[RFC] use optparse for option handling

[John Arbash Meinel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Aaron Bentley wrote:
> John Arbash Meinel wrote:
>>> I think you want 'store_false' not 'store_const'. I think by doing
>>> 'store_const' it is using the default 'const=None', so you get None
>>> back, not False.
> 
> Normally, yes, but there are bits of the codebase that misbehave if the
> value exists but is set to False.  This way imitates the way the system
> currently works most closely.  Naturally, I'd prefer to fix the code
> that's misbehaving when the value's set to False.
> 

...

> What do you mean by 'weird coupling'?

Just places where the English natural negative is: --non-foo rather than
- --no-foo

Like --no-verbose, --non-verbose is more grammatical. But probably
consistency out-weighs trying to make it sound lnice.

> 
>>>>>>> it's probably good to show both in the help.
>>>>> Okay, so that means we won't want to use optparse to generate our option
>>>>> help, because it wants help for every flag name, and we don't provide
>>>>> text for the inverted help.
>>> Actually, just add the inverted option first. Then you get:
>>>
>>> --other-option	help on other-option
>>> --no-foo
>>> --foo 		help on foo
>>>
>>> I do that all the time.
> 
> Okay, that's an option.  I was thinking more like:
> --other-option	help on other-option
> --foo, --no-foo	help on foo
> 
> Aaron

I don't believe optparse would let you do that. But it lets you do the
former, which is pretty close. Especially since if the line is long you
would end up with:

- --my-long-option, --no-my-long-option
                        help on my-long option
Anyway, ending up with:
- --no-my-long-option
- --my-long-option        help on my-long option

seems okay.

John
=:->

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFEtwoGJdeBCYSNAAMRAgNJAJwJGVzFy5wkMPq9kaKkeJHF04TpvQCfd4dR
hH7ZRWRZI23HNhkbnTGsow8=
=aj+m
-----END PGP SIGNATURE-----