1;
+ $tag = array(
+ 'href' => $this->parser->tagAttributes['src'],
+ 'linkID' => $link_id,
+ 'title' => $this->parser->tagAttributes['title']
+ );
+ array_push($this->stack['a'], $tag);
+ }
+
+ $this->out('!['.$this->parser->tagAttributes['alt'].']['.$link_id.']', true);
+*/
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_code() {
+ if ($this->hasParent('pre')) {
+ # ignore code blocks inside
+ return;
+ }
+ if ($this->parser->isStartTag) {
+ $this->buffer();
+ } else {
+ $buffer = $this->unbuffer();
+ # use as many backticks as needed
+ preg_match_all('#`+#', $buffer, $matches);
+ if (!empty($matches[0])) {
+ rsort($matches[0]);
+
+ $ticks = '`';
+ while (true) {
+ if (!in_array($ticks, $matches[0])) {
+ break;
+ }
+ $ticks .= '`';
+ }
+ } else {
+ $ticks = '`';
+ }
+ if ($buffer[0] == '`' || substr($buffer, -1) == '`') {
+ $buffer = ' '.$buffer.' ';
+ }
+ $this->out($ticks.$buffer.$ticks, true);
+ }
+ }
+ /**
+ * handle tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_pre() {
+ if ($this->keepHTML && $this->parser->isStartTag) {
+ # check if a simple follows
+ if (!preg_match('#^\s* tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_blockquote() {
+ $this->indent('> ');
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_ul() {
+ if ($this->parser->isStartTag) {
+ $this->stack();
+ if (!$this->keepHTML && $this->lastClosedTag == $this->parser->tagName) {
+ $this->out("\n".$this->indent.''."\n".$this->indent."\n".$this->indent);
+ }
+ } else {
+ $this->unstack();
+ if ($this->parent() != 'li' || preg_match('#^\s*(\s*
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_ol() {
+ # same as above
+ $this->parser->tagAttributes['num'] = 0;
+ $this->handleTag_ul();
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_hr() {
+ if (!$this->parser->isStartTag) {
+ return; # just to be sure this really is an empty tag
+ }
+ $this->out('* * *', true);
+ $this->setLineBreaks(2);
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_br() {
+ $this->out(" \n".$this->indent, true);
+ $this->parser->html = ltrim($this->parser->html);
+ }
+ /**
+ * node stack, e.g. for and tags
+ *
+ * @var array -
)
+ *
+ * @param int $level 1-6
+ * @return void
+ */
+ function handleHeader($level) {
+ static $id = null;
+ if ($this->parser->isStartTag) {
+ if (isset($this->parser->tagAttributes['id'])) {
+ $id = $this->parser->tagAttributes['id'];
+ }
+ } else {
+ if (!is_null($id)) {
+ $this->out(' {#'.$id.'}');
+ $id = null;
+ }
+ }
+ parent::handleHeader($level);
+ }
+ /**
+ * handle tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_abbr() {
+ if ($this->parser->isStartTag) {
+ $this->stack();
+ $this->buffer();
+ } else {
+ $tag = $this->unstack();
+ $tag['text'] = $this->unbuffer();
+ $add = true;
+ foreach ($this->stack['abbr'] as $stacked) {
+ if ($stacked['text'] == $tag['text']) {
+ /** TODO: differing abbr definitions, i.e. different titles for same text **/
+ $add = false;
+ break;
+ }
+ }
+ $this->out($tag['text']);
+ if ($add) {
+ array_push($this->stack['abbr'], $tag);
+ }
+ }
+ }
+ /**
+ * flush stacked abbr tags
+ *
+ * @param void
+ * @return void
+ */
+ function flushStacked_abbr() {
+ $out = array();
+ foreach ($this->stack['abbr'] as $k => $tag) {
+ if (!isset($tag['unstacked'])) {
+ array_push($out, ' *['.$tag['text'].']: '.$tag['title']);
+ $tag['unstacked'] = true;
+ $this->stack['abbr'][$k] = $tag;
+ }
+ }
+ if (!empty($out)) {
+ $this->out("\n\n".implode("\n", $out));
+ }
+ }
+ /**
+ * handle
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_table() {
+ if ($this->parser->isStartTag) {
+ # check if upcoming table can be converted
+ if ($this->keepHTML) {
+ if (preg_match($this->tableLookaheadHeader, $this->parser->html, $matches)) {
+ # header seems good, now check body
+ # get align & number of cols
+ preg_match_all('#
tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_tr() {
+ if ($this->parser->isStartTag) {
+ $this->col = -1;
+ } else {
+ $this->row++;
+ }
+ }
+ /**
+ * handle tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_td() {
+ if ($this->parser->isStartTag) {
+ $this->col++;
+ if (!isset($this->table['col_widths'][$this->col])) {
+ $this->table['col_widths'][$this->col] = 0;
+ }
+ $this->buffer();
+ } else {
+ $buffer = trim($this->unbuffer());
+ $this->table['col_widths'][$this->col] = max($this->table['col_widths'][$this->col], $this->strlen($buffer));
+ $this->table['rows'][$this->row][$this->col] = $buffer;
+ }
+ }
+ /**
+ * handle tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_th() {
+ if (!$this->keepHTML && !isset($this->table['rows'][1]) && !isset($this->table['aligns'][$this->col+1])) {
+ if (isset($this->parser->tagAttributes['align'])) {
+ $this->table['aligns'][$this->col+1] = $this->parser->tagAttributes['align'];
+ } else {
+ $this->table['aligns'][$this->col+1] = '';
+ }
+ }
+ $this->handleTag_td();
+ }
+ /**
+ * handle tags
+ *
+ * @param void
+ * @return void
+ */
+ function handleTag_dl() {
+ if (!$this->parser->isStartTag) {
+ $this->setLineBreaks(2);
+ }
+ }
+ /**
+ * handle
+ #
+ #
+ #
+ #
\s*\s*(.+)\s*
\s*
) set this to true as well!
+ *
+ * @var bool | null
+ */
+ var $isStartTag = null;
+ /**
+ * wether current node is an empty tag (
) or not ()
+ *
+ * @var bool | null
+ */
+ var $isEmptyTag = null;
+ /**
+ * tag name
+ *
+ * @var string | null
+ */
+ var $tagName = '';
+ /**
+ * attributes of current tag
+ *
+ * @var array (attribName=>value) | null
+ */
+ var $tagAttributes = null;
+ /**
+ * wether the current tag is a block element
+ *
+ * @var bool | null
+ */
+ var $isBlockElement = null;
+
+ /**
+ * keep whitespace
+ *
+ * @var int
+ */
+ var $keepWhitespace = 0;
+ /**
+ * list of open tags
+ * count this to get current depth
+ *
+ * @var array
+ */
+ var $openTags = array();
+ /**
+ * list of block elements
+ *
+ * @var array
+ * TODO: what shall we do with and ?!
+ */
+ var $blockElements = array (
+ # tag name => or contents
+ $this->keepWhitespace++;
+ }
+ }
+
+ if ($this->html[0] == '<') {
+ $token = substr($this->html, 0, 9);
+ if (substr($token, 0, 2) == 'html, '>');
+ $this->setNode('pi', $pos + 1);
+ return true;
+ }
+ if (substr($token, 0, 4) == '');
+ if ($pos === false) {
+ # could not find a closing -->, use next gt instead
+ # this is firefox' behaviour
+ $pos = strpos($this->html, '>') + 1;
+ } else {
+ $pos += 3;
+ }
+ $this->setNode('comment', $pos);
+
+ $skipWhitespace = true;
+ return true;
+ }
+ if ($token == 'setNode('doctype', strpos($this->html, '>')+1);
+
+ $skipWhitespace = true;
+ return true;
+ }
+ if ($token == 'html = substr($this->html, 9);
+
+ $this->setNode('text', strpos($this->html, ']]>')+3);
+
+ # remove trailing ]]> and trim
+ $this->node = substr($this->node, 0, -3);
+ $this->handleWhitespaces();
+
+ $skipWhitespace = true;
+ return true;
+ }
+ if ($this->parseTag()) {
+ # seems to be a tag
+ # handle whitespaces
+ if ($this->isBlockElement) {
+ $skipWhitespace = true;
+ } else {
+ $skipWhitespace = false;
+ }
+ return true;
+ }
+ }
+ if ($this->keepWhitespace) {
+ $skipWhitespace = false;
+ }
+ # when we get here it seems to be a text node
+ $pos = strpos($this->html, '<');
+ if ($pos === false) {
+ $pos = strlen($this->html);
+ }
+ $this->setNode('text', $pos);
+ $this->handleWhitespaces();
+ if ($skipWhitespace && $this->node == ' ') {
+ return $this->nextNode();
+ }
+ $skipWhitespace = false;
+ return true;
+ }
+ /**
+ * parse tag, set tag name and attributes, see if it's a closing tag and so forth...
+ *
+ * @param void
+ * @return bool
+ */
+ function parseTag() {
+ static $a_ord, $z_ord, $special_ords;
+ if (!isset($a_ord)) {
+ $a_ord = ord('a');
+ $z_ord = ord('z');
+ $special_ords = array(
+ ord(':'), // for xml:lang
+ ord('-'), // for http-equiv
+ );
+ }
+
+ $tagName = '';
+
+ $pos = 1;
+ $isStartTag = $this->html[$pos] != '/';
+ if (!$isStartTag) {
+ $pos++;
+ }
+ # get tagName
+ while (isset($this->html[$pos])) {
+ $pos_ord = ord(strtolower($this->html[$pos]));
+ if (($pos_ord >= $a_ord && $pos_ord <= $z_ord) || (!empty($tagName) && is_numeric($this->html[$pos]))) {
+ $tagName .= $this->html[$pos];
+ $pos++;
+ } else {
+ $pos--;
+ break;
+ }
+ }
+
+ $tagName = strtolower($tagName);
+ if (empty($tagName) || !isset($this->blockElements[$tagName])) {
+ # something went wrong => invalid tag
+ $this->invalidTag();
+ return false;
+ }
+ if ($this->noTagsInCode && end($this->openTags) == 'code' && !($tagName == 'code' && !$isStartTag)) {
+ # we supress all HTML tags inside code tags
+ $this->invalidTag();
+ return false;
+ }
+
+ # get tag attributes
+ /** TODO: in html 4 attributes do not need to be quoted **/
+ $isEmptyTag = false;
+ $attributes = array();
+ $currAttrib = '';
+ while (isset($this->html[$pos+1])) {
+ $pos++;
+ # close tag
+ if ($this->html[$pos] == '>' || $this->html[$pos].$this->html[$pos+1] == '/>') {
+ if ($this->html[$pos] == '/') {
+ $isEmptyTag = true;
+ $pos++;
+ }
+ break;
+ }
+
+ $pos_ord = ord(strtolower($this->html[$pos]));
+ if ( ($pos_ord >= $a_ord && $pos_ord <= $z_ord) || in_array($pos_ord, $special_ords)) {
+ # attribute name
+ $currAttrib .= $this->html[$pos];
+ } elseif (in_array($this->html[$pos], array(' ', "\t", "\n"))) {
+ # drop whitespace
+ } elseif (in_array($this->html[$pos].$this->html[$pos+1], array('="', "='"))) {
+ # get attribute value
+ $pos++;
+ $await = $this->html[$pos]; # single or double quote
+ $pos++;
+ $value = '';
+ while (isset($this->html[$pos]) && $this->html[$pos] != $await) {
+ $value .= $this->html[$pos];
+ $pos++;
+ }
+ $attributes[$currAttrib] = $value;
+ $currAttrib = '';
+ } else {
+ $this->invalidTag();
+ return false;
+ }
+ }
+ if ($this->html[$pos] != '>') {
+ $this->invalidTag();
+ return false;
+ }
+
+ if (!empty($currAttrib)) {
+ # html 4 allows something like