LibXML error when trying to delete a node in DOM

[msivask]

Hi ,

Below is my requirement

I have an XML like

 <sys xmlns="urn:mavenir:ns:yang:wrg-sc:6.0.2.8">
<platform><contact>
    <name>Mavenir</name>
    <phone>214-616-3499</phone>
  </contact>
  <solution>
    <name>mOne</name>
    <fkcontact>Mavenir</fkcontact>
  </solution>
                <tmmTableFilenameMap>
                                <fileName>ALARM</fileName>
                                <tableId>1</tableId>
                        </tmmTableFilenameMap>
                <tmmTableFilenameMap>
                                <fileName>AVGOVLDPERF</fileName>
                                <tableId>7</tableId>
                        </tmmTableFilenameMap>
</platform>
</sys>

I want to delete all nodes with tmmTableFilenameMap

I tried the below code

#!/usr/bin/perl

use 5.010;
use strict;
use warnings;

use XML::LibXML;

my $filename = 'webrtcc_system_static_config_data_platremoved.xml';
my $a;

my $dom = XML::LibXML->load_xml(location => $filename);
my $book = $dom->documentElement;
my $del="tmmTableFilenameMap";

my @a = $book->findnodes("tmmTableFilenameMap");
$book->removeChild($a);
print $book->serialize();

It gives a strange error as below

XML::LibXML::Node::removeChild() -- node is not a blessed SV reference
+ at DeleteNodesXml.pl line 17.

Could some one help on this

2019-04-26 Athanasius added code and paragraph tags

[hippo]

$book->removeChild($a);

It's a bit hard to tell because you haven't used <code> tags around your code, but I don't see where you are assigning any value to $a so it is probably undef when you try to use it as an argument to removeChild().

Also, please don't use $a as a variable name - it has a special meaning for sort and can bite you if you are not careful.

[bliako]

after the ergonomic intervention of monk Athanasius, I can also see the use of $a instead of $a[0].

[choroba]

Your XML document has a XML namespace defined. Use XML::LibXML::XPathContext to handle namespaces in XML::LibXML.

Also, the correct XPath expression to find tmmTableFilenameMap anywhere in the document is //tmmTableFilenameMap, otherwise the searching only happens at the current node.

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

use XML::LibXML;

my $filename = shift;

my $xpc = 'XML::LibXML::XPathContext'->new;
$xpc->registerNs(m => 'urn:mavenir:ns:yang:wrg-sc:6.0.2.8');

my $dom = 'XML::LibXML'->load_xml(location => $filename);

for my $delete ($xpc->findnodes('//m:tmmTableFilenameMap', $dom)) {
    my ($parent) = $delete->findnodes('..');
    $parent->removeChild($delete);
}

print $dom->serialize;

It gets even shorter using XML::XSH2:

open webrtcc_system_static_config_data_platremoved.xml ;
register-namespace m urn:mavenir:ns:yang:wrg-sc:6.0.2.8 ;
delete //m:tmmTableFilenameMap ; 
save :b ;

[Jenda]

use strict;
use XML::Rules;

my $rules = XML::Rules->new(
    style => 'filter', 
    rules => {tmmTableFilenameMap => sub {}},
    namespaces => {'urn:mavenir:ns:yang:wrg-sc:6.0.2.8' => ''}
);

$rules->filter(\*DATA);

__DATA__
<sys xmlns="urn:mavenir:ns:yang:wrg-sc:6.0.2.8">
...

This doesn't load the whole file into memory which may be a big advantage.