X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=include%2FScrape.php;h=6ee3dabfca88c11c42576e2a5c75c5194a96283e;hb=143e1a4a836e1fe4d05ab5d1c6008aa0198d2db5;hp=7b72a199dfe1307a689dbe4d3a3321ac015686be;hpb=dbf51229d9014089eba0da41a55f23c507f7e2ea;p=friendica.git diff --git a/include/Scrape.php b/include/Scrape.php index 7b72a199df..6ee3dabfca 100644 --- a/include/Scrape.php +++ b/include/Scrape.php @@ -4,7 +4,7 @@ require_once('library/HTML5/Parser.php'); require_once('include/crypto.php'); if(! function_exists('scrape_dfrn')) { -function scrape_dfrn($url) { +function scrape_dfrn($url, $dont_probe = false) { $a = get_app(); @@ -14,16 +14,23 @@ function scrape_dfrn($url) { $s = fetch_url($url); - if(! $s) + if(! $s) return $ret; + if (!$dont_probe) { + $probe = probe_url($url); + + if (isset($probe["addr"])) + $ret["addr"] = $probe["addr"]; + } + $headers = $a->get_curl_headers(); logger('scrape_dfrn: headers=' . $headers, LOGGER_DEBUG); $lines = explode("\n",$headers); if(count($lines)) { - foreach($lines as $line) { + foreach($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; @@ -120,7 +127,7 @@ function scrape_meta($url) { $s = fetch_url($url); - if(! $s) + if(! $s) return $ret; $headers = $a->get_curl_headers(); @@ -128,7 +135,7 @@ function scrape_meta($url) { $lines = explode("\n",$headers); if(count($lines)) { - foreach($lines as $line) { + foreach($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; @@ -169,13 +176,13 @@ function scrape_vcard($url) { $s = fetch_url($url); - if(! $s) + if(! $s) return $ret; $headers = $a->get_curl_headers(); $lines = explode("\n",$headers); if(count($lines)) { - foreach($lines as $line) { + foreach($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; @@ -236,20 +243,20 @@ function scrape_feed($url) { logger('scrape_feed: returns: ' . $code . ' headers=' . $headers, LOGGER_DEBUG); if(! $s) { - logger('scrape_feed: no data returned for ' . $url); + logger('scrape_feed: no data returned for ' . $url); return $ret; } $lines = explode("\n",$headers); if(count($lines)) { - foreach($lines as $line) { + foreach($lines as $line) { if(stristr($line,'content-type:')) { if(stristr($line,'application/atom+xml') || stristr($s,'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 . $ret['feed_atom']; @@ -327,7 +327,7 @@ function scrape_feed($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. * */ @@ -335,7 +335,7 @@ function scrape_feed($url) { define ( 'PROBE_NORMAL', 0); define ( 'PROBE_DIASPORA', 1); -function probe_url($url, $mode = PROBE_NORMAL) { +function probe_url($url, $mode = PROBE_NORMAL, $level = 1) { require_once('include/email.php'); $result = array(); @@ -343,6 +343,12 @@ function probe_url($url, $mode = PROBE_NORMAL) { if(! $url) return $result; + $result = Cache::get("probe_url:".$mode.":".$url); + if (!is_null($result)) { + $result = unserialize($result); + return $result; + } + $network = null; $diaspora = false; $diaspora_base = ''; @@ -350,13 +356,21 @@ function probe_url($url, $mode = PROBE_NORMAL) { $diaspora_key = ''; $has_lrdd = false; $email_conversant = false; + $connectornetworks = false; + $appnet = false; + + if (strpos($url,'twitter.com')) { + $connectornetworks = true; + $network = NETWORK_TWITTER; + } - $twitter = ((strpos($url,'twitter.com') !== false) ? true : false); + // Twitter is deactivated since twitter closed its old API + //$twitter = ((strpos($url,'twitter.com') !== false) ? true : false); $lastfm = ((strpos($url,'last.fm/user') !== false) ? true : false); $at_addr = ((strpos($url,'@') !== false) ? true : false); - if((! $twitter) && (! $lastfm)) { + if((!$appnet) && (!$lastfm) && !$connectornetworks) { if(strpos($url,'mailto:') !== false && $at_addr) { $url = str_replace('mailto:','',$url); @@ -400,13 +414,16 @@ function probe_url($url, $mode = PROBE_NORMAL) { $pubkey = $diaspora_key; $diaspora = true; } + if(($link['@attributes']['rel'] === 'http://ostatus.org/schema/1.0/subscribe') AND ($mode == PROBE_NORMAL)) { + $diaspora = false; + } } // Status.Net can have more than one profile URL. We need to match the profile URL // to a contact on incoming messages to prevent spam, and we won't know which one // to match. So in case of two, one of them is stored as an alias. Only store URL's // and not webfinger user@host aliases. If they've got more than two non-email style - // aliases, let's hope we're lucky and get one that matches the feed author-uri because + // aliases, let's hope we're lucky and get one that matches the feed author-uri because // otherwise we're screwed. foreach($links as $link) { @@ -421,6 +438,10 @@ function probe_url($url, $mode = PROBE_NORMAL) { } } } + + // If the profile is different from the url then the url is abviously an alias + if (($alias == "") AND ($profile != "") AND !$at_addr AND (normalise_link($profile) != normalise_link($url))) + $alias = $url; } elseif($mode == PROBE_NORMAL) { @@ -488,6 +509,7 @@ function probe_url($url, $mode = PROBE_NORMAL) { } if($mode == PROBE_NORMAL) { + if(strlen($zot)) { $s = fetch_url($zot); if($s) { @@ -495,8 +517,8 @@ function probe_url($url, $mode = PROBE_NORMAL) { if($j) { $network = NETWORK_ZOT; $vcard = array( - 'fn' => $j->fullname, - 'nick' => $j->nickname, + 'fn' => $j->fullname, + 'nick' => $j->nickname, 'photo' => $j->photo ); $profile = $j->url; @@ -507,8 +529,9 @@ function probe_url($url, $mode = PROBE_NORMAL) { } } + if(strlen($dfrn)) { - $ret = scrape_dfrn(($hcard) ? $hcard : $dfrn); + $ret = scrape_dfrn(($hcard) ? $hcard : $dfrn, true); if(is_array($ret) && x($ret,'dfrn-request')) { $network = NETWORK_DFRN; $request = $ret['dfrn-request']; @@ -536,8 +559,12 @@ function probe_url($url, $mode = PROBE_NORMAL) { if($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,'@')) + $addr = str_replace('acct:', '', $url); + $priority = 0; if($hcard && ! $vcard) { @@ -561,7 +588,7 @@ function probe_url($url, $mode = PROBE_NORMAL) { $vcard['nick'] = $addr_parts[0]; } - if($twitter) { + /* if($twitter) { logger('twitter: setup'); $tid = basename($url); $tapi = 'https://api.twitter.com/1/statuses/user_timeline.rss'; @@ -574,7 +601,7 @@ function probe_url($url, $mode = PROBE_NORMAL) { $vcard['photo'] = 'https://api.twitter.com/1/users/profile_image?screen_name=' . $tid . '&size=bigger'; $vcard['nick'] = $tid; $vcard['fn'] = $tid; - } + } */ if($lastfm) { $profile = $url; @@ -596,27 +623,31 @@ function probe_url($url, $mode = PROBE_NORMAL) { // Will leave it to others to figure out how to grab the avatar, which is on the $url page in the open graph meta links } - if($twitter || ! $poll) + if($appnet || ! $poll) $check_feed = true; if((! isset($vcard)) || (! x($vcard,'fn')) || (! $profile)) $check_feed = true; if(($at_addr) && (! count($links))) $check_feed = false; + if ($connectornetworks) + $check_feed = false; + if($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'])); - if(! x($vcard)) + if(! x($vcard)) $vcard = array(); } if(x($feedret,'photo') && (! x($vcard,'photo'))) $vcard['photo'] = $feedret['photo']; require_once('library/simplepie/simplepie.inc'); - $feed = new SimplePie(); + $feed = new SimplePie(); $xml = fetch_url($poll); logger('probe_url: fetch feed: ' . $poll . ' returns: ' . $xml, LOGGER_DATA); @@ -624,40 +655,52 @@ function probe_url($url, $mode = PROBE_NORMAL) { logger('probe_url: scrape_feed: headers: ' . $a->get_curl_headers(), LOGGER_DATA); - $feed->set_raw_data($xml); + // Don't try and parse an empty string + $feed->set_raw_data(($xml) ? $xml : ''); - $feed->init(); - if($feed->error()) + $feed->init(); + if($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(); $author = $feed->get_author(); - if($author) { + if($author) { $vcard['fn'] = unxmlify(trim($author->get_name())); if(! $vcard['fn']) $vcard['fn'] = trim(unxmlify($author->get_email())); if(strpos($vcard['fn'],'@') !== false) $vcard['fn'] = substr($vcard['fn'],0,strpos($vcard['fn'],'@')); + $email = unxmlify($author->get_email()); if(! $profile && $author->get_link()) $profile = trim(unxmlify($author->get_link())); if(! $vcard['photo']) { $rawtags = $feed->get_feed_tags( SIMPLEPIE_NAMESPACE_ATOM_10, 'author'); - if($rawtags) { + if($rawtags) { $elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10]; if((x($elems,'link')) && ($elems['link'][0]['attribs']['']['rel'] === 'photo')) $vcard['photo'] = $elems['link'][0]['attribs']['']['href']; - } + } + } + // Fetch fullname via poco:displayName + $pocotags = $feed->get_feed_tags(SIMPLEPIE_NAMESPACE_ATOM_10, 'author'); + if ($pocotags) { + $elems = $pocotags[0]['child']['http://portablecontacts.net/spec/1.0']; + if (isset($elems["displayName"])) + $vcard['fn'] = $elems["displayName"][0]["data"]; + if (isset($elems["preferredUsername"])) + $vcard['nick'] = $elems["preferredUsername"][0]["data"]; } } else { $item = $feed->get_item(0); if($item) { $author = $item->get_author(); - if($author) { + if($author) { $vcard['fn'] = trim(unxmlify($author->get_name())); if(! $vcard['fn']) $vcard['fn'] = trim(unxmlify($author->get_email())); @@ -674,22 +717,66 @@ function probe_url($url, $mode = PROBE_NORMAL) { } if(! $vcard['photo']) { $rawtags = $item->get_item_tags( SIMPLEPIE_NAMESPACE_ATOM_10, 'author'); - if($rawtags) { + if($rawtags) { $elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10]; if((x($elems,'link')) && ($elems['link'][0]['attribs']['']['rel'] === 'photo')) $vcard['photo'] = $elems['link'][0]['attribs']['']['href']; - } + } } } } + // Workaround for misconfigured Friendica servers + if (($network == "") AND (strstr($url, "/profile/"))) { + $noscrape = str_replace("/profile/", "/noscrape/", $url); + $noscrapejson = fetch_url($noscrape); + if ($noscrapejson) { + + $network = NETWORK_DFRN; + + $poco = str_replace("/profile/", "/poco/", $url); + + $noscrapedata = json_decode($noscrapejson, true); + + if (isset($noscrapedata["addr"])) + $addr = $noscrapedata["addr"]; + + if (isset($noscrapedata["fn"])) + $vcard["fn"] = $noscrapedata["fn"]; + + if (isset($noscrapedata["key"])) + $pubkey = $noscrapedata["key"]; + + if (isset($noscrapedata["photo"])) + $vcard["photo"] = $noscrapedata["photo"]; + + if (isset($noscrapedata["dfrn-request"])) + $request = $noscrapedata["dfrn-request"]; + + if (isset($noscrapedata["dfrn-confirm"])) + $confirm = $noscrapedata["dfrn-confirm"]; + + if (isset($noscrapedata["dfrn-notify"])) + $notify = $noscrapedata["dfrn-notify"]; + + if (isset($noscrapedata["dfrn-poll"])) + $poll = $noscrapedata["dfrn-poll"]; + + } + } + if((! $vcard['photo']) && strlen($email)) $vcard['photo'] = avatar_img($email); if($poll === $profile) $lnk = $feed->get_permalink(); if(isset($lnk) && strlen($lnk)) - $profile = $lnk; + $profile = $lnk; + if(! $network) { + $network = NETWORK_FEED; + // If it is a feed, don't take the author name as feed name + unset($vcard['fn']); + } if(! (x($vcard,'fn'))) $vcard['fn'] = notags($feed->get_title()); if(! (x($vcard,'fn'))) @@ -704,8 +791,6 @@ function probe_url($url, $mode = PROBE_NORMAL) { if(strpos($vcard['nick'],' ')) $vcard['nick'] = trim(substr($vcard['nick'],0,strpos($vcard['nick'],' '))); } - if(! $network) - $network = NETWORK_FEED; if(! $priority) $priority = 2; } @@ -713,7 +798,7 @@ function probe_url($url, $mode = PROBE_NORMAL) { if(! x($vcard,'photo')) { $a = get_app(); - $vcard['photo'] = $a->get_baseurl() . '/images/person-175.jpg' ; + $vcard['photo'] = $a->get_baseurl() . '/images/person-175.jpg' ; } if(! $profile) @@ -724,9 +809,28 @@ function probe_url($url, $mode = PROBE_NORMAL) { if(($network === NETWORK_FEED) && ($poll) && (! x($vcard,'fn'))) $vcard['fn'] = $url; + if (($notify != "") AND ($poll != "")) { + $baseurl = matching(normalise_link($notify), normalise_link($poll)); + + $baseurl2 = matching($baseurl, normalise_link($profile)); + if ($baseurl2 != "") + $baseurl = $baseurl2; + } + + if (($baseurl == "") AND ($notify != "")) + $baseurl = matching(normalise_link($profile), normalise_link($notify)); + + if (($baseurl == "") AND ($poll != "")) + $baseurl = matching(normalise_link($profile), normalise_link($poll)); + + $baseurl = rtrim($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'])); - + $result['name'] = $vcard['fn']; $result['nick'] = $vcard['nick']; $result['url'] = $profile; @@ -742,8 +846,49 @@ function probe_url($url, $mode = PROBE_NORMAL) { $result['network'] = $network; $result['alias'] = $alias; $result['pubkey'] = $pubkey; + $result['baseurl'] = $baseurl; logger('probe_url: ' . print_r($result,true), LOGGER_DEBUG); + if ($level == 1) { + // Trying if it maybe a diaspora account + if (($result['network'] == NETWORK_FEED) OR ($result['addr'] == "")) { + require_once('include/bbcode.php'); + $address = GetProfileUsername($url, "", true); + $result2 = probe_url($address, $mode, ++$level); + if ($result2['network'] != "") + $result = $result2; + } + + // Maybe it's some non standard GNU Social installation (Single user, subfolder or no uri rewrite) + if (($result['network'] == NETWORK_FEED) AND ($result['baseurl'] != "") AND ($result['nick'] != "")) { + $addr = $result['nick'].'@'.str_replace("http://", "", $result['baseurl']); + $result2 = probe_url($addr, $mode, ++$level); + if (($result2['network'] != "") AND ($result2['network'] != NETWORK_FEED)) + $result = $result2; + } + } + + // Only store into the cache if the value seems to be valid + if ($result['network'] != NETWORK_PHANTOM) + Cache::set("probe_url:".$mode.":".$url,serialize($result), CACHE_DAY); + return $result; } + +function matching($part1, $part2) { + $len = min(strlen($part1), strlen($part2)); + + $match = ""; + $matching = true; + $i = 0; + while (($i <= $len) AND $matching) { + if (substr($part1, $i, 1) == substr($part2, $i, 1)) + $match .= substr($part1, $i, 1); + else + $matching = false; + + $i++; + } + return($match); +}