[PATCH] commit: block amend while histedit is in progress (issue4800)

[timeless]

Sent

On Fri, Feb 26, 2016 at 1:09 PM, Pierre-Yves David
<pierre-yves.david at ens-lyon.org> wrote:
>
>
> On 02/14/2016 07:14 PM, Pierre-Yves David wrote:
>>
>>
>>
>> On 02/14/2016 07:36 AM, timeless wrote:
>>>
>>> # HG changeset patch
>>> # User timeless <timeless at mozdev.org>
>>> # Date 1455435350 0
>>> #      Sun Feb 14 07:35:50 2016 +0000
>>> # Node ID 800740823f9d3c5ac6cd349139405b83e35d5da9
>>> # Parent  a036e1ae1fbe88ab99cb861ebfc2e4da7a3912ca
>>> commit: block amend while histedit is in progress (issue4800)
>>>
>>> Currently histedit gets confused if an amend happens while histedit
>>> is in progress. Since we have a checkunfinished command, we are
>>> temporarily honoring it.
>>>
>>> Note: eventually this guard will be removed. Please do not expect
>>> this behavior to remain.
>>
>>
>> This seems a sensible reaction until we have a better fix. But this is
>> quite a usuability bummer as, for example it will prevent people to
>> amend their -new- commit during an edit.
>>
>> I'm CCing Kostia who is working on a fix for issue4800. Maybe it will be
>> ready soon enough so that we don't need this patch ?
>>
>> As more long terms vision, we should probably grow "checkrewrite" that
>> would check if it sensible to rewrite a changeset or not.|
>
>
> So, Kostia landed a better fix (using obsolescence history to update
> replacement instead of crashing) for evolution user. But we still need
> something for user without evolution. Disabling commit --amend as a first
> move make sense here. Ideally we would check if the amend is touching
> something "forbiden" by reading the histedit state file, but this is much
> more complicated. So disabling ci --amend here seems sensible
>
> Can you send a V2 ?
>
> Cheers,
>
>
> --
> Pierre-Yves David
> _______________________________________________
> Mercurial-devel mailing list
> Mercurial-devel at mercurial-scm.org
> https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel