Handling getopt for option without optional argument value

[lisa-asket]

From: Greg Wooledge <greg@wooledge.org>
To: help-bash@gnu.org
Subject: Re: Handling getopt for option without optional argument value
Date: 23/07/2021 22:48:30 Europe/Paris

On Fri, Jul 23, 2021 at 10:03:06PM +0200, lisa-asket@perso.be wrote:
> You are also steadfast on stopping supporting getopt so that we cannot use it 
> anymore.  

Assuming you mean getopt(1), it was never supported to begin with.
It's not part of POSIX at all. It's also not part of bash. It's not
even part of GNU. There's no "support" to "stop".





--- It is a capability that people use.  But for you it is broken because you 
say 

it is a linux thing.





At this point, I don't know what your goal is. It seems to be to complain
as loudly and as frequently as you can until someone decides to spend
a few hundred person-hours designing and implementing and testing a
backward-compatible bash getopts that satisfies all of your needs, without
you having to put in *any* of the work, including stating precisely
what your needs *are*.



--- To satisfy Chet's needs, to be clear.  I simply suggested advantage

of having built in long option capability for your users.  Only to get a barrage

of insults, and do it for us.  Then saying you need 5000+ hours to pass options.

More than the three years time to get a phd studying black holes.



--- Many problems here are partly to do with the unforgiving nature of 
criticizing

and run-in with some gnu developers.  In short, it appears that people do not 
pay

much attention, and basically in the end ask me to submit changes to them. 

It's quite frustrating trying to contribute.



--- Who are the main developers of FSF software these days? Mostly, they are 
either

paid as FSF employee, or students still trying to break out their craft in 
programing, 

or semi-retired programmers who otherwise don't do anything. Those willing and 
able,

start their own projects or get decent salary in commercial corps. 




If that's the goal, I estimate your chances of success to be extremely low.

If you'd like to pursue some more attainable goal, like "get someone to
write the option parsing section of my script for me, and by the way, here
are my requirements", that's got a better chance of succeeding. But you
need to cooperate with people instead of constantly fighting with them.



---

I do not have a problem myself, simply made a suggestion that others would

find using long options as a built-in feature of bash quite handy and the code 

simpler.  As for myself you can stay as you are.  



You could start by stating what you're trying to do.