Merge pull request #2016 from fabrixxm/template_vars_hook

[friendica.git] / include / Scrape.php

diff --git a/include/Scrape.php b/include/Scrape.php
index 90fb1a9e3dd99e992d391a6f09f1ff61618c9f9c..ae9331ff7af8fc389ea67aa2e24837e84f8d2e58 100644 (file)
--- a/include/Scrape.php
+++ b/include/Scrape.php
@@ -249,7 +249,7 @@ function scrape_feed($url) {
                                        $ret['feed_atom'] = $url;
                                        return $ret;
                                }
-                               if(stristr($line,'application/rss+xml') || stristr($s,'<rss')) {
+                               if(stristr($line,'application/rss+xml') || stristr($s,'<rss')) {
                                        $ret['feed_rss'] = $url;
                                        return $ret;
                                }
@@ -262,47 +262,40 @@ function scrape_feed($url) {
                }
        }
 
-       try {
-               $dom = HTML5_Parser::parse($s);
-       } catch (DOMException $e) {
-               logger('scrape_feed: parse error: ' . $e);
-       }
+       $basename = implode('/', array_slice(explode('/',$url),0,3)) . '/';
 
-       if(! $dom) {
-               logger('scrape_feed: failed to parse.');
-               return $ret;
-       }
+       $doc = new DOMDocument();
+       @$doc->loadHTML($s);
+       $xpath = new DomXPath($doc);
 
+       $base = $xpath->query("//base");
+       foreach ($base as $node) {
+               $attr = array();
 
-       $head = $dom->getElementsByTagName('base');
-       if($head) {
-               foreach($head as $head0) {
-                       $basename = $head0->getAttribute('href');
-                       break;
-               }
+               if ($node->attributes->length)
+                       foreach ($node->attributes as $attribute)
+                               $attr[$attribute->name] = $attribute->value;
+
+               if ($attr["href"] != "")
+                       $basename = $attr["href"] ;
        }
-       if(! $basename)
-               $basename = implode('/', array_slice(explode('/',$url),0,3)) . '/';
 
-       $items = $dom->getElementsByTagName('link');
+       $list = $xpath->query("//link");
+       foreach ($list as $node) {
+               $attr = array();
 
-       // get Atom/RSS link elements, take the first one of either.
+               if ($node->attributes->length)
+                       foreach ($node->attributes as $attribute)
+                               $attr[$attribute->name] = $attribute->value;
 
-       if($items) {
-               foreach($items as $item) {
-                       $x = $item->getAttribute('rel');
-                       if(($x === 'alternate') && ($item->getAttribute('type') === 'application/atom+xml')) {
-                               if(! x($ret,'feed_atom'))
-                                       $ret['feed_atom'] = $item->getAttribute('href');
-                       }
-                       if(($x === 'alternate') && ($item->getAttribute('type') === 'application/rss+xml')) {
-                               if(! x($ret,'feed_rss'))
-                                       $ret['feed_rss'] = $item->getAttribute('href');
-                       }
-               }
+               if (($attr["rel"] == "alternate") AND ($attr["type"] == "application/atom+xml"))
+                       $ret["feed_atom"] = $attr["href"];
+
+               if (($attr["rel"] == "alternate") AND ($attr["type"] == "application/rss+xml"))
+                       $ret["feed_rss"] = $attr["href"];
        }
 
-       // Drupal and perhaps others only provide relative URL's. Turn them into absolute.
+       // Drupal and perhaps others only provide relative URLs. Turn them into absolute.
 
        if(x($ret,'feed_atom') && (! strstr($ret['feed_atom'],'://')))
                $ret['feed_atom'] = $basename . $ret['feed_atom'];
@@ -327,7 +320,7 @@ function scrape_feed($url) {
  * PROBE_DIASPORA has a bias towards returning Diaspora information
  * while PROBE_NORMAL has a bias towards dfrn/zot - in the case where
  * an address (such as a Friendica address) supports more than one type
- * of network. 
+ * of network.
  *
  */
 
@@ -414,7 +407,7 @@ function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
                                                $pubkey = $diaspora_key;
                                        $diaspora = true;
                                }
-                               if($link['@attributes']['rel'] === 'http://ostatus.org/schema/1.0/subscribe') {
+                               if(($link['@attributes']['rel'] === 'http://ostatus.org/schema/1.0/subscribe') AND ($mode == PROBE_NORMAL)) {
                                        $diaspora = false;
                                }
                        }
@@ -509,6 +502,7 @@ function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
        }
 
        if($mode == PROBE_NORMAL) {
+
                if(strlen($zot)) {
                        $s = fetch_url($zot);
                        if($s) {
@@ -528,6 +522,7 @@ function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
                        }
                }
 
+
                if(strlen($dfrn)) {
                        $ret = scrape_dfrn(($hcard) ? $hcard : $dfrn);
                        if(is_array($ret) && x($ret,'dfrn-request')) {
@@ -557,7 +552,7 @@ function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
        if($network !== NETWORK_ZOT && $network !== NETWORK_DFRN && $network !== NETWORK_MAIL) {
                if($diaspora)
                        $network = NETWORK_DIASPORA;
-               elseif($has_lrdd)
+               elseif($has_lrdd AND ($notify))
                        $network  = NETWORK_OSTATUS;
 
                if(strpos($url,'@'))
@@ -634,6 +629,7 @@ function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
                if($check_feed) {
 
                        $feedret = scrape_feed(($poll) ? $poll : $url);
+
                        logger('probe_url: scrape_feed ' . (($poll)? $poll : $url) . ' returns: ' . print_r($feedret,true), LOGGER_DATA);
                        if(count($feedret) && ($feedret['feed_atom'] || $feedret['feed_rss'])) {
                                $poll = ((x($feedret,'feed_atom')) ? unamp($feedret['feed_atom']) : unamp($feedret['feed_rss']));
@@ -653,12 +649,13 @@ function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
                        logger('probe_url: scrape_feed: headers: ' . $a->get_curl_headers(), LOGGER_DATA);
 
                        // Don't try and parse an empty string
-                       $feed->set_raw_data(($xml) ? $xml : '<?xml version="1.0" encoding="utf-8" ?><xml></xml>');
+                       $feed->set_raw_data(($xml) ? $xml : '<?xml version="1.0" encoding="utf-8" ?><xml></xml>');
 
                        $feed->init();
-                       if($feed->error())
+                       if($feed->error()) {
                                logger('probe_url: scrape_feed: Error parsing XML: ' . $feed->error());
-
+                               $network = NETWORK_PHANTOM;
+                       }
 
                        if(! x($vcard,'photo'))
                                $vcard['photo'] = $feed->get_image_url();
@@ -782,6 +779,9 @@ function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
 
        $baseurl = rtrim($baseurl, "/");
 
+       if(strpos($url,'@') AND ($addr == "") AND ($network == NETWORK_DFRN))
+               $addr = str_replace('acct:', '', $url);
+
        $vcard['fn'] = notags($vcard['fn']);
        $vcard['nick'] = str_replace(' ','',notags($vcard['nick']));
 
@@ -823,7 +823,9 @@ function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
                }
        }
 
-       Cache::set("probe_url:".$mode.":".$url,serialize($result));
+       // Only store into the cache if the value seems to be valid
+       if ($result['network'] != NETWORK_FEED)
+               Cache::set("probe_url:".$mode.":".$url,serialize($result), CACHE_DAY);
 
        return $result;
 }