Re: Help with autopkgtest for package andi
- To: debian-med@lists.debian.org
- Subject: Re: Help with autopkgtest for package andi
- From: Andreas Tille <andreas@an3as.eu>
- Date: Sun, 5 Apr 2020 23:36:45 +0200
- Message-id: <[🔎] 20200405213645.GV7599@an3as.eu>
- In-reply-to: <07a18219-9cca-c641-4b64-68347ce0d85b@evolbio.mpg.de>
- References: <CAOAkg3x0_u5vuBY2rsmmEZ=axuEqD5kAYBNQj6bjEVGa93rGdQ@mail.gmail.com> <CAJN192-j_BtNgVsQ_kDWNK2bji68owZnp81RArsjL=Y+oPt-dw@mail.gmail.com> <f7d95ff1-3f94-5ae9-abfc-231e215a9469@evolbio.mpg.de> <CAOAkg3zTYH7Zx0DOyCLG0Dw+P5W8H=_fxQTxnZsRBh69fF4GnQ@mail.gmail.com> <07a18219-9cca-c641-4b64-68347ce0d85b@evolbio.mpg.de>
Hi Fabian,
sponsored - but since you are upstream please fix
I: andi: spelling-error-in-binary usr/bin/andi exlusive exclusive
in your next release.
Thank you for your preparation
Andreas.
On Mon, Mar 23, 2020 at 04:27:33PM +0100, Fabian Klötzl wrote:
> Hi Samyak,
>
> By accident I noticed that you opened a pull request on Salsa. Note that
> most communication in this team happens via mail. Thus, your changes
> flew under the radar.
>
> The update to v0.13 looks fine to me. You can push that directly to the
> repository and ask for a sponsorship on the mailing list. As I have some
> free time on my hand now, I will look closer at the autopkgtests. We can
> then add them in a -2.
>
> Best,
> Fabian
>
>
>
> On 07.03.20 20:12, Samyak Jain wrote:
> > Hey Fabian,
> > Thank you for the follow-ups :D
> >
> > On Sat, Mar 7, 2020 at 9:56 PM Fabian Klötzl <kloetzl@evolbio.mpg.de
> > <mailto:kloetzl@evolbio.mpg.de>> wrote:
> >
> > Hi,
> >
> > Adding autopackagetests for andi has been on my TODO list for a long
> > time, as I am both its upstream author and maintainer. However, as my
> > dissertation is due in two weeks, that currently takes priority.
> >
> > If you want to go forward with the tests, here are some pointers: The
> > package comes with eight tests, all located in the test directory. Of
> > these test_esa, test_process, and test_seq are compile time only. Don't
> > worry about them.
> >
> >
> > The shell scripts verify andi in different conditions. To that end, andi
> > has to be fed simulated sequences as I didn't want to check in large
> > data into git. So the shell scripts use a program called test_fasta,
> > which basically simulates genetic sequences. So you first have to figure
> > out how to build that simulator during the autopackgetest. Next, run the
> > shell scripts. Finally, make the tests reproducible, as they can
> > randomly fail. Check test_random.sh for some documentation on that. You
> > probably also have to patch the path to andi currently hard coded in the
> > scripts.
> >
> > Feel free to ask any question you like; I am happy to assist you.
> >
> >
> > I went through the files, andi-manual.pdf and went through the
> > test/directory.
> > For that, I have added a file (debian/tests/run-unit-test) which is as
> > follows:
> >
> > #!/bin/bash
> > set -e
> >
> > pkg=andi
> >
> > export LC_ALL=C.UTF-8
> > if [ "${AUTOPKGTEST_TMP}" = "" ] ; then
> > AUTOPKGTEST_TMP=$(mktemp -d /tmp/${pkg}-test.XXXXXX)
> > # Double quote below to expand the temporary directory variable now versus
> > # later is on purpose.
> > # shellcheck disable=SC2064
> > trap "rm -rf ${AUTOPKGTEST_TMP}" 0 INT QUIT ABRT PIPE TERM
> > fi
> >
> > mkdir ${AUTOPKGTEST_TMP}/test
> > cp -a test/* "${AUTOPKGTEST_TMP}/test"
> >
> > cd "${AUTOPKGTEST_TMP}"
> >
> > ./test/*.sh
> >
> > Following error was encountered:
> >
> > autopkgtest [00:00:12]: test run-unit-test: [-----------------------
> > Triggering low homology failed
> > a_low.fa:>S0 (base_seed: 744030346)
> > a_low.fa:>S1 (base_seed: 744030346)
> > b_low.fa:>S0 (base_seed: 2613208344)
> > b_low.fa:>S1 (base_seed: 2613208344)
> > both_low.fa:>S0 (base_seed: 2592210513)
> > both_low.fa:>S1 (base_seed: 2592210513)
> > autopkgtest [00:00:13]: test run-unit-test: -----------------------]
> > autopkgtest [00:00:13]: test run-unit-test: - - - - - - - - - - results - - - - - - - - - -
> > run-unit-test FAIL non-zero exit status 1
> >
> >
> > I'm not really sure about it, can you please help me with it?
> >
> > Thanks and regards,
> > Samyak Jain.
>
>
--
http://fam-tille.de
Reply to: