Re: [Bug report] git diff stat shows unrelated diff

From: Philip Oakley <philipoakley@iee.org>
Date: Feb 17, 2019, 23:34 UTC
Message-ID: <d2127f77-9124-0e01-e96b-b6a1650ebc59@iee.org>
In-Reply-To: <xmqq1s473x63.fsf@gitster-ct.c.googlers.com>

Hi Junio,

On 17/02/2019 03:34, Junio C Hamano wrote:

Show 21 quoted lines

> Philip Oakley <philipoakley@iee.org> writes:
>
>>> Those who do *not* opt into that "early warning" configuration dance
>>> would eventually be warned whenever they type "diff A..B", and the
>>> timing for that eventuality is not under their control, so quite
>>> honestly, I do not see much point in "giving users the chance".
>> With the opposite hat on, not giving users the choice does seem unfair
>> to those that are trying to keep up. If we are warning (in the release
>> notes) of an upcoming deprecation (in the code) then it does seem
>> helpful that users could buy into the deprecation early, and at their
>> convenience, to assist in the unlearning of an old habit, which can be
>> much harder than learning a new habit, hence my comment.
>>
>> You are right that those who neither notice nor care will be surprised
>> later, but we shouldn't let that limit others.
> I still do not quite get where you are coming from.  Are you saying
> that those who do not opt into the early warning may get 2 cycles
> (just picked out of thin-air) of deprecation period, and with an
> optional early warning feature, those who feel that 2 cycles is not
> long enough to train their fingers would spend 3 cycles and they
> will be helped than without?

It was my understanding that the end point would be total removal of any options and the typing of the double dot would be an error. Given that hard end point I was looking to ensure that users of double dots have a manageable route to unlearning old bad habits. Thus the first phase would be opt-in, the second phase opt-out, and on the third final phase it would be a non-optional error (assuming your first comment in [1]).

Show 6 quoted lines

>
> That line of thinking sounds somewhat ridiculous---where does it
> end?  If those who opt into would find it sufficient to have 3
> cycles to train, there may still be people who feel 3 is not enough
> and want to have 4.  As we make it longer, we'd cover more people
> and at some point we'd reach the point of diminishing returns.

The length of the phase 1 is your choice, but having a zero length (as some discussions implied) felt too short. For me, one cycle of users 'opting-in' to do their testing and training given a deprecation notification would be sufficient.

Show 6 quoted lines

>
> Wouldn't it be even better, and far simper, to just extend the
> deprecation period to that many cycles to make it long enough for
> majority of users' needs, without any early warning option?
>
> The thing is, once you train your fingers,

To train the fingers, and to check local scripts and aliases, the user needs feedback, preferably at a time of their convenience (as opposed to being a time of inconvenience), so assuming they have been paying moderate attention to the release notes, providing the opt-in phase gives them that.

Show 6 quoted lines

>   it does not matter to you
> if the deprecation warning is still there, as you'd not be typing
> "diff A..B" at that point.  So I am not sure who you are trying to
> help by the early warning option.
>
> Thanks.

I do note that you had indicated at the end of [1]: "I am not sure if it is worth the deprecation cost, though.", so this may be a bit of a mute point anyway.

Philip

[1] <xmqqmumy6mxe.fsf@gitster-ct.c.googlers.com> "then eventually error out when two-dot notation is used"