don't try to scrape atom/rss feeds. missing param in contact store (mod_follow)

[friendica.git] / include / Scrape.php

diff --git a/include/Scrape.php b/include/Scrape.php
index 10ec54d1347aa968ec35c3d9b1b9185ca61976dc..bb42c3bdd3474946db9325ad4fb6846141c5571e 100644 (file)
--- a/include/Scrape.php
+++ b/include/Scrape.php
@@ -5,12 +5,25 @@ require_once('library/HTML5/Parser.php');
 if(! function_exists('scrape_dfrn')) {
 function scrape_dfrn($url) {
 
+       $a = get_app();
+
        $ret = array();
        $s = fetch_url($url);
 
        if(! $s) 
                return $ret;
 
+       $headers = $a->get_curl_headers();
+       $lines = explode("\n",$headers);
+       if(count($lines)) {
+               foreach($lines as $line) {                              
+                       // don't try and run feeds through the html5 parser
+                       if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
+                               return ret;
+               }
+       }
+
+
        $dom = HTML5_Parser::parse($s);
 
        if(! $dom)
@@ -22,6 +35,8 @@ function scrape_dfrn($url) {
 
        foreach($items as $item) {
                $x = $item->getAttribute('rel');
+               if(($x === 'alternate') && ($item->getAttribute('type') === 'application/atom+xml'))
+                       $ret['feed_atom'] = $item->getAttribute('href');
                if(substr($x,0,5) == "dfrn-")
                        $ret[$x] = $item->getAttribute('href');
                if($x === 'lrdd') {
@@ -75,12 +90,26 @@ function validate_dfrn($a) {
 if(! function_exists('scrape_meta')) {
 function scrape_meta($url) {
 
+       $a = get_app();
+
        $ret = array();
        $s = fetch_url($url);
 
        if(! $s) 
                return $ret;
 
+       $headers = $a->get_curl_headers();
+       $lines = explode("\n",$headers);
+       if(count($lines)) {
+               foreach($lines as $line) {                              
+                       // don't try and run feeds through the html5 parser
+                       if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
+                               return ret;
+               }
+       }
+
+
+
        $dom = HTML5_Parser::parse($s);
 
        if(! $dom)
@@ -103,12 +132,24 @@ function scrape_meta($url) {
 if(! function_exists('scrape_vcard')) {
 function scrape_vcard($url) {
 
+       $a = get_app();
+
        $ret = array();
        $s = fetch_url($url);
 
        if(! $s) 
                return $ret;
 
+       $headers = $a->get_curl_headers();
+       $lines = explode("\n",$headers);
+       if(count($lines)) {
+               foreach($lines as $line) {                              
+                       // don't try and run feeds through the html5 parser
+                       if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
+                               return ret;
+               }
+       }
+
        $dom = HTML5_Parser::parse($s);
 
        if(! $dom)
@@ -135,3 +176,52 @@ function scrape_vcard($url) {
 
        return $ret;
 }}
+
+
+if(! function_exists('scrape_feed')) {
+function scrape_feed($url) {
+
+       $a = get_app();
+
+       $ret = array();
+       $s = fetch_url($url);
+
+       if(! $s) 
+               return $ret;
+
+       $headers = $a->get_curl_headers();
+       $lines = explode("\n",$headers);
+       if(count($lines)) {
+               foreach($lines as $line) {                              
+                       if(stristr($line,'content-type:')) {
+                               if(stristr($line,'application/atom+xml')) {
+                                       $ret['feed_atom'] = $url;
+                                       return $ret;
+                               }
+                               if(stristr($line,'application/rss+xml')) {
+                                       $ret['feed_rss'] = $url;
+                                       return ret;
+                               }
+                       }
+               }
+       }
+
+       $dom = HTML5_Parser::parse($s);
+
+       if(! $dom)
+               return $ret;
+
+       $items = $dom->getElementsByTagName('link');
+
+       // get Atom link elements
+
+       foreach($items as $item) {
+               $x = $item->getAttribute('rel');
+               if(($x === 'alternate') && ($item->getAttribute('type') === 'application/atom+xml'))
+                       $ret['feed_atom'] = $item->getAttribute('href');
+               if(($x === 'alternate') && ($item->getAttribute('type') === 'application/rss+xml'))
+                       $ret['feed_rss'] = $item->getAttribute('href');
+       }
+
+       return $ret;
+}}
\ No newline at end of file