Re: [PATCH] RFC: Introduce '.gitorderfile'

From: Stefan Beller <sbeller@google.com>
Date: Jul 12, 2017, 21:55 UTC
Message-ID: <CAGZ79kZuMirZVrWTQ5eU7izYWyM8YtvDzMh4gRzvGXXVp=NMkQ@mail.gmail.com>
In-Reply-To: <xmqqzic9z69i.fsf@gitster.mtv.corp.google.com>

On Wed, Jul 12, 2017 at 2:37 PM, Junio C Hamano <gitster@pobox.com> wrote:

Show 16 quoted lines

> Stefan Beller <sbeller@google.com> writes:
>
>> 2. collaboration.
>>     When I want to review a patch from the mailing list,
>>     I could (a) download the patch, apply locally, see the diff
>>     formatted nicely according to diff.orderFile.
>
> If you are not doing a review of a patch with complex changes that
> benefits by a local diff.orderfile (i.e. only in the mail client
> without applying and viewing the changes in wider context), then you
> either (1) have a much greater memory than I do and know all the
> code outside the patch context by heart, or (2) not reviewing them
> properly in context.
>
> I tend to suspect that it is the latter case, so that argument does
> not sound convincing at least to me.

Note that this request originated from https://public-inbox.org/git/cover.1499800530.git.jonathantanmy@google.com/

There are different sorts of patches to review:
(a) major new features, introducing radically new concepts.
    Example given above.

    In this case I neither need (1) nor (2). I want to get the abstract
    design and then decide if it is worth my time to pursue reviewing the
    details of the patch.

    The order file makes an impact!

(b) minor new features (in the big picture of a major feature), refactorings
    Examples: 96dc883b3c, repository: enable initialization of submodules
    lb/status-stash-count

    Reviewing these patches requires more diving into code, but you'd
    still want to make a call early on whether to reject the design
    before calling out the memory leak that you found after applying
    the patch.

    The order file *may* be useful.

(c) updating existing things
    (bug fixes, documentation, improving performance)
    Example: jk/reflog-walk-maint

    When looking at these changes, (2) is the answer.
    I look at surrounding code (there may be no need to apply
    the patch though, it depends)

    The order file is useless, IMHO, but also does not produce harm.

Show 8 quoted lines

> No, I do not apply all patches before commenting from my mailbox; a
> one or two-pager patch can often be viewed and judged without much
> surrounding context, and can be answered in the mail client, perhaps
> while running "less" on some related files that may or may not be
> touched by the patch in another terminal, without applying the
> patch.  But such a one or two-pager patch can be viewed in any
> presentation order and do not behefit much fro diff.orderfile
> anyway.

This sounds like (c), which we have a lot more of than (a) or (b).

Thanks for your thoughts, Stefan