ansaurus

Question

Recursively reading an xml document and using regex to get contents

Answer 1

+4 A:

A basic recursion (not tested but I think it's ok):

private void Caller(String filepath)
{
    XPathDocument oDoc = new XPathDocument(filepath);
    Readnodes( oDoc.CreateNavigator() );
}

private void ReadNodes(XPathNavigator nav)
{
    XPathNodeIterator nodes = nav.Select("menuitem");
    while (nodes.MoveNext())
    {
     //A - read the attribute
     string url = nodes.Current.GetAttribute("navigateurl", string.Empty);

        //B - do something with the data

     //C - recurse
     ReadNodes(nodes.Current);
    }
}

...works because an XPathNodeIterator's Current property is also an XPathNavigator. Obviously you'd need to extend this to push data to a dictionary or keep track of depth or whatever.

annakata 2009-05-05 15:48:29

Doh! You beat me to it, and with an example, too.

ZombieSheep 2009-05-05 15:49:52

heh - but now I'm utterly baffled as to what the question *is* anymore :)

annakata 2009-05-05 15:51:12

Thanks! I'll give that a try. There's some new classes for me to learn :)

dotnetdev 2009-05-05 15:55:53

Also of note is the fact that you can create XPath expressions that handle the recursion, so you don't need to make the function recurse.For a quick example: http://msdn.microsoft.com/en-us/library/ms256086.aspx

ee 2009-05-05 16:07:22

i.e. nav.Select("//menuitem") should get all menu items recursively

ee 2009-05-05 16:11:12

Yeah, I really should have mentioned that - really good call, and likely an even simpler, better answer in practice - but the reason I've provided this as is is to demonstrate recursion. Also worth noting that "//menuitem" won't give you the option of additional logic i.e. around depth checks.

annakata 2009-05-05 16:34:55

Answer 2

+1 A:

Why use Regex for this when XPath is (to me, at least) the natural choice? That's basically what XSLT should implement...

ZombieSheep 2009-05-05 15:49:21

Simply because regex I don't have much experience of.

dotnetdev 2009-05-05 16:24:23

Answer 3

A:

Any particular reason you're using a regex? Have you tried using XPath for this? Here are some examples of how to use XPath. http://www.w3schools.com/XPath/xpath_examples.asp

Conrad 2009-05-05 15:51:48

Answer 4

A:

Use xpath, //menuitem[@navigateurl]/@navigateurl .

This xpath will grab all the menu items which have an attribute naviagate url and return a node-list (xpath 1.0) or sequence (xpath 2.0) of navigateurl values. By having the navigateurl attribute predicate, that ensures that only the leaf menu items are fetched.

JavaRocky 2009-05-06 12:50:00

Your xpath is wrong for leaves actually - "//menuitem[not(menuitem)]" would capture all the leaves only (not that this is what the OP requests). I still like this as a solution generally, but the OP's position on recursion has not been clarified.

annakata 2009-05-06 13:21:48

ansaurus

tags:

views:

answers:

Recursively reading an xml document and using regex to get contents

related questions