"
ej...@pacbell.net" <
ej...@pacbell.net> writes:
> On Friday, May 28, 2021 at 12:26:20 PM UTC-7, Janis Papanagnou wrote:
>> On 28.05.2021 21:04,
ej...@pacbell.net wrote:
>> >
>> > Heres the sequence of events
>> > open a terminal
>> > cd to a specific path
>> > load various modules <- for this happen, the paths needs to be set by reading the bashrc ( just like opening a terminal )
>> > this terminal is available for user input
>> What's wrong about putting all the stuff (cd, source rc-file, load
>> module, AND a final bash call) in a shell executable file and start
>> that file from the 'terminal -e' call. If the environment variables
>> are exported they should be available in the interactive shell then.
>> (Not that I'd think that it is a good idea to handle that through
>> an implicitly launched terminal application.)
>
> it would certainly be easier but the parent GUI is in perl and thats the request.
Are you saying that you need to write some Perl code that will be
incorporated into an existing Perl script or module? Your original post
showed a standalone Perl script.
My reaction to that was that using Perl adds no value. As a user, if I
invoke a command, I seldom care whether it's a shell script, a Perl
script. or a binary executable. What you're doing can be done more
cleanly in pure shell.
Of course you could write a bash script and have your script invoked
from Perl using system(). Is there some reason that wouldn't meet the
requirement?
--
Keith Thompson (The_Other_Keith)
Keith.S.T...@gmail.com
Working, but not speaking, for Philips Healthcare
void Void(void) { Void(); } /* The recursive call of the void */