clones/friendica
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
friendica/mod/parse_url.php

353 lines
9.0 KiB
PHP
Raw Normal View History

ajaxify the wall poster
2010-07-23 01:41:45 -04:00
<?php
parse_url: Just added a comment for future elements that should be added as well.
2012-07-18 15:06:38 -04:00
/* To-Do
https://developers.google.com/+/plugins/snippet/
<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>
*/
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
if(!function_exists('deletenode')) {
function deletenode(&$doc, $node)
{
$xpath = new DomXPath($doc);
$list = $xpath->query("//".$node);
foreach ($list as $child)
$child->parentNode->removeChild($child);
}
}
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
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"];
update writable for dfrn contacts when using local delivery
2012-09-05 19:26:11 -04:00
if(strpos($urlarr['path'],'/') !== 0)
$complete .= '/';
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
$complete .= $urlarr["path"];
if ($urlarr["query"] != "")
$complete .= "?".$urlarr["query"];
if ($urlarr["fragment"] != "")
$complete .= "#".$urlarr["fragment"];
return($complete);
}
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
function parseurl_getsiteinfo($url) {
$siteinfo = array();
ajaxify the wall poster
2010-07-23 01:41:45 -04:00
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
$ch = curl_init();
curl_setopt($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);
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
// Fetch the first mentioned charset. Can be in body or header
if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
$charset = trim(array_pop($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);
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
//$list = $xpath->query("head/title");
$list = $xpath->query("//title");
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
foreach ($list as $node)
$siteinfo["title"] = html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
//$list = $xpath->query("head/meta[@name]");
$list = $xpath->query("//meta[@name]");
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
foreach ($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"] = $attr["content"];
break;
}
}
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
//$list = $xpath->query("head/meta[@property]");
$list = $xpath->query("//meta[@property]");
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
foreach ($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"] == "") {
allow comments on other folks wall-to-wall postings
2012-09-05 21:24:06 -04:00
$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;
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
$src = completeurl($attr["src"], $url);
allow comments on other folks wall-to-wall postings
2012-09-05 21:24:06 -04:00
$photodata = @getimagesize($src);
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
allow comments on other folks wall-to-wall postings
2012-09-05 21:24:06 -04:00
if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
if ($photodata[0] > 300) {
parse_url: Problem when resizing images
2012-07-14 07:59:42 -04:00
$photodata[1] = round($photodata[1] * (300 / $photodata[0]));
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
$photodata[0] = 300;
}
if ($photodata[1] > 300) {
parse_url: Problem when resizing images
2012-07-14 07:59:42 -04:00
$photodata[0] = round($photodata[0] * (300 / $photodata[1]));
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
$photodata[1] = 300;
}
$siteinfo["images"][] = array("src"=>$src,
"width"=>$photodata[0],
"height"=>$photodata[1]);
parse_url: Fetch multiple pictures so that the user can decide what to take.
2012-07-12 02:20:27 -04:00
}
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
allow comments on other folks wall-to-wall postings
2012-09-05 21:24:06 -04:00
}
} else {
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
$src = completeurl($siteinfo["image"], $url);
parse_url: Fixed a problem in the detection if there was an image or not. Additionally removed some test content.
2012-08-02 04:25:08 -04:00
unset($siteinfo["image"]);
allow comments on other folks wall-to-wall postings
2012-09-05 21:24:06 -04:00
$photodata = @getimagesize($src);
parse_url: Fetch multiple pictures so that the user can decide what to take.
2012-07-12 02:20:27 -04:00
allow comments on other folks wall-to-wall postings
2012-09-05 21:24:06 -04:00
if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
$siteinfo["images"][] = array("src"=>$src,
"width"=>$photodata[0],
"height"=>$photodata[1]);
parse_url: Fetch multiple pictures so that the user can decide what to take.
2012-07-12 02:20:27 -04:00
}
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
if ($siteinfo["text"] == "") {
$text = "";
$list = $xpath->query("//div[@class='article']");
foreach ($list as $node)
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
if (strlen($node->nodeValue) > 40)
$text .= " ".trim($node->nodeValue);
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
if ($text == "") {
$list = $xpath->query("//div[@class='content']");
foreach ($list as $node)
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
if (strlen($node->nodeValue) > 40)
$text .= " ".trim($node->nodeValue);
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
}
// If none text was found then take the paragraph content
if ($text == "") {
$list = $xpath->query("//p");
foreach ($list as $node)
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
if (strlen($node->nodeValue) > 40)
$text .= " ".trim($node->nodeValue);
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
}
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($siteinfo);
}
begin plugin api
2010-12-20 22:38:34 -05:00
social bookmark API, oexchange enhancements
2011-09-20 19:31:45 -04:00
function arr_add_hashes(&$item,$k) {
$item = '#' . $item;
}
ajaxify the wall poster
2010-07-23 01:41:45 -04:00
function parse_url_content(&$a) {
enhance url parse
2010-07-23 02:17:41 -04:00
allow setting all the bookmarked URL fields via script/API
2011-09-20 01:21:55 -04:00
$text = null;
social bookmark API, oexchange enhancements
2011-09-20 19:31:45 -04:00
$str_tags = '';
allow setting all the bookmarked URL fields via script/API
2011-09-20 01:21:55 -04:00
more fixes for textmode
2012-02-07 03:13:16 -05:00
$textmode = false;
if(local_user() && intval(get_pconfig(local_user(),'system','plaintext')))
$textmode = true;
fix "insert web link" for those without richtext feature enabled
2012-12-03 19:31:28 -05:00
if(local_user() && (! feature_enabled(local_user(),'richtext')))
$textmode = true;
parse_url: Fetch multiple pictures so that the user can decide what to take.
2012-07-12 02:20:27 -04:00
//if($textmode)
$br = (($textmode) ? "\n" : '<br />');
more fixes for textmode
2012-02-07 03:13:16 -05:00
allow setting all the bookmarked URL fields via script/API
2011-09-20 01:21:55 -04:00
if(x($_GET,'binurl'))
$url = trim(hex2bin($_GET['binurl']));
else
$url = trim($_GET['url']);
if($_GET['title'])
$title = strip_tags(trim($_GET['title']));
bug #37, ampersand + everything following in url is stripped when using link icon to share link
2011-04-10 06:36:12 -04:00
social bookmark API, oexchange enhancements
2011-09-20 19:31:45 -04:00
if($_GET['description'])
$text = strip_tags(trim($_GET['description']));
if($_GET['tags']) {
$arr_tags = str_getcsv($_GET['tags']);
if(count($arr_tags)) {
array_walk($arr_tags,'arr_add_hashes');
more fixes for textmode
2012-02-07 03:13:16 -05:00
$str_tags = $br . implode(' ',$arr_tags) . $br;
social bookmark API, oexchange enhancements
2011-09-20 19:31:45 -04:00
}
}
bug #37, ampersand + everything following in url is stripped when using link icon to share link
2011-04-10 06:36:12 -04:00
logger('parse_url: ' . $url);
ajaxify the wall poster
2010-07-23 01:41:45 -04:00
more fixes for textmode
2012-02-07 03:13:16 -05:00
if($textmode)
$template = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
else
$template = "<br /><a class=\"bookmark\" href=\"%s\" >%s</a>%s<br />";
ajaxify the wall poster
2010-07-23 01:41:45 -04:00
more plugin hooks
2010-12-25 18:01:02 -05:00
$arr = array('url' => $url, 'text' => '');
call_hooks('parse_link', $arr);
if(strlen($arr['text'])) {
echo $arr['text'];
killme();
}
better handling of remote urls - especially FB opengraph
2011-10-28 05:50:00 -04:00
allow setting all the bookmarked URL fields via script/API
2011-09-20 01:21:55 -04:00
if($url && $title && $text) {
more fixes for textmode
2012-02-07 03:13:16 -05:00
if($textmode)
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
$text = $br . '[quote]' . trim($text) . '[/quote]' . $br;
more fixes for textmode
2012-02-07 03:13:16 -05:00
else
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
$text = '<br /><blockquote>' . trim($text) . '</blockquote><br />';
more fixes for textmode
2012-02-07 03:13:16 -05:00
allow setting all the bookmarked URL fields via script/API
2011-09-20 01:21:55 -04:00
$title = str_replace(array("\r","\n"),array('',''),$title);
social bookmark API, oexchange enhancements
2011-09-20 19:31:45 -04:00
$result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
allow setting all the bookmarked URL fields via script/API
2011-09-20 01:21:55 -04:00
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
logger('parse_url (unparsed): returns: ' . $result);
allow setting all the bookmarked URL fields via script/API
2011-09-20 01:21:55 -04:00
echo $result;
killme();
}
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
$siteinfo = parseurl_getsiteinfo($url);
allow setting all the bookmarked URL fields via script/API
2011-09-20 01:21:55 -04:00
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
if($siteinfo["title"] == "") {
parse_url: Fixed a problem in the detection if there was an image or not. Additionally removed some test content.
2012-08-02 04:25:08 -04:00
echo sprintf($template,$url,$url,'') . $str_tags;
ajaxify the wall poster
2010-07-23 01:41:45 -04:00
killme();
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
} else {
$text = $siteinfo["text"];
$title = $siteinfo["title"];
ajaxify the wall poster
2010-07-23 01:41:45 -04:00
}
parse_url: Fetch multiple pictures so that the user can decide what to take.
2012-07-12 02:20:27 -04:00
$image = "";
typo on previous checkin - set max number of images for an inserted link, default 2.
2012-09-14 02:01:10 -04:00
if(sizeof($siteinfo["images"]) > 0){
/* Execute below code only if image is present in siteinfo */
$total_images = 0;
$max_images = get_config('system','max_bookmark_images');
if($max_images === false)
$max_images = 2;
parse_url: Fetch multiple pictures so that the user can decide what to take.
2012-07-12 02:20:27 -04:00
else
typo on previous checkin - set max number of images for an inserted link, default 2.
2012-09-14 02:01:10 -04:00
$max_images = intval($max_images);
foreach ($siteinfo["images"] as $imagedata) {
if($textmode)
$image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
else
$image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
$total_images ++;
if($max_images && $max_images >= $total_images)
break;
Access image part of siteinfo only if $siteinfo["images"] is actually set. The current code didn't check this so I was getting an error Warning: Invalid argument supplied for foreach() in /home/vasudev/Public/friendica/mod/parse_url.php on line 305 Implemented a check before accessing $siteinfo["images"] with this commit and now warning doesn't come when giving URL of a site which doesn't have image from link button of post area
2012-07-22 03:11:07 -04:00
}
typo on previous checkin - set max number of images for an inserted link, default 2.
2012-09-14 02:01:10 -04:00
}
parse_url: Fetch multiple pictures so that the user can decide what to take.
2012-07-12 02:20:27 -04:00
enhance url parse
2010-07-23 02:17:41 -04:00
if(strlen($text)) {
more fixes for textmode
2012-02-07 03:13:16 -05:00
if($textmode)
parse_url: Fetch multiple pictures so that the user can decide what to take.
2012-07-12 02:20:27 -04:00
$text = $br.'[quote]'.trim($text).'[/quote]'.$br ;
more fixes for textmode
2012-02-07 03:13:16 -05:00
else
parse_url: Fetch multiple pictures so that the user can decide what to take.
2012-07-12 02:20:27 -04:00
$text = '<br /><blockquote>'.trim($text).'</blockquote><br />';
ajaxify the wall poster
2010-07-23 01:41:45 -04:00
}
better handling of remote urls - especially FB opengraph
2011-10-28 05:50:00 -04:00
if($image) {
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
$text = $br.$br.$image.$text;
better handling of remote urls - especially FB opengraph
2011-10-28 05:50:00 -04:00
}
allow setting all the bookmarked URL fields via script/API
2011-09-20 01:21:55 -04:00
$title = str_replace(array("\r","\n"),array('',''),$title);
add timezone convert module, several other minor or in progress fixes
2011-09-04 03:48:45 -04:00
social bookmark API, oexchange enhancements
2011-09-20 19:31:45 -04:00
$result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
bookmarks + bug #140
2011-09-04 22:58:03 -04:00
parse_url: complete new code for fetching website information
2012-07-11 19:17:33 -04:00
logger('parse_url: returns: ' . $result);
bookmarks + bug #140
2011-09-04 22:58:03 -04:00
parse_url: Further improvements of the new method to fetch page data
2012-07-12 17:41:04 -04:00
echo trim($result);
ajaxify the wall poster
2010-07-23 01:41:45 -04:00
killme();
Add oembed user option for use oembed instead of standard yotube embedding. Remove global option and parse_url use of oembed.
2011-05-23 04:37:09 -04:00
}
Reference in New Issue Copy Permalink