Re: Bio-Linux vs Debian-Med package lists [update]

[Tony Travis]

On 11/03/18 13:28, Andreas Tille wrote:
> Hi Tony,
> 
> it would be nice if you could add some clarifying comments to some
> question marks I added to the list.  Meanwhile I ITPed and uploaded the
> missing R packages - at least those than can be packaged for licensing
> reasons.  If I do not hear from you I will soon remove those packages
> from the list that are obvious false positives.

Hi, Andreas.

I compared the Bio-Linux 'master' list of packages added to Ubuntu in
the Bio-Linux .iso with your Debian-Med/Bio* tasks. The false positives
appear to be packages installed from the main Ubuntu/Debian repositories
and are not part of your Debian-Med/Bio* tasks e.g. bioperl-run:

> ajt@beluga:~$ apt-cache policy bioperl-run
> bioperl-run:
>   Installed: 1.6.9-2
>   Candidate: 1.6.9-2
>   Version table:
>  *** 1.6.9-2 0
> 500 http://gb.archive.ubuntu.com/ubuntu/ trusty/universe amd64 
> Packages
> 100 /var/lib/dpkg/status

I'll check your '???' comments and report back here.

Bye,

  Tony.

-- 
Minke Informatics Limited, Registered in Scotland - Company No. SC419028
Registered Office: 3 Donview, Bridge of Alford, AB33 8QJ, Scotland (UK)
tel. +44(0)19755 63548http://minke-informatics.co.uk
mob. +44(0)7985 078324mailto:tony.tra...@minke-informatics.co.uk

Re: Outreachy round 16 | Applying to Debian

[Andreas Tille]

Dear Rupal,

On Mon, Mar 12, 2018 at 11:23:34PM +0530, Rupal Jain wrote:
> 
> That's great.
> To confirm, by "teaching" you're referring to MoM program, right?

Well, that's kind of a formalised teaching method but we have taught
newcomers also without this formalism.

> You can let me know what time suits you when we can schedule our first
> discussion. We'll try to keep it as early as possible.

So lets start right now. 

> Any medium suits me.

I prefer e-mail.

> > In our last sprint we started developing a new list about packages that
> > are in Bio-Linux but not in Debian[1].  As a random pick I would think
> > that cap3[2] might be simple enough for a start.  Disclaimer:  I have
> > not done a single look into the source at all.  The most frequent
> > blocker for this kind of software is a missing license - feel free to
> > have a look here.
> 
> It's cool with me. We can start with any med package with which you're
> smooth and comfortable teaching me.
> And, which is the link that you're referring for license? Couldn't get it.

The link to cap3 is at the end of this mail[2].  If I would have a link
to the license I would have provided it.  The page contains a source
code archive but I have not seen any link to a license.  Part of the
Debian developer work is to do research (may be by contacting the
authors) to find out the license.
 
> > That's fine.  I'd recommend you get your hands on a Debian installation
> > which should you give some practical understanding.  There you can do
> > for instance
> >
> >  sudo apt install med-bio
> >
> > (and let us know if you have trouble to come to this point!)
> 
> Yeah. On it.

Let me know about success or failure.
 
> > Than you should create a login on salsa.debian.org and apply for
> > membership in the Debian Med team.
> 
> Done.

I did not received any application message via the Gitlab interface
but I've manually added you to the project.

> > As I said I'm happy about your interest and will support your attempt to
> > start with the packaging.
> 
> I am grateful for your willingness to help me.

You are welcome.  I think the first task is to create a Git archive for
cap3 now and inject the source archive.  Please let me know whether you
see any blockers here.

Kind regards

   Andreas.

[1] 
https://salsa.debian.org/med-team/community/bio-linux/blob/master/debian-med_missings.txt
[2] http://seq.cs.iastate.edu/cap3.html

-- 
http://fam-tille.de

Re: Outreachy round 16 | Applying to Debian

[Rupal Jain]

‌
On Sun, Mar 11, 2018 at 2:16 PM, Andreas Tille  wrote:

> Dear Rupal,
>
> thanks for your introduction and your interest in the Debian Med project.
>
> On Sun, Mar 11, 2018 at 01:17:41AM +0530, Rupal Jain wrote:
> > ...
> > In addition to just sounding cool, the work looks challenging too. Being
> a
> > total beginner to Debian packaging or concepts along this line, the
> project
> > surely has a good learning curve for me.
>
> The Debian Med project as quite some record in teaching packaging (as
> you are writing below).  To get you starting for the GSoC project we
> should start with this soon.
>

That's great.
To confirm, by "teaching" you're referring to MoM program, right?
You can let me know what time suits you when we can schedule our first
discussion. We'll try to keep it as early as possible.
Any medium suits me.

>
> > 1. First and foremost, I have subscribed to your mailing list (have also
> > cc'ed it here as I have read that you guys prefer to keep most of your
> > conversations public, which does make sense to me).
>
> Very good - so I just write to the list (as per Debian list policy).
>
> > 3. I have gone through this
> >  policy document
> as
> > I saw that you have redirected many individuals to this document as an
> > essential reading. I got to know about Debian Pure Blend, motivation and
> > aim of the "Debian Med project", and also reading like, "Developers
> > reference" and "New Maintainers guide", although I couldn't understand
> the
> > technical portion too well.
>
> It seems you did quite well in studying the available documents. :-)
>
> > 4. I also came across the *MoM *program here
> > , I want to be a part of it and
> in
> > the process of learning Debian packaging during MoM, I can also make
> > appropriate contributions to the project for Outreachy.
>
> Very good. :-)
>
> > 5. On this  tasks page, it shows a
> > whole list of Debian Med packages, so is there a particular package on
> > which I will be working on in the beginning or is it still to be decided?
>
> In principle I prefer if you might have used some software in your own
> work that is not yet packaged since you know the software.  Since in
> your current state of education you might not have stumbled upon such a
> candidate you can pick freely from the list according to your interest.
>

Yeah, you rightly mentioned that as per my current exposure I haven't used
any such software in my work yet.

In our last sprint we started developing a new list about packages that
> are in Bio-Linux but not in Debian[1].  As a random pick I would think
> that cap3[2] might be simple enough for a start.  Disclaimer:  I have
> not done a single look into the source at all.  The most frequent
> blocker for this kind of software is a missing license - feel free to
> have a look here.


It's cool with me. We can start with any med package with which you're
smooth and comfortable teaching me.
And, which is the link that you're referring for license? Couldn't get it.


>
> > 6. Is this  the right link for me to
> head
> > towards setting up Debian environment?
>
> Yes.
>
> > I have just looked up to the internet to know what do we *mean *by
> > packaging in Debian, to get a gist of what is there to be done, but this
> > gave me a very broad view/glimpse of what it means, but only
> *theoretically.
>
> Many Linux distributions consists of a set of prebuilded binaries in
> ready to install archives that are featuring metainformation about
> dependencies between each others - so called packages.  There are two
> wide spread packaging formats: The Debian format which is used by Debian
> and its derivatives (Ubuntu, Mint and lots of others) and the Redhat
> package format which is used by RedHat, SuSE and others.  These packages
> are allowing one click installation and maintenance of very complex
> software systems.  The job of a Debian maintainer is to download a
> package source and create some files in a predefined directory debian/
> that contain the needed information to build the Debian package.  How
> this is done is explained in the newbie documentation you have mentioned
> in your initial mail.  I hope this is some sufficient information what
> "packaging" might mean.


> The sense of the Debian Med project is to package free software that is
> relevant in life science and medicine.  For instance the packaging of
> this software enables setting up a bioinformaticans workstation with a
> single command line / mouse click (when using a graphical installer).
>

Yup. I can make sense out of what you explained.


>
> > *But to get into actual coding and building packages and understanding
> the
> > *workflow*, without which there wouldn't be much of a point to go on
> > solving bugs, I'll need help (as a beginner would do). Like MoM program,
> > 

Fwd: jellyfish_2.2.8-2_source.changes REJECTED

[Michael Crusoe]

Hi hi,

Can I be given permissions to upload new packages for Jellyfish?

dcut dm --uid 724D609337113C710550D7473C26763F6C67E6E2 --allow jellyfish

Thanks!
-- Forwarded message --
From: "Debian FTP Masters" 
Date: Mar 12, 2018 14:54
Subject: jellyfish_2.2.8-2_source.changes REJECTED
To: "Michael R. Crusoe" , "Debian Med Packaging
Team" 
Cc:


>
> ACL dm: not allowed to upload source package 'jellyfish'
>
>
>
> ===
>
> Please feel free to respond to this email if you don't understand why
> your files were rejected, or if you upload new files which address our
> concerns.
>
>

Re: RFS: ChromHMM -- Chromatin state discovery and characterization

[Dylan Aïssi]

Hi Andreas,

2018-03-12 10:00 GMT+01:00 Andreas Tille :

> I've asked upstream for release tags[1].  May be upstream will set those
> tags and you do not need a get-orig-source script any more but if not it
> makes sense to use such a script.  I've provided an example that even
> removes Files-Excluded in our packaging example[2] (at least if you
> specify the files to exclude in a single line separated by spaces).

Thanks for this. The other point here is about example files, the
github archive doesn't provides these files whereas the archive on
their main website provides them.
Of course, I can put these files into debian/ but I prefer to not have
these ~30Mo in the debian/ folder.
I will try to convince upstream to put these files in github to have a
clean archive.

> Regarding the installation of ChromHMM.jar:  I'm not very picky about
> this but I'm usually installing JAR files to /usr/share/java but I
> accept your decision to use /usr/share/chromhmm/ instead.  Just want
> to make sure that this is a decision you have drawn intentionally.

Honestly, I used the same method as I did before for beagle packaging
without asking myself more questions (and I don't remember why I did
this for beagle).
Thanks for pointing this, I will fix this for the next upload.

> I've run `cme fix dpkg-control` and uploaded the current state of Git to
> NEW.

Oups, sorry to forgot the "cme fix".

Thanks for the upload.

Best,
Dylan

Re: RFS: ChromHMM -- Chromatin state discovery and characterization

[Andreas Tille]

Hi Dylan,

On Sun, Mar 11, 2018 at 11:55:46PM +0100, Dylan Aïssi wrote:
> Could someone upload it? Thanks.

I've asked upstream for release tags[1].  May be upstream will set those
tags and you do not need a get-orig-source script any more but if not it
makes sense to use such a script.  I've provided an example that even
removes Files-Excluded in our packaging example[2] (at least if you
specify the files to exclude in a single line separated by spaces).

Regarding the installation of ChromHMM.jar:  I'm not very picky about
this but I'm usually installing JAR files to /usr/share/java but I
accept your decision to use /usr/share/chromhmm/ instead.  Just want
to make sure that this is a decision you have drawn intentionally.

I've run `cme fix dpkg-control` and uploaded the current state of Git to
NEW.

Thanks for your work on this

   Andreas.

[1] https://github.com/jernst98/ChromHMM/issues/11
[2] 
https://salsa.debian.org/med-team/community/package_template/blob/master/debian/get-orig-source_Files-Excluded

-- 
http://fam-tille.de