G'day vinoth.ree,

"perlrun - how to execute the Perl interpreter" has information on running Perl programs on different OSes which may be of interest.

I use Tkx myself, but the Tk toolkit is very nice once you wrap your head around it. I've heard good things about wxwidgits as well, but have not taken time to play with it yet.

It might well be easier to talk about what would not be operating-system independent.   All of the core features of the language are implemented in an OS-independent (or, transparently OS-specific) way.   Many (but not all ...) CPAN packages have OS-independence (and there exists a “CPAN Testing Service = CPANTS” which specifically tests them against a very broad gamut of architectures that a package developer might not have access to.)   There are also packages, such as File::Temp, which are specifically designed to let you write things that otherwise would be OS-dependent, in an OS-independent (more or less ...) way.

Problems will arise in your own applications when you write code “by hand,” e.g. using system() calls directly, and so on.   Or even in more obscure ways, as with case-sensitive vs. case-insensitive filenames.   There is a great deal of information already on the Web in a number of places, not just here.   For example, Google "portable perl code".

Finally, “the operating system type” is just one specific example of “enviornment dependency.”   You wouldn’t believe how many systems simply outgrow a small server-farm comprised of sneezy, wheezy, and doc, (or gandalf, frodo, and bilbo), and those server-names (and lots of other stuff that was hastily built with no regard for the future) are friggin’ everywhere . . .   It is best to be thinking of portability (and expansion) from the earliest possible moment in a project.

Just write it, sticking to the core functions, avoid "system()" lines as said, test it on different platforms and Operative systems and if you find troubles ask a specific question here.

There is a gazillion ways to write a "perl script"; without knowing what problem you want to solve is complicated to know the best way to write and fix it.

The hardest bit IMO isn't writing the perl code, it's making it executable on all the bizarro platforms that perl runs on. That is, to get the #! line right on Unix-a-likes, but make it a valid .BAT file on Windows, and whatever is needed on VMS.

Thankfully, there's a tool to do that.

ExtUtils::MakeMaker and, presumably, Module::Build, have the smarts to edit your scripts appropriately at install time so you don't have to worry about it. If for some reason you can't use either of those tools, then the maximally portable preamble for Unix-a-likes is:

#!/bin/sh
exec perl -x $0 "$@"
#!perl

# your perl code goes here
[download]

but I have no idea what to do, or even if you can, make that also portable to other platforms.

For Windows, there is a really simple .cmd file that I use to launch a plain Perl script with the same name:

@echo off
perl -w %~dpn0.pl
[download]

For a file test.cmd, this will launch perl -w test.pl in the same directory as test.cmd.

Is %~dpn0 some bizarro shell magic then?