[Reproducible-builds] Bug#808809: diffoscope: tests will fail in future

[Holger Levsen]

source: diffoscope
version: 44
severity: important

Hi,

when building diffoscope in the future, the tests fail like this:

=================================== FAILURES ===================================
_________________________________ test_listing _________________________________

differences = [<Difference file list -- file list []>, <Difference dir/link -- dir/link []>, <Difference 
dir/text -- dir/text []>]

    @pytest.mark.skipif(tool_missing('cpio'), reason='missing cpio')
    def test_listing(differences):
        expected_diff = open(os.path.join(os.path.dirname(__file__), 
'../data/cpio_listing_expected_diff')).read()
>       assert differences[0].unified_diff == expected_diff
E       assert '@@ -1,4 +1,4...4  2015 dir\n' == '@@ -1,4 +1,4 ...4 15:11 dir\n'
E         Skipping 56 identical leading characters in diff, use -v to show
E         -  6 Jun 24  2015 dir/link -> broken
E         ?           ---
E         +  6 Jun 24 15:10 dir/link -> broken
E         ?             +++
E         - +lrwxrwxrwx   1 1000     1000           13 Jun 24  2015 dir/link -> really-broken
E         ?                                                   ---
E         + +lrwxrwxrwx   1 1000     1000           13 Jun 24 15:11 dir/link -> really-broken
E         ?                                                     +++
E         Detailed information truncated (17 more lines), use "-vv" to show

tests/comparators/test_cpio.py:53: AssertionError
============== 1 failed, 164 passed, 41 skipped in 98.58 seconds ===============

<deki> hm, i guess the reason is that it's build in 2017. in this case it prints "13 Jun 24  2015" instead of 
"13 Jun 24 15:11"
<h01ger> | deki: hehe & wow. nice find.
<h01ger> | so we're finding real build failures in the future now. neato.
<deki> :)
<h01ger> | Lunar: ^^ i think thats a nice story for the weekly report too


I've not filed this as serious (though I think it is) to not block testing
migration and because I'm certain there will be quite some uploads in the next
months anway. But we definitly shouldn't release with this in stable… ;-)


cheers,
	Holger
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 828 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.alioth.debian.org/pipermail/reproducible-builds/attachments/20151223/ffdbad87/attachment.sig>