Re^5: How to parse xml with namespase vale in XMl:LibXML? (xmllint --shell setns / xpathtester)

[ravi06]

Hi Choroba,

Thanks a lot and sorry for being stupid. Actually, my objective was to get values of "xlink:href" and "xlink:title" for first occurrence of "book:locator" in Xpath context.

Any hint or tweak on that?

[choroba]

To use namespaces in the full depth, use XML::LibXML::XPathContext:

#!/usr/bin/perl
use warnings;
use strict;

use XML::LibXML;

my $xpc  = XML::LibXML::XPathContext->new();
my $xml  = XML::LibXML->load_xml(IO => *DATA);
my $root = $xml->documentElement;
$xpc->registerNs('xlink', 'http://www.w3.org/1999/xlink');
my $chapter_node = $xpc->findnodes('book:chapter', $root)->[0];
print $xpc->findvalue('book:locator/@xlink:title | book:locator/@xlink
+:href', $chapter_node) . "\n";

[Anonymous Monk]

Probably ought to store the context in the ROOT, oh well

#!/usr/bin/perl --
use strict;
use warnings;
use XML::LibXML;
use Data::Dump qw/ dd pp /;

Main( @ARGV );
exit( 0 );

sub ROOTNS {
    my( $self ) = @_;
    $::xpcroot = ROOT( $self );
    $::xpc     = XML::LibXML::XPathContext->new( );
    for my $node ( $::xpcroot->F('//*') ){
        for my $att(  $node->attributes() ){
            if( $att->isa( "XML::LibXML::Namespace" ) ){
                my $suffix = $att->localname;
                my $value = $att->value;
                print '# autoRegisterNs( ', pp( $suffix ), ' => ', pp(
+ $value ) , " );\n";
                $::xpc->registerNs( $suffix => $value );
            }
        }
    }
}
sub ROOT {
    my( $self ) = @_;
    my $parent = $self;
    while( my $newparent = $parent->getParentNode ){
        $parent = $newparent;
    }
    return $parent;
}
BEGIN {
    $::xpc = XML::LibXML::XPathContext->new();
    sub XML::LibXML::Node::F {
        my( $self, $xpath, $context ) = @_;
        unless( $::xpcroot and $::xpcroot == ROOT( $self ) ){
            ROOTNS( $self );
        }
        $::xpc->findnodes( $xpath, $context || $self );
    }
}

sub Main {
    my $dom = XML::LibXML->new(
        qw/ recover 2 /
    )->load_xml(
        string => q{<?xml version="1.0" ?>
<book xmlns:book="http://generated.ns/book" xmlns:xlink="http://www.w3
+.org/1999/xlink">
  <book:chapter id="bk444444ch1" type="CHAPTER">
    <book:locator xlink:href="/book/isbn/979-0-4444-1000-17/book-part/
+chapter/bk444444ch1?releaseStatus=RELEASED" xlink:title="Photonic cry
+stal light-emitting sources" xlink:type="locator"></book:locator>
    <book:locator xlink:href="/book/isbn/979-0-4444-1000-17/book-part/
+chapter/bk444444ch1?releaseStatus=RELEASED&amp;format=pdf" xlink:titl
+e="Photonic crystal light-emitting sources" xlink:type="locator"></bo
+ok:locator>
    <book:locator xlink:href="/book/isbn/979-0-4444-1000-17/book-part/
+chapter/bk444444ch1?releaseStatus=RELEASED&amp;format=epub" xlink:tit
+le="Photonic crystal light-emitting sources" xlink:type="locator"></b
+ook:locator>
  </book:chapter>
</book>},
    );
    print $_->nodePath,"\n" for $dom->F(q{//book:locator/@xlink:href})
+;
}

__END__
# autoRegisterNs( "book" => "http://generated.ns/book" );
# autoRegisterNs( "xlink" => "http://www.w3.org/1999/xlink" );
/book/book:chapter/book:locator[1]/@xlink:href
/book/book:chapter/book:locator[2]/@xlink:href
/book/book:chapter/book:locator[3]/@xlink:href

[Anonymous Monk]

A twist in the smae context. what could be the tweak if we want to caputue values of "xlink:href" and "xlink:title" if "xlink:href" ends with "releaseStatus=RELEASED"?

[Anonymous Monk]

FWIW, this autoRegisterNs is completely unneccessary with the xml included in the program, but if you remove   xmlns:book="http://generated.ns/book" xmlns:xlink="http://www.w3.org/1999/xlink" then it becomes neccessary