XPath and nodelist .. im confused :p

[Øyvind Jægtnes]

I'm playing around a bit with XPath and nodelist and i want to extract
some info from a RSS feed.
The one that i am testing at can be viewed at http://slashdot.org/index.rss

Ok.. heres the deal:
I load xml via httpwebrequest etc and put it in a XmlDocument.

xmlFeed.Load(xRes.GetResponseStream());

This works fine as i can use this stream with a XmlReader to parse the
document in an ugly way.

Now i try to get all the title elemelts:

nodeList = xmlFeed.DocumentElement.SelectNodes("RDF/item/title/text()");

This should (according to my browsing of xpath and thing and reading for
a few hrs) result in getting the titles ;)
But it don't.
System.Console.WriteLine("Nodes: " + nodeList.Count + "\n");
Outputs: "Nodes: 0"

So what part of my brain isnt working? :)
Anyone spot something horribly wrong?

---
Øyvind Jægtnes

[Iain A. Mcleod]

I'm no expert on xpath, but I think you need:
nodeList = xmlFeed.DocumentElement.SelectNodes("/RDF/item/title/text()");

The RDF node must be prefixed with a forward slash to show it is an absolute
path from the root of the document...

Try it and see?
"Øyvind Jægtnes" <oj****@start.no> wrote in message
news:0z********************@news4.e.nsc.no...

I'm playing around a bit with XPath and nodelist and i want to extract
some info from a RSS feed.
The one that i am testing at can be viewed at http://slashdot.org/index.rss
Ok.. heres the deal:
I load xml via httpwebrequest etc and put it in a XmlDocument.

xmlFeed.Load(xRes.GetResponseStream());

This works fine as i can use this stream with a XmlReader to parse the
document in an ugly way.

Now i try to get all the title elemelts:

nodeList = xmlFeed.DocumentElement.SelectNodes("RDF/item/title/text()");

This should (according to my browsing of xpath and thing and reading for
a few hrs) result in getting the titles ;)
But it don't.
System.Console.WriteLine("Nodes: " + nodeList.Count + "\n");
Outputs: "Nodes: 0"

So what part of my brain isnt working? :)
Anyone spot something horribly wrong?

---
Øyvind Jægtnes

[Øyvind Jægtnes]

Iain A. Mcleod wrote:

I'm no expert on xpath, but I think you need:
nodeList = xmlFeed.DocumentElement.SelectNodes("/RDF/item/title/text()");

The RDF node must be prefixed with a forward slash to show it is an absolute
path from the root of the document...

Try it and see?

Still same.
But when i debug and look at the XmlElement of <rdf:RDF> it sais:
LocalName is RDF and Name is rdf:RDF.

I then made this XPath string: "/rdf:RDF/item/title/text()"

it trows exception on that :p

System.Xml.XPath.XPathException: Namespace Manager or XsltContext
needed. This query has a prefix, variable, or user-defined function.

Which i cant really decode ;)
Do i need to use Namespace Manager to get a proper query?

---
Øyvind Jægtnes

[Øyvind Jægtnes]

Øyvind Jægtnes wrote:

Iain A. Mcleod wrote:

I'm no expert on xpath, but I think you need:
nodeList = xmlFeed.DocumentElement.SelectNodes("/RDF/item/title/text()");

The RDF node must be prefixed with a forward slash to show it is an
absolute
path from the root of the document...

Try it and see?

Still same.
But when i debug and look at the XmlElement of <rdf:RDF> it sais:
LocalName is RDF and Name is rdf:RDF.

I then made this XPath string: "/rdf:RDF/item/title/text()"

it trows exception on that :p

Actually i've been poking around a bit and i think it has to do with
rdf:RDF is its name, but SelectNodes think that it is a prefix for a
namespace when i use it. Question then is, is it posible to "escape" the
: in someway in the query?

---
Øyvind Jægtnes

[Iain Mcleod]

Sorry man, way over my head!

Hope someone else can help you :)

"Øyvind Jægtnes" <oj****@start.no> wrote in message
news:H3********************@news2.e.nsc.no...

Øyvind Jægtnes wrote:

Iain A. Mcleod wrote:

I'm no expert on xpath, but I think you need:
nodeList = xmlFeed.DocumentElement.SelectNodes("/RDF/item/title/text()");
The RDF node must be prefixed with a forward slash to show it is an
absolute
path from the root of the document...

Try it and see?

Still same.
But when i debug and look at the XmlElement of <rdf:RDF> it sais:
LocalName is RDF and Name is rdf:RDF.

I then made this XPath string: "/rdf:RDF/item/title/text()"

it trows exception on that :p

Actually i've been poking around a bit and i think it has to do with
rdf:RDF is its name, but SelectNodes think that it is a prefix for a
namespace when i use it. Question then is, is it posible to "escape" the
: in someway in the query?

---
Øyvind Jægtnes

[Oleg Tkachenko [MVP]]

Øyvind Jægtnes wrote:

I'm playing around a bit with XPath and nodelist and i want to extract
some info from a RSS feed.
The one that i am testing at can be viewed at http://slashdot.org/index.rss

Ok.. heres the deal:
I load xml via httpwebrequest etc and put it in a XmlDocument.

xmlFeed.Load(xRes.GetResponseStream());

This works fine as i can use this stream with a XmlReader to parse the
document in an ugly way.

Now i try to get all the title elemelts:

nodeList = xmlFeed.DocumentElement.SelectNodes("RDF/item/title/text()");

This should (according to my browsing of xpath and thing and reading for
a few hrs) result in getting the titles ;)

Nope, you forgot about namespaces. It should be something like

XmlNamespaceManager nm = new XmlNamespaceManager(xmlFeed.NameTable);
nm.AddNamespace("rdf", "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
nm.AddNamespace("rss", "http://purl.org/rss/1.0/");
nodeList =
xmlFeed.DocumentElement.SelectNodes("rdf:RDF//rss:item/title/text()");

Read "XML Namespaces and How They Affect XPath and XSLT" at
http://msdn.microsoft.com/library/en...ml05202002.asp

--
Oleg Tkachenko [XML MVP]
http://blog.tkachenko.com

[Øyvind Jægtnes]

Oleg Tkachenko [MVP] wrote:

Øyvind Jægtnes wrote:

Now i try to get all the title elemelts:

nodeList = xmlFeed.DocumentElement.SelectNodes("RDF/item/title/text()");

Nope, you forgot about namespaces. It should be something like

XmlNamespaceManager nm = new XmlNamespaceManager(xmlFeed.NameTable);
nm.AddNamespace("rdf", "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
nm.AddNamespace("rss", "http://purl.org/rss/1.0/");
nodeList =
xmlFeed.DocumentElement.SelectNodes("rdf:RDF//rss:item/title/text()");

Hmm this actually throws exception ;)
But with:
nodeList =
xmlFeed.DocumentElement.SelectNodes("/rdf:RDF//rss:item/title/text()", nm);

nodeList.Count is still 0.
When i use "//*" as XPath line it takes the whole document as it should.
But if i do "/*/*/title/text()", or "//*//*/title" or any variations of
that i still get 0 count...
So im really confused ;)
Starting to wander if the best way to use XML is to iterate the nodes
and do the searching manually.. but XPath looks cool if it would work :p
Read "XML Namespaces and How They Affect XPath and XSLT" at
http://msdn.microsoft.com/library/en...ml05202002.asp

Will do..

---
Øyvind Jægtnes

[Øyvind Jægtnes]

Oleg Tkachenko [MVP] wrote:

Read "XML Namespaces and How They Affect XPath and XSLT" at
http://msdn.microsoft.com/library/en...ml05202002.asp

Ok, solved ;)
I finally got them extrated.. Now its time to see if i can actually use
the namespaces to check the document version (rss 0.9, 1.0, 2.0). That
would be nice :)
(correct xpath was "/rdf:RDF//rss:item/rss:title/text()")

---

[Øyvind Jægtnes]

Øyvind Jægtnes wrote:

I'm playing around a bit with XPath and nodelist and i want to extract
some info from a RSS feed.
The one that i am testing at can be viewed at http://slashdot.org/index.rss

Ah finally i have something semi working here ;)
While project can be seen at http://home.no.net/~ojaegt/Rss.rar if
someone are interested.

Still todo: make it so you can save a list of feeds and select between
them (GUI) and add support for ATOM feeds (library).

---
Øyvind Jægtnes