Re: RFC: Hiijacking Patches

Am 25.09.20 um 16:42 schrieb James Cook:
> Is hijacking patches a common operation for anyone? The least
> surprising behaviour would be to always ask the user what to do if
> they haven't explicitly chosen an option on the command line; I'd
> certainly prefer that. But I've never needed to hijack a patch, so
> take my opinion with a grain of salt.

It usually comes up when someone sends a patch and there is a minor
detail I would like to be changed. Ideally one would request the author
to amend their patch, but sometimes the communication overhead to
explain what to change and wait for their response etc is just too much
trouble. A variation of that is when a conflicting patch is pushed
before the external contribution gets proper attention; since this is
not their fault it seems a bit unfair to request that they rebase their
change, so I sometimes do that myself.

Another instance that came up lately for me is when I accidentally
recorded a patch with a wrong identity and then wanted to amend that patch.

I agree that these are probably exceptional situations. So perhaps it is
better to keep the default as is (option (a), i.e. prompt when the user
is about to hijack a patch, but don't change the author). The only
rationale for changing the default behavior is to make it more
consistent with darcs record, but I guess that's pretty weak.

Cheers
Ben