Re: [PATCH v4 0/10] Long names for `git log -S` and `git log -G`

From: Junio C Hamano <gitster@pobox.com>
Date: Feb 11, 2025, 18:07 UTC
Message-ID: <xmqqwmdwxsmz.fsf@gitster.g>
In-Reply-To: <20250211085028.3923875-1-illia.bobyr@gmail.com>

Illia Bobyr <illia.bobyr@gmail.com> writes:

Show 19 quoted lines

> I've split the big change from v3 [1] into multiple, mostly independent patches
> to make it easier to review and merge each one separately.
>
> [1] https://lore.kernel.org/git/20250206014324.1839232-1-illia.bobyr@gmail.com/
>
> Patches 1 through 4 are fixing minor bugs and inconsistencies.
>
> Patch 5 contains updates gitdiffcore to use same placeholder names as the rest
> of the code.
>
> Patch 6 contains a minimum change to add long versions of -S and -G.
>
> Patch 7 adds bash completion support.
>
> Patches 8 through 10 increase usage of the long argument versions in tests, CLI
> help and docs respectively.
>
> Please, let me know if you prefer it split in a different way, or reorder the
> changes.

When you base your patch on a different base than 'master' (or if the previous iteration of the topic has already been queued in my tree, then the commit used as the base to queue the topic), please make sure you state it clearly.

This iteration seems to apply on none of bc204b74 (The seventh batch, 2025-02-03), on top of which the previous round dcc02caba2 (ib/diff-S-G-with-longhand) has been queued, or any of the recent tips of 'master', like 388218fa (The ninth batch, 2025-02-10) or 9520f7d9 (The eighth batch, 2025-02-06), so I cannot look at it.

> I was not sure if I should include a reference to the previous version of the
> patch into the next reroll.  It seems that
> `Documentation/MyFirstContribution.adoc` suggests so.  But it creates very long
> threads.  And I've noticed that not everyone is doing it.

Almost everybody does so, actually.

Taking a topic that has 5 iterations, each about ~20 patches, as an example:

https://lore.kernel.org/git/20250207-pks-reftable-drop-git-compat-util-v5-0-ba2adc79110f@pks.im/

it is perfectly clear and easy to nagivate from the list of messages what discussions we had in previous iterations.

> Reply to review notes ...

It is more customary to Reply-all directly to review messages, instead of sending new round of patches. When the cover letter of a new iteration is sent as a response to the cover letter of the previous iteration, readers can find the previous discussion messages.

Thanks.