git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] t1304,t2007: quell output to stdout and stderr
@ 2010-04-21 13:55 Michael J Gruber
  2010-04-21 14:45 ` Jonathan Nieder
  0 siblings, 1 reply; 6+ messages in thread
From: Michael J Gruber @ 2010-04-21 13:55 UTC (permalink / raw)
  To: git; +Cc: Junio C Hamano

These tests send output to stdout or stderr even without -v. This is
distracting because unexpected output flashing by during make test
usually indicates problems

Shut them up uncoditionally: In both cases, the output was due to
intermediate commands in between the actual test cases.

Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
---
 t/t1304-default-acl.sh      |    2 +-
 t/t2007-checkout-symlink.sh |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/t/t1304-default-acl.sh b/t/t1304-default-acl.sh
index 055ad00..b26d2e8 100755
--- a/t/t1304-default-acl.sh
+++ b/t/t1304-default-acl.sh
@@ -15,7 +15,7 @@ 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
+if ! setfacl -m u:root:rwx . 2>/dev/null; then
     say "Skipping ACL tests: unable to use setfacl"
     test_done
 fi
diff --git a/t/t2007-checkout-symlink.sh b/t/t2007-checkout-symlink.sh
index 20f3343..fc5db05 100755
--- a/t/t2007-checkout-symlink.sh
+++ b/t/t2007-checkout-symlink.sh
@@ -45,7 +45,7 @@ test_expect_success 'switch from symlink to dir' '
 '
 
 rm -fr frotz xyzzy nitfol &&
-git checkout -f master || exit
+git checkout -q -f master || exit
 
 test_expect_success 'switch from dir to symlink' '
 
-- 
1.7.1.rc1.248.gcefbb

^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] t1304,t2007: quell output to stdout and stderr
  2010-04-21 13:55 [PATCH] t1304,t2007: quell output to stdout and stderr Michael J Gruber
@ 2010-04-21 14:45 ` Jonathan Nieder
  2010-04-21 14:53   ` Michael J Gruber
  0 siblings, 1 reply; 6+ messages in thread
From: Jonathan Nieder @ 2010-04-21 14:45 UTC (permalink / raw)
  To: Michael J Gruber; +Cc: git, Junio C Hamano

Hi Michael,

Michael J Gruber wrote:

> --- a/t/t1304-default-acl.sh
> +++ b/t/t1304-default-acl.sh
> @@ -15,7 +15,7 @@ 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
> +if ! setfacl -m u:root:rwx . 2>/dev/null; then
>      say "Skipping ACL tests: unable to use setfacl"
>      test_done
>  fi
[and a similar suppression of ‘git checkout’ output]

In the spirit of commit 4a45f7d (Use test_expect_success for test
setups, 2010-03-20), might it make sense to turn these into tests?

I am imagining something like this.

diff --git a/t/t1304-default-acl.sh b/t/t1304-default-acl.sh
index b26d2e8..8b3ff7a 100755
--- a/t/t1304-default-acl.sh
+++ b/t/t1304-default-acl.sh
@@ -15,7 +15,15 @@ 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 . 2>/dev/null; then
+test_expect_success 'Setup: try to set an ACL' '
+	if setfacl -m u:root:rwx .
+	then
+		test_set_prereq ACL
+	fi
+'
+
+if ! test_have_prereq ACL
+then
     say "Skipping ACL tests: unable to use setfacl"
     test_done
 fi
diff --git a/t/t2007-checkout-symlink.sh b/t/t2007-checkout-symlink.sh
index fc5db05..f8f40e5 100755
--- a/t/t2007-checkout-symlink.sh
+++ b/t/t2007-checkout-symlink.sh
@@ -44,11 +44,12 @@ test_expect_success 'switch from symlink to dir' '
 
 '
 
-rm -fr frotz xyzzy nitfol &&
-git checkout -q -f master || exit
-
 test_expect_success 'switch from dir to symlink' '
 
+	rm -fr frotz xyzzy nitfol &&
+	git rm -fr . &&
+	git checkout -f master &&
+
 	git checkout side
 
 '

^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] t1304,t2007: quell output to stdout and stderr
  2010-04-21 14:45 ` Jonathan Nieder
@ 2010-04-21 14:53   ` Michael J Gruber
  2010-04-21 15:12     ` Jonathan Nieder
  0 siblings, 1 reply; 6+ messages in thread
From: Michael J Gruber @ 2010-04-21 14:53 UTC (permalink / raw)
  To: Jonathan Nieder; +Cc: git, Junio C Hamano

Jonathan Nieder venit, vidit, dixit 21.04.2010 16:45:
> Hi Michael,
> 
> Michael J Gruber wrote:
> 
>> --- a/t/t1304-default-acl.sh
>> +++ b/t/t1304-default-acl.sh
>> @@ -15,7 +15,7 @@ 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
>> +if ! setfacl -m u:root:rwx . 2>/dev/null; then
>>      say "Skipping ACL tests: unable to use setfacl"
>>      test_done
>>  fi
> [and a similar suppression of ‘git checkout’ output]
> 
> In the spirit of commit 4a45f7d (Use test_expect_success for test
> setups, 2010-03-20), might it make sense to turn these into tests?

In principle yes, of course, I've done this in other cases. But note
that here, on both occasions, the test script wants to exit if
prerequisites are not met. I don't think it's OK to exit or test_done
from within test_expect_something, is it?

> I am imagining something like this.
> 
> diff --git a/t/t1304-default-acl.sh b/t/t1304-default-acl.sh
> index b26d2e8..8b3ff7a 100755
> --- a/t/t1304-default-acl.sh
> +++ b/t/t1304-default-acl.sh
> @@ -15,7 +15,15 @@ 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 . 2>/dev/null; then
> +test_expect_success 'Setup: try to set an ACL' '
> +	if setfacl -m u:root:rwx .
> +	then
> +		test_set_prereq ACL
> +	fi
> +'
> +
> +if ! test_have_prereq ACL
> +then
>      say "Skipping ACL tests: unable to use setfacl"
>      test_done
>  fi
> diff --git a/t/t2007-checkout-symlink.sh b/t/t2007-checkout-symlink.sh
> index fc5db05..f8f40e5 100755
> --- a/t/t2007-checkout-symlink.sh
> +++ b/t/t2007-checkout-symlink.sh
> @@ -44,11 +44,12 @@ test_expect_success 'switch from symlink to dir' '
>  
>  '
>  
> -rm -fr frotz xyzzy nitfol &&
> -git checkout -q -f master || exit
> -
>  test_expect_success 'switch from dir to symlink' '
>  
> +	rm -fr frotz xyzzy nitfol &&
> +	git rm -fr . &&
> +	git checkout -f master &&
> +
>  	git checkout side
>  
>  '

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] t1304,t2007: quell output to stdout and stderr
  2010-04-21 14:53   ` Michael J Gruber
@ 2010-04-21 15:12     ` Jonathan Nieder
  2010-04-22 20:45       ` [PATCH v2] " Michael J Gruber
  0 siblings, 1 reply; 6+ messages in thread
From: Jonathan Nieder @ 2010-04-21 15:12 UTC (permalink / raw)
  To: Michael J Gruber; +Cc: git, Junio C Hamano

Michael J Gruber wrote:

> I don't think it's OK to exit or test_done
> from within test_expect_something, is it?

exit: the consequences are the same whether you’re in a test or not.
Either way, it’s bad.  I think the intent of that ‘exit’ was to make a
good effort to set up for the next test and skip the test if that
didn’t work, which is a bit old-fashioned.  Emptying the index before
the checkout would make it like that the setup should not fail even if
some related bug resurfaces.

test_done: not sure what would happen.  Indeed, it doesn’t sound like
a very sane thing to do. :)

Jonathan

^ permalink raw reply	[flat|nested] 6+ messages in thread

* [PATCH v2] t1304,t2007: quell output to stdout and stderr
  2010-04-21 15:12     ` Jonathan Nieder
@ 2010-04-22 20:45       ` Michael J Gruber
  2010-05-26  9:08         ` Michael J Gruber
  0 siblings, 1 reply; 6+ messages in thread
From: Michael J Gruber @ 2010-04-22 20:45 UTC (permalink / raw)
  To: git; +Cc: Junio C Hamano, Jonathan Nieder

These tests send output to stdout or stderr even without -v. This is
distracting because unexpected output flashing by during make test
usually indicates problems.

Shut them up unconditionally by integrating them in test code: In both
cases, the output was due to intermediate commands in between the actual
test cases.

Helped-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
---
 t/t1304-default-acl.sh      |   10 +++++++++-
 t/t2007-checkout-symlink.sh |    5 ++---
 2 files changed, 11 insertions(+), 4 deletions(-)

diff --git a/t/t1304-default-acl.sh b/t/t1304-default-acl.sh
index 055ad00..8b3ff7a 100755
--- a/t/t1304-default-acl.sh
+++ b/t/t1304-default-acl.sh
@@ -15,7 +15,15 @@ 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
+test_expect_success 'Setup: try to set an ACL' '
+	if setfacl -m u:root:rwx .
+	then
+		test_set_prereq ACL
+	fi
+'
+
+if ! test_have_prereq ACL
+then
     say "Skipping ACL tests: unable to use setfacl"
     test_done
 fi
diff --git a/t/t2007-checkout-symlink.sh b/t/t2007-checkout-symlink.sh
index 20f3343..e98ec4c 100755
--- a/t/t2007-checkout-symlink.sh
+++ b/t/t2007-checkout-symlink.sh
@@ -44,11 +44,10 @@ test_expect_success 'switch from symlink to dir' '
 
 '
 
-rm -fr frotz xyzzy nitfol &&
-git checkout -f master || exit
-
 test_expect_success 'switch from dir to symlink' '
 
+	rm -fr frotz xyzzy nitfol &&
+	git checkout -f master &&
 	git checkout side
 
 '
-- 
1.7.1.rc1.248.gcefbb

^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH v2] t1304,t2007: quell output to stdout and stderr
  2010-04-22 20:45       ` [PATCH v2] " Michael J Gruber
@ 2010-05-26  9:08         ` Michael J Gruber
  0 siblings, 0 replies; 6+ messages in thread
From: Michael J Gruber @ 2010-05-26  9:08 UTC (permalink / raw)
  Cc: git, Junio C Hamano, Jonathan Nieder

Michael J Gruber venit, vidit, dixit 22.04.2010 22:45:
> These tests send output to stdout or stderr even without -v. This is
> distracting because unexpected output flashing by during make test
> usually indicates problems.
> 
> Shut them up unconditionally by integrating them in test code: In both
> cases, the output was due to intermediate commands in between the actual
> test cases.
> 
> Helped-by: Jonathan Nieder <jrnieder@gmail.com>
> Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
> ---

ping...

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2010-05-26  9:08 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-04-21 13:55 [PATCH] t1304,t2007: quell output to stdout and stderr Michael J Gruber
2010-04-21 14:45 ` Jonathan Nieder
2010-04-21 14:53   ` Michael J Gruber
2010-04-21 15:12     ` Jonathan Nieder
2010-04-22 20:45       ` [PATCH v2] " Michael J Gruber
2010-05-26  9:08         ` Michael J Gruber

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).