Re: difflame

From: Jeff King <peff@peff.net>
Date: Jan 28, 2017, 03:53 UTC
Message-ID: <20170128035321.yrcqwkg2fiwadxj4@sigill.intra.peff.net>
In-Reply-To: <CAOc6etaCk=OEyarMNhorM94MBnYRscCkJBM-K08snv1ecmOaPQ@mail.gmail.com>

On Tue, Jan 17, 2017 at 11:24:02PM -0600, Edmundo Carmona Antoranz wrote:

> For a very long time I had wanted to get the output of diff to include
> blame information as well (to see when something was added/removed).

This is something I've wanted, too. The trickiest part, though, is blaming deletions, because git-blame only tracks the origin of content, not the origin of a change.

For example, try this case:

  git init
  for i in $(seq 1 10); do
    echo $i >>file
    git add file
    git commit -m "add $i"
  done
  sed 4d <file >tmp && mv tmp file
  git commit -am "drop 4"

Running "difflame HEAD~5 HEAD" produces this output:

  diff --git a/file b/file
  index b414108..051c298 100644
  --- a/file
  +++ b/file
  @@ -1,6 +1,9 @@
   ^2b028ff (Jeff King 2017-01-27 22:44:10 -0500 1) 1
   ed056366 (Jeff King 2017-01-27 22:44:10 -0500 2) 2
   771030d8 (Jeff King 2017-01-27 22:44:10 -0500 3) 3
  -89c09c82 (Jeff King 2017-01-27 22:44:10 -0500 4) 4
   b619039c (Jeff King 2017-01-27 22:44:10 -0500 4) 5
   6a7aa0e5 (Jeff King 2017-01-27 22:44:10 -0500 5) 6
  +39bc9dc4 (Jeff King 2017-01-27 22:44:10 -0500 6) 7
  +f253cc8f (Jeff King 2017-01-27 22:44:10 -0500 7) 8
  +85c10f46 (Jeff King 2017-01-27 22:44:10 -0500 8) 9
  +89c09c82 (Jeff King 2017-01-27 22:44:10 -0500 9) 10

The last 4 lines are right; they correspond to the addition commits. But the line taking away 4 is wrong. You can see even without looking at its patch, because it is blamed to the same commit that added "10", which is wrong.

Sorry I don't have a solution. I think it's an open problem with git-blame, though you could probably script something around "git blame --reverse". See the commit message of 85af7929e (git-blame --reverse, 2008-04-02) for some discussion.

-Peff