http://pear.php.net/bugs/search.php?cmd=display&package_name%5B0%5D=XML_Feed_Parser Search Results en-us pear-webmaster@lists.php.net pear-webmaster@lists.php.net hourly 1 2000-01-01T12:00+00:00 http://pear.php.net/gifs/pearsmall.gif http://pear.php.net/bugs http://pear.php.net/bugs/16416 XML_Feed_Parser Bug Reported by crimson 2009-07-09T04:59:12+00:00 PHP: 5.2.5 OS: Debian Lenny Package Version: 1.0.3 Description: ------------ The RSS 2.0 spec specifies that you should use <link> however some sites such as http://workbench.cadenhead.org/news/3284/adding-atomlink-your-rss-feed recommend adding <atom:link /> also. getElementsByTagName will return both these links when searching for 'link'. If the atom:link is first then $link->nodeValue is an empty string and this eventually calls combineBases passing $element->baseURI (which in my case ends up being the folder of the file) An example of the kind of feed that can trigger this is http://blog.mudrak.name/feed/?tag=mahara I have attached a diff of my change to fix this issue using xpath to only match <link> and not <atom:link> (or any other namespaced link attributes). Actual result: -------------- ben@tifa:Parser/$ git diff diff --git a/htdocs/lib/pear/XML/Feed/Parser/RSS2.php b/htdocs/lib/pear/XML/Feed/Parser/RSS2.php index c5d79d1..308a4ab 100644 --- a/htdocs/lib/pear/XML/Feed/Parser/RSS2.php +++ b/htdocs/lib/pear/XML/Feed/Parser/RSS2.php @@ -321,7 +321,8 @@ class XML_Feed_Parser_RSS2 extends XML_Feed_Parser_Type */ function getLink($offset, $attribute = 'href', $params = array()) { - $links = $this->model->getElementsByTagName('link'); + $xPath = new DOMXPath($this->model); + $links = $xPath->query('//link'); if ($links->length <= $offset) { return false; @@ -331,4 +332,4 @@ class XML_Feed_Parser_RSS2 extends XML_Feed_Parser_Type } } -?> \ No newline at end of file +?>]]> XML_Feed_Parser Bug Reported by crimson 2009-07-09T04:59:12+00:00 PHP: 5.2.5 OS: Debian Lenny Package Version: 1.0.3 Description: ------------ The RSS 2.0 spec specifies that you should use <link> however some sites such as http://workbench.cadenhead.org/news/3284/adding-atomlink-your-rss-feed recommend adding <atom:link /> also. getElementsByTagName will return both these links when searching for 'link'. If the atom:link is first then $link->nodeValue is an empty string and this eventually calls combineBases passing $element->baseURI (which in my case ends up being the folder of the file) An example of the kind of feed that can trigger this is http://blog.mudrak.name/feed/?tag=mahara I have attached a diff of my change to fix this issue using xpath to only match <link> and not <atom:link> (or any other namespaced link attributes). Actual result: -------------- ben@tifa:Parser/$ git diff diff --git a/htdocs/lib/pear/XML/Feed/Parser/RSS2.php b/htdocs/lib/pear/XML/Feed/Parser/RSS2.php index c5d79d1..308a4ab 100644 --- a/htdocs/lib/pear/XML/Feed/Parser/RSS2.php +++ b/htdocs/lib/pear/XML/Feed/Parser/RSS2.php @@ -321,7 +321,8 @@ class XML_Feed_Parser_RSS2 extends XML_Feed_Parser_Type */ function getLink($offset, $attribute = 'href', $params = array()) { - $links = $this->model->getElementsByTagName('link'); + $xPath = new DOMXPath($this->model); + $links = $xPath->query('//link'); if ($links->length <= $offset) { return false; @@ -331,4 +332,4 @@ class XML_Feed_Parser_RSS2 extends XML_Feed_Parser_Type } } -?> \ No newline at end of file +?>]]> 2009-07-09T04:59:12+00:00 ben &#x61;&#116; crimson &#x64;&#111;&#x74; net &#x64;&#111;&#x74; nz XML_Feed_Parser Bug http://pear.php.net/bugs/15366 XML_Feed_Parser Bug Reported by doconnor 2008-12-20T21:12:10+00:00 PHP: 5.2.6 OS: Package Version: CVS Description: ------------ This is: http://bugzilla.gnome.org/show_activity.cgi?id=565219 for validating against atom.rng]]> XML_Feed_Parser Bug Reported by doconnor 2008-12-20T21:12:10+00:00 PHP: 5.2.6 OS: Package Version: CVS Description: ------------ This is: http://bugzilla.gnome.org/show_activity.cgi?id=565219 for validating against atom.rng]]> 2008-12-21T00:12:44+00:00 daniel &#x64;&#111;&#x74; oconnor &#x61;&#116; gmail &#x64;&#111;&#x74; com XML_Feed_Parser Bug http://pear.php.net/bugs/12916 XML_Feed_Parser Bug Reported by sunfish 2008-01-16T15:19:56+00:00 PHP: 5.2.3 OS: Package Version: 1.0.2 Description: ------------ German umlauts (ÄÖÜäöüß) from an atom xml file (encoding="iso-8859-1") are displayed wrong after parsing, eg. http://www.keine-gentechnik.de/news-regionen.xml Test script: --------------- This solution works for me Type.php, Line 352 // remove htmlentities, because they prevent utf8_decode conversion [$content .= htmlentities($node->nodeValue);] $content .= $node->nodeValue; Atom.php, line 263 // add utf8_decode [return $content->nodeValue;] return utf8_decode($content->nodeValue);]]> XML_Feed_Parser Bug Reported by sunfish 2008-01-16T15:19:56+00:00 PHP: 5.2.3 OS: Package Version: 1.0.2 Description: ------------ German umlauts (ÄÖÜäöüß) from an atom xml file (encoding="iso-8859-1") are displayed wrong after parsing, eg. http://www.keine-gentechnik.de/news-regionen.xml Test script: --------------- This solution works for me Type.php, Line 352 // remove htmlentities, because they prevent utf8_decode conversion [$content .= htmlentities($node->nodeValue);] $content .= $node->nodeValue; Atom.php, line 263 // add utf8_decode [return $content->nodeValue;] return utf8_decode($content->nodeValue);]]> 2009-05-23T18:59:34+00:00 ulrich-fischer &#x61;&#116; gmx &#x64;&#111;&#x74; net XML_Feed_Parser Bug http://pear.php.net/bugs/10335 XML_Feed_Parser Feature/Change Request Reported by ezka1209a 2007-03-12T02:13:07+00:00 PHP: 5.1.6 OS: FreeBSD6.1 Package Version: 1.0.1 Description: ------------ Now I'm trying to get nodeValue within a certain namespace.But currently,this package doesn't support this feature.Do you have any plan to support namespaces? For example,here is the xml file.and I want to get element like this,"$element->{ex:rank}". <?xml version="1.0" encoding="utf-8" ?> <rss version="2.0" xmlns:ex="http://www.example.com/xml/xmlns# "> <channel> <title>Sample</title> <link>http://www.example.com/</link> <description>This is sample xml file.</description> <item> <title>Foo</title> <link>http://www.example.com/bar.html</link> <ex:rank>1</ex:rank> </item> (snip)... </channel> </rss> Now I get like this: $namespace = 'http://www.example.com/xml/xmlns#'; $feed = new XML_Feed_Parser($xml_source); foreach ($feed as $entry) { echo $ranks = $entry->model->getElementsByTagNameNS($namespace, 'rank')->item(0)->nodeValue; } But this way is too long for me.And `echo $e->{ex:rank};' would be intuitively.I wrote the patch to support this notation.Can others agree with my idea?If you have better idea,I want to know. == patch == diff -u Parser.org/RSS2.php Parser/RSS2.php --- Parser.org/RSS2.php Fri Mar 2 14:47:40 2007 +++ Parser/RSS2.php Fri Mar 2 14:47:57 2007 @@ -101,7 +101,7 @@ 'date' => array('pubDate'), 'author' => array('managingEditor')); - protected $namespaces = array( + public $namespaces = array( 'dc' => 'http://purl.org/rss/1.0/modules/dc/', 'content' => 'http://purl.org/rss/1.0/modules/content/' diff -u Parser.org/Type.php Parser/Type.php --- Parser.org/Type.php Wed Feb 28 14:09:01 2007 +++ Parser/Type.php Fri Mar 2 15:03:51 2007 @@ -84,6 +84,10 @@ } } } + if( strstr($call,":") ){ + list($ns_prefix,$method) = explode(":",$call); + return $this->getTextNS($ns_prefix,$method,$arguments); + } if (empty($this->map[$call])) { return false; @@ -241,6 +245,17 @@ } return false; } + protected function getTextNS($ns_prefix,$method, $arguments = array()) + { + $namespace = $this->parent->namespaces[$ns_prefix]; + + $tags = $this->model->getElementsByTagNameNS($namespace,$method); + if ($tags->length > 0) { + $value = $tags->item(0)->nodeValue; + return $value; + } + return false; + } /** * Apply various rules to retrieve category data. --- Parser.php.org Fri Mar 2 15:00:32 2007 +++ Parser.php Fri Mar 2 15:02:18 2007 @@ -181,6 +181,14 @@ /* Instantiate feed object */ $this->feed = new $class($this->model, $strict); + /* Find third party's namespaces */ + if( preg_match_all("|xmlns:(.+)=['\"]([^>]+)['\"]|i",$feed,$matches) ){ + foreach( $matches[1] as $key=>$prefix ){ + if( empty($this->feed->namespaces[$prefix]) ){ + $this->feed->namespaces[$prefix] = $matches[2][$key]; + } + } + } } /** Test script: --------------- $namespace = 'http://www.example.com/xml/xmlns#'; $feed = new XML_Feed_Parser($xml_source); foreach ($feed as $entry) { echo $entry->{"ex:rank"}; } Expected result: ---------------- 1 Actual result: -------------- cannot get any value.]]> XML_Feed_Parser Feature/Change Request Reported by ezka1209a 2007-03-12T02:13:07+00:00 PHP: 5.1.6 OS: FreeBSD6.1 Package Version: 1.0.1 Description: ------------ Now I'm trying to get nodeValue within a certain namespace.But currently,this package doesn't support this feature.Do you have any plan to support namespaces? For example,here is the xml file.and I want to get element like this,"$element->{ex:rank}". <?xml version="1.0" encoding="utf-8" ?> <rss version="2.0" xmlns:ex="http://www.example.com/xml/xmlns# "> <channel> <title>Sample</title> <link>http://www.example.com/</link> <description>This is sample xml file.</description> <item> <title>Foo</title> <link>http://www.example.com/bar.html</link> <ex:rank>1</ex:rank> </item> (snip)... </channel> </rss> Now I get like this: $namespace = 'http://www.example.com/xml/xmlns#'; $feed = new XML_Feed_Parser($xml_source); foreach ($feed as $entry) { echo $ranks = $entry->model->getElementsByTagNameNS($namespace, 'rank')->item(0)->nodeValue; } But this way is too long for me.And `echo $e->{ex:rank};' would be intuitively.I wrote the patch to support this notation.Can others agree with my idea?If you have better idea,I want to know. == patch == diff -u Parser.org/RSS2.php Parser/RSS2.php --- Parser.org/RSS2.php Fri Mar 2 14:47:40 2007 +++ Parser/RSS2.php Fri Mar 2 14:47:57 2007 @@ -101,7 +101,7 @@ 'date' => array('pubDate'), 'author' => array('managingEditor')); - protected $namespaces = array( + public $namespaces = array( 'dc' => 'http://purl.org/rss/1.0/modules/dc/', 'content' => 'http://purl.org/rss/1.0/modules/content/' diff -u Parser.org/Type.php Parser/Type.php --- Parser.org/Type.php Wed Feb 28 14:09:01 2007 +++ Parser/Type.php Fri Mar 2 15:03:51 2007 @@ -84,6 +84,10 @@ } } } + if( strstr($call,":") ){ + list($ns_prefix,$method) = explode(":",$call); + return $this->getTextNS($ns_prefix,$method,$arguments); + } if (empty($this->map[$call])) { return false; @@ -241,6 +245,17 @@ } return false; } + protected function getTextNS($ns_prefix,$method, $arguments = array()) + { + $namespace = $this->parent->namespaces[$ns_prefix]; + + $tags = $this->model->getElementsByTagNameNS($namespace,$method); + if ($tags->length > 0) { + $value = $tags->item(0)->nodeValue; + return $value; + } + return false; + } /** * Apply various rules to retrieve category data. --- Parser.php.org Fri Mar 2 15:00:32 2007 +++ Parser.php Fri Mar 2 15:02:18 2007 @@ -181,6 +181,14 @@ /* Instantiate feed object */ $this->feed = new $class($this->model, $strict); + /* Find third party's namespaces */ + if( preg_match_all("|xmlns:(.+)=['\"]([^>]+)['\"]|i",$feed,$matches) ){ + foreach( $matches[1] as $key=>$prefix ){ + if( empty($this->feed->namespaces[$prefix]) ){ + $this->feed->namespaces[$prefix] = $matches[2][$key]; + } + } + } } /** Test script: --------------- $namespace = 'http://www.example.com/xml/xmlns#'; $feed = new XML_Feed_Parser($xml_source); foreach ($feed as $entry) { echo $entry->{"ex:rank"}; } Expected result: ---------------- 1 Actual result: -------------- cannot get any value.]]> 2007-11-06T09:44:35+00:00 ezka1209 &#x61;&#116; gmail &#x64;&#111;&#x74; com XML_Feed_Parser Feature/Change Request