Re: [GSoC][PATCH v3 2/3] rebase -i: rewrite setup_reflog_action() in C

[Junio C Hamano <gitster@pobox.com>]

Alban Gruin <alban.gruin@gmail.com> writes:

>>>> I do not think "base_commit" is a good name, either, though.  When I
>>>> hear 'base' in the context of 'rebase', I would imagine that the
>>>> speaker is talking about the bottom of the range of the commits to
>>>> be rebased (i.e. "rebase --onto ONTO BASE BRANCH", which replays
>>>> commits BASE..BRANCH on top of ONTO and then points BRANCH at the
>>>> result), not the top of the range or the branch these commits are
>>>> taken from.
>>> ...
> Now I really don’t know how to call this function.
> checkout_top_of_range(), perhaps?

If this is a straight rewrite of setup_reflog_action, i.e. the
division of labor between its caller and this function is such that
the caller blindly calls it without even checking if it got the
optional "branch to be rebased" argument and this function is
responsible to decide if the preparatory checkout of the named
branch is necessary, then any name that does not even hint that
checkout is done conditionally would not work well.

How about callilng it "prepare_branch_to_be_rebased()"?  This
function, at least the original setup_reflog_action, responds to

	git rebase [--onto ONTO] UPSTREAM

by doing nothing (i.e. the result of preparation is to do nothing
because we are rebasing the commits between UPSTREAM and currently
checked out state on top of ONTO, or UPSTREAM if ONTO is not
given) and it responds to

	git rebase [--onto ONTO] UPSTREAM BRANCH

by checking out BRANCH (most importantly, when given a concrete
branch name, it checks the branch out, and does not detach at the
commit at the tip of the branch), because that is the first
preparatory step to rebase the BRANCH.