Re: [PATCH 01/12] t: prepare `test_match_signal ()` calls for `set -e`
- From
Patrick Steinhardt <ps@pks.im>
- Date
- Apr 14, 2026, 07:23 UTC
- Message-ID
- <ad3rbjDYad8rTybG@pks.im>
- In-Reply-To
- <xmqqldeqkesa.fsf@gitster.g>
On Mon, Apr 13, 2026 at 09:26:13AM -0700, Junio C Hamano wrote:
Show 15 quoted lines
> Patrick Steinhardt <ps@pks.im> writes: > > > We have a couple of calls to `test_match_signal ()` where we execute a > > Git command and expect it to die with a specific signal. These calls > > will essentially execute the process in a subshell via `foo; echo $?`, > > but as we expect `foo` to fail this will cause the overall subshell to > > fail once we `set -e`. > > > > Fix this issue by using `foo || echo $?` instead. > > Hmph, if the 'foo' command that is expected to fail succeeds by a > bug, we won't see 0 on the standard output anymore. Comparing 13 > with 0 or 13 with an empty string "" would fail either way, so that > may not give us a practical difference, but this somehow leaves a > foul small in my nose.
Hm, true, it is a wee bit ugly.
Show 27 quoted lines
> The technique used in <20260325062114.2067946-2-gitster@pobox.com>
> would give us something ugly like
>
> - OUT=$( ((large_git; echo $? 1>&3) | :) 3>&1 ) &&
> + OUT=$( ((st=0; large_git || st=$?; echo $st 1>&3) | :) 3>&1 ) &&
>
> and it ensures that we give 0 when we succeed, but I am not sure if
> this is worth doing.
>
> Or perhaps
>
> - OUT=$( ((large_git; echo $? 1>&3) | :) 3>&1 ) &&
> + OUT=$( ((large_git || echo $? 1>&3) | :) 3>&1 ) &&
> + OUT=${OUT:-0} &&
>
> I dunno.
>
> If "large_git" somehow writes into file descriptor #3, then OUT will
> be contaminated with something other than $?, so the additional "if
> empty substitute with 0" would not work without doing something like
>
> - OUT=$( ((large_git; echo $? 1>&3) | :) 3>&1 ) &&
> + OUT=$( ((large_git 3>/dev/null || echo $? 1>&3) | :) 3>&1 ) &&
> + OUT=${OUT:-0} &&
>
> I guess. For that matter, if large_git spits out "13" to its file
> descriptor #3, I do not know what would have happened ;-)How about the below patch? It's a mouthful, but the intent is quite explicit and it's overall not too bad, if you ask me.
Patrick
diff --git a/t/t0005-signals.sh b/t/t0005-signals.sh index 74c2a27972..84319cf169 100755 --- a/t/t0005-signals.sh +++ b/t/t0005-signals.sh @@ -42,12 +42,12 @@ test_expect_success 'create blob' ' ' test_expect_success !MINGW 'a constipated git dies with SIGPIPE' ' - OUT=$( ((large_git || echo $? 1>&3) | :) 3>&1 ) && + OUT=$( ((large_git && echo 0 1>&3 || echo $? 1>&3) | :) 3>&1 ) && test_match_signal 13 "$OUT" ' test_expect_success !MINGW 'a constipated git dies with SIGPIPE even if parent ignores it' ' - OUT=$( ((trap "" PIPE && large_git || echo $? 1>&3) | :) 3>&1 ) && + OUT=$( ((trap "" PIPE && large_git && echo 0 1>&3 || echo $? 1>&3) | :) 3>&1 ) && test_match_signal 13 "$OUT" ' diff --git a/t/t3600-rm.sh b/t/t3600-rm.sh index 8e68a00dcb..a371ea690e 100755 --- a/t/t3600-rm.sh +++ b/t/t3600-rm.sh @@ -260,7 +260,7 @@ test_expect_success 'choking "git rm" should not let it die with cruft (induce S test_expect_success !MINGW 'choking "git rm" should not let it die with cruft (induce and check SIGPIPE)' ' choke_git_rm_setup && - OUT=$( ((trap "" PIPE && git rm -n "some-file-*" || echo $? 1>&3) | :) 3>&1 ) && + OUT=$( ((trap "" PIPE && git rm -n "some-file-*" && echo 0 1>&3 || echo $? 1>&3) | :) 3>&1 ) && test_match_signal 13 "$OUT" && test_path_is_missing .git/index.lock '