RE: [ANNOUNCE] Git v2.16.0-rc2 - breakages in t1308 and 1404

RE: [ANNOUNCE] Git v2.16.0-rc2 - breakages in t1308 and 1404

[Randall S. Becker]

On January 11, 2018 11:11 PM, Junio C Hamano wrote:
> A release candidate Git v2.16.0-rc2 is now available for testing at the usual
> places.  It is comprised of 483 non-merge commits since v2.15.0, contributed
> by 80 people, 23 of which are new faces.
<snip>

Please forgive my not knowing the proper way to report this ... we have a few breakages on the NonStop port (which has finally caught up to 2.16, yay!). The test run takes a very long time to run, so this is preliminary. I'm also not sure who should try to resolve these.

For 1308:

Value not found for "foo.bar"
not ok 23 - proper error on directory "files"
#
#               echo "Error (-1) reading configuration file a-directory." >expect &&
#               mkdir a-directory &&
#               test_expect_code 2 test-config configset_get_value foo.bar a-directory 2>output &&
#               grep "^warning:" output &&
#               grep "^Error" output >actual &&
#               test_cmp expect actual
test_expect_code: command exited with 1, we wanted 2 test-config configset_get_value foo.bar a-directory

It looks like the exit code is coming back as 1 not 2. There is also a file except vs expect.
./trash directory.t1308-config-set: ls
a-directory  actual  config2  except  expect  output  result

For 1404: It looks like the error was detected, but the test case was not structured to handle where the detection occurred. Here's what I see:

not ok 52 - delete fails cleanly if packed-refs file is locked
#
#               prefix=refs/locked-packed-refs &&
#               # Set up a reference with differing loose and packed versions:
#               git update-ref $prefix/foo $C &&
#               git pack-refs --all &&
#               git update-ref $prefix/foo $D &&
#               git for-each-ref $prefix >unchanged &&
#               # Now try to delete it while the `packed-refs` lock is held:
#               : >.git/packed-refs.lock &&
#               test_when_finished "rm -f .git/packed-refs.lock" &&
#               test_must_fail git update-ref -d $prefix/foo >out 2>err &&
#               git for-each-ref $prefix >actual &&
#               test_i18ngrep "Unable to create $Q.*packed-refs.lock$Q: File exists" err &&
#               test_cmp unchanged actual
#

# failed 1 among 52 test(s)
1..52
./trash directory.t1404-update-ref-errors: ls
actual       commands  expected      expected-refs  out         unchanged
actual-refs  err       expected-err  input          output.err
./trash directory.t1404-update-ref-errors: cat err
error: Unable to create '/home/git/git/t/trash directory.t1404-update-ref-errors/.git/packed-refs.lock': File already exists.

Another git process seems to be running in this repository, e.g.
an editor opened by 'git commit'. Please make sure all processes
are terminated then try again. If it still fails, a git process
may have crashed in this repository earlier:
remove the file manually to continue.
./trash directory.t1404-update-ref-errors: cat out

Sincerely,
Randall

-- Brief whoami:
 NonStop developer since approximately 211288444200000000
 UNIX developer since approximately 421664400
-- In my real life, I talk too much.

Re: [ANNOUNCE] Git v2.16.0-rc2 - breakages in t1308 and 1404

[Junio C Hamano]

"Randall S. Becker" <rsbecker@nexbridge.com> writes:

> It looks like the exit code is coming back as 1 not 2. There is
> also a file except vs expect.
> ./trash directory.t1308-config-set: ls
> a-directory  actual  config2  except  expect  output  result

The test that leaves "except" does look wrong.  The relevant part
looks like this:

    test_expect_success 'find value with highest priority from a configset' '
            echo hask >expect &&
            test-config configset_get_value case.baz config2 .git/config >actual &&
            test_cmp expect actual
    '

    test_expect_success 'find value_list for a key from a configset' '
            cat >except <<-\EOF &&
            sam
            ...
            EOF
            test-config configset_get_value case.baz config2 .git/config >actual &&
            test_cmp expect actual
    '

The invocations of test-config in these two tests look exactly the
same, at least to me, so whatever comes out of the command and
stored in 'actual' must match what the first test stored in 'expect'
and compared the same as 'actual' in that test.

So the second test is probably asking a wrong question to test-config
but because it prepares an expected answer in a wrong file, it did
not even notice that it is asking a wrong question X-<.

The wrong test comes from 4c715ebb ("test-config: add tests for the
config_set API", 2014-07-28); I do not know offhand if the author of
that change is still around.

Having said all that, I suspect that the "except" thing may not have
anything to do with what you are observing; sorry for not ending up
to be very helpful X-<.

RE: [ANNOUNCE] Git v2.16.0-rc2 - breakages in t1308 and 1404

[Randall S. Becker]

On January 12, 2018 5:08 PM, Junio C Hamano wrote:
> "Randall S. Becker" <rsbecker@nexbridge.com> writes:
> > It looks like the exit code is coming back as 1 not 2. There is also a
> > file except vs expect.
> > ./trash directory.t1308-config-set: ls a-directory  actual  config2
> > except  expect  output  result
> 
> The test that leaves "except" does look wrong.  The relevant part looks
like
> this:
> 
>     test_expect_success 'find value with highest priority from a
configset' '
>             echo hask >expect &&
>             test-config configset_get_value case.baz config2 .git/config
>actual &&
>             test_cmp expect actual
>     '
> 
>     test_expect_success 'find value_list for a key from a configset' '
>             cat >except <<-\EOF &&
>             sam
>             ...
>             EOF
>             test-config configset_get_value case.baz config2 .git/config
>actual &&
>             test_cmp expect actual
>     '
> 
> The invocations of test-config in these two tests look exactly the same,
at
> least to me, so whatever comes out of the command and stored in 'actual'
> must match what the first test stored in 'expect'
> and compared the same as 'actual' in that test.
> 
> So the second test is probably asking a wrong question to test-config but
> because it prepares an expected answer in a wrong file, it did not even
> notice that it is asking a wrong question X-<.
> 
> The wrong test comes from 4c715ebb ("test-config: add tests for the
> config_set API", 2014-07-28); I do not know offhand if the author of that
> change is still around.
> 
> Having said all that, I suspect that the "except" thing may not have
anything
> to do with what you are observing; sorry for not ending up to be very
helpful
> X-<.

Sadly, fixing the "except" thing causes the test to break now. I can't make
sense of the intent of this, unless the test is to find unmatching keys and
it really does mean "except"
--- expect      2018-01-12 23:03:41 +0000
+++ actual      2018-01-12 23:03:41 +0000
@@ -1,5 +1 @@
-sam
-bat
 hask
-lama
-ball

[case]
baz = lama
[my]
new = silk
[case]
baz = ball

Re: [ANNOUNCE] Git v2.16.0-rc2 - breakages in t1308 and 1404

[Junio C Hamano]

"Randall S. Becker" <rsbecker@nexbridge.com> writes:

> Sadly, fixing the "except" thing causes the test to break now.

That is exactly what I wanted to say.  If you want to "fix" it,
you'd need to figure out what the author of the "except" thing
wanted to test, adjust the args given to test-config (it cannot be
the same as the test-config invocation of the previous test), and
then fix the typo s/except/expect/.  Changing the typo alone *will*
of course make the test fail, because then the file with the
corrected name, i.e. "expect", has bogus lines that does not match
how the current invocation of "test-config" command is expected to
output.

> sense of the intent of this, unless the test is to find unmatching keys and
> it really does mean "except"
> --- expect      2018-01-12 23:03:41 +0000
> +++ actual      2018-01-12 23:03:41 +0000
> @@ -1,5 +1 @@
> -sam
> -bat
>  hask
> -lama
> -ball
>
> [case]
> baz = lama
> [my]
> new = silk
> [case]
> baz = ball

Re: [ANNOUNCE] Git v2.16.0-rc2 - breakages in t1308 and 1404

[Tanay Abhra]

On Fri, Jan 12, 2018 at 5:27 PM, Junio C Hamano <gitster@pobox.com> wrote:
>
> "Randall S. Becker" <rsbecker@nexbridge.com> writes:
>
> > Sadly, fixing the "except" thing causes the test to break now.
>
> That is exactly what I wanted to say.  If you want to "fix" it,
> you'd need to figure out what the author of the "except" thing
> wanted to test, adjust the args given to test-config (it cannot be
> the same as the test-config invocation of the previous test), and
> then fix the typo s/except/expect/.  Changing the typo alone *will*
> of course make the test fail, because then the file with the
> corrected name, i.e. "expect", has bogus lines that does not match
> how the current invocation of "test-config" command is expected to
> output.


Hi Guys,

I was the original author of the test, I am sorry about the typo.

I will submit a patch fixing the test. The fix can be checked at
https://github.com/git/git/pull/451.

'configset_get_value' will be changed to 'configset_get_value_multi'
since the test expects a list of values instead of a single value.

Thanks,
Tanay Abhra

Re: [ANNOUNCE] Git v2.16.0-rc2 - breakages in t1308 and 1404

[Junio C Hamano]

Tanay Abhra <tanayabh@gmail.com> writes:

> I was the original author of the test, I am sorry about the typo.
> ...
> 'configset_get_value' will be changed to 'configset_get_value_multi'
> since the test expects a list of values instead of a single value.

Thanks for a quick reaction!!