[PATCH v4 0/5] TAP support for Git

[PATCH v4 0/5] TAP support for Git

[Ævar Arnfjörð Bjarmason]

This is v4 of the TAP series for Git. This version fixes the failure
Junio had with it (see <7v6319duky.fsf@alter.siamese.dyndns.org>).

It does this by reverting the removal of the test_external_*
functions. Now they're made TAP aware instead with a global
`test_external_has_tap' that can be set by tests who wish to output
their own TAP.

Thus t9700-perl-git.sh will again fail if it has any output on stderr,
but since it sets `test_external_has_tap=1' we'll let it finalize its
own TAP plan.

Not removing the test_external_* makes the impact of this series
smaller, and preserves compatibility with other topics in-flight, such
as the Gettext series.

This is the diff --stat from v3:

 t/t9700-perl-git.sh |   11 ++++--
 t/test-lib.sh       |   97 ++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 101 insertions(+), 7 deletions(-)

Most of the additions are the restorations of the test_external_*
functions, in slightly altered form.

Ævar Arnfjörð Bjarmason (5):
  test-lib: Adjust output to be valid TAP format
  test-lib: Make the test_external_* functions TAP-aware
  test-lib: output a newline before "ok" under a TAP harness
  tests: Skip tests in a way that makes sense under TAP
  tests: Say "pass" rather than "ok" on empty lines for TAP

 t/README                               |   55 ++++++++++++++++-------
 t/lib-git-svn.sh                       |    4 +-
 t/lib-httpd.sh                         |    3 +-
 t/t1020-subdirectory.sh                |   12 +++---
 t/t1304-default-acl.sh                 |    9 +++-
 t/t1509-root-worktree.sh               |    6 +-
 t/t2007-checkout-symlink.sh            |    2 +-
 t/t2102-update-index-symlinks.sh       |    2 +-
 t/t3300-funny-names.sh                 |    2 +-
 t/t3302-notes-index-expensive.sh       |    2 +-
 t/t3600-rm.sh                          |    2 +-
 t/t3700-add.sh                         |   12 +++---
 t/t3701-add-interactive.sh             |    4 +-
 t/t3902-quoted.sh                      |    2 +-
 t/t4004-diff-rename-symlink.sh         |    2 +-
 t/t4011-diff-symlink.sh                |    2 +-
 t/t4016-diff-quote.sh                  |    2 +-
 t/t4023-diff-rename-typechange.sh      |    2 +-
 t/t4114-apply-typechange.sh            |    2 +-
 t/t4115-apply-symlink.sh               |    2 +-
 t/t4122-apply-symlink-inside.sh        |    2 +-
 t/t5302-pack-index.sh                  |    2 +-
 t/t5503-tagfollow.sh                   |    2 +-
 t/t5522-pull-symlink.sh                |    2 +-
 t/t5540-http-push.sh                   |    2 +-
 t/t5541-http-push.sh                   |    2 +-
 t/t5550-http-fetch.sh                  |    2 +-
 t/t5551-http-fetch.sh                  |    2 +-
 t/t5561-http-backend.sh                |    2 +-
 t/t5705-clone-2gb.sh                   |    2 +-
 t/t6035-merge-dir-to-symlink.sh        |    2 +-
 t/t7004-tag.sh                         |    2 +-
 t/t7006-pager.sh                       |    2 +-
 t/t7800-difftool.sh                    |    2 +-
 t/t9001-send-email.sh                  |    4 +-
 t/t9100-git-svn-basic.sh               |    2 +-
 t/t9119-git-svn-info.sh                |    2 +-
 t/t9129-git-svn-i18n-commitencoding.sh |    2 +-
 t/t9143-git-svn-gc.sh                  |    2 +-
 t/t9200-git-cvsexportcommit.sh         |    4 +-
 t/t9400-git-cvsserver-server.sh        |    6 +-
 t/t9401-git-cvsserver-crlf.sh          |    6 +-
 t/t9600-cvsimport.sh                   |    2 +-
 t/t9700-perl-git.sh                    |    7 ++-
 t/t9700/test.pl                        |   11 +++++
 t/test-lib.sh                          |   76 +++++++++++++++++++++++--------
 46 files changed, 178 insertions(+), 103 deletions(-)

[PATCH v4 1/5] test-lib: Adjust output to be valid TAP format

[Ævar Arnfjörð Bjarmason]

TAP, the Test Anything Protocol, is a simple text-based interface
between testing modules in a test harness. test-lib.sh's output was
already very close to being valid TAP. This change brings it all the
way there. Before:

   $ ./t0005-signals.sh
   *   ok 1: sigchain works
   * passed all 1 test(s)

And after:

   $ ./t0005-signals.sh
   ok 1 - sigchain works
   # passed all 1 test(s)
   1..1

The advantage of using TAP is that any program that reads the format
(a "test harness") can run the tests. The most popular of these is the
prove(1) utility that comes with Perl. It can run tests in parallel,
display colored output, format the output to console, file, HTML etc.,
and much more. An example:

   $ prove ./t0005-signals.sh
   ./t0005-signals.sh .. ok
   All tests successful.
   Files=1, Tests=1,  0 wallclock secs ( 0.03 usr  0.00 sys +  0.01 cusr  0.02 csys =  0.06 CPU)
   Result: PASS

prove(1) gives you human readable output without being too
verbose. Running the test suite in parallel with `make test -j15`
produces a flood of text. Running them with `prove -j 15 ./t[0-9]*.sh`
makes it easy to follow what's going on.

All this patch does is re-arrange the output a bit so that it conforms
with the TAP spec, everything that the test suite did before continues
to work. That includes aggregating results in t/test-results/, the
--verbose, --debug and other options for tests, and the test color
output.

TAP harnesses ignore everything that they don't know about, so running
the tests with --verbose works:

    $ prove ./t0005-signals.sh :: --verbose --debug
    ./t0005-signals.sh .. Terminated
    ./t0005-signals.sh .. ok
    All tests successful.
    Files=1, Tests=1,  0 wallclock secs ( 0.02 usr  0.01 sys +  0.01 cusr  0.01 csys =  0.05 CPU)
    Result: PASS

Just supply the -v option to prove itself to get all the verbose
output that it suppresses:

    $ prove -v ./t0005-signals.sh :: --verbose --debug
    ./t0005-signals.sh ..
    Initialized empty Git repository in /home/avar/g/git/t/trash directory.t0005-signals/.git/
    expecting success:
            test-sigchain >actual
            case "$?" in
            143) true ;; # POSIX w/ SIGTERM=15
              3) true ;; # Windows
              *) false ;;
            esac &&
            test_cmp expect actual
    Terminated
    ok 1 - sigchain works
    # passed all 1 test(s)
    1..1
    ok
    All tests successful.
    Files=1, Tests=1,  0 wallclock secs ( 0.02 usr  0.00 sys +  0.01 cusr  0.01 csys =  0.04 CPU)
    Result: PASS

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
---
 t/README      |   55 +++++++++++++++++++++++++++++++++++++++----------------
 t/test-lib.sh |   30 ++++++++++++++++++------------
 2 files changed, 57 insertions(+), 28 deletions(-)

diff --git a/t/README b/t/README
index 0e4e8d8..aa4ed28 100644
--- a/t/README
+++ b/t/README
@@ -18,25 +18,48 @@ The easiest way to run tests is to say "make".  This runs all
 the tests.
 
     *** t0000-basic.sh ***
-    *   ok 1: .git/objects should be empty after git-init in an empty repo.
-    *   ok 2: .git/objects should have 256 subdirectories.
-    *   ok 3: git-update-index without --add should fail adding.
+    ok 1 - .git/objects should be empty after git init in an empty repo.
+    ok 2 - .git/objects should have 3 subdirectories.
+    ok 3 - success is reported like this
     ...
-    *   ok 23: no diff after checkout and git-update-index --refresh.
-    * passed all 23 test(s)
-    *** t0100-environment-names.sh ***
-    *   ok 1: using old names should issue warnings.
-    *   ok 2: using old names but having new names should not issue warnings.
-    ...
-
-Or you can run each test individually from command line, like
-this:
-
-    $ sh ./t3001-ls-files-killed.sh
-    *   ok 1: git-update-index --add to add various paths.
-    *   ok 2: git-ls-files -k to show killed files.
-    *   ok 3: validate git-ls-files -k output.
-    * passed all 3 test(s)
+    ok 43 - very long name in the index handled sanely
+    # fixed 1 known breakage(s)
+    # still have 1 known breakage(s)
+    # passed all remaining 42 test(s)
+    1..43
+    *** t0001-init.sh ***
+    ok 1 - plain
+    ok 2 - plain with GIT_WORK_TREE
+    ok 3 - plain bare
+
+Since the tests all output TAP (see http://testanything.org) they can
+be run with any TAP harness. Here's an example of paralell testing
+powered by a recent version of prove(1):
+
+    $ prove --timer --jobs 15 ./t[0-9]*.sh
+    [19:17:33] ./t0005-signals.sh ................................... ok       36 ms
+    [19:17:33] ./t0022-crlf-rename.sh ............................... ok       69 ms
+    [19:17:33] ./t0024-crlf-archive.sh .............................. ok      154 ms
+    [19:17:33] ./t0004-unwritable.sh ................................ ok      289 ms
+    [19:17:33] ./t0002-gitfile.sh ................................... ok      480 ms
+    ===(     102;0  25/?  6/?  5/?  16/?  1/?  4/?  2/?  1/?  3/?  1... )===
+
+prove and other harnesses come with a lot of useful options. The
+--state option in particular is very useful:
+
+    # Repeat until no more failures
+    $ prove -j 15 --state=failed,save ./t[0-9]*.sh
+
+You can also run each test individually from command line, like this:
+
+    $ sh ./t3010-ls-files-killed-modified.sh
+    ok 1 - git update-index --add to add various paths.
+    ok 2 - git ls-files -k to show killed files.
+    ok 3 - validate git ls-files -k output.
+    ok 4 - git ls-files -m to show modified files.
+    ok 5 - validate git ls-files -m output.
+    # passed all 5 test(s)
+    1..5
 
 You can pass --verbose (or -v), --debug (or -d), and --immediate
 (or -i) command line argument to the test, or by setting GIT_TEST_OPTS
diff --git a/t/test-lib.sh b/t/test-lib.sh
index 367f053..e97645c 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -160,7 +160,7 @@ if test -n "$color"; then
 			*) test -n "$quiet" && return;;
 		esac
 		shift
-		printf "* %s" "$*"
+		printf "%s" "$*"
 		tput sgr0
 		echo
 		)
@@ -169,7 +169,7 @@ else
 	say_color() {
 		test -z "$1" && test -n "$quiet" && return
 		shift
-		echo "* $*"
+		echo "$*"
 	}
 fi
 
@@ -339,25 +339,25 @@ test_have_prereq () {
 
 test_ok_ () {
 	test_success=$(($test_success + 1))
-	say_color "" "  ok $test_count: $@"
+	say_color "" "ok $test_count - $@"
 }
 
 test_failure_ () {
 	test_failure=$(($test_failure + 1))
-	say_color error "FAIL $test_count: $1"
+	say_color error "not ok - $test_count $1"
 	shift
-	echo "$@" | sed -e 's/^/	/'
+	echo "$@" | sed -e 's/^/#	/'
 	test "$immediate" = "" || { GIT_EXIT_OK=t; exit 1; }
 }
 
 test_known_broken_ok_ () {
 	test_fixed=$(($test_fixed+1))
-	say_color "" "  FIXED $test_count: $@"
+	say_color "" "ok $test_count - $@ # TODO known breakage"
 }
 
 test_known_broken_failure_ () {
 	test_broken=$(($test_broken+1))
-	say_color skip "  still broken $test_count: $@"
+	say_color skip "not ok $test_count - $@ # TODO known breakage"
 }
 
 test_debug () {
@@ -390,7 +390,7 @@ test_skip () {
 	case "$to_skip" in
 	t)
 		say_color skip >&3 "skipping test: $@"
-		say_color skip "skip $test_count: $1"
+		say_color skip "ok $test_count: # skip $1"
 		: true
 		;;
 	*)
@@ -620,18 +620,22 @@ test_done () {
 
 	if test "$test_fixed" != 0
 	then
-		say_color pass "fixed $test_fixed known breakage(s)"
+		say_color pass "# fixed $test_fixed known breakage(s)"
 	fi
 	if test "$test_broken" != 0
 	then
-		say_color error "still have $test_broken known breakage(s)"
+		say_color error "# still have $test_broken known breakage(s)"
 		msg="remaining $(($test_count-$test_broken)) test(s)"
 	else
 		msg="$test_count test(s)"
 	fi
 	case "$test_failure" in
 	0)
-		say_color pass "passed all $msg"
+		# Maybe print SKIP message
+		[ -z "$skip_all" ] || skip_all=" # SKIP $skip_all"
+
+		say_color pass "# passed all $msg"
+		say "1..$test_count$skip_all"
 
 		test -d "$remove_trash" &&
 		cd "$(dirname "$remove_trash")" &&
@@ -640,7 +644,9 @@ test_done () {
 		exit 0 ;;
 
 	*)
-		say_color error "failed $test_failure among $msg"
+		say_color error "# failed $test_failure among $msg"
+		say "1..$test_count"
+
 		exit 1 ;;
 
 	esac
-- 
1.7.1.251.g92a7

[PATCH v4 2/5] test-lib: Make the test_external_* functions TAP-aware

[Ævar Arnfjörð Bjarmason]

Before TAP we just ran the Perl test and assumed that it failed if
nothing was printed on STDERR. Continue doing that, but introduce a
`test_external_has_tap' variable which tests can set to indicate that
they're outputting TAP.

If it's set we won't output a test plan, but trust the external test
to do so. That way we can make external tests work with a TAP harness,
but still maintain compatibility with test-lib's own way of tracking
tests through the test-results directory.

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
---
 t/t9700-perl-git.sh |    3 +++
 t/t9700/test.pl     |   11 +++++++++++
 t/test-lib.sh       |   51 +++++++++++++++++++++++++++++++++++++++------------
 3 files changed, 53 insertions(+), 12 deletions(-)

diff --git a/t/t9700-perl-git.sh b/t/t9700-perl-git.sh
index 8686086..0b92726 100755
--- a/t/t9700-perl-git.sh
+++ b/t/t9700-perl-git.sh
@@ -46,6 +46,9 @@ test_expect_success \
      git config --add test.int 2k
      '
 
+# The external test will outputs its own plan
+test_external_has_tap=1
+
 test_external_without_stderr \
     'Perl API' \
     "$PERL_PATH" "$TEST_DIRECTORY"/t9700/test.pl
diff --git a/t/t9700/test.pl b/t/t9700/test.pl
index 666722d..e5d4b03 100755
--- a/t/t9700/test.pl
+++ b/t/t9700/test.pl
@@ -7,6 +7,13 @@ use strict;
 
 use Test::More qw(no_plan);
 
+BEGIN {
+	# t9700-perl-git.sh kicks off our testing, so we have to go from
+	# there.
+	$Test::Builder::Test->{Curr_Test} = 1;
+	$Test::Builder::Test->{No_Ending} = 1;
+}
+
 use Cwd;
 use File::Basename;
 
@@ -105,3 +112,7 @@ my $last_commit = $r2->command_oneline(qw(rev-parse --verify HEAD));
 like($last_commit, qr/^[0-9a-fA-F]{40}$/, 'rev-parse returned hash');
 my $dir_commit = $r2->command_oneline('log', '-n1', '--pretty=format:%H', '.');
 isnt($last_commit, $dir_commit, 'log . does not show last commit');
+
+printf "1..%d\n", $Test::Builder::Test->{Curr_Test};
+
+exit($Test::Builder::Test->{Is_Passing} ? 0 : 1);
diff --git a/t/test-lib.sh b/t/test-lib.sh
index e97645c..7077210 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -206,6 +206,8 @@ test_fixed=0
 test_broken=0
 test_success=0
 
+test_external_has_tap=0
+
 die () {
 	code=$?
 	if test -n "$GIT_EXIT_OK"
@@ -456,7 +458,7 @@ test_expect_code () {
 # test_external runs external test scripts that provide continuous
 # test output about their progress, and succeeds/fails on
 # zero/non-zero exit code.  It outputs the test output on stdout even
-# in non-verbose mode, and announces the external script with "* run
+# in non-verbose mode, and announces the external script with "# run
 # <n>: ..." before running it.  When providing relative paths, keep in
 # mind that all scripts run in "trash directory".
 # Usage: test_external description command arguments...
@@ -471,7 +473,7 @@ test_external () {
 	then
 		# Announce the script to reduce confusion about the
 		# test output that follows.
-		say_color "" " run $test_count: $descr ($*)"
+		say_color "" "# run $test_count: $descr ($*)"
 		# Export TEST_DIRECTORY, TRASH_DIRECTORY and GIT_TEST_LONG
 		# to be able to use them in script
 		export TEST_DIRECTORY TRASH_DIRECTORY GIT_TEST_LONG
@@ -481,9 +483,19 @@ test_external () {
 		"$@" 2>&4
 		if [ "$?" = 0 ]
 		then
-			test_ok_ "$descr"
+			if test $test_external_has_tap -eq 0; then
+				test_ok_ "$descr"
+			else
+				say_color "" "# test_external test $descr was ok"
+				test_success=$(($test_success + 1))
+			fi
 		else
-			test_failure_ "$descr" "$@"
+			if test $test_external_has_tap -eq 0; then
+				test_failure_ "$descr" "$@"
+			else
+				say_color error "# test_external test $descr failed: $@"
+				test_failure=$(($test_failure + 1))
+			fi
 		fi
 	fi
 }
@@ -499,19 +511,30 @@ test_external_without_stderr () {
 	[ -f "$stderr" ] || error "Internal error: $stderr disappeared."
 	descr="no stderr: $1"
 	shift
-	say >&3 "expecting no stderr from previous command"
+	say >&3 "# expecting no stderr from previous command"
 	if [ ! -s "$stderr" ]; then
 		rm "$stderr"
-		test_ok_ "$descr"
+
+		if test $test_external_has_tap -eq 0; then
+			test_ok_ "$descr"
+		else
+			say_color "" "# test_external_without_stderr test $descr was ok"
+			test_success=$(($test_success + 1))
+		fi
 	else
 		if [ "$verbose" = t ]; then
-			output=`echo; echo Stderr is:; cat "$stderr"`
+			output=`echo; echo "# Stderr is:"; cat "$stderr"`
 		else
 			output=
 		fi
 		# rm first in case test_failure exits.
 		rm "$stderr"
-		test_failure_ "$descr" "$@" "$output"
+		if test $test_external_has_tap -eq 0; then
+			test_failure_ "$descr" "$@" "$output"
+		else
+			say_color error "# test_external_without_stderr test $descr failed: $@: $output"
+			test_failure=$(($test_failure + 1))
+		fi
 	fi
 }
 
@@ -634,8 +657,10 @@ test_done () {
 		# Maybe print SKIP message
 		[ -z "$skip_all" ] || skip_all=" # SKIP $skip_all"
 
-		say_color pass "# passed all $msg"
-		say "1..$test_count$skip_all"
+		if test $test_external_has_tap -eq 0; then
+			say_color pass "# passed all $msg"
+			say "1..$test_count$skip_all"
+		fi
 
 		test -d "$remove_trash" &&
 		cd "$(dirname "$remove_trash")" &&
@@ -644,8 +669,10 @@ test_done () {
 		exit 0 ;;
 
 	*)
-		say_color error "# failed $test_failure among $msg"
-		say "1..$test_count"
+		if test $test_external_has_tap -eq 0; then
+			say_color error "# failed $test_failure among $msg"
+			say "1..$test_count"
+		fi
 
 		exit 1 ;;
 
-- 
1.7.1.251.g92a7

[PATCH v4 3/5] test-lib: output a newline before "ok" under a TAP harness

[Ævar Arnfjörð Bjarmason]

Some tests in the testsuite will emit a line that doesn't end with a
newline, right before we're about to output "ok" or "not ok". This
breaks the TAP output with "Tests out of sequence" errors since a TAP
harness can't understand this:

    ok 1 - A test
    [some output here]ok 2 - Another test
    ok 3 - Yet another test

Work around it by emitting an empty line before we're about to say
"ok" or "not ok", but only if we're running under --verbose and
HARNESS_ACTIVE=1 is set, which'll only be the case when running under
a harnesses like prove(1).

I think it's better to do this than fix each tests by adding `&& echo'
everywhere. More tests might be added that break TAP in the future,
and a human isn't going to look at the extra whitespace, since
HARNESS_ACTIVE=1 always means a harness is reading it.

The tests that had issues were:

   t1007, t3410, t3413, t3409, t3414, t3415, t3416, t3412, t3404,
   t5407, t7402, t7003, t9001

With this workaround the entire test suite runs without errors under:

    prove -j 10 ./t[0-9]*.sh :: --verbose

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
---
 t/test-lib.sh |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/t/test-lib.sh b/t/test-lib.sh
index 7077210..ac496aa 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -371,6 +371,9 @@ test_run_ () {
 	eval >&3 2>&4 "$1"
 	eval_ret=$?
 	eval >&3 2>&4 "$test_cleanup"
+	if test "$verbose" = "t" && test -n "$HARNESS_ACTIVE"; then
+		echo ""
+	fi
 	return 0
 }
 
-- 
1.7.1.251.g92a7

[PATCH v4 4/5] tests: Skip tests in a way that makes sense under TAP

[Ævar Arnfjörð Bjarmason]

SKIP messages are now part of the TAP plan. A TAP harness now knows
why a particular test was skipped and can report that information. The
non-TAP harness built into Git's test-lib did nothing special with
these messages, and is unaffected by these changes.

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
---
 t/lib-git-svn.sh                       |    4 ++--
 t/lib-httpd.sh                         |    3 +--
 t/t1304-default-acl.sh                 |    9 ++++++---
 t/t1509-root-worktree.sh               |    6 +++---
 t/t2007-checkout-symlink.sh            |    2 +-
 t/t3300-funny-names.sh                 |    2 +-
 t/t3302-notes-index-expensive.sh       |    2 +-
 t/t3600-rm.sh                          |    2 +-
 t/t3701-add-interactive.sh             |    4 ++--
 t/t3902-quoted.sh                      |    2 +-
 t/t4004-diff-rename-symlink.sh         |    2 +-
 t/t4011-diff-symlink.sh                |    2 +-
 t/t4016-diff-quote.sh                  |    2 +-
 t/t4023-diff-rename-typechange.sh      |    2 +-
 t/t4114-apply-typechange.sh            |    2 +-
 t/t4115-apply-symlink.sh               |    2 +-
 t/t4122-apply-symlink-inside.sh        |    2 +-
 t/t5302-pack-index.sh                  |    2 +-
 t/t5503-tagfollow.sh                   |    2 +-
 t/t5522-pull-symlink.sh                |    2 +-
 t/t5540-http-push.sh                   |    2 +-
 t/t5541-http-push.sh                   |    2 +-
 t/t5550-http-fetch.sh                  |    2 +-
 t/t5551-http-fetch.sh                  |    2 +-
 t/t5561-http-backend.sh                |    2 +-
 t/t5705-clone-2gb.sh                   |    2 +-
 t/t6035-merge-dir-to-symlink.sh        |    2 +-
 t/t7004-tag.sh                         |    2 +-
 t/t7006-pager.sh                       |    2 +-
 t/t7800-difftool.sh                    |    2 +-
 t/t9001-send-email.sh                  |    4 ++--
 t/t9100-git-svn-basic.sh               |    2 +-
 t/t9119-git-svn-info.sh                |    2 +-
 t/t9129-git-svn-i18n-commitencoding.sh |    2 +-
 t/t9143-git-svn-gc.sh                  |    2 +-
 t/t9200-git-cvsexportcommit.sh         |    4 ++--
 t/t9400-git-cvsserver-server.sh        |    6 +++---
 t/t9401-git-cvsserver-crlf.sh          |    6 +++---
 t/t9600-cvsimport.sh                   |    2 +-
 t/t9700-perl-git.sh                    |    4 ++--
 40 files changed, 56 insertions(+), 54 deletions(-)

diff --git a/t/lib-git-svn.sh b/t/lib-git-svn.sh
index 0f7f35c..344785d 100644
--- a/t/lib-git-svn.sh
+++ b/t/lib-git-svn.sh
@@ -5,11 +5,11 @@ git_svn_id=git""-svn-id
 
 if test -n "$NO_SVN_TESTS"
 then
-	say 'skipping git svn tests, NO_SVN_TESTS defined'
+	skip_all='skipping git svn tests, NO_SVN_TESTS defined'
 	test_done
 fi
 if ! test_have_prereq PERL; then
-	say 'skipping git svn tests, perl not available'
+	skip_all='skipping git svn tests, perl not available'
 	test_done
 fi
 
diff --git a/t/lib-httpd.sh b/t/lib-httpd.sh
index da4b8d5..a0944d6 100644
--- a/t/lib-httpd.sh
+++ b/t/lib-httpd.sh
@@ -5,8 +5,7 @@
 
 if test -z "$GIT_TEST_HTTPD"
 then
-	say "skipping test, network testing disabled by default"
-	say "(define GIT_TEST_HTTPD to enable)"
+	skip_all="Network testing disabled (define GIT_TEST_HTTPD to enable)"
 	test_done
 fi
 
diff --git a/t/t1304-default-acl.sh b/t/t1304-default-acl.sh
index 055ad00..97ab02a 100755
--- a/t/t1304-default-acl.sh
+++ b/t/t1304-default-acl.sh
@@ -15,9 +15,12 @@ umask 077
 # is a good candidate: exists on all unices, and it has permission
 # anyway, so we don't create a security hole running the testsuite.
 
-if ! setfacl -m u:root:rwx .; then
-    say "Skipping ACL tests: unable to use setfacl"
-    test_done
+setfacl_out="$(setfacl -m u:root:rwx . 2>&1)"
+setfacl_ret=$?
+
+if [ $setfacl_ret != 0 ]; then
+	skip_all="Skipping ACL tests: unable to use setfacl (output: '$setfacl_out'; return code: '$setfacl_ret')"
+	test_done
 fi
 
 check_perms_and_acl () {
diff --git a/t/t1509-root-worktree.sh b/t/t1509-root-worktree.sh
index 5322a3b..7f60fd0 100755
--- a/t/t1509-root-worktree.sh
+++ b/t/t1509-root-worktree.sh
@@ -99,17 +99,17 @@ test_foobar_foobar() {
 }
 
 if ! test_have_prereq POSIXPERM || ! [ -w / ]; then
-	say "Dangerous test skipped. Read this test if you want to execute it"
+	skip_all="Dangerous test skipped. Read this test if you want to execute it"
 	test_done
 fi
 
 if [ "$IKNOWWHATIAMDOING" != "YES" ]; then
-	say "You must set env var IKNOWWHATIAMDOING=YES in order to run this test"
+	skip_all="You must set env var IKNOWWHATIAMDOING=YES in order to run this test"
 	test_done
 fi
 
 if [ "$UID" = 0 ]; then
-	say "No you can't run this with root"
+	skip_all="No you can't run this with root"
 	test_done
 fi
 
diff --git a/t/t2007-checkout-symlink.sh b/t/t2007-checkout-symlink.sh
index 27e2127..05cc8fd 100755
--- a/t/t2007-checkout-symlink.sh
+++ b/t/t2007-checkout-symlink.sh
@@ -8,7 +8,7 @@ test_description='git checkout to switch between branches with symlink<->dir'
 
 if ! test_have_prereq SYMLINKS
 then
-	say "symbolic links not supported - skipping tests"
+	skip_all="symbolic links not supported - skipping tests"
 	test_done
 fi
 
diff --git a/t/t3300-funny-names.sh b/t/t3300-funny-names.sh
index db46d53..a99e4d8 100755
--- a/t/t3300-funny-names.sh
+++ b/t/t3300-funny-names.sh
@@ -26,7 +26,7 @@ echo 'Foo Bar Baz' >"$p2"
 
 test -f "$p1" && cmp "$p0" "$p1" || {
 	# since FAT/NTFS does not allow tabs in filenames, skip this test
-	say 'Your filesystem does not allow tabs in filenames, test skipped.'
+	skip_all='Your filesystem does not allow tabs in filenames, test skipped.'
 	test_done
 }
 
diff --git a/t/t3302-notes-index-expensive.sh b/t/t3302-notes-index-expensive.sh
index ee84fc4..361a10a 100755
--- a/t/t3302-notes-index-expensive.sh
+++ b/t/t3302-notes-index-expensive.sh
@@ -8,7 +8,7 @@ test_description='Test commit notes index (expensive!)'
 . ./test-lib.sh
 
 test -z "$GIT_NOTES_TIMING_TESTS" && {
-	say Skipping timing tests
+	skip_all="Skipping timing tests"
 	test_done
 	exit
 }
diff --git a/t/t3600-rm.sh b/t/t3600-rm.sh
index 0aaf0ad..b514cbb 100755
--- a/t/t3600-rm.sh
+++ b/t/t3600-rm.sh
@@ -39,7 +39,7 @@ if test -f test-file
 then
 	test_set_prereq RO_DIR
 else
-	say 'skipping removal failure test (perhaps running as root?)'
+	skip_all='skipping removal failure test (perhaps running as root?)'
 fi
 chmod 775 .
 rm -f test-file
diff --git a/t/t3701-add-interactive.sh b/t/t3701-add-interactive.sh
index b6eba6a..7ad8465 100755
--- a/t/t3701-add-interactive.sh
+++ b/t/t3701-add-interactive.sh
@@ -4,7 +4,7 @@ test_description='add -i basic tests'
 . ./test-lib.sh
 
 if ! test_have_prereq PERL; then
-	say 'skipping git add -i tests, perl not available'
+	skip_all='skipping git add -i tests, perl not available'
 	test_done
 fi
 
@@ -154,7 +154,7 @@ rm -f .gitignore
 
 if test "$(git config --bool core.filemode)" = false
 then
-	say 'skipping filemode tests (filesystem does not properly support modes)'
+	say '# skipping filemode tests (filesystem does not properly support modes)'
 else
 	test_set_prereq FILEMODE
 fi
diff --git a/t/t3902-quoted.sh b/t/t3902-quoted.sh
index 29103f6..147e634 100755
--- a/t/t3902-quoted.sh
+++ b/t/t3902-quoted.sh
@@ -17,7 +17,7 @@ DQ='"'
 echo foo 2>/dev/null > "Name and an${HT}HT"
 test -f "Name and an${HT}HT" || {
 	# since FAT/NTFS does not allow tabs in filenames, skip this test
-	say 'Your filesystem does not allow tabs in filenames, test skipped.'
+	skip_all='Your filesystem does not allow tabs in filenames, test skipped.'
 	test_done
 }
 
diff --git a/t/t4004-diff-rename-symlink.sh b/t/t4004-diff-rename-symlink.sh
index a4da119..1a09e8d 100755
--- a/t/t4004-diff-rename-symlink.sh
+++ b/t/t4004-diff-rename-symlink.sh
@@ -14,7 +14,7 @@ by an edit for them.
 
 if ! test_have_prereq SYMLINKS
 then
-	say 'Symbolic links not supported, skipping tests.'
+	skip_all='Symbolic links not supported, skipping tests.'
 	test_done
 fi
 
diff --git a/t/t4011-diff-symlink.sh b/t/t4011-diff-symlink.sh
index e12fbea..918a21a 100755
--- a/t/t4011-diff-symlink.sh
+++ b/t/t4011-diff-symlink.sh
@@ -11,7 +11,7 @@ test_description='Test diff of symlinks.
 
 if ! test_have_prereq SYMLINKS
 then
-	say 'Symbolic links not supported, skipping tests.'
+	skip_all='Symbolic links not supported, skipping tests.'
 	test_done
 fi
 
diff --git a/t/t4016-diff-quote.sh b/t/t4016-diff-quote.sh
index 55eb5f8..34e5144 100755
--- a/t/t4016-diff-quote.sh
+++ b/t/t4016-diff-quote.sh
@@ -14,7 +14,7 @@ P2='pathname with SP'
 P3='pathname
 with LF'
 : 2>/dev/null >"$P1" && test -f "$P1" && rm -f "$P1" || {
-	say 'Your filesystem does not allow tabs in filenames, test skipped.'
+	skip_all='Your filesystem does not allow tabs in filenames, test skipped.'
 	test_done
 }
 
diff --git a/t/t4023-diff-rename-typechange.sh b/t/t4023-diff-rename-typechange.sh
index 9bdf659..40a95a1 100755
--- a/t/t4023-diff-rename-typechange.sh
+++ b/t/t4023-diff-rename-typechange.sh
@@ -6,7 +6,7 @@ test_description='typechange rename detection'
 
 if ! test_have_prereq SYMLINKS
 then
-	say 'Symbolic links not supported, skipping tests.'
+	skip_all='Symbolic links not supported, skipping tests.'
 	test_done
 fi
 
diff --git a/t/t4114-apply-typechange.sh b/t/t4114-apply-typechange.sh
index 99ec13d..164d58c 100755
--- a/t/t4114-apply-typechange.sh
+++ b/t/t4114-apply-typechange.sh
@@ -11,7 +11,7 @@ test_description='git apply should not get confused with type changes.
 
 if ! test_have_prereq SYMLINKS
 then
-	say 'Symbolic links not supported, skipping tests.'
+	skip_all='Symbolic links not supported, skipping tests.'
 	test_done
 fi
 
diff --git a/t/t4115-apply-symlink.sh b/t/t4115-apply-symlink.sh
index b852e58..aff4348 100755
--- a/t/t4115-apply-symlink.sh
+++ b/t/t4115-apply-symlink.sh
@@ -11,7 +11,7 @@ test_description='git apply symlinks and partial files
 
 if ! test_have_prereq SYMLINKS
 then
-	say 'Symbolic links not supported, skipping tests.'
+	skip_all='Symbolic links not supported, skipping tests.'
 	test_done
 fi
 
diff --git a/t/t4122-apply-symlink-inside.sh b/t/t4122-apply-symlink-inside.sh
index 0d3c1d5..923fcab 100755
--- a/t/t4122-apply-symlink-inside.sh
+++ b/t/t4122-apply-symlink-inside.sh
@@ -5,7 +5,7 @@ test_description='apply to deeper directory without getting fooled with symlink'
 
 if ! test_have_prereq SYMLINKS
 then
-	say 'Symbolic links not supported, skipping tests.'
+	skip_all='Symbolic links not supported, skipping tests.'
 	test_done
 fi
 
diff --git a/t/t5302-pack-index.sh b/t/t5302-pack-index.sh
index 4360e77..fb3a270 100755
--- a/t/t5302-pack-index.sh
+++ b/t/t5302-pack-index.sh
@@ -74,7 +74,7 @@ if msg=$(git verify-pack -v "test-3-${pack3}.pack" 2>&1) ||
 then
 	test_set_prereq OFF64_T
 else
-	say "skipping tests concerning 64-bit offsets"
+	say "# skipping tests concerning 64-bit offsets"
 fi
 
 test_expect_success OFF64_T \
diff --git a/t/t5503-tagfollow.sh b/t/t5503-tagfollow.sh
index d5db75d..bab1a53 100755
--- a/t/t5503-tagfollow.sh
+++ b/t/t5503-tagfollow.sh
@@ -6,7 +6,7 @@ test_description='test automatic tag following'
 
 case $(uname -s) in
 *MINGW*)
-	say "GIT_DEBUG_SEND_PACK not supported - skipping tests"
+	skip_all="GIT_DEBUG_SEND_PACK not supported - skipping tests"
 	test_done
 esac
 
diff --git a/t/t5522-pull-symlink.sh b/t/t5522-pull-symlink.sh
index 7206817..298200f 100755
--- a/t/t5522-pull-symlink.sh
+++ b/t/t5522-pull-symlink.sh
@@ -6,7 +6,7 @@ test_description='pulling from symlinked subdir'
 
 if ! test_have_prereq SYMLINKS
 then
-	say 'Symbolic links not supported, skipping tests.'
+	skip_all='Symbolic links not supported, skipping tests.'
 	test_done
 fi
 
diff --git a/t/t5540-http-push.sh b/t/t5540-http-push.sh
index 37fe875..a266ca5 100755
--- a/t/t5540-http-push.sh
+++ b/t/t5540-http-push.sh
@@ -11,7 +11,7 @@ This test runs various sanity checks on http-push.'
 
 if git http-push > /dev/null 2>&1 || [ $? -eq 128 ]
 then
-	say "skipping test, USE_CURL_MULTI is not defined"
+	skip_all="skipping test, USE_CURL_MULTI is not defined"
 	test_done
 fi
 
diff --git a/t/t5541-http-push.sh b/t/t5541-http-push.sh
index 17e1bdc..504884b 100755
--- a/t/t5541-http-push.sh
+++ b/t/t5541-http-push.sh
@@ -7,7 +7,7 @@ test_description='test smart pushing over http via http-backend'
 . ./test-lib.sh
 
 if test -n "$NO_CURL"; then
-	say 'skipping test, git built without http support'
+	skip_all='skipping test, git built without http support'
 	test_done
 fi
 
diff --git a/t/t5550-http-fetch.sh b/t/t5550-http-fetch.sh
index fc675b5..2fb48d0 100755
--- a/t/t5550-http-fetch.sh
+++ b/t/t5550-http-fetch.sh
@@ -4,7 +4,7 @@ test_description='test dumb fetching over http via static file'
 . ./test-lib.sh
 
 if test -n "$NO_CURL"; then
-	say 'skipping test, git built without http support'
+	skip_all='skipping test, git built without http support'
 	test_done
 fi
 
diff --git a/t/t5551-http-fetch.sh b/t/t5551-http-fetch.sh
index 7faa31a..fd19121 100755
--- a/t/t5551-http-fetch.sh
+++ b/t/t5551-http-fetch.sh
@@ -4,7 +4,7 @@ test_description='test smart fetching over http via http-backend'
 . ./test-lib.sh
 
 if test -n "$NO_CURL"; then
-	say 'skipping test, git built without http support'
+	skip_all='skipping test, git built without http support'
 	test_done
 fi
 
diff --git a/t/t5561-http-backend.sh b/t/t5561-http-backend.sh
index 8c6d0b2..b5d7fbc 100755
--- a/t/t5561-http-backend.sh
+++ b/t/t5561-http-backend.sh
@@ -4,7 +4,7 @@ test_description='test git-http-backend'
 . ./test-lib.sh
 
 if test -n "$NO_CURL"; then
-	say 'skipping test, git built without http support'
+	skip_all='skipping test, git built without http support'
 	test_done
 fi
 
diff --git a/t/t5705-clone-2gb.sh b/t/t5705-clone-2gb.sh
index 8afbdd4..e4d1b6a 100755
--- a/t/t5705-clone-2gb.sh
+++ b/t/t5705-clone-2gb.sh
@@ -4,7 +4,7 @@ test_description='Test cloning a repository larger than 2 gigabyte'
 . ./test-lib.sh
 
 test -z "$GIT_TEST_CLONE_2GB" &&
-say "Skipping expensive 2GB clone test; enable it with GIT_TEST_CLONE_2GB=t" &&
+skip_all="Skipping expensive 2GB clone test; enable it with GIT_TEST_CLONE_2GB=t" &&
 test_done &&
 exit
 
diff --git a/t/t6035-merge-dir-to-symlink.sh b/t/t6035-merge-dir-to-symlink.sh
index 3202e1d..cd3190c 100755
--- a/t/t6035-merge-dir-to-symlink.sh
+++ b/t/t6035-merge-dir-to-symlink.sh
@@ -5,7 +5,7 @@ test_description='merging when a directory was replaced with a symlink'
 
 if ! test_have_prereq SYMLINKS
 then
-	say 'Symbolic links not supported, skipping tests.'
+	skip_all='Symbolic links not supported, skipping tests.'
 	test_done
 fi
 
diff --git a/t/t7004-tag.sh b/t/t7004-tag.sh
index 73dbc43..ac943f5 100755
--- a/t/t7004-tag.sh
+++ b/t/t7004-tag.sh
@@ -583,7 +583,7 @@ test_expect_success \
 # subsequent tests require gpg; check if it is available
 gpg --version >/dev/null 2>/dev/null
 if [ $? -eq 127 ]; then
-	say "gpg not found - skipping tag signing and verification tests"
+	say "# gpg not found - skipping tag signing and verification tests"
 else
 	# As said here: http://www.gnupg.org/documentation/faqs.html#q6.19
 	# the gpg version 1.0.6 didn't parse trust packets correctly, so for
diff --git a/t/t7006-pager.sh b/t/t7006-pager.sh
index c2a3c8e..eb9651d 100755
--- a/t/t7006-pager.sh
+++ b/t/t7006-pager.sh
@@ -37,7 +37,7 @@ then
 	}
 	test_set_prereq TTY
 else
-	say no usable terminal, so skipping some tests
+	say "# no usable terminal, so skipping some tests"
 fi
 
 test_expect_success 'setup' '
diff --git a/t/t7800-difftool.sh b/t/t7800-difftool.sh
index 1de83ef..196827e 100755
--- a/t/t7800-difftool.sh
+++ b/t/t7800-difftool.sh
@@ -11,7 +11,7 @@ Testing basic diff tool invocation
 . ./test-lib.sh
 
 if ! test_have_prereq PERL; then
-	say 'skipping difftool tests, perl not available'
+	skip_all='skipping difftool tests, perl not available'
 	test_done
 fi
 
diff --git a/t/t9001-send-email.sh b/t/t9001-send-email.sh
index 0b8a591..37610db 100755
--- a/t/t9001-send-email.sh
+++ b/t/t9001-send-email.sh
@@ -4,7 +4,7 @@ test_description='git send-email'
 . ./test-lib.sh
 
 if ! test_have_prereq PERL; then
-	say 'skipping git send-email tests, perl not available'
+	skip_all='skipping git send-email tests, perl not available'
 	test_done
 fi
 
@@ -58,7 +58,7 @@ test_no_confirm () {
 # Exit immediately to prevent hang if a no-confirm test fails
 check_no_confirm () {
 	test -f no_confirm_okay || {
-		say 'No confirm test failed; skipping remaining tests to prevent hanging'
+		skip_all='confirm test failed; skipping remaining tests to prevent hanging'
 		test_done
 	}
 }
diff --git a/t/t9100-git-svn-basic.sh b/t/t9100-git-svn-basic.sh
index 570e035..13766ab 100755
--- a/t/t9100-git-svn-basic.sh
+++ b/t/t9100-git-svn-basic.sh
@@ -15,7 +15,7 @@ case "$GIT_SVN_LC_ALL" in
 	test_set_prereq UTF8
 	;;
 *)
-	say "UTF-8 locale not set, some tests skipped ($GIT_SVN_LC_ALL)"
+	say "# UTF-8 locale not set, some tests skipped ($GIT_SVN_LC_ALL)"
 	;;
 esac
 
diff --git a/t/t9119-git-svn-info.sh b/t/t9119-git-svn-info.sh
index a9a558d..5fb94fb 100755
--- a/t/t9119-git-svn-info.sh
+++ b/t/t9119-git-svn-info.sh
@@ -13,7 +13,7 @@ case $v in
 1.[456].*)
 	;;
 *)
-	say "skipping svn-info test (SVN version: $v not supported)"
+	skip_all="skipping svn-info test (SVN version: $v not supported)"
 	test_done
 	;;
 esac
diff --git a/t/t9129-git-svn-i18n-commitencoding.sh b/t/t9129-git-svn-i18n-commitencoding.sh
index 1e9a2eb..8cfdfe7 100755
--- a/t/t9129-git-svn-i18n-commitencoding.sh
+++ b/t/t9129-git-svn-i18n-commitencoding.sh
@@ -23,7 +23,7 @@ if test -n "$a_utf8_locale"
 then
 	test_set_prereq UTF8
 else
-	say "UTF-8 locale not available, some tests are skipped"
+	say "# UTF-8 locale not available, some tests are skipped"
 fi
 
 compare_svn_head_with () {
diff --git a/t/t9143-git-svn-gc.sh b/t/t9143-git-svn-gc.sh
index 99f69c6..337ea59 100755
--- a/t/t9143-git-svn-gc.sh
+++ b/t/t9143-git-svn-gc.sh
@@ -43,7 +43,7 @@ then
 		 gunzip .git/svn/refs/remotes/git-svn/unhandled.log.gz
 		'
 else
-	say "Perl Compress::Zlib unavailable, skipping gunzip test"
+	say "# Perl Compress::Zlib unavailable, skipping gunzip test"
 fi
 
 test_expect_success 'git svn gc does not change unhandled.log files' '
diff --git a/t/t9200-git-cvsexportcommit.sh b/t/t9200-git-cvsexportcommit.sh
index 61bcb8f..ee39b36 100755
--- a/t/t9200-git-cvsexportcommit.sh
+++ b/t/t9200-git-cvsexportcommit.sh
@@ -7,14 +7,14 @@ test_description='Test export of commits to CVS'
 . ./test-lib.sh
 
 if ! test_have_prereq PERL; then
-	say 'skipping git cvsexportcommit tests, perl not available'
+	skip_all='skipping git cvsexportcommit tests, perl not available'
 	test_done
 fi
 
 cvs >/dev/null 2>&1
 if test $? -ne 1
 then
-    say 'skipping git cvsexportcommit tests, cvs not found'
+    skip_all='skipping git cvsexportcommit tests, cvs not found'
     test_done
 fi
 
diff --git a/t/t9400-git-cvsserver-server.sh b/t/t9400-git-cvsserver-server.sh
index 8639506..36c457e 100755
--- a/t/t9400-git-cvsserver-server.sh
+++ b/t/t9400-git-cvsserver-server.sh
@@ -11,17 +11,17 @@ cvs CLI client via git-cvsserver server'
 . ./test-lib.sh
 
 if ! test_have_prereq PERL; then
-	say 'skipping git cvsserver tests, perl not available'
+	skip_all='skipping git cvsserver tests, perl not available'
 	test_done
 fi
 cvs >/dev/null 2>&1
 if test $? -ne 1
 then
-    say 'skipping git-cvsserver tests, cvs not found'
+    skip_all='skipping git-cvsserver tests, cvs not found'
     test_done
 fi
 "$PERL_PATH" -e 'use DBI; use DBD::SQLite' >/dev/null 2>&1 || {
-    say 'skipping git-cvsserver tests, Perl SQLite interface unavailable'
+    skip_all='skipping git-cvsserver tests, Perl SQLite interface unavailable'
     test_done
 }
 
diff --git a/t/t9401-git-cvsserver-crlf.sh b/t/t9401-git-cvsserver-crlf.sh
index ed7b513..925bd0f 100755
--- a/t/t9401-git-cvsserver-crlf.sh
+++ b/t/t9401-git-cvsserver-crlf.sh
@@ -41,16 +41,16 @@ not_present() {
 cvs >/dev/null 2>&1
 if test $? -ne 1
 then
-    say 'skipping git-cvsserver tests, cvs not found'
+    skip_all='skipping git-cvsserver tests, cvs not found'
     test_done
 fi
 if ! test_have_prereq PERL
 then
-    say 'skipping git-cvsserver tests, perl not available'
+    skip_all='skipping git-cvsserver tests, perl not available'
     test_done
 fi
 "$PERL_PATH" -e 'use DBI; use DBD::SQLite' >/dev/null 2>&1 || {
-    say 'skipping git-cvsserver tests, Perl SQLite interface unavailable'
+    skip_all='skipping git-cvsserver tests, Perl SQLite interface unavailable'
     test_done
 }
 
diff --git a/t/t9600-cvsimport.sh b/t/t9600-cvsimport.sh
index b572ce3..2eff9cd 100755
--- a/t/t9600-cvsimport.sh
+++ b/t/t9600-cvsimport.sh
@@ -4,7 +4,7 @@ test_description='git cvsimport basic tests'
 . ./lib-cvs.sh
 
 if ! test_have_prereq PERL; then
-	say 'skipping git cvsimport tests, perl not available'
+	skip_all='skipping git cvsimport tests, perl not available'
 	test_done
 fi
 
diff --git a/t/t9700-perl-git.sh b/t/t9700-perl-git.sh
index 0b92726..3787186 100755
--- a/t/t9700-perl-git.sh
+++ b/t/t9700-perl-git.sh
@@ -7,12 +7,12 @@ test_description='perl interface (Git.pm)'
 . ./test-lib.sh
 
 if ! test_have_prereq PERL; then
-	say 'skipping perl interface tests, perl not available'
+	skip_all='skipping perl interface tests, perl not available'
 	test_done
 fi
 
 "$PERL_PATH" -MTest::More -e 0 2>/dev/null || {
-	say "Perl Test::More unavailable, skipping test"
+	skip_all="Perl Test::More unavailable, skipping test"
 	test_done
 }
 
-- 
1.7.1.251.g92a7

[PATCH v4 5/5] tests: Say "pass" rather than "ok" on empty lines for TAP

[Ævar Arnfjörð Bjarmason]

Lines that begin with "ok" confuse the TAP harness because it can't
distinguish them from a test counter. Work around the issue by saying
"pass" instead, which isn't a reserved TAP word.

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
---
 t/t1020-subdirectory.sh          |   12 ++++++------
 t/t2102-update-index-symlinks.sh |    2 +-
 t/t3700-add.sh                   |   12 ++++++------
 3 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/t/t1020-subdirectory.sh b/t/t1020-subdirectory.sh
index 210e594..5687499 100755
--- a/t/t1020-subdirectory.sh
+++ b/t/t1020-subdirectory.sh
@@ -24,18 +24,18 @@ test_expect_success 'update-index and ls-files' '
 	cd "$HERE" &&
 	git update-index --add one &&
 	case "`git ls-files`" in
-	one) echo ok one ;;
+	one) echo pass one ;;
 	*) echo bad one; exit 1 ;;
 	esac &&
 	cd dir &&
 	git update-index --add two &&
 	case "`git ls-files`" in
-	two) echo ok two ;;
+	two) echo pass two ;;
 	*) echo bad two; exit 1 ;;
 	esac &&
 	cd .. &&
 	case "`git ls-files`" in
-	dir/two"$LF"one) echo ok both ;;
+	dir/two"$LF"one) echo pass both ;;
 	*) echo bad; exit 1 ;;
 	esac
 '
@@ -58,17 +58,17 @@ test_expect_success 'diff-files' '
 	echo a >>one &&
 	echo d >>dir/two &&
 	case "`git diff-files --name-only`" in
-	dir/two"$LF"one) echo ok top ;;
+	dir/two"$LF"one) echo pass top ;;
 	*) echo bad top; exit 1 ;;
 	esac &&
 	# diff should not omit leading paths
 	cd dir &&
 	case "`git diff-files --name-only`" in
-	dir/two"$LF"one) echo ok subdir ;;
+	dir/two"$LF"one) echo pass subdir ;;
 	*) echo bad subdir; exit 1 ;;
 	esac &&
 	case "`git diff-files --name-only .`" in
-	dir/two) echo ok subdir limited ;;
+	dir/two) echo pass subdir limited ;;
 	*) echo bad subdir limited; exit 1 ;;
 	esac
 '
diff --git a/t/t2102-update-index-symlinks.sh b/t/t2102-update-index-symlinks.sh
index 1ed44ee..4d0d0a3 100755
--- a/t/t2102-update-index-symlinks.sh
+++ b/t/t2102-update-index-symlinks.sh
@@ -24,7 +24,7 @@ git update-index symlink'
 test_expect_success \
 'the index entry must still be a symbolic link' '
 case "`git ls-files --stage --cached symlink`" in
-120000" "*symlink) echo ok;;
+120000" "*symlink) echo pass;;
 *) echo fail; git ls-files --stage --cached symlink; (exit 1);;
 esac'
 
diff --git a/t/t3700-add.sh b/t/t3700-add.sh
index 525c9a8..6f031af 100755
--- a/t/t3700-add.sh
+++ b/t/t3700-add.sh
@@ -26,7 +26,7 @@ test_expect_success \
 	 chmod 755 xfoo1 &&
 	 git add xfoo1 &&
 	 case "`git ls-files --stage xfoo1`" in
-	 100644" "*xfoo1) echo ok;;
+	 100644" "*xfoo1) echo pass;;
 	 *) echo fail; git ls-files --stage xfoo1; (exit 1);;
 	 esac'
 
@@ -35,7 +35,7 @@ test_expect_success SYMLINKS 'git add: filemode=0 should not get confused by sym
 	ln -s foo xfoo1 &&
 	git add xfoo1 &&
 	case "`git ls-files --stage xfoo1`" in
-	120000" "*xfoo1) echo ok;;
+	120000" "*xfoo1) echo pass;;
 	*) echo fail; git ls-files --stage xfoo1; (exit 1);;
 	esac
 '
@@ -47,7 +47,7 @@ test_expect_success \
 	 chmod 755 xfoo2 &&
 	 git update-index --add xfoo2 &&
 	 case "`git ls-files --stage xfoo2`" in
-	 100644" "*xfoo2) echo ok;;
+	 100644" "*xfoo2) echo pass;;
 	 *) echo fail; git ls-files --stage xfoo2; (exit 1);;
 	 esac'
 
@@ -56,7 +56,7 @@ test_expect_success SYMLINKS 'git add: filemode=0 should not get confused by sym
 	ln -s foo xfoo2 &&
 	git update-index --add xfoo2 &&
 	case "`git ls-files --stage xfoo2`" in
-	120000" "*xfoo2) echo ok;;
+	120000" "*xfoo2) echo pass;;
 	*) echo fail; git ls-files --stage xfoo2; (exit 1);;
 	esac
 '
@@ -67,7 +67,7 @@ test_expect_success SYMLINKS \
 	 ln -s xfoo2 xfoo3 &&
 	 git update-index --add xfoo3 &&
 	 case "`git ls-files --stage xfoo3`" in
-	 120000" "*xfoo3) echo ok;;
+	 120000" "*xfoo3) echo pass;;
 	 *) echo fail; git ls-files --stage xfoo3; (exit 1);;
 	 esac'
 
@@ -172,7 +172,7 @@ test_expect_success 'git add --refresh' '
 	test -z "`git diff-index HEAD -- foo`" &&
 	git read-tree HEAD &&
 	case "`git diff-index HEAD -- foo`" in
-	:100644" "*"M	foo") echo ok;;
+	:100644" "*"M	foo") echo pass;;
 	*) echo fail; (exit 1);;
 	esac &&
 	git add --refresh -- foo &&
-- 
1.7.1.251.g92a7

Re: [PATCH v4 1/5] test-lib: Adjust output to be valid TAP format

[Jakub Narebski]

Ævar Arnfjörð Bjarmason  <avarab@gmail.com> writes:

> TAP, the Test Anything Protocol, is a simple text-based interface
> between testing modules in a test harness. test-lib.sh's output was
> already very close to being valid TAP. This change brings it all the
> way there. Before:
> 
>    $ ./t0005-signals.sh
>    *   ok 1: sigchain works
>    * passed all 1 test(s)
> 
> And after:
> 
>    $ ./t0005-signals.sh
>    ok 1 - sigchain works
>    # passed all 1 test(s)
>    1..1

This is incomplete information.  It only tells us how the output is
changed for the case when all test_expect_success tests passes.  This
commit message doesn't tell us how failing tests looks like, and how
test_expect_failure results gets translated.

Take for example a following test script:
-- >8 --
#!/bin/sh

test_description='this is a sample test.

This test is here to see various test outputs.'

. ./test-lib.sh

say 'diagnostic message'

test_expect_success 'true  test' 'true'
test_expect_success 'false test' 'false'

test_expect_failure 'true  test (todo)' 'true'
test_expect_failure 'false test (todo)' 'false'

test_debug 'echo "debug message"'

test_done
-- 8< --

This test script output looks like the following (the comments are not
part of output, but denote color of given line of test output):
----
* diagnostic message                      # yellow
*   ok 1: true  test
* FAIL 2: false test                      # bold red
        false
*   FIXED 3: true  test (todo)
*   still broken 4: false test (todo)     # bold green
* fixed 1 known breakage(s)               # green
* still have 1 known breakage(s)          # bold red
* failed 1 among remaining 3 test(s)      # bold red
----

How would the output of this test look like after TAP-ification?
Would it still provide color output when run on terminal?

This test also generates summary of test run in a file in
test-results/ subdirectory.  Currently such file can be e.g. named
test-results/test_test-8030, and consist of
-- >8 --
total 4
success 1
fixed 1
broken 1
failed 1

-- 8< --

Would TAP-ification change that?  Even if there is no change, this is
worth a sentence or a few words in a commit message.

P.S. With current output pass / fail results of test_expect_failure
are justified on ':' separator.  This is not the case of TAP output.
-- 
Jakub Narebski
Poland
ShadeHawk on #git

Re: [PATCH v4 1/5] test-lib: Adjust output to be valid TAP format

[Ævar Arnfjörð Bjarmason]

On Thu, Jun 24, 2010 at 18:39, Jakub Narebski <jnareb@gmail.com> wrote:
> Ævar Arnfjörð Bjarmason  <avarab@gmail.com> writes:
>
>> TAP, the Test Anything Protocol, is a simple text-based interface
>> between testing modules in a test harness. test-lib.sh's output was
>> already very close to being valid TAP. This change brings it all the
>> way there. Before:
>>
>>    $ ./t0005-signals.sh
>>    *   ok 1: sigchain works
>>    * passed all 1 test(s)
>>
>> And after:
>>
>>    $ ./t0005-signals.sh
>>    ok 1 - sigchain works
>>    # passed all 1 test(s)
>>    1..1
>
> This is incomplete information.  It only tells us how the output is
> changed for the case when all test_expect_success tests passes.  This
> commit message doesn't tell us how failing tests looks like, and how
> test_expect_failure results gets translated.
>
> Take for example a following test script:
> -- >8 --
> #!/bin/sh
>
> test_description='this is a sample test.
>
> This test is here to see various test outputs.'
>
> . ./test-lib.sh
>
> say 'diagnostic message'
>
> test_expect_success 'true  test' 'true'
> test_expect_success 'false test' 'false'
>
> test_expect_failure 'true  test (todo)' 'true'
> test_expect_failure 'false test (todo)' 'false'
>
> test_debug 'echo "debug message"'
>
> test_done
> -- 8< --
>
> This test script output looks like the following (the comments are not
> part of output, but denote color of given line of test output):
> ----
> * diagnostic message                      # yellow
> *   ok 1: true  test
> * FAIL 2: false test                      # bold red
>        false
> *   FIXED 3: true  test (todo)
> *   still broken 4: false test (todo)     # bold green
> * fixed 1 known breakage(s)               # green
> * still have 1 known breakage(s)          # bold red
> * failed 1 among remaining 3 test(s)      # bold red
> ----
>
> How would the output of this test look like after TAP-ification?
> Would it still provide color output when run on terminal?
>
> This test also generates summary of test run in a file in
> test-results/ subdirectory.  Currently such file can be e.g. named
> test-results/test_test-8030, and consist of
> -- >8 --
> total 4
> success 1
> fixed 1
> broken 1
> failed 1
>
> -- 8< --
>
> Would TAP-ification change that?  Even if there is no change, this is
> worth a sentence or a few words in a commit message.

To all of the above: I could be more verbose in the commit message in
needed, but after you raised similar questions in a previous version
of this series I added this to the above commit message:

    All this patch does is re-arrange the output a bit so that it conforms
    with the TAP spec, everything that the test suite did before continues
    to work. That includes aggregating results in t/test-results/, the
    --verbose, --debug and other options for tests, and the test color
    output.

I.e. aside from turning "FAIL" into "not ok" and removing the "* "
prefix from lines everything else works exactly as before, that
includes color output and the aggregation in t/test-results/.

I can expand on that paragraph with more examples if you think it's necessary.

Here's how the output from the above test script looks under TAP:

    diagnostic message
    ok 1 - true  test
    not ok - 2 false test
    #       false
    ok 3 - true  test (todo) # TODO known breakage
    not ok 4 - false test (todo) # TODO known breakage
    # fixed 1 known breakage(s)
    # still have 1 known breakage(s)
    # failed 1 among remaining 3 test(s)
    1..4

> P.S. With current output pass / fail results of test_expect_failure
> are justified on ':' separator.  This is not the case of TAP output.

TAP output is conventionally not justified, unlike the existing
output. I don't know if anything would break if it were, i.e.:

    ok 1       - foo
    not ok 2   - bar

[PATCH v5 1/5] test-lib: Adjust output to be valid TAP format

[Ævar Arnfjörð Bjarmason]

TAP, the Test Anything Protocol, is a simple text-based interface
between testing modules in a test harness. test-lib.sh's output was
already very close to being valid TAP. This change brings it all the
way there. Before:

    $ ./t0005-signals.sh
    *   ok 1: sigchain works
    * passed all 1 test(s)

And after:

    $ ./t0005-signals.sh
    ok 1 - sigchain works
    # passed all 1 test(s)
    1..1

The advantage of using TAP is that any program that reads the format
(a "test harness") can run the tests. The most popular of these is the
prove(1) utility that comes with Perl. It can run tests in parallel,
display colored output, format the output to console, file, HTML etc.,
and much more. An example:

    $ prove ./t0005-signals.sh
    ./t0005-signals.sh .. ok
    All tests successful.
    Files=1, Tests=1,  0 wallclock secs ( 0.03 usr  0.00 sys +  0.01 cusr  0.02 csys =  0.06 CPU)
    Result: PASS

prove(1) gives you human readable output without being too
verbose. Running the test suite in parallel with `make test -j15`
produces a flood of text. Running them with `prove -j 15 ./t[0-9]*.sh`
makes it easy to follow what's going on.

All this patch does is re-arrange the output a bit so that it conforms
with the TAP spec, everything that the test suite did before continues
to work. That includes aggregating results in t/test-results/, the
--verbose, --debug and other options for tests, and the test color
output.

TAP harnesses ignore everything that they don't know about, so running
the tests with --verbose works:

    $ prove ./t0005-signals.sh :: --verbose --debug
    ./t0005-signals.sh .. Terminated
    ./t0005-signals.sh .. ok
    All tests successful.
    Files=1, Tests=1,  0 wallclock secs ( 0.02 usr  0.01 sys +  0.01 cusr  0.01 csys =  0.05 CPU)
    Result: PASS

Just supply the -v option to prove itself to get all the verbose
output that it suppresses:

    $ prove -v ./t0005-signals.sh :: --verbose --debug
    ./t0005-signals.sh ..
    Initialized empty Git repository in /home/avar/g/git/t/trash directory.t0005-signals/.git/
    expecting success:
            test-sigchain >actual
            case "$?" in
            143) true ;; # POSIX w/ SIGTERM=15
              3) true ;; # Windows
              *) false ;;
            esac &&
            test_cmp expect actual
    Terminated
    ok 1 - sigchain works
    # passed all 1 test(s)
    1..1
    ok
    All tests successful.
    Files=1, Tests=1,  0 wallclock secs ( 0.02 usr  0.00 sys +  0.01 cusr  0.01 csys =  0.04 CPU)
    Result: PASS

As a further example, consider this test script that uses a lot of
test-lib.sh features by Jakub Narebski:

    #!/bin/sh

    test_description='this is a sample test.

    This test is here to see various test outputs.'

    . ./test-lib.sh

    say 'diagnostic message'

    test_expect_success 'true  test' 'true'
    test_expect_success 'false test' 'false'

    test_expect_failure 'true  test (todo)' 'true'
    test_expect_failure 'false test (todo)' 'false'

    test_debug 'echo "debug message"'

    test_done

The output of that was previously:

    * diagnostic message                      # yellow
    *   ok 1: true  test
    * FAIL 2: false test                      # bold red
            false
    *   FIXED 3: true  test (todo)
    *   still broken 4: false test (todo)     # bold green
    * fixed 1 known breakage(s)               # green
    * still have 1 known breakage(s)          # bold red
    * failed 1 among remaining 3 test(s)      # bold red

But is now:

    diagnostic message                                    # yellow
    ok 1 - true  test
    not ok - 2 false test                                 # bold red
    #       false
    ok 3 - true  test (todo) # TODO known breakage
    not ok 4 - false test (todo) # TODO known breakage    # bold green
    # fixed 1 known breakage(s)                           # green
    # still have 1 known breakage(s)                      # bold red
    # failed 1 among remaining 3 test(s)                  # bold red
    1..4

All the coloring is preserved when the test is run manually. Under
prove(1) the test performs as expected, even with --debug and
--verbose options:

    $ prove ./example.sh :: --debug --verbose
    ./example.sh .. Dubious, test returned 1 (wstat 256, 0x100)
    Failed 1/4 subtests
            (1 TODO test unexpectedly succeeded)

    Test Summary Report
    -------------------
    ./example.sh (Wstat: 256 Tests: 4 Failed: 1)
      Failed test:  2
      TODO passed:   3
      Non-zero exit status: 1
    Files=1, Tests=4,  0 wallclock secs ( 0.02 usr  0.00 sys +  0.00 cusr  0.01 csys =  0.03 CPU)
    Result: FAIL

The TAP harness itself doesn't get confused by the color output, they
aren't used by test-lib.sh stdout isn't open to a terminal (test -t 1).

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
---

> To all of the above: I could be more verbose in the commit message in
> needed.

Here's v5 of this patch which does that. I used the example Jakub
provided and explained how it works with regards to colors and
interaction with a TAP harness.

 t/README      |   55 +++++++++++++++++++++++++++++++++++++++----------------
 t/test-lib.sh |   30 ++++++++++++++++++------------
 2 files changed, 57 insertions(+), 28 deletions(-)

diff --git a/t/README b/t/README
index 0e4e8d8..aa4ed28 100644
--- a/t/README
+++ b/t/README
@@ -18,25 +18,48 @@ The easiest way to run tests is to say "make".  This runs all
 the tests.
 
     *** t0000-basic.sh ***
-    *   ok 1: .git/objects should be empty after git-init in an empty repo.
-    *   ok 2: .git/objects should have 256 subdirectories.
-    *   ok 3: git-update-index without --add should fail adding.
+    ok 1 - .git/objects should be empty after git init in an empty repo.
+    ok 2 - .git/objects should have 3 subdirectories.
+    ok 3 - success is reported like this
     ...
-    *   ok 23: no diff after checkout and git-update-index --refresh.
-    * passed all 23 test(s)
-    *** t0100-environment-names.sh ***
-    *   ok 1: using old names should issue warnings.
-    *   ok 2: using old names but having new names should not issue warnings.
-    ...
-
-Or you can run each test individually from command line, like
-this:
-
-    $ sh ./t3001-ls-files-killed.sh
-    *   ok 1: git-update-index --add to add various paths.
-    *   ok 2: git-ls-files -k to show killed files.
-    *   ok 3: validate git-ls-files -k output.
-    * passed all 3 test(s)
+    ok 43 - very long name in the index handled sanely
+    # fixed 1 known breakage(s)
+    # still have 1 known breakage(s)
+    # passed all remaining 42 test(s)
+    1..43
+    *** t0001-init.sh ***
+    ok 1 - plain
+    ok 2 - plain with GIT_WORK_TREE
+    ok 3 - plain bare
+
+Since the tests all output TAP (see http://testanything.org) they can
+be run with any TAP harness. Here's an example of paralell testing
+powered by a recent version of prove(1):
+
+    $ prove --timer --jobs 15 ./t[0-9]*.sh
+    [19:17:33] ./t0005-signals.sh ................................... ok       36 ms
+    [19:17:33] ./t0022-crlf-rename.sh ............................... ok       69 ms
+    [19:17:33] ./t0024-crlf-archive.sh .............................. ok      154 ms
+    [19:17:33] ./t0004-unwritable.sh ................................ ok      289 ms
+    [19:17:33] ./t0002-gitfile.sh ................................... ok      480 ms
+    ===(     102;0  25/?  6/?  5/?  16/?  1/?  4/?  2/?  1/?  3/?  1... )===
+
+prove and other harnesses come with a lot of useful options. The
+--state option in particular is very useful:
+
+    # Repeat until no more failures
+    $ prove -j 15 --state=failed,save ./t[0-9]*.sh
+
+You can also run each test individually from command line, like this:
+
+    $ sh ./t3010-ls-files-killed-modified.sh
+    ok 1 - git update-index --add to add various paths.
+    ok 2 - git ls-files -k to show killed files.
+    ok 3 - validate git ls-files -k output.
+    ok 4 - git ls-files -m to show modified files.
+    ok 5 - validate git ls-files -m output.
+    # passed all 5 test(s)
+    1..5
 
 You can pass --verbose (or -v), --debug (or -d), and --immediate
 (or -i) command line argument to the test, or by setting GIT_TEST_OPTS
diff --git a/t/test-lib.sh b/t/test-lib.sh
index 367f053..e97645c 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -160,7 +160,7 @@ if test -n "$color"; then
 			*) test -n "$quiet" && return;;
 		esac
 		shift
-		printf "* %s" "$*"
+		printf "%s" "$*"
 		tput sgr0
 		echo
 		)
@@ -169,7 +169,7 @@ else
 	say_color() {
 		test -z "$1" && test -n "$quiet" && return
 		shift
-		echo "* $*"
+		echo "$*"
 	}
 fi
 
@@ -339,25 +339,25 @@ test_have_prereq () {
 
 test_ok_ () {
 	test_success=$(($test_success + 1))
-	say_color "" "  ok $test_count: $@"
+	say_color "" "ok $test_count - $@"
 }
 
 test_failure_ () {
 	test_failure=$(($test_failure + 1))
-	say_color error "FAIL $test_count: $1"
+	say_color error "not ok - $test_count $1"
 	shift
-	echo "$@" | sed -e 's/^/	/'
+	echo "$@" | sed -e 's/^/#	/'
 	test "$immediate" = "" || { GIT_EXIT_OK=t; exit 1; }
 }
 
 test_known_broken_ok_ () {
 	test_fixed=$(($test_fixed+1))
-	say_color "" "  FIXED $test_count: $@"
+	say_color "" "ok $test_count - $@ # TODO known breakage"
 }
 
 test_known_broken_failure_ () {
 	test_broken=$(($test_broken+1))
-	say_color skip "  still broken $test_count: $@"
+	say_color skip "not ok $test_count - $@ # TODO known breakage"
 }
 
 test_debug () {
@@ -390,7 +390,7 @@ test_skip () {
 	case "$to_skip" in
 	t)
 		say_color skip >&3 "skipping test: $@"
-		say_color skip "skip $test_count: $1"
+		say_color skip "ok $test_count: # skip $1"
 		: true
 		;;
 	*)
@@ -620,18 +620,22 @@ test_done () {
 
 	if test "$test_fixed" != 0
 	then
-		say_color pass "fixed $test_fixed known breakage(s)"
+		say_color pass "# fixed $test_fixed known breakage(s)"
 	fi
 	if test "$test_broken" != 0
 	then
-		say_color error "still have $test_broken known breakage(s)"
+		say_color error "# still have $test_broken known breakage(s)"
 		msg="remaining $(($test_count-$test_broken)) test(s)"
 	else
 		msg="$test_count test(s)"
 	fi
 	case "$test_failure" in
 	0)
-		say_color pass "passed all $msg"
+		# Maybe print SKIP message
+		[ -z "$skip_all" ] || skip_all=" # SKIP $skip_all"
+
+		say_color pass "# passed all $msg"
+		say "1..$test_count$skip_all"
 
 		test -d "$remove_trash" &&
 		cd "$(dirname "$remove_trash")" &&
@@ -640,7 +644,9 @@ test_done () {
 		exit 0 ;;
 
 	*)
-		say_color error "failed $test_failure among $msg"
+		say_color error "# failed $test_failure among $msg"
+		say "1..$test_count"
+
 		exit 1 ;;
 
 	esac
-- 
1.7.1.251.g92a7

Re: [PATCH v5 1/5] test-lib: Adjust output to be valid TAP format

[Junio C Hamano]

Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:

> diff --git a/t/README b/t/README
> index 0e4e8d8..aa4ed28 100644
> --- a/t/README
> +++ b/t/README
> @@ -18,25 +18,48 @@ The easiest way to run tests is to say "make".  This runs all
> ...

This describes the general framework of tests from the point of view of a
person who runs them and reads their output well.  I'll replace ab/tap
topic with these five patches.

I'd however like to ask you to update the sections in t/README to help
test writers.  You have gained enough experience with patches 4 and 5 (and
with patch 2 for Perl tests) to know what kind of script constructs, used
by people who are used to the old way of writing our tests, would cause
the TAP stuff to break, and are in the ideal position to rewrite that
section.  Two examples of what is drastically different from the old world
order are "do not print to the standard output strings that begin with XX
(or partial lines)" and "here is how to skip the remainder of the tests",
but you may have a handful of more rules.  I expect that "Writing Tests"
through "Test harness library" sections would be affected.

Earlier I kept wondering if this series is worth the trouble.  Having to
learn and follow the set of these new rules is exactly what I meant by
"trouble".  Let's make sure that the new rules are clear, small, and easy
to follow for other contributors.

Thanks.

Re: [PATCH v5 1/5] test-lib: Adjust output to be valid TAP format

[Ævar Arnfjörð Bjarmason]

On Fri, Jun 25, 2010 at 17:21, Junio C Hamano <gitster@pobox.com> wrote:
> Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:
>
>> diff --git a/t/README b/t/README
>> index 0e4e8d8..aa4ed28 100644
>> --- a/t/README
>> +++ b/t/README
>> @@ -18,25 +18,48 @@ The easiest way to run tests is to say "make".  This runs all
>> ...
>
> This describes the general framework of tests from the point of view of a
> person who runs them and reads their output well.  I'll replace ab/tap
> topic with these five patches.
>
> I'd however like to ask you to update the sections in t/README to help
> test writers.  You have gained enough experience with patches 4 and 5 (and
> with patch 2 for Perl tests) to know what kind of script constructs, used
> by people who are used to the old way of writing our tests, would cause
> the TAP stuff to break, and are in the ideal position to rewrite that
> section.  Two examples of what is drastically different from the old world
> order are "do not print to the standard output strings that begin with XX
> (or partial lines)" and "here is how to skip the remainder of the tests",
> but you may have a handful of more rules.  I expect that "Writing Tests"
> through "Test harness library" sections would be affected.
>
> Earlier I kept wondering if this series is worth the trouble.  Having to
> learn and follow the set of these new rules is exactly what I meant by
> "trouble".  Let's make sure that the new rules are clear, small, and easy
> to follow for other contributors.

I'll write those docs, but I probably won't have time in the next few
days so I'll submit a patch for that it separately if that's OK.

Re: [PATCH v5 1/5] test-lib: Adjust output to be valid TAP format

[Junio C Hamano]

Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:

> On Fri, Jun 25, 2010 at 17:21, Junio C Hamano <gitster@pobox.com> wrote:
>> Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:
>>
>>> diff --git a/t/README b/t/README
>>> index 0e4e8d8..aa4ed28 100644
>>> --- a/t/README
>>> +++ b/t/README
>>> @@ -18,25 +18,48 @@ The easiest way to run tests is to say "make".  This runs all
>>> ...
>>
>> This describes the general framework of tests from the point of view of a
>> person who runs them and reads their output well.  I'll replace ab/tap
>> topic with these five patches.
>> ...
>> I'd however like to ask you to update the sections in t/README to help
>> test writers....
>>
>> Earlier I kept wondering if this series is worth the trouble.  Having to
>> learn and follow the set of these new rules is exactly what I meant by
>> "trouble".  Let's make sure that the new rules are clear, small, and easy
>> to follow for other contributors.
>
> I'll write those docs, but I probably won't have time in the next few
> days so I'll submit a patch for that it separately if that's OK.

Oh, absolutely.  I meant to say that v4/v5 round already has a good
documentation for test users/readers; a follow-up patch to help test
writers would make the series more complete.

Thanks.

Re: [PATCH v5 1/5] test-lib: Adjust output to be valid TAP format

[Ævar Arnfjörð Bjarmason]

On Fri, Jun 25, 2010 at 21:36, Junio C Hamano <gitster@pobox.com> wrote:
> Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:
>> I'll write those docs, but I probably won't have time in the next few
>> days so I'll submit a patch for that it separately if that's OK.
>
> Oh, absolutely.  I meant to say that v4/v5 round already has a good
> documentation for test users/readers; a follow-up patch to help test
> writers would make the series more complete.

Sweet, yeah I'll make the t/README more awesomer later, good to hear
that v4/v5 is good enough.

Thanks!

[PATCH] t9700: Use Test::More->builder, not $Test::Builder::Test

[Ævar Arnfjörð Bjarmason]

$Test::Builder::Test was only made into an `our' variable in 0.94
released in September 2009, older distros are more likely to have 0.92
or earlier. Use the singleton Test::More->builder constructor instead,
the test now works on 0.92 too.

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
---

This patch fixes the TAP output in pu for older Test::More
libraries. It could be applied as a fixup to "test-lib: Make the
test_external_* functions TAP-aware", but it's probably useful to note
this in the version history.

 t/t9700/test.pl |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/t/t9700/test.pl b/t/t9700/test.pl
index e5d4b03..3e6a27c 100755
--- a/t/t9700/test.pl
+++ b/t/t9700/test.pl
@@ -10,8 +10,8 @@ use Test::More qw(no_plan);
 BEGIN {
 	# t9700-perl-git.sh kicks off our testing, so we have to go from
 	# there.
-	$Test::Builder::Test->{Curr_Test} = 1;
-	$Test::Builder::Test->{No_Ending} = 1;
+	Test::More->builder->{Curr_Test} = 1;
+	Test::More->builder->{No_Ending} = 1;
 }
 
 use Cwd;
@@ -113,6 +113,6 @@ like($last_commit, qr/^[0-9a-fA-F]{40}$/, 'rev-parse returned hash');
 my $dir_commit = $r2->command_oneline('log', '-n1', '--pretty=format:%H', '.');
 isnt($last_commit, $dir_commit, 'log . does not show last commit');
 
-printf "1..%d\n", $Test::Builder::Test->{Curr_Test};
+printf "1..%d\n", Test::More->builder->{Curr_Test};
 
-exit($Test::Builder::Test->{Is_Passing} ? 0 : 1);
+exit(Test::More->builder->{Is_Passing} ? 0 : 1);
-- 
1.7.1.251.g92a7

[PATCH v2] t9700: Use Test::More->builder, not $Test::Builder::Test

[Ævar Arnfjörð Bjarmason]

$Test::Builder::Test was only made into an `our' variable in 0.94
released in September 2009, older distros are more likely to have 0.92
or earlier. Use the singleton Test::More->builder constructor instead.

The exit() call was also unportable to <0.94. Just output a meaningful
exit code if the ->is_passing method exists. The t9700-perl-git.sh
test only cares about stderr output, so this doesn't affect test
results when using older Test::More modules.

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
---

The previous patch didn't take into account that the exit() call was
also unportable. Oops.

 t/t9700/test.pl |    9 +++++----
 1 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/t/t9700/test.pl b/t/t9700/test.pl
index e5d4b03..f2820d2 100755
--- a/t/t9700/test.pl
+++ b/t/t9700/test.pl
@@ -10,8 +10,8 @@ use Test::More qw(no_plan);
 BEGIN {
 	# t9700-perl-git.sh kicks off our testing, so we have to go from
 	# there.
-	$Test::Builder::Test->{Curr_Test} = 1;
-	$Test::Builder::Test->{No_Ending} = 1;
+	Test::More->builder->{Curr_Test} = 1;
+	Test::More->builder->{No_Ending} = 1;
 }
 
 use Cwd;
@@ -113,6 +113,7 @@ like($last_commit, qr/^[0-9a-fA-F]{40}$/, 'rev-parse returned hash');
 my $dir_commit = $r2->command_oneline('log', '-n1', '--pretty=format:%H', '.');
 isnt($last_commit, $dir_commit, 'log . does not show last commit');
 
-printf "1..%d\n", $Test::Builder::Test->{Curr_Test};
+printf "1..%d\n", Test::More->builder->{Curr_Test};
 
-exit($Test::Builder::Test->{Is_Passing} ? 0 : 1);
+my $is_passing = eval { Test::More->is_passing };
+exit($is_passing ? 0 : 1) unless $@ =~ /Can't locate object method/;
-- 
1.7.1.251.g92a7

Re: [PATCH v2] t9700: Use Test::More->builder, not $Test::Builder::Test

[Junio C Hamano]

Ævar Arnfjörð Bjarmason  <avarab@gmail.com> writes:

> $Test::Builder::Test was only made into an `our' variable in 0.94
> released in September 2009, older distros are more likely to have 0.92
> or earlier. Use the singleton Test::More->builder constructor instead.

Ah, that is what it was.  I was wondering what it was that showed 1..1 and
then 1..40.

Thanks; will queue on top of ab/tap.