Shell Programming and Scripting

Good afternoon!

I have an XML file from which I want to extract only certain elements contained within each line. The problem is that the format of each line is not exactly the same (though similiar). For example, oa_var will be in each line, however, there may be no value or other variables present in some lines (but not others).

So, given the example lines below:

<ex1 oa_var="some_var_name1">val1</ex1>
<ex2 oa_var="some_var_name2" abc="dont_care">val2</ex2>
<ex3 oa_var="some_var_name3" />
<ex4 oa_var="some_var_name4">val4</ex4>

I want to only extract the bolded items (note ex3 doesn't have a value).

some_var_name1 val1
some_var_name2 val2
some_var_name3
some_var_name4 val4

I am open to using shell | awk | sed | perl | python (in that order).

Hi, see if this works:

Code:

awk -F '["<>]' '{print $3 (NF>6?OFS $(NF-2):x)}' infile

Bingo, man dingo!! Thanks for the help. It is greatly appreciated!!

It is safer to use an XML tool to extract the information. Just install perl-XML-XPath

Code:

#!/usr/bin/perl

use XML::XPath;

$filename = 'stdin';
$xpath = XML::XPath->new(ioref => \*STDIN);
$nodes = $xpath->find('//*[@oa_var]');
foreach my $node ($nodes->get_nodelist) {
        print $node->getAttribute('oa_var'), " ";
        $text=$node->getFirstChild();
        if ( $text ) {
                print $text->getData();
        }
        print "\n";
}