phpxpath-users

Retrieving a node list

[Eric A. <eri...@co...>]

All my XPath hacking has previously been with Perl using XML::XPath, but
I am working on a PHP project and I like using XPath retrieve config
info in little XML files. I came across this project and thought that it
looked pretty good, but I am having trouble figuring out how to do a
XPath query and get back a list of nodes. So for example if I have a XML
file as follows:

<root>
	<tag>item1</tag>
	<tag>item2</tag>
	<tag>item3</tag>
</root>

I want to be able to say something like:

$xPath = new XPath();
$xPath->importFromFile( $filename );
$list = $xPath->find( '/root/tag' );
foreach ($list as $item) {
	echo $item->getText() . '<br>';
}

and have it output:
item1
item2
item3

This is somewhat like the Perl syntax (mixed with a bit of your
libraries). But things that are missing are a method that takes a
XPathQuery and returns a list of nodes that match that query, and a
method on the node that will return the text inside the node.

I could probably achieve the second needed method by using the join
function on 'textParts' attribute of the Node, but that seems a bit
messy.

Basically I am wondering what is the best way of reading a list of nodes
that are returned from a XPathQuery. Everything in PHP.XPath seems to be
based around giving an absolute XPathQuery or a requiring that only one
node being returned by a query. How do you get a list of nodes that
match a path?

Thanks for your help and the great library. XPath greatly simplifies my
code, and I will be real happy if I can use it in this PHP project.

-- 
Eric Anderson

There are 10 types of people in the word
Those who can read binary, and those who can't

Re: Retrieving a node list

[Peter R. <php...@pe...>]

On Sunday 16 Mar 2003 02:33, Eric Anderson wrote:
>  I am having
> trouble figuring out how to do a XPath query and get back a list of
> nodes. So for example if I have a XML file as follows:
>
> <root>
> 	<tag>item1</tag>
> 	<tag>item2</tag>
> 	<tag>item3</tag>
> </root>
>
> I want to be able to say something like:
>
> $xPath = new XPath();
> $xPath->importFromFile( $filename );
> $list = $xPath->find( '/root/tag' );
> foreach ($list as $item) {
> 	echo $item->getText() . '<br>';
> }

replace 'find' with 'evaluate' (or 'match') and 'getText' with 
'getData' and that should do what you're looking for (though you 
might want to change your location path to e.g. //tag to fetch all 
the descendant nodes)

Re: Retrieving a node list

[Eric A. <eri...@co...>]

On Mon, 2003-03-17 at 06:15, Peter Robins wrote:
> replace 'find' with 'evaluate' (or 'match') and 'getText' with 
> 'getData' and that should do what you're looking for (though you 
> might want to change your location path to e.g. //tag to fetch all 
> the descendant nodes)

But according to the documentation and my experience with the library,
'evaluate' and 'match' will only resolve a XPathQuery if it resolves to
one node. If it resolves to a collection of nodes (or no nodes), then an
error is generated.

-- 
Eric Anderson

There are 10 types of people in the word
Those who can read binary, and those who can't

Re: Retrieving a node list

[Peter R. <php...@pe...>]

On Monday 17 Mar 2003 11:38, Eric Anderson wrote:
> On Mon, 2003-03-17 at 06:15, Peter Robins wrote:
> > replace 'find' with 'evaluate' (or 'match') and 'getText' with
> > 'getData' and that should do what you're looking for (though you
> > might want to change your location path to e.g. //tag to fetch
> > all the descendant nodes)
>
> But according to the documentation and my experience with the
> library, 'evaluate' and 'match' will only resolve a XPathQuery if
> it resolves to one node. If it resolves to a collection of nodes
> (or no nodes), then an error is generated.

not on the copy I have!
from the documentation for match:
"Matches (evaluates) an XPath query . . . 
Return Value: The result of the XPath expression. Either: node-set (an 
ordered collection of nodes without duplicates) . . ."

Though it doesn't actually say so, what it returns is an array of 
paths, which you can then use in your for loop (or whatever) to 
getData (or whatever). So in your example //tag will return
/root[1]/tag[1]
/root[1]/tag[2]
/root[1]/tag[3]