git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH] sequencer: preserve commit messages

From
Michael J Gruber <git@drmicha.warpmail.net>
Date
Feb 24, 2015, 15:29 UTC
Message-ID
<54EC98BD.7060100@drmicha.warpmail.net>
In-Reply-To
<xmqqsidwtq4i.fsf@gitster.dls.corp.google.com>
Junio C Hamano venit, vidit, dixit 23.02.2015 19:54:
Show 15 quoted lines
> Michael J Gruber <git@drmicha.warpmail.net> writes:
> 
>> sequencer calls "commit" with default options, which implies
>> "--cleanup=default" unless the user specified something else in their
>> config. This leads to cherry-picked commits getting a cleaned up commit
>> message, which is usually not an intended side-effect.
>>
>> Make the sequencer use "--cleanup=verbatim" so that it preserves commit
>> messages independent of the defaults and user config for "commit".
> 
> Hmm, wouldn't it introduce a grave regression for users who
> explicitly ask to clean crufty messages up (by setting their own
> commit.cleanup configuration) if you unconditionally force
> "--cleanup=verbatim" here?
> 
That's what I meant by possible side-effects below.
There are no side-effects which our tests would catch.

I don't know sequencer.c well enough to know whether run_git_commit() would be run with a user-edited commit message at all. My (possibly wrong) understanding is that it is called only when a cherry-pick succeeds without any conflicts, so that it is called with a commit message from a pre-existing commit, i.e. a message after cleanup which is to be preserved as is.

In case of a conflict, resolution is left to be done by the user. But I guess I'm overlooking --edit and --continue here.

But git cherry-pick without conflict should no re-cleanup the commit message either, should it?

Show 79 quoted lines
>> Reported-by: Christoph Anton Mitterer <calestyo@scientia.net>
>> Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
>> ---
>>
>> Notes:
>>     All tests run fine with this changed behavior. I don't know
>>     whether this may have any side-effects on other (untested)
>>     uses of the sequencer.
>>
>>  sequencer.c              |  1 +
>>  t/t3511-cherry-pick-x.sh | 28 ++++++++++++++++++++++++++++
>>  2 files changed, 29 insertions(+)
>>
>> diff --git a/sequencer.c b/sequencer.c
>> index 77a1266..35fe9d9 100644
>> --- a/sequencer.c
>> +++ b/sequencer.c
>> @@ -377,6 +377,7 @@ static int run_git_commit(const char *defmsg, struct replay_opts *opts,
>>  	argv_array_init(&array);
>>  	argv_array_push(&array, "commit");
>>  	argv_array_push(&array, "-n");
>> +	argv_array_push(&array, "--cleanup=verbatim");
> 
> 
> 
>>  
>>  	if (opts->gpg_sign)
>>  		argv_array_pushf(&array, "-S%s", opts->gpg_sign);
>> diff --git a/t/t3511-cherry-pick-x.sh b/t/t3511-cherry-pick-x.sh
>> index f977279..b7dff09 100755
>> --- a/t/t3511-cherry-pick-x.sh
>> +++ b/t/t3511-cherry-pick-x.sh
>> @@ -36,6 +36,20 @@ mesg_with_cherry_footer="$mesg_with_footer_sob
>>  (cherry picked from commit da39a3ee5e6b4b0d3255bfef95601890afd80709)
>>  Tested-by: C.U. Thor <cuthor@example.com>"
>>  
>> +mesg_unclean="$mesg_one_line
>> +
>> +
>> +leading empty lines
>> +
>> +
>> +consecutive empty lines
>> +
>> +# hash tag comment
>> +
>> +trailing empty lines
>> +
>> +
>> +"
>>  
>>  test_expect_success setup '
>>  	git config advice.detachedhead false &&
>> @@ -53,6 +67,10 @@ test_expect_success setup '
>>  	test_commit "$mesg_with_footer_sob" foo b mesg-with-footer-sob &&
>>  	git reset --hard initial &&
>>  	test_commit "$mesg_with_cherry_footer" foo b mesg-with-cherry-footer &&
>> +	git reset --hard initial &&
>> +	test_config commit.cleanup verbatim &&
>> +	test_commit "$mesg_unclean" foo b mesg-unclean &&
>> +	test_unconfig commit.cleanup &&
>>  	pristine_detach initial &&
>>  	test_commit conflicting unrelated
>>  '
>> @@ -216,4 +234,14 @@ test_expect_success 'cherry-pick -x -s treats "(cherry picked from..." line as p
>>  	test_cmp expect actual
>>  '
>>  
>> +test_expect_success 'cherry-pick preserves commit message' '
>> +	pristine_detach initial &&
>> +	printf "$mesg_unclean" >expect &&
>> +	git log -1 --pretty=format:%B mesg-unclean >actual &&
>> +	test_cmp expect actual &&
>> +	git cherry-pick mesg-unclean &&
>> +	git log -1 --pretty=format:%B >actual &&
>> +	test_cmp expect actual
>> +'
>> +
>>  test_done
Previous: Junio C HamanoNext: Junio C Hamano
Message 4 of 14 in “[BUG] git mangles up commit messages on rebase”
  1. Christoph Anton MittererFeb 21, 2015
  2. sequencer: preserve commit messagesMichael J Gruber, Feb 23, 2015
  3. Junio C HamanoFeb 23, 2015
  4. Michael J GruberFeb 24, 2015
  5. Junio C HamanoFeb 24, 2015
  6. Michael J GruberFeb 25, 2015
  7. Junio C HamanoFeb 25, 2015
  8. Michael J GruberFeb 26, 2015
  9. Junio C HamanoFeb 26, 2015
  10. Michael J GruberFeb 27, 2015
  11. Junio C HamanoFeb 27, 2015
  12. [PATCHv2] sequencer: preserve commit messagesMichael J Gruber, Mar 6, 2015
  13. Junio C HamanoMar 6, 2015
  14. Christoph Anton MittererFeb 25, 2015

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.