in reply to Trouble in text manipulation

I did try to use the two versions of code in the OP, but neither of them produced any output at all for the sample data you provided. And I'm not about to figure out why -- the time would be better spent starting over with a better approach.

I think this bit from the OP is the "goal", but I'm not quite sure what you mean by this:

Am trying to club the <text entries together those having same y & page attribute values from the below input.

Do you mean: if there are two (or three, or more) "text" elements within one "font" element, and they all have the same attribute values for "y" and "page", you want them to be collapsed together into a single "text" element? If that's what you mean, that's an intriguing task, which I was able to solve using XML::LibXML. (Other monks could probably do it more neatly -- "I am just an egg" when it comes to DOM manipulation...)

In order to get that data to work with XML::LibXML (or any XML parser), I needed to add a "root" element around the set of "font" elements. Here's the code with the parsable version of the data attached: 

#!/usr/bin/perl

use strict;
use warnings;
use XML::LibXML;

my $xmlstring;
{
    local $/;
    $xmlstring = <DATA>;
}
my $xml = XML::LibXML->new;
my $doc = $xml->parse_string( $xmlstring );
   # you can do $xml->parse_file( "filename" ) instead

for my $font_node ( $doc->findnodes( "//font" )) {
    my %attr_val;
    my $matched_text_nodes = 1;
    my $matched_content;
    my @text_nodes = $font_node->findnodes( "./text" );
    for my $tnode ( @text_nodes ) {
        my @atts = $tnode->attributes;
        my $y_indx = grep { $atts[$_]->nodeName eq 'y' } 0 .. $#atts;
        my $p_indx = grep { $atts[$_]->nodeName eq 'page' } 0 .. $#att
+s;

        if ( ! keys %attr_val ) {  # first text_node
            $attr_val{y} = $atts[$y_indx]->textContent;
            $attr_val{p} = $atts[$p_indx]->textContent;
        }
        elsif ( $attr_val{y} ne $atts[$y_indx]->textContent or
                $attr_val{p} ne $atts[$p_indx]->textContent ) {
            $matched_text_nodes = 0;
        }
        if ( $matched_text_nodes ) {
            $matched_content .= $tnode->textContent . " ";
        }
    }
    if ( $matched_text_nodes ) {
        $text_nodes[0]->firstChild->setData( $matched_content );
        $font_node->removeChild( $_ ) for ( @text_nodes[1..$#text_node
+s] );
    }
}
print $doc->toString;

__DATA__
<doc>
<font size="12" face="IJCINN+AvantGarde-Bold" color="#1BADEB">
<text x="198" y="200" width="32" height="12" page="vii">Part I</text>
<text x="242" y="200" width="75" height="12" page="vii">Introduction</
+text>
<text x="329" y="200" width="7" height="12" page="vii">2</text>
</font>
<font size="9" face="IJCINN+AvantGarde-Bold" color="#231F20">
<text x="183" y="221" width="47" height="9" page="vii">Chapter 1</text
+>
</font>
<font size="10" face="IJCIOP+Frutiger-Light" color="#231F20">
<text x="242" y="220" width="121" height="10" page="vii">Managers and 
+Management</text>
<text x="373" y="220" width="6" height="10" page="vii">2</text>
</font>
<font size="9" face="IJCINN+AvantGarde-Bold" color="#231F20">
<text x="198" y="234" width="32" height="9" page="vii">History</text>
<text x="195" y="246" width="35" height="9" page="vii">Module</text>
</font>
<font size="12" face="IJCINN+AvantGarde-Bold" color="#1BADEB">
<text x="194" y="292" width="36" height="12" page="vii">Part II</text>
<text x="242" y="292" width="54" height="12" page="vii">Planning</text
+>
<text x="308" y="292" width="15" height="12" page="vii">56</text>
</font>
</doc>

[download]
(update: streamlined the logic for concatenating textContent values onto $matched_content)
When I ran that, it printed the following results to STDOUT -- but of course I don't know if this is what you really want... 

<?xml version="1.0"?>
<doc>
<font size="12" face="IJCINN+AvantGarde-Bold" color="#1BADEB">
<text x="198" y="200" width="32" height="12" page="vii">Part I Introdu
+ction 2 </text>


</font>
<font size="9" face="IJCINN+AvantGarde-Bold" color="#231F20">
<text x="183" y="221" width="47" height="9" page="vii">Chapter 1 </tex
+t>
</font>
<font size="10" face="IJCIOP+Frutiger-Light" color="#231F20">
<text x="242" y="220" width="121" height="10" page="vii">Managers and 
+Management 2 </text>

</font>
<font size="9" face="IJCINN+AvantGarde-Bold" color="#231F20">
<text x="198" y="234" width="32" height="9" page="vii">History</text>
<text x="195" y="246" width="35" height="9" page="vii">Module</text>
</font>
<font size="12" face="IJCINN+AvantGarde-Bold" color="#1BADEB">
<text x="194" y="292" width="36" height="12" page="vii">Part II Planni
+ng 56 </text>


</font>
</doc>

[download]

Replies are listed 'Best First'.
Re^2: Trouble in text manipulation
by thirilog (Acolyte) on Aug 06, 2010 at 11:00 UTC
    Great you are. Hats off Graff!
    You're just like that covered my requirement, am going to use your code straight away by adding a small computation. In additional I need to sum the x axis values of 'text" elements get collapsed.
    As am new for perl modules bit hard to understand your logic; will manipulate the code per my req.
    The hardest part is: Even I loaded the XML::LibXML module its showing error like cant locate the loadable object for module XML::LibXML...
    Can't locate loadable object for module XML::LibXML in @INC (@INC cont
+ains: C:/P
erl/lib C:/Perl/site/lib .) at C:/Perl/lib/DynaLoader.pm line 153
BEGIN failed--compilation aborted at C:/Perl/lib/XML/LibXML.pm line 15
+3.
Compilation failed in require at tocgen_v3.pl line 4.
BEGIN failed--compilation aborted at tocgen_v3.pl line 4.

    [download]
    Don't know what the problem is, keep on trying...
    And...regarding the necessity of root element in input XML; I have it already; I was missed this to provide.
    Many Thanks to All
[reply]
[d/l]
      Looks like a problem with how XML::LibXML was installed on your machine, maybe involving a dependency that is external to the module -- like maybe the GNU LibXML package is missing, or was installed in a "non-standard" path?

      GNU LibXML (a separate thing from cpan modules) needs to be installed first, and when you (re)install the cpan XML::LibXML modules, that installation process needs to know where to find the GNU LibXML stuff.

      It could be that your current XML::LibXML installation was done "by force", which would put all the XML/LibXML/*.pm files where they belong, even though there's no actual linkage to the GNU library (hence making the modules unusable). Good luck with (re)installing...

[reply]

In Section Seekers of Perl Wisdom