# [PATCH] bring description of git diff --cc up to date 2 messages from 2008-07-21 to 2008-07-22. Participants: Jonathan Nieder, Junio C Hamano. Thread: https://gitlist.dev/t/14573 ## Jonathan Nieder, 2008-07-21 18:27 Subject: [PATCH] bring description of git diff --cc up to date Message-ID: URL: https://gitlist.dev/e/Pine.GSO.4.62.0807211316530.2841%40harper.uchicago.edu ``` In February 2006 [1], the behavior of diff --cc was changed to fit a more appropriate notion of "interesting hunks" in Octopus merges. This patch updates documentation accordingly. [1] commit bf1c32bdec8223785c779779d0a660a099f69a63 combine-diff: update --cc "uninteresting hunks" logic Signed-off-by: Jonathan Nieder --- Hello, all. A while ago when someone was asking about the diff --cc output, I noticed some mistakes in the documentation but never got around to fixing them. So here's a small fix. After this patch, many issues remain: - The user would usually look to git-diff(1) or git-log(1) for an explanation of the --cc option, not git-diff-tree(1) or git-rev-list(1). - diff-format.txt contains the cryptic text "Note that 'combined diff' lists only files which were modified from all parents." I have no idea what that means. - There is some duplication of text between git-diff-tree(1) and git-rev-list(1) here. I don't think that's such a big deal, though. - My proposed text here is not very intuitive or clear, and it does not suggest very strongly what --cc is intended to do (to point out "evil merges" and other merges where the integrator is likely to have made a mistake). Alternate wordings welcome! At least this patch would not makes matters worse. Documentation/git-diff-tree.txt | 12 +++++++----- Documentation/rev-list-options.txt | 9 +++++---- 2 files changed, 12 insertions(+), 9 deletions(-) diff --git a/Documentation/git-diff-tree.txt b/Documentation/git-diff-tree.txt index 0e45b58..5a81d5d 100644 --- a/Documentation/git-diff-tree.txt +++ b/Documentation/git-diff-tree.txt @@ -92,12 +92,14 @@ include::pretty-options.txt[] --cc:: This flag changes the way a merge commit patch is displayed, in a similar way to the '-c' option. It implies the '-c' - and '-p' options and further compresses the patch output - by omitting hunks that show differences from only one - parent, or show the same change from all but one parent - for an Octopus merge. When this optimization makes all + and '-p' options and makes the patch output + even more compact by omitting uninteresting hunks. A hunk is + considered interesting only if either (a) it shows changes from + all parents or (b) in an Octopus merge, it shows different changes + from at least three different parents. + When this optimization makes all hunks disappear, the commit itself and the commit log - message is not shown, just like in any other "empty diff" case. + message are not shown, just like in any other "empty diff" case. --always:: Show the commit itself and the commit log message even diff --git a/Documentation/rev-list-options.txt b/Documentation/rev-list-options.txt index b6f5d87..a399e2b 100644 --- a/Documentation/rev-list-options.txt +++ b/Documentation/rev-list-options.txt @@ -111,10 +111,11 @@ options may be given. See linkgit:git-diff-files[1] for more options. --cc:: - This flag implies the '-c' options and further compresses the - patch output by omitting hunks that show differences from only - one parent, or show the same change from all but one parent for - an Octopus merge. + This flag implies the '-c' option and makes the patch output + even more compact by omitting uninteresting hunks. A hunk is + considered interesting only if either (a) it shows changes from + all parents or (b) in an Octopus merge, it shows different changes + from at least three different parents. -r:: -- 1.5.6.3.549.g8ca11 ``` ## Junio C Hamano, 2008-07-22 06:41 Subject: Re: [PATCH] bring description of git diff --cc up to date Message-ID: <7vfxq2pgik.fsf@gitster.siamese.dyndns.org> URL: https://gitlist.dev/e/7vfxq2pgik.fsf%40gitster.siamese.dyndns.org In-Reply-To: ``` Jonathan Nieder writes: > + This flag implies the '-c' option and makes the patch output > + even more compact by omitting uninteresting hunks. A hunk is > + considered interesting only if either (a) it shows changes from > + all parents or (b) in an Octopus merge, it shows different changes > + from at least three different parents. I am not sure where that "at lesta three different parents" comes from. It might be that what the logic does can be expressed that way, but that was not the guiding principle why the code does what it currently does. These two might make it clearer what hunks are considered interesting: http://thread.gmane.org/gmane.comp.version-control.git/15098/focus=15109 http://thread.gmane.org/gmane.comp.version-control.git/15486/focus=15491 Especially the latter thread, not just the focused one but the whole thing, is quite amusing. This one gives an insight into the logical consequence of the hunk selection criteria: http://thread.gmane.org/gmane.comp.version-control.git/15486/focus=15600 it is worth a read after understanding why some hunks are omitted and some hunks are included by reading other articles in the thread. ```