in reply to Re^5: layout/configuration of deployed files
in thread layout/configuration of deployed files

I'm trying to envision a way to do that that's less verbose than putting the boilerplate into each script to begin with. :-)

For example, let's say each script has its own wrapper... That is, "script1" is really just a wrapper that sets up PERL5LIB and then invokes "script1.real". In that case, you've got 2x the number of scripts (one wrapper, one real) and each wrapper effectively has the boilerplate in it.

Even with a virtualized wrapper (i.e. one that you symlink to, that decides what to exec on the basis of $0), you've got N scripts and N symlinks.

Is this what you meant, or did I misunderstand?

  • Comment on Re^6: layout/configuration of deployed files

Replies are listed 'Best First'.
Re^7: layout/configuration of deployed files
by Eliya (Vicar) on Mar 02, 2011 at 22:36 UTC

    Personally, I'd prefer having one additional wrapper plus a bunch of symlinks — in particular as it helps keep the installation/usage specifics out of the scripts or modules. So, if you should have to deploy them in another environment with different needs, you would just have to adapt one wrapper (in case it's required at all in that environment).

    YMMV, though :)