X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=include%2Ffeed.php;h=fb97c14ba126aa1875d1202df05a5caa8aad64e7;hb=86d3adaa5084639a1221fc74ac44d5508add25f4;hp=dd360a07b58bf46df256c3143e97329435617a1f;hpb=a3221945d0235d99ae74a86ec96c5b9b7619a1cf;p=friendica.git
diff --git a/include/feed.php b/include/feed.php
index dd360a07b5..fb97c14ba1 100644
--- a/include/feed.php
+++ b/include/feed.php
@@ -2,63 +2,129 @@
require_once("include/html2bbcode.php");
require_once("include/items.php");
-function feed_import($xml,$importer,&$contact, &$hub) {
+/**
+ * @brief Read a RSS/RDF/Atom feed and create an item entry for it
+ *
+ * @param string $xml The feed data
+ * @param array $importer The user record of the importer
+ * @param array $contact The contact record of the feed
+ * @param string $hub Unused dummy value for compatibility reasons
+ * @param bool $simulate If enabled, no data is imported
+ *
+ * @return array In simulation mode it returns the header and the first item
+ */
+function feed_import($xml,$importer,&$contact, &$hub, $simulate = false) {
$a = get_app();
- logger("Import Atom/RSS feed", LOGGER_DEBUG);
-
- if ($xml == "")
+ if (!$simulate) {
+ logger("Import Atom/RSS feed '".$contact["name"]."' (Contact ".$contact["id"].") for user ".$importer["uid"], LOGGER_DEBUG);
+ } else {
+ logger("Test Atom/RSS feed", LOGGER_DEBUG);
+ }
+ if ($xml == "") {
+ logger('XML is empty.', LOGGER_DEBUG);
return;
+ }
+
+ if (!empty($contact['poll'])) {
+ $basepath = $contact['poll'];
+ } elseif (!empty($contact['url'])) {
+ $basepath = $contact['url'];
+ } else {
+ $basepath = '';
+ }
$doc = new DOMDocument();
- @$doc->loadXML($xml);
+ @$doc->loadXML(trim($xml));
$xpath = new DomXPath($doc);
- $xpath->registerNamespace('atom', "http://www.w3.org/2005/Atom");
+ $xpath->registerNamespace('atom', NAMESPACE_ATOM1);
$xpath->registerNamespace('dc', "http://purl.org/dc/elements/1.1/");
$xpath->registerNamespace('content', "http://purl.org/rss/1.0/modules/content/");
$xpath->registerNamespace('rdf', "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
$xpath->registerNamespace('rss', "http://purl.org/rss/1.0/");
$xpath->registerNamespace('media', "http://search.yahoo.com/mrss/");
+ $xpath->registerNamespace('poco', NAMESPACE_POCO);
$author = array();
// Is it RDF?
if ($xpath->query('/rdf:RDF/rss:channel')->length > 0) {
- //$author["author-link"] = $xpath->evaluate('/rdf:RDF/rss:channel/rss:link/text()')->item(0)->nodeValue;
+ $author["author-link"] = $xpath->evaluate('/rdf:RDF/rss:channel/rss:link/text()')->item(0)->nodeValue;
$author["author-name"] = $xpath->evaluate('/rdf:RDF/rss:channel/rss:title/text()')->item(0)->nodeValue;
- if ($author["author-name"] == "")
+ if ($author["author-name"] == "") {
$author["author-name"] = $xpath->evaluate('/rdf:RDF/rss:channel/rss:description/text()')->item(0)->nodeValue;
-
+ }
$entries = $xpath->query('/rdf:RDF/rss:item');
}
// Is it Atom?
- if ($xpath->query('/atom:feed/atom:entry')->length > 0) {
- //$self = $xpath->query("/atom:feed/atom:link[@rel='self']")->item(0)->attributes;
- //if (is_object($self))
- // foreach($self AS $attributes)
- // if ($attributes->name == "href")
- // $author["author-link"] = $attributes->textContent;
-
- //if ($author["author-link"] == "") {
- // $alternate = $xpath->query("/atom:feed/atom:link[@rel='alternate']")->item(0)->attributes;
- // if (is_object($alternate))
- // foreach($alternate AS $attributes)
- // if ($attributes->name == "href")
- // $author["author-link"] = $attributes->textContent;
- //}
+ if ($xpath->query('/atom:feed')->length > 0) {
+ $alternate = $xpath->query("atom:link[@rel='alternate']")->item(0)->attributes;
+ if (is_object($alternate)) {
+ foreach ($alternate AS $attributes) {
+ if ($attributes->name == "href") {
+ $author["author-link"] = $attributes->textContent;
+ }
+ }
+ }
+
+ if ($author["author-link"] == "") {
+ $author["author-link"] = $author["author-id"];
+ }
+ if ($author["author-link"] == "") {
+ $self = $xpath->query("atom:link[@rel='self']")->item(0)->attributes;
+ if (is_object($self)) {
+ foreach ($self AS $attributes) {
+ if ($attributes->name == "href") {
+ $author["author-link"] = $attributes->textContent;
+ }
+ }
+ }
+ }
+
+ if ($author["author-link"] == "") {
+ $author["author-link"] = $xpath->evaluate('/atom:feed/atom:id/text()')->item(0)->nodeValue;
+ }
+ $author["author-avatar"] = $xpath->evaluate('/atom:feed/atom:logo/text()')->item(0)->nodeValue;
$author["author-name"] = $xpath->evaluate('/atom:feed/atom:title/text()')->item(0)->nodeValue;
- if ($author["author-name"] == "")
+ if ($author["author-name"] == "") {
$author["author-name"] = $xpath->evaluate('/atom:feed/atom:subtitle/text()')->item(0)->nodeValue;
-
- if ($author["author-name"] == "")
+ }
+ if ($author["author-name"] == "") {
$author["author-name"] = $xpath->evaluate('/atom:feed/atom:author/atom:name/text()')->item(0)->nodeValue;
+ }
+ $value = $xpath->evaluate('atom:author/poco:displayName/text()')->item(0)->nodeValue;
+ if ($value != "") {
+ $author["author-name"] = $value;
+ }
+ if ($simulate) {
+ $author["author-id"] = $xpath->evaluate('/atom:feed/atom:author/atom:uri/text()')->item(0)->nodeValue;
- //$author["author-avatar"] = $xpath->evaluate('/atom:feed/atom:logo/text()')->item(0)->nodeValue;
+ $value = $xpath->evaluate('atom:author/poco:preferredUsername/text()')->item(0)->nodeValue;
+ if ($value != "") {
+ $author["author-nick"] = $value;
+ }
+ $value = $xpath->evaluate('atom:author/poco:address/poco:formatted/text()', $context)->item(0)->nodeValue;
+ if ($value != "") {
+ $author["author-location"] = $value;
+ }
+ $value = $xpath->evaluate('atom:author/poco:note/text()')->item(0)->nodeValue;
+ if ($value != "") {
+ $author["author-about"] = $value;
+ }
+ $avatar = $xpath->evaluate("atom:author/atom:link[@rel='avatar']")->item(0)->attributes;
+ if (is_object($avatar)) {
+ foreach ($avatar AS $attributes) {
+ if ($attributes->name == "href") {
+ $author["author-avatar"] = $attributes->textContent;
+ }
+ }
+ }
+ }
$author["edited"] = $author["created"] = $xpath->query('/atom:feed/atom:updated/text()')->item(0)->nodeValue;
@@ -69,16 +135,17 @@ function feed_import($xml,$importer,&$contact, &$hub) {
// Is it RSS?
if ($xpath->query('/rss/channel')->length > 0) {
- //$author["author-link"] = $xpath->evaluate('/rss/channel/link/text()')->item(0)->nodeValue;
+ $author["author-link"] = $xpath->evaluate('/rss/channel/link/text()')->item(0)->nodeValue;
+
$author["author-name"] = $xpath->evaluate('/rss/channel/title/text()')->item(0)->nodeValue;
- //$author["author-avatar"] = $xpath->evaluate('/rss/channel/image/url/text()')->item(0)->nodeValue;
+ $author["author-avatar"] = $xpath->evaluate('/rss/channel/image/url/text()')->item(0)->nodeValue;
- if ($author["author-name"] == "")
+ if ($author["author-name"] == "") {
$author["author-name"] = $xpath->evaluate('/rss/channel/copyright/text()')->item(0)->nodeValue;
-
- if ($author["author-name"] == "")
+ }
+ if ($author["author-name"] == "") {
$author["author-name"] = $xpath->evaluate('/rss/channel/description/text()')->item(0)->nodeValue;
-
+ }
$author["edited"] = $author["created"] = $xpath->query('/rss/channel/pubDate/text()')->item(0)->nodeValue;
$author["app"] = $xpath->evaluate('/rss/channel/generator/text()')->item(0)->nodeValue;
@@ -86,18 +153,18 @@ function feed_import($xml,$importer,&$contact, &$hub) {
$entries = $xpath->query('/rss/channel/item');
}
- //if ($author["author-link"] == "")
+ if (!$simulate) {
$author["author-link"] = $contact["url"];
- if ($author["author-name"] == "")
- $author["author-name"] = $contact["name"];
-
- //if ($author["author-avatar"] == "")
+ if ($author["author-name"] == "") {
+ $author["author-name"] = $contact["name"];
+ }
$author["author-avatar"] = $contact["thumb"];
- $author["owner-link"] = $contact["url"];
- $author["owner-name"] = $contact["name"];
- $author["owner-avatar"] = $contact["thumb"];
+ $author["owner-link"] = $contact["url"];
+ $author["owner-name"] = $contact["name"];
+ $author["owner-avatar"] = $contact["thumb"];
+ }
$header = array();
$header["uid"] = $importer["uid"];
@@ -112,103 +179,116 @@ function feed_import($xml,$importer,&$contact, &$hub) {
$header["contact-id"] = $contact["id"];
- if(!strlen($contact["notify"])) {
+ if (!strlen($contact["notify"])) {
// one way feed - no remote comment ability
$header["last-child"] = 0;
}
- if (!is_object($entries))
+ if (!is_object($entries)) {
+ logger("There are no entries in this feed.", LOGGER_DEBUG);
return;
+ }
- foreach ($entries AS $entry)
- $entrylist[] = $entry;
+ $items = array();
+ $entrylist = array();
+
+ foreach ($entries AS $entry) {
+ $entrylist[] = $entry;
+ }
foreach (array_reverse($entrylist) AS $entry) {
$item = array_merge($header, $author);
- $item["title"] = $xpath->evaluate('atom:title/text()', $entry)->item(0)->nodeValue;
-
- if ($item["title"] == "")
- $item["title"] = $xpath->evaluate('title/text()', $entry)->item(0)->nodeValue;
-
- if ($item["title"] == "")
- $item["title"] = $xpath->evaluate('rss:title/text()', $entry)->item(0)->nodeValue;
-
$alternate = $xpath->query("atom:link[@rel='alternate']", $entry)->item(0)->attributes;
- if (!is_object($alternate))
+ if (!is_object($alternate)) {
$alternate = $xpath->query("atom:link", $entry)->item(0)->attributes;
-
- if (is_object($alternate))
- foreach($alternate AS $attributes)
- if ($attributes->name == "href")
+ }
+ if (is_object($alternate)) {
+ foreach ($alternate AS $attributes) {
+ if ($attributes->name == "href") {
$item["plink"] = $attributes->textContent;
-
- if ($item["plink"] == "")
+ }
+ }
+ }
+ if ($item["plink"] == "") {
$item["plink"] = $xpath->evaluate('link/text()', $entry)->item(0)->nodeValue;
-
- if ($item["plink"] == "")
+ }
+ if ($item["plink"] == "") {
$item["plink"] = $xpath->evaluate('rss:link/text()', $entry)->item(0)->nodeValue;
-
- $item["plink"] = original_url($item["plink"]);
+ }
$item["uri"] = $xpath->evaluate('atom:id/text()', $entry)->item(0)->nodeValue;
- if ($item["uri"] == "")
+ if ($item["uri"] == "") {
$item["uri"] = $xpath->evaluate('guid/text()', $entry)->item(0)->nodeValue;
-
- if ($item["uri"] == "")
+ }
+ if ($item["uri"] == "") {
$item["uri"] = $item["plink"];
+ }
+
+ $orig_plink = $item["plink"];
+
+ $item["plink"] = original_url($item["plink"]);
$item["parent-uri"] = $item["uri"];
+ if (!$simulate) {
+ $r = q("SELECT `id` FROM `item` WHERE `uid` = %d AND `uri` = '%s' AND `network` IN ('%s', '%s')",
+ intval($importer["uid"]), dbesc($item["uri"]), dbesc(NETWORK_FEED), dbesc(NETWORK_DFRN));
+ if (dbm::is_result($r)) {
+ logger("Item with uri ".$item["uri"]." for user ".$importer["uid"]." already existed under id ".$r[0]["id"], LOGGER_DEBUG);
+ continue;
+ }
+ }
+
+ $item["title"] = $xpath->evaluate('atom:title/text()', $entry)->item(0)->nodeValue;
+
+ if ($item["title"] == "") {
+ $item["title"] = $xpath->evaluate('title/text()', $entry)->item(0)->nodeValue;
+ }
+ if ($item["title"] == "") {
+ $item["title"] = $xpath->evaluate('rss:title/text()', $entry)->item(0)->nodeValue;
+ }
$published = $xpath->query('atom:published/text()', $entry)->item(0)->nodeValue;
- if ($published == "")
+ if ($published == "") {
$published = $xpath->query('pubDate/text()', $entry)->item(0)->nodeValue;
-
- if ($published == "")
+ }
+ if ($published == "") {
$published = $xpath->query('dc:date/text()', $entry)->item(0)->nodeValue;
-
+ }
$updated = $xpath->query('atom:updated/text()', $entry)->item(0)->nodeValue;
- if ($updated == "")
+ if ($updated == "") {
$updated = $published;
-
- if ($published != "")
+ }
+ if ($published != "") {
$item["created"] = $published;
-
- if ($updated != "")
+ }
+ if ($updated != "") {
$item["edited"] = $updated;
-
+ }
$creator = $xpath->query('author/text()', $entry)->item(0)->nodeValue;
- if ($creator == "")
+ if ($creator == "") {
$creator = $xpath->query('atom:author/atom:name/text()', $entry)->item(0)->nodeValue;
-
- if ($creator == "")
+ }
+ if ($creator == "") {
$creator = $xpath->query('dc:creator/text()', $entry)->item(0)->nodeValue;
-
- if ($creator != "")
+ }
+ if ($creator != "") {
$item["author-name"] = $creator;
-
- if ($pubDate != "")
+ }
+ if ($pubDate != "") {
$item["edited"] = $item["created"] = $pubDate;
-
+ }
$creator = $xpath->query('dc:creator/text()', $entry)->item(0)->nodeValue;
- if ($creator != "")
+ if ($creator != "") {
$item["author-name"] = $creator;
-
- //$item["object"] = $xml;
-
- $r = q("SELECT `id` FROM `item` WHERE `uid` = %d AND `uri` = '%s'",
- intval($importer["uid"]), dbesc($item["uri"]));
- if ($r) {
- logger("Item with uri ".$item["uri"]." for user ".$importer["uid"]." already existed under id ".$r[0]["id"], LOGGER_DEBUG);
- continue;
}
- // To-Do?
+ /// @TODO ?
// Ausland
//
@@ -221,63 +301,115 @@ function feed_import($xml,$importer,&$contact, &$hub) {
$type = "";
$title = "";
- foreach($enclosure->attributes AS $attributes) {
- if ($attributes->name == "url")
+ foreach ($enclosure->attributes AS $attributes) {
+ if ($attributes->name == "url") {
$href = $attributes->textContent;
- elseif ($attributes->name == "length")
+ } elseif ($attributes->name == "length") {
$length = $attributes->textContent;
- elseif ($attributes->name == "type")
+ } elseif ($attributes->name == "type") {
$type = $attributes->textContent;
+ }
}
- if(strlen($item["attach"]))
+ if (strlen($item["attach"])) {
$item["attach"] .= ',';
+ }
$attachments[] = array("link" => $href, "type" => $type, "length" => $length);
$item["attach"] .= '[attach]href="'.$href.'" length="'.$length.'" type="'.$type.'"[/attach]';
}
+ $body = trim($xpath->evaluate('atom:content/text()', $entry)->item(0)->nodeValue);
+
+ if ($body == "") {
+ $body = trim($xpath->evaluate('content:encoded/text()', $entry)->item(0)->nodeValue);
+ }
+ if ($body == "") {
+ $body = trim($xpath->evaluate('description/text()', $entry)->item(0)->nodeValue);
+ }
+ if ($body == "") {
+ $body = trim($xpath->evaluate('atom:summary/text()', $entry)->item(0)->nodeValue);
+ }
+
+ // remove the content of the title if it is identically to the body
+ // This helps with auto generated titles e.g. from tumblr
+ if (title_is_body($item["title"], $body)) {
+ $item["title"] = "";
+ }
+ $item["body"] = html2bbcode($body, $basepath);
+
+ if (($item["body"] == '') && ($item["title"] != '')) {
+ $item["body"] = $item["title"];
+ $item["title"] = '';
+ }
+
if ($contact["fetch_further_information"]) {
$preview = "";
// Handle enclosures and treat them as preview picture
- foreach ($attachments AS $attachment)
- if ($attachment["type"] == "image/jpeg")
+ foreach ($attachments AS $attachment) {
+ if ($attachment["type"] == "image/jpeg") {
$preview = $attachment["link"];
+ }
+ }
- $item["body"] = $item["title"].add_page_info($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]);
- $item["tag"] = add_page_keywords($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]);
+ // Remove a possible link to the item itself
+ $item["body"] = str_replace($item["plink"], '', $item["body"]);
+ $item["body"] = preg_replace('/\[url\=\](\w+.*?)\[\/url\]/i', '', $item["body"]);
+
+ // Replace the content when the title is longer than the body
+ $replace = (strlen($item["title"]) > strlen($item["body"]));
+
+ // Replace it, when there is an image in the body
+ if (strstr($item["body"], '[/img]')) {
+ $replace = true;
+ }
+
+ // Replace it, when there is a link in the body
+ if (strstr($item["body"], '[/url]')) {
+ $replace = true;
+ }
+
+ if ($replace) {
+ $item["body"] = $item["title"];
+ }
+ // We always strip the title since it will be added in the page information
$item["title"] = "";
+ $item["body"] = $item["body"].add_page_info($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]);
+ $item["tag"] = add_page_keywords($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]);
$item["object-type"] = ACTIVITY_OBJ_BOOKMARK;
unset($item["attach"]);
} else {
- $body = trim($xpath->evaluate('atom:content/text()', $entry)->item(0)->nodeValue);
+ if (!strstr($item["body"], '[url') && ($item['plink'] != '')) {
+ $item["body"] .= "[hr][url]".$item['plink']."[/url]";
+ }
+ }
- if ($body == "")
- $body = trim($xpath->evaluate('content:encoded/text()', $entry)->item(0)->nodeValue);
+ if (!$simulate) {
+ logger("Stored feed: ".print_r($item, true), LOGGER_DEBUG);
- if ($body == "")
- $body = trim($xpath->evaluate('description/text()', $entry)->item(0)->nodeValue);
+ $notify = item_is_remote_self($contact, $item);
- if ($body == "")
- $body = trim($xpath->evaluate('atom:summary/text()', $entry)->item(0)->nodeValue);
+ // Distributed items should have a well formatted URI.
+ // Additionally we have to avoid conflicts with identical URI between imported feeds and these items.
+ if ($notify) {
+ $item['guid'] = uri_to_guid($orig_plink, $a->get_hostname());
+ unset($item['uri']);
+ unset($item['parent-uri']);
+ }
- // remove the content of the title if it is identically to the body
- // This helps with auto generated titles e.g. from tumblr
- if (title_is_body($item["title"], $body))
- $item["title"] = "";
+ $id = item_store($item, false, $notify);
- $item["body"] = html2bbcode($body);
+ logger("Feed for contact ".$contact["url"]." stored under id ".$id);
+ } else {
+ $items[] = $item;
}
+ if ($simulate) {
+ break;
+ }
+ }
- logger("Stored feed: ".print_r($item, true), LOGGER_DEBUG);
-
- $notify = item_is_remote_self($contact, $item);
- $id = item_store($item, false, $notify);
-
- //print_r($item);
-
- logger("Feed for contact ".$contact["url"]." stored under id ".$id);
+ if ($simulate) {
+ return array("header" => $author, "items" => $items);
}
}
-?>