[PATCH] git-commit: exit non-zero if we fail to commit the index
Show 14 quoted lines
> Brandon Casey <casey@nrlssc.navy.mil> writes:
>
>>> We would need to tell the user that the index is not where
>>> it is when we detect the error, though.
>> The new index we are trying to rename will be deleted.
>> Are you saying we should
>> warn the user that the index is now out of sync?
>
> Yeah, something like that. But I think that once this happens
> there is no easy and sane recovery path for the user, as the
> most likely cause of the failure there would be the user running
> out of quota, so "git reset HEAD" which may be the way to
> recover from that failure would not have enough room to create a
> new index file anyway.
If you're interested, here's a patch.
-brandon
builtin-commit.c | 15 +++++++++++----
1 files changed, 11 insertions(+), 4 deletions(-)
diff --git a/builtin-commit.c b/builtin-commit.c
index 0227936..d8deb1a 100644
--- a/builtin-commit.c
+++ b/builtin-commit.c
@@ -122,19 +122,23 @@ static void rollback_index_files(void)
}
}
-static void commit_index_files(void)
+static int commit_index_files(void)
{
+ int err = 0;
+
switch (commit_style) {
case COMMIT_AS_IS:
break; /* nothing to do */
case COMMIT_NORMAL:
- commit_lock_file(&index_lock);
+ err = commit_lock_file(&index_lock);
break;
case COMMIT_PARTIAL:
- commit_lock_file(&index_lock);
+ err = commit_lock_file(&index_lock);
rollback_lock_file(&false_lock);
break;
}
+
+ return err;
}
/*
@@ -926,7 +930,10 @@ int cmd_commit(int argc, const char **argv, const char *prefix)
unlink(git_path("MERGE_HEAD"));
unlink(git_path("MERGE_MSG"));
- commit_index_files();
+ if (commit_index_files())
+ die ("Repository has been updated, but unable to write\n"
+ "new_index file. Check that disk is not full or quota is\n"
+ "not exceeded, and then \"git reset HEAD\" to recover.");
rerere();
run_hook(get_index_file(), "post-commit", NULL);
--
1.5.4.rc4.16.gdd591
Brandon Casey <casey@nrlssc.navy.mil>