Re: [PATCH RFC v2 02/19] rebase -i: reword complains about empty commit despite --keep-empty

2014-07-09 Thread Fabian Ruch
Junio C Hamano writes:
> Fabian Ruch  writes:
>> Subject: rebase -i: reword complains about empty commit despite --keep-empty
> 
> I had to read the title and then the log twice before understanding
> that the above is not "change the complaint message" (i.e. "reword
> complaints" spelled incorrectly) but is a description of the current
> behaviour (i.e. "the command complains when 'reword' is used on an
> empty commit") that is not accompanied by an evaluation ("ok, it
> complains; are you saying it is a good thing or a bad thing?") or an
> action ("if it is a bad thing, what are you doing about it?").
> 
> Perhaps
> 
>   rebase -i: allow rewording an empty commit
> 
> or something?

I thought "...despite --keep-empty" would already imply that "reword
complains about empty commit" is not supposed to happen, the action
would have been obvious. However, I understand that --keep-empty is
first of all concerned with which commits of $upstream...$orig_head end
up on the initial to-do list and the git-rebase manual page doesn't
mention that it picks commits using the --allow-empty option. It is
simply a necessity of a script not to complain about something it put on
the to-do list itself.

The subject reads now

rebase -i: allow rewording an empty commit without complaints

trying to convey that this is not a new feature but a bug fix.

   Fabian
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html


Re: [PATCH RFC v2 02/19] rebase -i: reword complains about empty commit despite --keep-empty

2014-07-08 Thread Junio C Hamano
Fabian Ruch  writes:

> Subject: rebase -i: reword complains about empty commit despite --keep-empty

I had to read the title and then the log twice before understanding
that the above is not "change the complaint message" (i.e. "reword
complaints" spelled incorrectly) but is a description of the current
behaviour (i.e. "the command complains when 'reword' is used on an
empty commit") that is not accompanied by an evaluation ("ok, it
complains; are you saying it is a good thing or a bad thing?") or an
action ("if it is a bad thing, what are you doing about it?").

Perhaps

rebase -i: allow rewording an empty commit

or something?

> The to-do list command `reword` replays a commit like `pick` but lets
> the user also edit the commit's log message. If `--keep-empty` is
> passed as option to git-rebase--interactive, empty commits ought to
> be replayed without complaints. However, if the users chooses to
> reword an empty commit by changing the respective to-do list entry
> from
>
> pick fa1afe1 Empty commit
>
> to
>
> reword fa1afe1 Empty commit
>
> , then git-rebase--interactive suddenly fails to replay the empty
> commit. This is especially counterintuitive because `reword` is
> thought of as a `pick` that alters the log message in some way but
> nothing more and the unchanged to-do list entry would not fail.
>
> Handle `reword` by cherry-picking the named commit and editing the
> log message using
>
> git commit --allow-empty --amend
>
> instead of
>
> git commit --amend.
>
> Add test.
>
> Signed-off-by: Fabian Ruch 
> ---
>  git-rebase--interactive.sh| 2 +-
>  t/t3404-rebase-interactive.sh | 8 
>  2 files changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/git-rebase--interactive.sh b/git-rebase--interactive.sh
> index e733d7f..689400e 100644
> --- a/git-rebase--interactive.sh
> +++ b/git-rebase--interactive.sh
> @@ -504,7 +504,7 @@ do_next () {
>  
>   mark_action_done
>   do_pick $sha1 "$rest"
> - git commit --amend --no-post-rewrite 
> ${gpg_sign_opt:+"$gpg_sign_opt"} || {
> + git commit --allow-empty --amend --no-post-rewrite 
> ${gpg_sign_opt:+"$gpg_sign_opt"} || {
>   warn "Could not amend commit after successfully picking 
> $sha1... $rest"
>   exit_with_patch $sha1 1
>   }
> diff --git a/t/t3404-rebase-interactive.sh b/t/t3404-rebase-interactive.sh
> index 8197ed2..9931143 100755
> --- a/t/t3404-rebase-interactive.sh
> +++ b/t/t3404-rebase-interactive.sh
> @@ -75,6 +75,14 @@ test_expect_success 'rebase --keep-empty' '
>   test_line_count = 6 actual
>  '
>  
> +test_expect_success 'rebase --keep-empty' '
> + git checkout emptybranch &&
> + set_fake_editor &&
> + FAKE_LINES="1 reword 2" git rebase --keep-empty -i HEAD~2 &&
> + git log --oneline >actual &&
> + test_line_count = 6 actual
> +'
> +
>  test_expect_success 'rebase -i with the exec command' '
>   git checkout master &&
>   (
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[PATCH RFC v2 02/19] rebase -i: reword complains about empty commit despite --keep-empty

2014-07-02 Thread Fabian Ruch
The to-do list command `reword` replays a commit like `pick` but lets
the user also edit the commit's log message. If `--keep-empty` is
passed as option to git-rebase--interactive, empty commits ought to
be replayed without complaints. However, if the users chooses to
reword an empty commit by changing the respective to-do list entry
from

pick fa1afe1 Empty commit

to

reword fa1afe1 Empty commit

, then git-rebase--interactive suddenly fails to replay the empty
commit. This is especially counterintuitive because `reword` is
thought of as a `pick` that alters the log message in some way but
nothing more and the unchanged to-do list entry would not fail.

Handle `reword` by cherry-picking the named commit and editing the
log message using

git commit --allow-empty --amend

instead of

git commit --amend.

Add test.

Signed-off-by: Fabian Ruch 
---
 git-rebase--interactive.sh| 2 +-
 t/t3404-rebase-interactive.sh | 8 
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/git-rebase--interactive.sh b/git-rebase--interactive.sh
index e733d7f..689400e 100644
--- a/git-rebase--interactive.sh
+++ b/git-rebase--interactive.sh
@@ -504,7 +504,7 @@ do_next () {
 
mark_action_done
do_pick $sha1 "$rest"
-   git commit --amend --no-post-rewrite 
${gpg_sign_opt:+"$gpg_sign_opt"} || {
+   git commit --allow-empty --amend --no-post-rewrite 
${gpg_sign_opt:+"$gpg_sign_opt"} || {
warn "Could not amend commit after successfully picking 
$sha1... $rest"
exit_with_patch $sha1 1
}
diff --git a/t/t3404-rebase-interactive.sh b/t/t3404-rebase-interactive.sh
index 8197ed2..9931143 100755
--- a/t/t3404-rebase-interactive.sh
+++ b/t/t3404-rebase-interactive.sh
@@ -75,6 +75,14 @@ test_expect_success 'rebase --keep-empty' '
test_line_count = 6 actual
 '
 
+test_expect_success 'rebase --keep-empty' '
+   git checkout emptybranch &&
+   set_fake_editor &&
+   FAKE_LINES="1 reword 2" git rebase --keep-empty -i HEAD~2 &&
+   git log --oneline >actual &&
+   test_line_count = 6 actual
+'
+
 test_expect_success 'rebase -i with the exec command' '
git checkout master &&
(
-- 
2.0.0

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html