Re: [PATCH v2 0/4] git add --resolved
- From
Michael Montalbo <mmontalbo@gmail.com>
- Date
- Aug 2, 2026, 00:22 UTC
- Message-ID
- <CAC2Qwm+6e8ZW36--iPmrCLT2tS7R4O_TDbx4S1MaSMb9bg=epg@mail.gmail.com>
- In-Reply-To
- <xmqqfr10lvhx.fsf@gitster.g>
On Thu, Jul 30, 2026 at 1:41 AM Junio C Hamano <gitster@pobox.com> wrote:
> > Sure, but I do not think of a reason why the posted patches cannot > later be extended in that direction if somebody wanted to... >
Yes, that is fair.
> Have no idea how the user recovers after performing an operation > like that. >
My thought was that a partial set of resolved files without markers would still be added and files with markers would be available for a subsequent add once they were resolved too. My understanding may be incorrect, though.
Show 6 quoted lines
> The naming is very much deliberate. "git am --resolved" tells the > command "I am done with them, so please continue". It is not like I > want to add anything that are unmerged. I dealt with unmerged stuff > and turned them into the resolved state, and the command is a request > to record that fact. It is a regression to call it "--unmerged". >
Thank you for the pointer to "git am --resolved", the context helps me see what you mean that the naming is a regression in the workflow. The name makes sense to me, especially in relation to the existing --resolved flag.