[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: I would like to submit a package to debian-med



Hi Andreas, all,


On Mon, Dec 2, 2013 at 1:29 PM, Andreas Tille <andreas@an3as.eu> wrote:
Well, if you are reading our mailing list, provide information what
should be packaged or how we could enhancing packaging thit would be
really cool.  Moreover becoming a member of the Alioth team and commit
packaging code.  Serving as proxy in case of things we would like to
ask upstream at Sanger and so on.


I'm happy to help in that role.
Also, now with Sascha Steinbiss (working by my side and helping a lot in the creation of the snp-sites package), there might be a chance of creating critical mass here at Sanger.
Let's keep this line of conversation open.
 
> The Sanger Institute is composed of several independent faculties I fall
> under one of these.

Sure - you probably can not deal with all Sanger faculties.


Maybe we could actually create a group here, establish a mailing list and this would in turn advertise the existence of a Debian group on site to the other faculties.
 
> But it would be very interesting for someone to have more of a central role
> in packaging Sangers mature code for Debian.

There used to be official Debian developers at Sanger but they did not
played an important role here inside the Debian Med team
(unforutunately). 

I guess they were extremely busy with other tasks... :(

:)


> > I guess you are talking about
> >
> >    https://github.com/sanger-pathogens/snp_sites
> >Just do some

    git tag 0.1
    git push --tags

would be perfectly sufficient (If I'm not totally wrong - I do not use
github).  can you publish your work you mentioned at some place to let
us have a look?


I've done that already, but I still need to decide with the upstream developer which version does he want to be the first.
He had already tagged it as version 1.0.
 

It might be worth digging in git-buildpackage where you can leave the
.git structure.  See Debian Med policy for this.


I will. Thank you.
 
> My initial procedure was, as I said above, to strip away all git related
> stuff and build the package as simply as possible.
> I also had a look at gitbuilder procedures, but didn't follow this line as
> I was advised to keep it simple.

I agree that making it the most simple is a good way to start.  So in
principle there is nothing wrong with your approach.


Brilliant.
I actually have a binary debian package with me at the moment.
Again, Sascha has pointed me in the right direction many times for the past couple of days.
 
> I am going through these documents and even if not for this package, I
> would be quite keen to go through the mentoring of the month method later
> in February.
> Possibly after the Stonehaven meet up which I am attending.

That's very cool if we can meet in Stonehaven!!!


It will indeed. :)
 
As for the package, I have a couple of questions.
As instructed by Sascha, I have run lintian over the snp-sites package.

This is the output I get (notice the new version number on the package name. This will all be made uniform before official deployment):

js21@builder:~/tinker/snp_sites-1$ lintian /tmp/snp-sites_1-1_amd64.deb
W: snp-sites: package-name-doesnt-match-sonames libsnp-sites1
W: snp-sites: hardening-no-relro usr/bin/snp_sites
W: snp-sites: hardening-no-relro usr/lib/libsnp_sites.so.1.0.0
W: snp-sites: hardening-no-fortify-functions usr/lib/libsnp_sites.so.1.0.0
W: snp-sites: wrong-bug-number-in-closes l3:#nnnn
W: snp-sites: new-package-should-close-itp-bug
E: snp-sites: helper-templates-in-copyright
W: snp-sites: copyright-has-url-from-dh_make-boilerplate
E: snp-sites: copyright-contains-dh_make-todo-boilerplate
W: snp-sites: readme-debian-contains-debmake-template
E: snp-sites: description-synopsis-is-duplicated
E: snp-sites: file-directly-in-usr-share usr/share/alignment_file.h
E: snp-sites: file-directly-in-usr-share usr/share/fasta_of_snp_sites.h
E: snp-sites: file-directly-in-usr-share usr/share/kseq.h
E: snp-sites: file-directly-in-usr-share usr/share/parse_phylip.h
E: snp-sites: file-directly-in-usr-share usr/share/phylib_of_snp_sites.h
E: snp-sites: file-directly-in-usr-share usr/share/snp_sites.h
E: snp-sites: file-directly-in-usr-share usr/share/string_cat.h
E: snp-sites: file-directly-in-usr-share usr/share/vcf.h
W: snp-sites: binary-without-manpage usr/bin/snp_sites
W: snp-sites: non-dev-pkg-with-shlib-symlink usr/lib/libsnp_sites.so.1.0.0 usr/lib/libsnp_sites.so

I was wondering if it would be better to define:

   /usr/share/snp-sites/

Has the place holder for the header files rather than bluntly on /usr/share/

I would like to hear/read your thoughts on this.

Sascha has advised me to use asciidoc as a step in the rules file to convert the .txt manpages ( that I haven't written yet ), into actual manpages. Again this will be adressed before official release.

I have also tried to run piuparts on the recently crerated snp-sites_1-1_amd64.deb in this fashion:

js21@builder:~/tinker/snp_sites-1$ sudo piuparts -b /var/cache/pbuilder/sid-amd64-base.tgz /tmp/snp-sites_1-1_amd64.deb

But it keeps complaining about the lack of a CD-ROM. Here's the output I got:

js21@builder:~/tinker/snp_sites-1$ sudo piuparts -b /var/cache/pbuilder/sid-amd64-base.tgz /tmp/snp-sites_1-1_amd64.deb
[sudo] password for js21:
Guessed: debian
0m0.0s INFO: ------------------------------------------------------------------------------
0m0.0s INFO: To quickly glance what went wrong, scroll down to the bottom of this logfile.
0m0.0s INFO: FAQ available at http://wiki.debian.org/piuparts/FAQ
0m0.0s INFO: ------------------------------------------------------------------------------
0m0.0s INFO: piuparts version 0.45 starting up.
0m0.0s INFO: Command line arguments: /usr/sbin/piuparts -b /var/cache/pbuilder/sid-amd64-base.tgz /tmp/snp-sites_1-1_amd64.deb
0m0.0s INFO: Running on: Linux builder 3.2.0-4-amd64 #1 SMP Debian 3.2.51-1 x86_64
0m0.0s DEBUG: Starting command: ['dpkg', '--info', '/tmp/snp-sites_1-1_amd64.deb']
0m0.2s DUMP:
   new debian package, version 2.0.
   size 20144 bytes: control archive=1084 bytes.
       396 bytes,    11 lines      control             
       982 bytes,    16 lines      md5sums             
       135 bytes,     7 lines   *  postinst             #!/bin/sh
       132 bytes,     7 lines   *  postrm               #!/bin/sh
        25 bytes,     1 lines      shlibs              
   Package: snp-sites
   Version: 1-1
   Architecture: amd64
   Maintainer: Jorge Soares <js21@sanger.ac.uk>
   Installed-Size: 107
   Depends: libc6 (>= 2.14), zlib1g (>= 1:1.1.4)
   Section: science
   Priority: extra
   Homepage: https://github.com/sanger-pathogens/snp_sites
   Description: <This program finds snp sites from a multi fasta alignment file>
    <This program finds snp sites from a multi fasta alignment file>
0m0.2s DEBUG: Command ok: ['dpkg', '--info', '/tmp/snp-sites_1-1_amd64.deb']
0m0.2s DEBUG: Created temporary directory /tmp/tmpf0i3XK
0m0.2s DEBUG: Unpacking /var/cache/pbuilder/sid-amd64-base.tgz into /tmp/tmpf0i3XK
0m0.2s DEBUG: Starting command: ['tar', '-C', '/tmp/tmpf0i3XK', '-zxf', '/var/cache/pbuilder/sid-amd64-base.tgz']
0m10.1s DEBUG: Command ok: ['tar', '-C', '/tmp/tmpf0i3XK', '-zxf', '/var/cache/pbuilder/sid-amd64-base.tgz']
0m10.1s DEBUG: Starting command: ['chroot', '/tmp/tmpf0i3XK', 'mount', '-t', 'proc', 'proc', '/proc']
0m10.2s DEBUG: Command ok: ['chroot', '/tmp/tmpf0i3XK', 'mount', '-t', 'proc', 'proc', '/proc']
0m10.5s DEBUG: Created policy-rc.d and chmodded it.
0m10.5s DEBUG: Starting command: ['chroot', '/tmp/tmpf0i3XK', 'apt-get', 'update']
0m10.7s DUMP:
  Ign cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy InRelease
  Ign cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy Release.gpg
  Ign cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy Release
  Err cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy/main amd64 Packages
    Please use apt-cdrom to make this CD-ROM recognized by APT. apt-get update cannot be used to add new CD-ROMs
  Ign cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy/main Translation-en
  W: Failed to fetch cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23]/dists/wheezy/main/binary-amd64/Packages  Please use apt-cdrom to make this CD-ROM recognized by APT. apt-get update cannot be used to add new CD-ROMs
 
  E: Some index files failed to download. They have been ignored, or old ones used instead.
0m10.7s ERROR: Command failed (status=100): ['chroot', '/tmp/tmpf0i3XK', 'apt-get', 'update']
  Ign cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy InRelease
  Ign cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy Release.gpg
  Ign cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy Release
  Err cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy/main amd64 Packages
    Please use apt-cdrom to make this CD-ROM recognized by APT. apt-get update cannot be used to add new CD-ROMs
  Ign cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23] wheezy/main Translation-en
  W: Failed to fetch cdrom://[Debian sid 7 _Wheezy_ - Official Snapshot amd64 LIVE/INSTALL Binary 20131013-23:23]/dists/wheezy/main/binary-amd64/Packages  Please use apt-cdrom to make this CD-ROM recognized by APT. apt-get update cannot be used to add new CD-ROMs
 
  E: Some index files failed to download. They have been ignored, or old ones used instead.
 
0m13.8s DEBUG: Starting command: ['chroot', '/tmp/tmpf0i3XK', 'umount', '/proc']
0m13.9s DEBUG: Command ok: ['chroot', '/tmp/tmpf0i3XK', 'umount', '/proc']
0m14.6s DEBUG: Removed directory tree at /tmp/tmpf0i3XK


Currently my Debian dist is running as a VM and I am behind a proxy.
I have HTTP_PROXY set and pbuilderrc was configured accordingly.
My .bash_aliases also sets this ENV variable.

I wonder if I need to do something else to get piuparts to run properly on this package.

If this is not the proper channel for all this, I am sorry.

Kind regards,

Jorge




Reply to: