1
1
Fork 0

Merge pull request #400 from annando/master

New bb2markdown, parse_url improved
This commit is contained in:
friendica 2012-07-18 16:30:12 -07:00
commit 082ad499c6
9 changed files with 420 additions and 261 deletions

View file

@ -156,6 +156,7 @@
//echo "<pre>"; var_dump($r); die();
}
}
header("HTTP/1.1 404 Not Found");
logger('API call not implemented: '.$a->query_string." - ".print_r($_REQUEST,true));
$r = '<status><error>not implemented</error></status>';
switch($type){
@ -490,7 +491,8 @@
$_REQUEST['type'] = 'wall';
$_REQUEST['profile_uid'] = local_user();
$_REQUEST['api_source'] = true;
$txt = urldecode(requestdata('status'));
$txt = requestdata('status');
//$txt = urldecode(requestdata('status'));
require_once('library/HTMLPurifier.auto.php');
require_once('include/html2bbcode.php');
@ -554,7 +556,8 @@
}
else
$_REQUEST['body'] = urldecode(requestdata('status'));
$_REQUEST['body'] = requestdata('status');
//$_REQUEST['body'] = urldecode(requestdata('status'));
$parent = requestdata('in_reply_to_status_id');
if(ctype_digit($parent))
@ -1725,4 +1728,6 @@ notifications/follow
notifications/leave
blocks/exists
blocks/blocking
lists
*/

View file

@ -190,63 +190,40 @@ function diaspora_ol($s) {
}
function bb2diaspora($Text,$preserve_nl = false) {
function bb2diaspora($Text,$preserve_nl = false, $fordiaspora = true) {
//////////////////////
// An attempt was made to convert bbcode to html and then to markdown
// consisting of the following lines.
// I'm undoing this as we have a lot of bbcode constructs which
// were simply getting lost, for instance bookmark, vimeo, video, youtube, events, etc.
// We can try this again, but need a very good test sequence to verify
// all the major bbcode constructs that we use are getting through.
//////////////////////
/*
// bbcode() will convert "[*]" into "<li>" with no closing "</li>"
// Markdownify() is unable to handle these, as it makes each new
// "<li>" into a deeper nested element until it crashes. So pre-format
// the lists as Diaspora lists before sending the $Text to bbcode()
//
// Note that to get nested lists to work for Diaspora, we would need
// to define the closing tag for the list elements. So nested lists
// are going to be flattened out in Diaspora for now
// Re-enabling the converter again.
// The bbcode parser now handles youtube-links (and the other stuff) correctly.
// Additionally the html code is now fixed so that lists are now working.
$endlessloop = 0;
while ((((strpos($Text, "[/list]") !== false) && (strpos($Text, "[list") !== false)) ||
((strpos($Text, "[/ol]") !== false) && (strpos($Text, "[ol]") !== false)) ||
((strpos($Text, "[/ul]") !== false) && (strpos($Text, "[ul]") !== false))) && (++$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);
}
// 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]', $Text);
*/
// the following was added on 10-January-2012 due to an inability of Diaspora's
// new javascript markdown processor to handle links with images as the link "text"
// It is not optimal and may be removed if this ability is restored in the future
if ($fordiaspora)
$Text = preg_replace("/\[url\=([^\[\]]*)\]\s*\[img\](.*?)\[\/img\]\s*\[\/url\]/ism",
"[url]$1[/url]\n[img]$2[/img]", $Text);
// Convert it to HTML - don't try oembed
// $Text = bbcode($Text, $preserve_nl, false);
$Text = bbcode($Text, $preserve_nl, false);
// Now convert HTML to Markdown
// $md = new Markdownify(false, false, false);
// $Text = $md->parseString($Text);
$md = new Markdownify(false, false, false);
$Text = $md->parseString($Text);
// If the text going into bbcode() has a plain URL in it, i.e.
// with no [url] tags around it, it will come out of parseString()
// looking like: <http://url.com>, which gets removed by strip_tags().
// So take off the angle brackets of any such URL
// $Text = preg_replace("/<http(.*?)>/is", "http$1", $Text);
$Text = preg_replace("/<http(.*?)>/is", "http$1", $Text);
// Remove all unconverted tags
// $Text = strip_tags($Text);
//////
// end of bb->html->md conversion attempt
//////
$Text = strip_tags($Text);
/* Old routine
$ev = bbtoevent($Text);
@ -422,6 +399,7 @@ function bb2diaspora($Text,$preserve_nl = false) {
$Text = preg_replace_callback('/\[(.*?)\]\((.*?)\)/ism','unescape_underscores_in_links',$Text);
*/
// Remove any leading or trailing whitespace, as this will mess up
// the Diaspora signature verification and cause the item to disappear

View file

@ -115,6 +115,10 @@ function bbcode($Text,$preserve_nl = false, $tryoembed = true) {
$a = get_app();
// Move all spaces out of the tags
$Text = preg_replace("/\[(\w*)\](\s*)/ism", '$2[$1]', $Text);
$Text = preg_replace("/(\s*)\[\/(\w*)\]/ism", '[/$2]$1', $Text);
// Hide all [noparse] contained bbtags by spacefying them
// POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
@ -313,21 +317,25 @@ function bbcode($Text,$preserve_nl = false, $tryoembed = true) {
$Text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . t('Image/photo') . '" />', $Text);
$Text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4))\[\/video\]/ism", '<video src="$1" controls="controls" width="425" height="350"><a href="$1">$1</a></video>', $Text);
$Text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3))\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $Text);
// Try to Oembed
if ($tryoembed) {
$Text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4))\[\/video\]/ism", '<video src="$1" controls="controls" width="425" height="350"><a href="$1">$1</a></video>', $Text);
$Text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3))\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $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", '<iframe src="$1" width="425" height="350"><a href="$1">$1</a></iframe>', $Text);
if ($tryoembed)
$Text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<iframe src="$1" width="425" height="350"><a href="$1">$1</a></iframe>', $Text);
else
$Text = preg_replace("/\[iframe\](.*?)\[\/iframe\]/ism", '<a href="$1">$1</a>', $Text);
// Youtube extensions
if ($tryoembed) {
@ -340,7 +348,10 @@ function bbcode($Text,$preserve_nl = false, $tryoembed = true) {
$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", '<iframe width="425" height="350" src="http://www.youtube.com/embed/$1" frameborder="0" ></iframe>', $Text);
if ($tryoembed)
$Text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", '<iframe width="425" height="350" src="http://www.youtube.com/embed/$1" frameborder="0" ></iframe>', $Text);
else
$Text = preg_replace("/\[youtube\]([A-Za-z0-9\-_=]+)(.*?)\[\/youtube\]/ism", "http://www.youtube.com/watch?v=$1", $Text);
if ($tryoembed) {
@ -349,8 +360,12 @@ function bbcode($Text,$preserve_nl = false, $tryoembed = true) {
}
$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", '<iframe width="425" height="350" src="http://player.vimeo.com/video/$1" frameborder="0" ></iframe>', $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", '<iframe width="425" height="350" src="http://player.vimeo.com/video/$1" frameborder="0" ></iframe>', $Text);
else
$Text = preg_replace("/\[vimeo\]([0-9]+)(.*?)\[\/vimeo\]/ism", "http://vimeo.com/$1", $Text);
// $Text = preg_replace("/\[youtube\](.*?)\[\/youtube\]/", '<object width="425" height="350" type="application/x-shockwave-flash" data="http://www.youtube.com/v/$1" ><param name="movie" value="http://www.youtube.com/v/$1"></param><!--[if IE]><embed src="http://www.youtube.com/v/$1" type="application/x-shockwave-flash" width="425" height="350" /><![endif]--></object>', $Text);
@ -358,6 +373,9 @@ function bbcode($Text,$preserve_nl = false, $tryoembed = true) {
// oembed tag
$Text = oembed_bbcode2html($Text);
// Avoid triple linefeeds through oembed
$Text = str_replace("<br style='clear:left'></span><br /><br />", "<br style='clear:left'></span><br />", $Text);
// If we found an event earlier, strip out all the event code and replace with a reformatted version.
// Replace the event-start section with the entire formatted event. The other bbcode is stripped.
// Summary (e.g. title) is required, earlier revisions only required description (in addition to
@ -391,6 +409,25 @@ function bbcode($Text,$preserve_nl = false, $tryoembed = true) {
if($saved_image)
$Text = bb_replace_images($Text, $saved_image);
// Clean up the HTML by loading and saving the HTML with the DOM
// Only do it when it has to be done - for performance reasons
if (!$tryoembed) {
$doc = new DOMDocument();
$doc->preserveWhiteSpace = false;
$Text = mb_convert_encoding($Text, 'HTML-ENTITIES', "UTF-8");
$doctype = '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">';
@$doc->loadHTML($doctype."<html><body>".$Text."</body></html>");
$Text = $doc->saveHTML();
$Text = str_replace(array("<html><body>", "</body></html>", $doctype), array("", "", ""), $Text);
$Text = str_replace('<br></li>','</li>', $Text);
$Text = mb_convert_encoding($Text, "UTF-8", 'HTML-ENTITIES');
}
call_hooks('bbcode',$Text);
return $Text;

View file

@ -71,22 +71,32 @@ class dba {
}
public function q($sql) {
global $a;
if((! $this->db) || (! $this->connected))
return false;
$this->error = '';
//if (get_config("system", "db_log") != "")
// @file_put_contents(get_config("system", "db_log"), datetime_convert().':'.session_id(). ' Start '.$sql."\n", FILE_APPEND);
if ($a->config["system"]["db_log"] != "")
$stamp1 = microtime(true);
if($this->mysqli)
$result = @$this->db->query($sql);
else
$result = @mysql_query($sql,$this->db);
//if (get_config("system", "db_log") != "")
// @file_put_contents(get_config("system", "db_log"), datetime_convert().':'.session_id(). ' Stop '."\n", FILE_APPEND);
if ($a->config["system"]["db_log"] != "") {
$stamp2 = microtime(true);
$duration = round($stamp2-$stamp1, 3);
if ($duration > $a->config["system"]["db_loglimit"]) {
$backtrace = debug_backtrace(DEBUG_BACKTRACE_IGNORE_ARGS);
@file_put_contents($a->config["system"]["db_log"], $duration."\t".
basename($backtrace[1]["file"])."\t".
$backtrace[1]["line"]."\t".$backtrace[2]["function"]."\t".
substr($sql, 0, 2000)."\n", FILE_APPEND);
}
}
if($this->mysqli) {
if($this->db->errno)

View file

@ -374,6 +374,29 @@ function limit_body_size($body) {
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 == $body);
}
@ -400,6 +423,11 @@ function get_atom_elements($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
@ -418,7 +446,7 @@ function get_atom_elements($feed,$item) {
$res['author-avatar'] = unxmlify($link['attribs']['']['href']);
}
}
}
}
$rawactor = $item->get_item_tags(NAMESPACE_ACTIVITY, 'actor');
@ -450,7 +478,7 @@ function get_atom_elements($feed,$item) {
$res['author-avatar'] = unxmlify($link['attribs']['']['href']);
}
}
}
}
$rawactor = $feed->get_feed_tags(NAMESPACE_ACTIVITY, 'subject');
@ -475,7 +503,7 @@ function get_atom_elements($feed,$item) {
$res['app'] = strip_tags(unxmlify($apps[0]['attribs']['']['source']));
if($res['app'] === 'web')
$res['app'] = 'OStatus';
}
}
// base64 encoded json structure representing Diaspora signature
@ -618,7 +646,7 @@ function get_atom_elements($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']);
}
}
@ -758,10 +786,41 @@ function get_atom_elements($feed,$item) {
$res['target'] .= '</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 $res;
}
@ -1717,7 +1776,7 @@ function consume_feed($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 = ' . $feed->get_item_quantity());
@ -1730,7 +1789,7 @@ function consume_feed($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'])) {
@ -1744,11 +1803,10 @@ function consume_feed($xml,$importer,&$contact, &$hub, $datedir = 0, $pass = 0)
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')) && ($contact['network'] != NETWORK_DFRN))
@ -3665,7 +3723,6 @@ function posted_date_widget($url,$uid,$wall) {
return $o;
}
function store_diaspora_retract_sig($item, $user, $baseurl) {
// Note that we can't add a target_author_signature
// if the comment was deleted by a remote user. That should be ok, because if a remote user is deleting

View file

@ -735,6 +735,13 @@ class Markdownify {
$this->parser->tagAttributes['src'] = $this->decode($this->parser->tagAttributes['src']);
}
// ![Alt text](/path/to/img.jpg "Optional title")
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'])) {
@ -759,6 +766,7 @@ class Markdownify {
}
$this->out('!['.$this->parser->tagAttributes['alt'].']['.$link_id.']', true);
*/
}
/**
* handle <code> tags
@ -1181,4 +1189,4 @@ class Markdownify {
function parent() {
return end($this->parser->openTags);
}
}
}

View file

@ -14,15 +14,16 @@ function fetch_url($url,$binary = false, &$redirects = 0, $timeout = 0, $accept_
return false;
@curl_setopt($ch, CURLOPT_HEADER, true);
if (!is_null($accept_content)){
curl_setopt($ch,CURLOPT_HTTPHEADER, array (
"Accept: " . $accept_content
));
}
@curl_setopt($ch, CURLOPT_RETURNTRANSFER,true);
@curl_setopt($ch, CURLOPT_USERAGENT, "Friendica");
//@curl_setopt($ch, CURLOPT_USERAGENT, "Friendica");
@curl_setopt($ch, CURLOPT_USERAGENT, "Mozilla/5.0 (compatible; Friendica)");
if(intval($timeout)) {
@ -59,7 +60,6 @@ function fetch_url($url,$binary = false, &$redirects = 0, $timeout = 0, $accept_
$base = $s;
$curl_info = @curl_getinfo($ch);
$http_code = $curl_info['http_code'];
// logger('fetch_url:' . $http_code . ' data: ' . $s);
$header = '';
@ -73,24 +73,22 @@ function fetch_url($url,$binary = false, &$redirects = 0, $timeout = 0, $accept_
}
if($http_code == 301 || $http_code == 302 || $http_code == 303 || $http_code == 307) {
$matches = array();
preg_match('/(Location:|URI:)(.*?)\n/', $header, $matches);
$newurl = trim(array_pop($matches));
$matches = array();
preg_match('/(Location:|URI:)(.*?)\n/', $header, $matches);
$newurl = trim(array_pop($matches));
if(strpos($newurl,'/') === 0)
$newurl = $url . $newurl;
$url_parsed = @parse_url($newurl);
if (isset($url_parsed)) {
$redirects++;
return fetch_url($newurl,$binary,$redirects,$timeout);
}
}
$url_parsed = @parse_url($newurl);
if (isset($url_parsed)) {
$redirects++;
return fetch_url($newurl,$binary,$redirects,$timeout);
}
}
$a->set_curl_code($http_code);
$body = substr($s,strlen($header));
$a->set_curl_headers($header);
@curl_close($ch);
return($body);
}}
@ -800,6 +798,9 @@ function scale_external_images($s, $include_link = true, $scale_replace = false)
$a = get_app();
// Picture addresses can contain special characters
$s = htmlspecialchars_decode($s);
$matches = null;
$c = preg_match_all('/\[img\](.*?)\[\/img\]/ism',$s,$matches,PREG_SET_ORDER);
if($c) {

View file

@ -1562,6 +1562,7 @@ function undo_post_tagging($s) {
function fix_mce_lf($s) {
$s = str_replace("\r\n","\n",$s);
$s = str_replace("\n\n","\n",$s);
return $s;
}

View file

@ -1,7 +1,224 @@
<?php
/* To-Do
https://developers.google.com/+/plugins/snippet/
require_once('library/HTML5/Parser.php');
require_once('library/HTMLPurifier.auto.php');
<meta itemprop="name" content="Toller Titel">
<meta itemprop="description" content="Eine tolle Beschreibung">
<meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
<body itemscope itemtype="http://schema.org/Product">
<h1 itemprop="name">Shiny Trinket</h1>
<img itemprop="image" src="{image-url}" />
<p itemprop="description">Shiny trinkets are shiny.</p>
</body>
*/
if(!function_exists('deletenode')) {
function deletenode(&$doc, $node)
{
$xpath = new DomXPath($doc);
$list = $xpath->query("//".$node);
foreach ($list as $child)
$child->parentNode->removeChild($child);
}
}
function completeurl($url, $scheme) {
$urlarr = parse_url($url);
if (isset($urlarr["scheme"]))
return($url);
$schemearr = parse_url($scheme);
$complete = $schemearr["scheme"]."://".$schemearr["host"];
if ($schemearr["port"] != "")
$complete .= ":".$schemearr["port"];
$complete .= $urlarr["path"];
if ($urlarr["query"] != "")
$complete .= "?".$urlarr["query"];
if ($urlarr["fragment"] != "")
$complete .= "#".$urlarr["fragment"];
return($complete);
}
function parseurl_getsiteinfo($url) {
$siteinfo = array();
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HEADER, 1);
curl_setopt($ch, CURLOPT_NOBODY, 0);
curl_setopt($ch, CURLOPT_TIMEOUT, 3);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch,CURLOPT_USERAGENT,'Opera/9.64(Windows NT 5.1; U; de) Presto/2.1.1');
$header = curl_exec($ch);
curl_close($ch);
// Fetch the first mentioned charset. Can be in body or header
if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
$charset = trim(array_pop($matches));
else
$charset = "utf-8";
$pos = strpos($header, "\r\n\r\n");
if ($pos)
$body = trim(substr($header, $pos));
else
$body = $header;
$body = mb_convert_encoding($body, "UTF-8", $charset);
$body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
$doc = new DOMDocument();
@$doc->loadHTML($body);
deletenode($doc, 'style');
deletenode($doc, 'script');
deletenode($doc, 'option');
deletenode($doc, 'h1');
deletenode($doc, 'h2');
deletenode($doc, 'h3');
deletenode($doc, 'h4');
deletenode($doc, 'h5');
deletenode($doc, 'h6');
deletenode($doc, 'ol');
deletenode($doc, 'ul');
$xpath = new DomXPath($doc);
//$list = $xpath->query("head/title");
$list = $xpath->query("//title");
foreach ($list as $node)
$siteinfo["title"] = html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
//$list = $xpath->query("head/meta[@name]");
$list = $xpath->query("//meta[@name]");
foreach ($list as $node) {
$attr = array();
if ($node->attributes->length)
foreach ($node->attributes as $attribute)
$attr[$attribute->name] = $attribute->value;
$attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
switch (strtolower($attr["name"])) {
case "fulltitle":
$siteinfo["title"] = $attr["content"];
break;
case "description":
$siteinfo["text"] = $attr["content"];
break;
case "dc.title":
$siteinfo["title"] = $attr["content"];
break;
case "dc.description":
$siteinfo["text"] = $attr["content"];
break;
}
}
//$list = $xpath->query("head/meta[@property]");
$list = $xpath->query("//meta[@property]");
foreach ($list as $node) {
$attr = array();
if ($node->attributes->length)
foreach ($node->attributes as $attribute)
$attr[$attribute->name] = $attribute->value;
$attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
switch (strtolower($attr["property"])) {
case "og:image":
$siteinfo["image"] = $attr["content"];
break;
case "og:title":
$siteinfo["title"] = $attr["content"];
break;
case "og:description":
$siteinfo["text"] = $attr["content"];
break;
}
}
if ($siteinfo["image"] == "") {
$list = $xpath->query("//img[@src]");
foreach ($list as $node) {
$attr = array();
if ($node->attributes->length)
foreach ($node->attributes as $attribute)
$attr[$attribute->name] = $attribute->value;
$src = completeurl($attr["src"], $url);
$photodata = getimagesize($src);
if (($photodata[0] > 150) and ($photodata[1] > 150)) {
if ($photodata[0] > 300) {
$photodata[1] = round($photodata[1] * (300 / $photodata[0]));
$photodata[0] = 300;
}
if ($photodata[1] > 300) {
$photodata[0] = round($photodata[0] * (300 / $photodata[1]));
$photodata[1] = 300;
}
$siteinfo["images"][] = array("src"=>$src,
"width"=>$photodata[0],
"height"=>$photodata[1]);
}
}
} else {
$src = completeurl($siteinfo["image"], $url);
$photodata = getimagesize($src);
if (($photodata[0] > 10) and ($photodata[1] > 10))
$siteinfo["images"][] = array("src"=>$src,
"width"=>$photodata[0],
"height"=>$photodata[1]);
}
if ($siteinfo["text"] == "") {
$text = "";
$list = $xpath->query("//div[@class='article']");
foreach ($list as $node)
if (strlen($node->nodeValue) > 40)
$text .= " ".trim($node->nodeValue);
if ($text == "") {
$list = $xpath->query("//div[@class='content']");
foreach ($list as $node)
if (strlen($node->nodeValue) > 40)
$text .= " ".trim($node->nodeValue);
}
// If none text was found then take the paragraph content
if ($text == "") {
$list = $xpath->query("//p");
foreach ($list as $node)
if (strlen($node->nodeValue) > 40)
$text .= " ".trim($node->nodeValue);
}
if ($text != "") {
$text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
while (strpos($text, " "))
$text = trim(str_replace(" ", " ", $text));
$siteinfo["text"] = html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...';
}
}
return($siteinfo);
}
function arr_add_hashes(&$item,$k) {
$item = '#' . $item;
@ -16,8 +233,8 @@ function parse_url_content(&$a) {
if(local_user() && intval(get_pconfig(local_user(),'system','plaintext')))
$textmode = true;
if($textmode)
$br = (($textmode) ? "\n" : '<br /?');
//if($textmode)
$br = (($textmode) ? "\n" : '<br />');
if(x($_GET,'binurl'))
$url = trim(hex2bin($_GET['binurl']));
@ -40,13 +257,11 @@ function parse_url_content(&$a) {
logger('parse_url: ' . $url);
if($textmode)
$template = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
else
$template = "<br /><a class=\"bookmark\" href=\"%s\" >%s</a>%s<br />";
$arr = array('url' => $url, 'text' => '');
call_hooks('parse_link', $arr);
@ -60,208 +275,55 @@ function parse_url_content(&$a) {
if($url && $title && $text) {
if($textmode)
$text = $br . $br . '[quote]' . $text . '[/quote]' . $br;
$text = $br . '[quote]' . trim($text) . '[/quote]' . $br;
else
$text = '<br /><br /><blockquote>' . $text . '</blockquote><br />';
$text = '<br /><blockquote>' . trim($text) . '</blockquote><br />';
$title = str_replace(array("\r","\n"),array('',''),$title);
$result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
logger('parse_url (unparsed): returns: ' . $result);
logger('parse_url (unparsed): returns: ' . $result);
echo $result;
killme();
}
$siteinfo = parseurl_getsiteinfo($url);
if($url) {
$s = fetch_url($url);
if($siteinfo["title"] == "") {
echo print_r($siteinfo, true);
//echo sprintf($template,$url,$url,'') . $str_tags;
killme();
} else {
echo '';
killme();
$text = $siteinfo["text"];
$title = $siteinfo["title"];
}
// logger('parse_url: data: ' . $s, LOGGER_DATA);
$image = "";
if(! $s) {
echo sprintf($template,$url,$url,'') . $str_tags;
killme();
}
$matches = '';
$c = preg_match('/\<head(.*?)\>(.*?)\<\/head\>/ism',$s,$matches);
if($c) {
// logger('parse_url: header: ' . $matches[2], LOGGER_DATA);
try {
$domhead = HTML5_Parser::parse($matches[2]);
} catch (DOMException $e) {
logger('scrape_dfrn: parse error: ' . $e);
}
if($domhead)
logger('parsed header');
}
if(! $title) {
if(strpos($s,'<title>')) {
$title = substr($s,strpos($s,'<title>')+7,64);
if(strpos($title,'<') !== false)
$title = strip_tags(substr($title,0,strpos($title,'<')));
}
}
$config = HTMLPurifier_Config::createDefault();
$config->set('Cache.DefinitionImpl', null);
$purifier = new HTMLPurifier($config);
$s = $purifier->purify($s);
// logger('purify_output: ' . $s);
try {
$dom = HTML5_Parser::parse($s);
} catch (DOMException $e) {
logger('scrape_dfrn: parse error: ' . $e);
}
if(! $dom) {
echo sprintf($template,$url,$url,'') . $str_tags;
killme();
}
$items = $dom->getElementsByTagName('title');
if($items) {
foreach($items as $item) {
$title = trim($item->textContent);
break;
}
}
if(! $text) {
$divs = $dom->getElementsByTagName('div');
if($divs) {
foreach($divs as $div) {
$class = $div->getAttribute('class');
if($class && (stristr($class,'article') || stristr($class,'content'))) {
$items = $div->getElementsByTagName('p');
if($items) {
foreach($items as $item) {
$text = $item->textContent;
if(stristr($text,'<script')) {
$text = '';
continue;
}
$text = strip_tags($text);
if(strlen($text) < 100) {
$text = '';
continue;
}
$text = substr($text,0,250) . '...' ;
break;
}
}
}
if($text)
break;
}
}
if(! $text) {
$items = $dom->getElementsByTagName('p');
if($items) {
foreach($items as $item) {
$text = $item->textContent;
if(stristr($text,'<script'))
continue;
$text = strip_tags($text);
if(strlen($text) < 100) {
$text = '';
continue;
}
$text = substr($text,0,250) . '...' ;
break;
}
}
}
}
if(! $text) {
logger('parsing meta');
$items = (isset($domhead) && is_object($domhead) ? $domhead->getElementsByTagName('meta') : null);
if($items) {
foreach($items as $item) {
$property = $item->getAttribute('property');
if($property && (stristr($property,':description'))) {
$text = $item->getAttribute('content');
if(stristr($text,'<script')) {
$text = '';
continue;
}
$text = strip_tags($text);
$text = substr($text,0,250) . '...' ;
}
if($property && (stristr($property,':image'))) {
$image = $item->getAttribute('content');
if(stristr($text,'<script')) {
$image = '';
continue;
}
$image = strip_tags($image);
$i = fetch_url($image);
if($i) {
require_once('include/Photo.php');
// guess mimetype from headers or filename
$type = guess_image_type($image,true);
$ph = new Photo($i, $type);
if($ph->is_valid()) {
if($ph->getWidth() > 300 || $ph->getHeight() > 300) {
$ph->scaleImage(300);
$new_width = $ph->getWidth();
$new_height = $ph->getHeight();
if($textmode)
$image = $br . $br . '[img=' . $new_width . 'x' . $new_height . ']' . $image . '[/img]';
else
$image = '<br /><br /><img height="' . $new_height . '" width="' . $new_width . '" src="' .$image . '" alt="photo" />';
}
else {
if($textmode)
$image = $br . $br . '[img]' . $image . '[/img]';
else
$image = '<br /><br /><img src="' . $image . '" alt="photo" />';
}
}
else
$image = '';
}
}
}
}
}
foreach ($siteinfo["images"] as $imagedata)
if($textmode)
$image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]';
else
$image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" />';
if(strlen($text)) {
if($textmode)
$text = $br .$br . '[quote]' . $text . '[/quote]' . $br ;
$text = $br.'[quote]'.trim($text).'[/quote]'.$br ;
else
$text = '<br /><br /><blockquote>' . $text . '</blockquote><br />';
$text = '<br /><blockquote>'.trim($text).'</blockquote><br />';
}
if($image) {
$text = $image . $br . $text;
$text = $br.$br.$image.$text;
}
$title = str_replace(array("\r","\n"),array('',''),$title);
$result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
logger('parse_url: returns: ' . $result);
logger('parse_url: returns: ' . $result);
echo $result;
echo trim($result);
killme();
}