From: Stefan Beller <sbeller@google.com>
Date: Fri, 17 Jun 2016 16:09:46 GMT
Subject: Re: [PATCH] diff compaction heuristic: favor shortest neighboring blank lines
Message-ID: <CAGZ79kZgdbqnWW9uqeBHNDbyGVfc0x5gTJbyD0Nyi1i3SNvENQ@mail.gmail.com>
In-Reply-To: <20160617153637.GA9314@sigill.intra.peff.net>

On Fri, Jun 17, 2016 at 8:36 AM, Jeff King <peff@peff.net> wrote:
>
> Did you want something that triggers the "bad" case with the existing
> compaction heuristic?
>
> I gave one in:
>
>   http://article.gmane.org/gmane.comp.version-control.git/296947

I found that out after sending the initial patch. Thanks for pointing out!


>
> I think the difference is that in my example, the diff (before
> compaction) has the blank line at the top (because we are adding a new
> entry at the bottom), whereas in yours, the blank line is at the bottom.
>
> This patch does make my "bad" case look better. Unfortunately, it
> re-breaks another case:

I think before spending more time on discussing and implementing new
(hopefully better) heuristics, I'd want to step back and try to be a bit more
systematic, i.e. I'll want to collect lots of test cases in different languages
and use cases. A mini test suite, maybe?

I think I'll include these examples in there, too.

Thanks,
Stefan