Re: [PATCH] xdiff/xpatience: support anchoring line(s)

From: Junio C Hamano <gitster@pobox.com>
Date: Nov 23, 2017, 02:47 UTC
Message-ID: <xmqqh8tlr9sp.fsf@gitster.mtv.corp.google.com>
In-Reply-To: <xmqq3755sprp.fsf@gitster.mtv.corp.google.com>

Junio C Hamano <gitster@pobox.com> writes:

Show 15 quoted lines

> What do we do when
>
>     git diff --histogram --patience
>
> is given?  Do we warn?  If we don't, perhaps it may not be too bad
> if 
>
>     git diff --histogram --patience-anchor=foo
>     git diff --patience-anchor=foo --histogram
>
> did not get any warning.  Instead we just implicitly turn any
> occurence of --patience-anchor=foo into --patience followed by the
> same option, and assume that the user wanted the usual "last one
> wins" semantics.  It would turn patience on for the former, and
> ignore the anchor for the latter and use historgram.

Thinking about this a bit more, I do like the basic idea of the UI even better. What we could do is to sell this to the end users as a new kind of diff algorithm choice (i.e. myers, patience, ... will gain a new friend) that internally happens to be implemented by piggybacking on patience (just like minimal is piggybacking on myers) and call it "anchor". Then just like this command line

    git diff --histogram --patience

makes the last one win without complaint, it is sane that these command lines

    git diff --histogram --anchored=<pattern>
    git diff --anchored=<pattern> --histogram

make the last one win without complaint, either.

Hmm?