The use of Inline::Files and \*STDOUT was simply for demo purposes; hence the "I'll leave you to replace Inline::Files and \*STDOUT with more appropriate I/O.".

I'm not sure whether you've got a handle on this. If you have, you probably won't need the following; if not, here's some hints and tips that may prove useful for your real code. Ask if you need more information, further explanation, etc.

I'd recommend the autodie pragma. This will save a lot of effort checking whether files could be opened, created, read, written and so on. Put it near the top of your code; I usually write this:

...
use strict;
use warnings;
use autodie;
...
[download]

Remove the 'use Inline::Files; line.

Use open to create your filehandles.

The output filehandle ($out_fh) is very straightforward:

open my $out_fh, '>', '/path/to/output_file';
[download]

The input filehandles are a little more complicated but still fairly easy. I wouldn't recommend creating all the filehandles in advance; instead, open and close one at a time. The \*FILE0, \*FILE1, etc. in my original code are filehandles for the Inline::Files: you'll see I didn't need to explicitly create those. Perhaps change:

my %doc_fh_for = (File0 => \*FILE0, File1 => \*FILE1, File2 => \*FILE2
+);
[download]

to something like this

my %source_file_for = (File0 => 'filename0', File1 => 'filename1', Fil
+e2 => 'filename2');
[download]

and then further down

for (@feature_files) {
    open my $in_fh, '<', $source_file_for{$_};
    write_xml_content($_, $in_fh, $out_fh, ' ' x 8);
    close $in_fh;
}
[download]

I don't know whether you have a predetermined list of source files or if you have to find them in one or more directories; if the latter, you may find readdir useful for this. Also note that may need to prefix the filenames with directory paths: while simple concatenation may suffice, I'd normally use File::Spec as it's portable (it's also a core module so there's no need to install it).

Thanks Ken. I can see some resutls now however the following issues exists:

(1)errors:
readline<> on unopened filehandle at line33
which is : while (<$in_fh>)

(2)my FILE0 is being skipped and only FILE1 and FILE2 are being processed

(3)Could you please give me quick explanation of whats going on within the subroutine write_xml_contents so that I can tweak my output as I desire.
here is my code:


#!/strawberry/perl/bin/perl

#use strict;
use warnings;
use autodie;

open my $out_fh, '>', 'out_data.txt';

my %source_file_for = (File0 => 'FILE0.xml', File1 => 'FILE1.xml', Fil
+e2 => 'FILE2.xml');

#my $out_fh = \*STDOUT;
my $key_file = 'File0';
my @feature_files = qw{File1 File2};

print $out_fh '<Key="1234">';
write_xml_content($key_file, $source_file_for{$key_file}, $out_fh, ' '
+ x 4);

print $out_fh "\n".'    <Status="in use"/>
    <Features>';

    for (@feature_files) {
    open my $in_fh, '<', $source_file_for{$_};
    write_xml_content($_, $in_fh, $out_fh, ' ' x 8);
    close $in_fh;
}
print $out_fh "\n".'    </Features>
    <Other_Status/>
</Key>';

sub write_xml_content {
    my ($file_id, $in_fh, $out_fh, $indent) = @_;
    
    while (<$in_fh>) {
        chomp;

        if (/^<\/?doc>$/) {
            s/doc/$file_id/;
        }
        else {
            if (/^<ABC id="(\d+)">$/) {
                my $id = $1;
                s/$id/$file_id-$id/;
            }
            $_ = ' ' x 4 . $_;
        } 

        print $out_fh $indent, $_;
    }
}
[download]

All issues stem from the same cause; namely, the change you made in the for loop needs to be also done for your File0. Change:

write_xml_content($key_file, $source_file_for{$key_file}, $out_fh, ' '
+ x 4);
[download]

to

open my $in_fh, '<', $source_file_for{$key_file};
write_xml_content($key_file, $in_fh, $out_fh, ' ' x 4);
close $in_fh;
[download]

-- Ken