Re: [PATCH 3/3] format-patch: learn --[no-]range-diff-notes

>
> On Sun, Sep 6, 2026, at 15:37, D. Ben Knoble wrote:
> > On Sun, Sep 6, 2026 at 3:23 AM Kristoffer Haugsbakk
> >> >[snip]
> >> > That this wasn’t clear is the fault of the doc here.
> >>
> >> Seeing as how the doc was unclear and did not spell out how you can
> >> build two separate list of notes, here’s a draft of a rewrite:
> >>
> >>     `--range-diff-notes[=<ref>]`::
> >>     `--no-range-diff-notes`::
> >>             Used with `--range-diff`, tweak what notes to display in the
> >>             range diff.
> >>     +
> >>     The default behavior is to display the same notes in the range diff as
> >>     on the patches; see `--notes`. But you can use these options to use a
> >>     different list of notes. For example, say you have given three notes
> >>     refs to `--notes`. At this point those same three notes will be
> >>     displayed in the range diff. But then you pass
> >>     `--range-diff-notes=<ref>`. Now the range diff will only display
> >>     _<ref>_. You can of course pass more refs to this option, just like
> >>     `--notes`. And you can also turn off all notes with
> >>     `--no-range-diff-notes`.
> >>     +
> >>     You may want to turn off this notes override behavior after it has been
> >
> > [nit: should we call this "no notes" override behavior? Otherwise I
> > think we are referring to --range-diff-notes=<ref> overriding
> > --notes=…]
>
> (I will shorten `range-diff` to `RD` for semi-brevity)
>
> What I mean here by “notes override behavior” is turning off all
> `--[no-]RD-notes` options. It means turning off `--RD-notes` as well as
> `--no-RD-notes`. And without the override you are back to the default
> behavior where `--notes` dictates the notes for the range diff.
>
> So that the utility is a bit more clear than these unmotivated examples,
> here’s an example alias (with forced linebreaks):
>
>     my-fp = format-patch --notes=review --notes=testing
>         --notes=attribution --notes=changelog
>         --range-diff-notes=changelog
>
> The patches will have four notes while the range diff will have one.
>
> But you may want to disregard that last `--RD-notes` and in turn get all
> of the notes in the range diff. But without repeating yourself. Then you
> can do this:
>
>     my-fp --no-range-diff-notes --range-diff-notes
>
> The option (the negation) is not sufficient since it would turn off all
> range diff notes. But this special meaning of `--RD-notes` allows you to
> go back to just regular `--notes` behavior. That `--RD-notes` has a
> special meaning when the list of range diff notes is empty does not lose
> anything since `--range-diff-notes` would just be a noöp otherwise.[1]

> Or to just not support it. ;)
>
> (my standard verbosity level might not be doing me any favors
> on this point.)
>
> ***
>
> That might seem like a lot of “power” for something as niche as
> overriding-then-reverting patch contra range diff notes. But code
> wise I don’t think the price is high... :)

> † 1: I just tested the behavior of `--notes` (no arg) on
>      `format-patch`. Yes, it does respect the default notes ref just
>      like git-log(1) does. So an alternative would be to have
>      `--RD-notes` do the same.
>
>      But I do not think some convenient default notes ref is good for a
>      command which is supposed to generate patches for email
>      sendout. For `log` you can make convenient notes to yourself and
>      conveniently display them. But `format-patch` should demand more
>      intentionality. (I also wrote about this on a bugfix for
>      `format-patch` behavior some years ago.)[2]
> † 2: I suspect there is a bug-looking like behavior in that
>      `format-patch` seems to use `notes.displayRef` for the default
>      notes (not just /refs/notes/commits). It should just respect
>      `format.notes`, I think. But I can look at that later.

> > A bit long, but easy to follow and understand the interactions, I
> > think. The examples are helpful.
>
> Thanks. I noticed the lines kept creeping up, but it is more involved
> than most options; an option for passing on to another command which
> also overrides the behavior of another option.