From: Jay Soffian <jaysoffian@gmail.com>
Date: Thu, 05 Mar 2009 21:04:44 GMT
Subject: Re: orthogonal cases of log --date option
Message-ID: <76718490903051304j6d8138f7qa5492ac15edd6460@mail.gmail.com>
In-Reply-To: <20090305104304.GA17760@coredump.intra.peff.net>

On Thu, Mar 5, 2009 at 5:43 AM, Jeff King <peff@peff.net> wrote:
>> But this patch may help you get started.
>
> FWIW, I think this is the wrong direction. You are working around the
> lack of orthogonality in the interface by tweaking things in the
> implementation. I think you are better to fix the interface, but support
> --date=local for historical reasons. IOW,
>
>  git log --local-dates --date=short
>
> with
>
>  git log --date=local
>
> as a historical synonym for
>
>  git log --local-dates --date=default
>
> This makes the interface simpler to understand: --date remains a
> selector, and --date=local is a special case that new people don't need
> to think about or understand.

I started to pick this up and I want to clarify what you meant by
interface. Was it the CLI you had an issue with? Because that I
understand and it's easy to support the CLI changes you outline above.

Or did you have a problem with how Junio was going about passing along
both bits (i.e. 1. date format; 2. local or not) in an enum? Because I
have to tell you, I started looking at what it would take to switch
the enum to something like:

struct date_mode {
	enum {
		DATE_NORMAL = 0,
		DATE_RELATIVE,
		DATE_SHORT,
		DATE_ISO8601,
		DATE_RFC2822,
		DATE_RAW
	} format;
	unsigned int local;
};

It's a significantly more invasive change.

j.