Re: [PATCH] builtin/config.c: don't print a newline with --color

[Taylor Blau <me@ttaylorr.com>]

On Tue, Mar 05, 2019 at 02:57:32PM +0900, Junio C Hamano wrote:
> Yup, that would be a very sensible first step, regardless of what
> the next step is.
>
> After that, choices are
>
>  (1) we'd introduce new inconsistency among --type=<type> by
>      matching what --type=color does to what --get-color does, to
>      allow us to revert that documentation update, or

I suppose... though I think that if others agree, I'd rather update the
documentation instead of introduce some inconsistency.

Yes, there's an argument to be made that if we're encouraging users to
go from '--get-color' -> '--type=color', that the two should behave the
same, but I don't think the cost we pay for behavioral equivalence
between the two is worth inconsistency among '--type=color' and all the
rest.

>  (2) we'd drop LF from all --type=<type>, that makes everything
>      consistent and risk breaking a few existing scripts while doing
>      so, and get yelled at by end users, or

As you indicate, I think that this option is one we should _not_ do. In
the interpolation example you shared earlier in the thread, script
writers most likely want and expect a trailing LF after each invocation
of 'git config'.

I'd argue that this case is more common than not wanting a LF when
interpolating with `--type=color`, so I agree it seems the tradeoff here
is not a good one.

>  (3) we stop at this documentation update and do nothing else.

To restate my response to (1), I think that the documentation update in
isolation makes the most sense here. I, too, was surprised in the same
way that Peff was when we stumbled upon this, but I think that
ultimately the consistency is most favorable.

Thanks all for your discussion and feedback.


Thanks,
Taylor