Re: [PATCH 0/8] builtin/config: canonicalize "value_regex" with `--type=bool[-or-int]`

From: Martin Ågren <martin.agren@gmail.com>
Date: Nov 14, 2019, 06:40 UTC
Message-ID: <CAN0heSrTtEHM7bEGMbokf1RJDGPobD=C73Wq1Ry5fgE69+QV-g@mail.gmail.com>
In-Reply-To: <xmqqeeyb9p4i.fsf@gitster-ct.c.googlers.com>

Hi Junio

On Thu, 14 Nov 2019 at 03:19, Junio C Hamano <gitster@pobox.com> wrote:

Show 12 quoted lines

>
> Martin Ågren <martin.agren@gmail.com> writes:
>
> > `--type=bool-or-int` gets the same treatment, except we need to to be
> > able to handle the ints and regexes matching particular ints that we
> > must expect.
>
> Hmm, so I can say 1024k or 1m and that would match 1048576?
>
> Doubly nice.
>
> Looking forward to reading it thru.

Maybe you already noticed, but no, I didn't get to canonicalizing integers like that. What I meant was, type=bool-or-int learns to handle booleans similar to what I did to type=bool.

I don't feel entirely satisfied by some of my commit message oneliners. They could make that a bit clearer, I think.

Not directly related to your question, but I realize now that with type=bool-or-int, I only added the first of these example usages below as a test. The second one is perhaps just as interesting.

$ ./git -c an.int=1 config --get --type=bool-or-int an.int 1 1 $ ./git -c an.int=1 config --get --type=bool-or-int an.int on 1

This last one emits "1". That's because by the time we've decided to
output the value, `format_config()` has some logic around
type=bool-or-int, but doesn't know about why exactly we selected this
an.int=1 in the first place
  (git_parse_maybe_bool("1") == git_parse_maybe_bool_TEXT("on")).

Just after thinking about this for a short while, I can't immediately say whether this second one should emit "1" or "true". My added documentation is actually vague enough to allow both of these to happen... I'll ponder this.

Martin