Re: RFS: ampliconnoise

Hi Andreas,

Thanks for clarifying. I forgot to account for the .dat file I included using ampliconnoise.examples. I just checked the data I added.

By the way, I ran lintian before pushing and didn't get any warnings. I've run lintian again on the version I pushed, inside the package directory, and here's what I get:

> git status
HEAD detached at 2973d08
nothing to commit, working tree clean

> lintian -I
E: ampliconnoise changes: bad-distribution-in-changes-file unstable
W: ampliconnoise-dbgsym: debug-file-with-no-debug-symbols usr/lib/debug/.build-id/bf/eb256be172f85624e78e0573d496ff575c8033.debug
I: ampliconnoise: hardening-no-fortify-functions usr/bin/FCluster
I: ampliconnoise: hardening-no-fortify-functions usr/bin/FastaUnique
I: ampliconnoise: hardening-no-fortify-functions usr/bin/NDist
I: ampliconnoise: hardening-no-fortify-functions ... use --no-tag-display-limit to see all (or pipe to a file/program)
I: ampliconnoise: manpage-without-executable usr/share/man/man1/ampliconnoise.1.gz

> lintian -I -T arch-dep-package-has-big-usr-share

> lintian -T big-usr-share
Unknown tag big-usr-share at /usr/share/perl5/Lintian/Profile.pm line 367.

Then I built with dpkg-buildpackage --no-sign and still didn't get any warnings:

> lintian -I ampliconnoise_1.29-9_amd64.changes
W: ampliconnoise-dbgsym: debug-file-with-no-debug-symbols usr/lib/debug/.build-id/f7/4c3b6563d86dfa4d9b1afc318b2417b09ffd4e.debug
I: ampliconnoise: manpage-without-executable usr/share/man/man1/ampliconnoise.1.gz
I: ampliconnoise source: out-of-date-standards-version 4.2.1 (released 2018-08-25) (current is 4.5.0)
I: ampliconnoise source: upstream-metadata-missing-bug-tracking
I: ampliconnoise source: upstream-metadata-missing-repository

> lintian -I ampliconnoise_1.29-9_amd64.deb
I: ampliconnoise: manpage-without-executable usr/share/man/man1/ampliconnoise.1.gz

> lintian -I ampliconnoise_1.29-9.dsc
I: ampliconnoise source: out-of-date-standards-version 4.2.1 (released 2018-08-25) (current is 4.5.0)
I: ampliconnoise source: upstream-metadata-missing-bug-tracking
I: ampliconnoise source: upstream-metadata-missing-repository

> lintian -I ampliconnoise-dbgsym_1.29-9_amd64.deb
W: ampliconnoise-dbgsym: debug-file-with-no-debug-symbols usr/lib/debug/.build-id/f7/4c3b6563d86dfa4d9b1afc318b2417b09ffd4e.debug
I: ampliconnoise-dbgsym: cannot-check-whether-usr-share-doc-symlink-points-to-foreign-package

Am I doing something wrong here?

Regards,

Pranav

On Fri, Jul 10, 2020 at 8:10 PM Andreas Tille <andreas@an3as.eu> wrote:

Hi Pranav,

(and list)

On Fri, Jul 10, 2020 at 07:35:44PM +0530, Pranav Ballaney wrote:
> Thanks for uploading it! Would you like me to make a new examples
> binary for every package from now?

Well, I've git the lintian warning "big-usr-share" and have seen that
the example data are exceeding the real program by one order of
magnitude (in the sense 6-digit number of binary and 7-digit number of
data).

> I'm not sure about when I should
> create a new binary. For ampliconnoise, the test data I added was only
> about 15% of the entire package - is that beyond the threshold?

$ ls -l ampl* | grep "\.deb$" | grep -v dbgsym
-rw-r--r-- 1 andreas admin 478084 9. Jul 17:20 ampliconnoise_1.29-9_amd64.deb
-rw-r--r-- 1 andreas admin 1705908 9. Jul 17:20 ampliconnoise-examples_1.29-9_all.deb

The relation of binary to data is more than factor 3. We always need to
consider that some users might like to craft slim containers or so. In
this case that extra workload might make them unhappy.

There is no real rule - but in this (admitedly borderline) case I
thought it would be a good idea.

Kind regards

Andreas.

> On Thu, Jul 9, 2020 at 9:11 PM Andreas Tille <andreas@an3as.eu> wrote:
>
> > Hi Pranav,
> >
> > On Thu, Jul 09, 2020 at 01:25:32PM +0530, Pranav Ballaney wrote:
> > > Hi,
> > > I've added autopkgtests to ampliconnoise. Please review and sponsor.
> > > https://salsa.debian.org/med-team/ampliconnoise
> >
> > I decided that its better to have an extra binary package for the
> > test data, created it and uploaded.
> >
> > Thanks for working on this
> >
> > Andreas.
> >
> > --
> > http://fam-tille.de
> >
> >

--
http://fam-tille.de