Re: [PATCH] Make git-commit cleverer - have it figure out whether it needs -a automatically
From: Junio C Hamano <hidden>
Date: 2016-08-11 20:12:26
Junio C Hamano [off-list ref] writes:
I think another exception should be needed. If the index does not match the working tree, it should not default to "-a". Otherwise,
I think there needs a bit of explanation and additional step
that happened here. This by the way is not a made-up example.
Everything, including the 20-minute away, were what happened
when I did the latest blame fix you saw a few days ago.
* I am still futzing with blame from time to time, and have
this change almost permanently in my working tree.
$ cat P.diff
diff --git a/builtin-blame.c b/builtin-blame.c
index dc3ffea..46ce45c 100644
--- a/builtin-blame.c
+++ b/builtin-blame.c
@@ -38,7 +38,7 @@ static int max_digits;
static int max_score_digits;
#ifndef DEBUG
-#define DEBUG 0
+#define DEBUG 1
#endif
* I also have the GIT-VERSION-GEN change in my working tree.
I want to fix another thing in pickaxe. $ edit builtin-blame.c
* Of course I did tests here.
My wife calls me. Away from desk for 20 minutes. Later I come back.
* And then reverted the DEBUG back to 0 in preparation for "checking into the index" $ edit builtin-blame.c
$ git update-index builtin-pickaxe.c
* And then I reverted it back for later futing. $ git apply P.diff ;# that is a permanent-temporary file.
I am so used to that name and did not realize that typo, and I was not paying too much attention. My wife calls me again. Away from desk and back in 20 minutes. $ git commit -m 'git-blame: Another fix.' Oops.
* Oops here is not just that builtin-blame.c would have been committed; I'd almost never do "commit -a" in this repository, because it would take that "DEBUG 1" change _and_ GIT-VERSION-GEN change into the commit.
So, please turn this "cleverness" off when the index does not match the working tree.