Re: [RFC PATCH 0/5] Add option to autostage changes when continuing a rebase

From: Phillip Wood <phillip.wood@talktalk.net>
Date: Aug 25, 2017, 11:54 UTC
Message-ID: <81909475-5801-c32b-efb5-a2a90c959a21@talktalk.net>
In-Reply-To: <xmqqtw0wx6b1.fsf@gitster.mtv.corp.google.com>

On 24/08/17 17:46, Junio C Hamano wrote:

Show 14 quoted lines

> Phillip Wood <phillip.wood@talktalk.net> writes:
> 
>> It could be expensive to check that the local modifications will not
>> interfere with the rebase - wouldn't it have to look at all the files
>> touched by each commit before starting? What do cherry-pick and am do
>> here when picking several commits?
> 
> Yes, so?
> 
> "I do not think of a way to implement it cheaply, so we forbid
> anybody from finding a clever optimization to implement it cheaply
> by adding a feature that will not work well with it when it
> happens?"
>

Ouch, I think that is a rather unfair characterization of what I said and certainly does not reflect what I was trying to say. I was asking since cherry-pick and am can pick/apply multiple commits/patches and apparently allow the user to start with a dirty tree what do they do about checking that the applied commits/patches do not interfere with any local changes. As for "adding a feature that will not work well with it when it happens?" I suggested a possible solution in the message.

Anyway you're not convinced this feature is a good idea and no one else has commented so I'll drop it.

Phil