Hi Aaron,
On Mon, Jun 22, 2020 at 05:29:53PM -0400, Aaron M. Ucko wrote:
> Andreas Tille <ti...@debian.org> writes:
>
> > ERR [1] -- [objects/genomecoll/gc_cli] test_gencoll_svc_ASSM (real
> > 0.04, user 0.03, sys 0.00)
>
> This is one of the tests that requires a network connection, and as such
> will fail on the autobuilders unless trimmed altogether. The available
> offline tests are
>
> OK -- [serial/datatool] datatool.sh
> SKP -- [serial/datatool] datatool_xml.sh
> OK -- [objtools/blast/gene_info_reader/unit_test] gene_info_unit_test
> OK -- [algo/blast/proteinkmer/unit_test] proteinkmer_unit_test
> OK -- [algo/blast/unit_tests/api] blastdiag_unit_test
> OK -- [misc/jsonwrapp/test] test_jsonwrapp
> OK -- [misc/jsonwrapp/test] test_jsonwrapp10
So your recommendation is to run only these tests - may be by patching
check.sh?
> NB: blastdiag_unit_test is in the same directory as aascan_unit_test,
> which requires a network connection, so should be pulled in by
> executable name rather than directory name.
I admit I do not understand this paragraph.
> > E: ncbi-magicblast source: license-problem-json-evil
> > c++/include/misc/jsonwrapp/rapidjson10/license.txt
> > E: ncbi-magicblast source: license-problem-json-evil
> > c++/include/misc/jsonwrapp/rapidjson11/license.txt
>
> Again
Hups, did I missed some message about this?
> these are false positives -- rapidjson here is trimmed even more
> aggressively than in #840333, and in particular doesn't include the
> problematic jsonchecker subtree.
Well, but there are these license files and ftpmaster will probably
reject the source if these are there. So would you mind talking to
upstream to clarify this? If the text contains the string
The Software shall be used for Good, not Evil.
it is not acceptable in Debian. If the code is not affected by this
license it should not be mentioned.
Kind regards
Andreas.
--
http://fam-tille.de
