> > There is no way I know of for a child process to modify the
> environment of
> > it's parent. You might try having your perl program create a script
> > somewhere in the path, then the parent could execute it.
> >
> > You can also get tricky by running your perl script in the current
> > environment using:
> > . program_name (space after the dot)
> > then your program could exit using exec('path/to/shell');
> >
> > -Mark
> >
>
> I am not sure I understood the OP but I would go with the
> wrapping approach in the other direction, that is, write a
> shell script that sets up the environment, etc. by calling
> the perl script within it. Then run the shell script rather
> than the perl script....
>
I don't understand the difference here. Even in a wrap, the perl script has
to be called in the same environment as the wrap, since it is the perl
script doing the mods, using the dot notation, if not, the perl script will
never modify the environment. Correct, or do I still not get it?
-Mark
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
