Jonathan Nieder writes ("Re: want <reason> option to git-rebase"):
> Oh, good catch. "git rebase" already generally does the right thing
> when GIT_REFLOG_ACTION is set (by only appending to it and never
> replacing it).
Great. I indeed did not know about this.
> Ian, does that work well for you? If so, any ideas where it should go
> in the documentation to be more discoverable for next time?
Thanks for asking exactly the right question :-).
I didn't make a record of exactly where I looked but I'm pretty sure I
looked at the manpages for git-reflog and git-rebase. I think I
probably also looked at git-update-ref; I have read git-update-ref a
number of times.
Right now in Debian unstable I see that none of these places document
this convention.
I think git-reflog ought to mention it, so that it says where the
information it provides comes from.
It also ought to be mentioned in git-update-ref, because all callers
of git-update-ref need to implement it !
Indeed, because I didn't know about this convention, dgit and
git-debrebase do not honour it. At least in my case, if it had been
in git-update-ref I would have implemented it myself and then I would
obviously have thought of making use of it myself.
Also, I have to say, the documentation for GIT_REFLOG_ACTION
in git(1) is very obscure. It sort of generally waffles around what
it is for, but it does not say:
* what does this variable contain
* who can and should set it
* who should consume it
* what the rules are for modifying it
I don't think simply adding a cross-reference to GIT_REFLOG_ACTION in
git(1) would be sufficient, without also improving this part.
The explanations provided by you and Johannes, here in these emails,
are much much better:
> >> "git rebase" sets this itself, so it doesn't solve your problem.
> >
> > If it does so unconditionally, then that is a bug. If a script
> > wants to set GIT_REFLOG_ACTION, but finds that it is already set,
> > then it must not change the value. set_reflog_action in
> > git-sh-setup does the right thing.
> >
> > So, if there is another script or application around git-rebase, then it
> > should just set GIT_REFLOG_ACTION (if it is not already set) and export the
> > environment variable to git-rebase.
>
> Oh, good catch. "git rebase" already generally does the right thing
> when GIT_REFLOG_ACTION is set (by only appending to it and never
> replacing it).
Maybe some of this prose, which explains things quite well, could be
reworked into a form suitable for the git docs. (Even though there
seems to be disagreement about whether a subcommand may *append* to
GIT_REFLOG_ACTION; which, ISTM, is a practice which ought to be
encouraged rather than discouraged.)
Regards,
Ian.
--
Ian Jackson <[email protected]> These opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
