We have several flags like "--verbose", "--verbose-only" or "-x" that
cause us to generate shell traces. The generated tracing output is split
up in these cases so that the test's stdout is printed to file
descriptor 3 whereas its stderr is printed to file descriptor 4.
Depending on which options have been given, we then end up either:
- Redirecting both file descriptors to a file.
- Redirecting them to stdout and stderr, respectively.
- Closing them in case we're running in none-verbose mode.
The second case causes problems though when passing output to a TAP
parser. We print the test's stdout to the console's stdout, and that
results in broken TAP output.
Fix the issue by instead redirecting the test's stdout to the shell's
stderr. This makes it impossible to discern stdout from stderr, but
going by my own experience I never came across a usecase where I would
have needed this distinction.
Signed-off-by: Patrick Steinhardt <ps@xxxxxx>
---
t/t0000-basic.sh | 35 +++++++++++++++++++----------------
t/test-lib.sh | 4 ++--
2 files changed, 21 insertions(+), 18 deletions(-)
diff --git a/t/t0000-basic.sh b/t/t0000-basic.sh
index 35c5c2b4f9b..16b785f3b91 100755
--- a/t/t0000-basic.sh
+++ b/t/t0000-basic.sh
@@ -219,41 +219,44 @@ test_expect_success 'subtest: --verbose option' '
test_expect_success "failing test" false
test_done
EOF
- mv t1234-verbose/out t1234-verbose/out+ &&
- grep -v "^Initialized empty" t1234-verbose/out+ >t1234-verbose/out &&
- check_sub_test_lib_test t1234-verbose <<-\EOF
- > expecting success of 1234.1 '\''passing test'\'': true
+ mv t1234-verbose/err t1234-verbose/err+ &&
+ grep -v "^Initialized empty" t1234-verbose/err+ >t1234-verbose/err &&
+ check_sub_test_lib_test_err t1234-verbose \
+ <<-\EOF_OUT 3<<-\EOF_ERR
> ok 1 - passing test
+ > ok 2 - test with output
+ > not ok 3 - failing test
+ > # false
+ > # failed 1 among 3 test(s)
+ > 1..3
+ EOF_OUT
+ > expecting success of 1234.1 '\''passing test'\'': true
> Z
> expecting success of 1234.2 '\''test with output'\'': echo foo
> foo
- > ok 2 - test with output
> Z
> expecting success of 1234.3 '\''failing test'\'': false
- > not ok 3 - failing test
- > # false
> Z
- > # failed 1 among 3 test(s)
- > 1..3
- EOF
+ EOF_ERR
'
test_expect_success 'subtest: --verbose-only option' '
run_sub_test_lib_test_err \
t1234-verbose \
--verbose-only=2 &&
- check_sub_test_lib_test t1234-verbose <<-\EOF
+ check_sub_test_lib_test_err t1234-verbose <<-\EOF_OUT 3<<-\EOF_ERR
> ok 1 - passing test
- > Z
- > expecting success of 1234.2 '\''test with output'\'': echo foo
- > foo
> ok 2 - test with output
- > Z
> not ok 3 - failing test
> # false
> # failed 1 among 3 test(s)
> 1..3
- EOF
+ EOF_OUT
+ > Z
+ > expecting success of 1234.2 '\''test with output'\'': echo foo
+ > foo
+ > Z
+ EOF_ERR
'
test_expect_success 'subtest: skip one with GIT_SKIP_TESTS' '
diff --git a/t/test-lib.sh b/t/test-lib.sh
index af722d383d9..6ce8570226c 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -707,7 +707,7 @@ then
exec 3>>"$GIT_TEST_TEE_OUTPUT_FILE" 4>&3
elif test "$verbose" = "t"
then
- exec 4>&2 3>&1
+ exec 4>&2 3>&2
else
exec 4>/dev/null 3>/dev/null
fi
@@ -949,7 +949,7 @@ maybe_setup_verbose () {
test -z "$verbose_only" && return
if match_pattern_list $test_count "$verbose_only"
then
- exec 4>&2 3>&1
+ exec 4>&2 3>&2
# Emit a delimiting blank line when going from
# non-verbose to verbose. Within verbose mode the
# delimiter is printed by test_expect_*. The choice
--
2.49.0.1045.g170613ef41.dirty
