As a quick remedy until someone who knows better can help you, try opening the output file before the loop, and then APPEND the structure to the file inside the loop. I believe from a quick glance at XML::Dumper that pl2xml($perl, $filename) does not append but overwrites whatever the current contents of file with $perl. If you omit the $filename, you will get the xml string back, e.g. my $xmlstr = pl2xml($perl); (untested). So, open the output file before the loop: open($fh, '>:encoding(UTF-8)', $filename) or die "file, $!"; Then inside the loop, after decode(): $perl = $tap3->structure; my $xmlstr = $dump->pl2xml($perl); print $fh $xmlstr; OR simpler: print $fh $dump->pl2xml($tap3->structure); And after the loop close $fh; But perhaps you want a new $tap3 object to be created inside the loop each iteration?

Is it possible to share the edited code with me,I didn't get unfortunately. yes, I need tap3. This code perfectly goes fine for one file, I just want to give a batch of similar code and input and get the outputs.

my $filename="Taps/Output.xml";
my $fh;
open($fh, '>:encoding(UTF-8)', $filename) or die "file, $!";

   foreach my $file(glob 'Taps/DA*')
   {
   my $tap3 = TAP3::Tap3edit->new();
   $files= basename($file); #<< are you sure this is needed/does it wo
+rk?
   $tap3->decode($files) || die $tap3->error;
   print $fh $dump->pl2xml($tap3->structure);  
    }
close $fh;
[download]

WARNING: This code reads just ONE tap file, decodes it and dumps it. It does that for all TAP files. Separately. I am not sure you will get correct output, because I am not sure you want to read and decode all files first or decode and dump each separately, as above code does.

What you really need is someone with experience of the specific Perl modules, which I do not have.

I did read the source of Tap3::Tap3edit though, and it looks to me like the decode() method replaces whatever was there before.

Of course, this means that at the end of the loop it would be the last file left in the $tap3 object, not the first. So something else must also be going on.

I also reiterate a previous poster's question of the use of basename(). As your code is written, the glob() finds all files named X* in directory Taps/ under your current directory. But by feeding decode() the basename of the file you find you are actually attempting to read a file in your current directory, not in Taps/. Now, you are checking the status of the operation, and though in general your check is subtly wrong (your habit should be to use ... or die ... rather than ... || die ...), in this specific case I do not think that the tighter binding of || versus or invalidates the test.

Given all this, I believe that if I were in your position I would dump the contents of $file, $files, and any relevant information about the parse immediately after the call to decode() and see if it was what I expected. Maybe initially put die 'Debugging'; after the dump so that you are only dealing with one mass of information at a time, instead of a whole directories' worth.

Irrelevantly to your question, I would like to note that for a self-confessed newbie you are doing basic things right. Using use strict; and use warnings; will spare you grief now and in the future.