RFS: concurrentqueue feels ready to upload

[Steffen Möller]

Upstream offers a series of additional tests. Not all are feasible to
perform at build time, anyway, so having removed some source trees that
go with a reduced set of tests seem fine to me.

Steffen

Re: mecat2 requires ncbi-blast+ sources (yet ncbi-blast+ is not a library)?

[Aaron M. Ucko]

Shayan Doust  writes:

> A simple grep we can get an idea of how extensively the ncbi_blast
> headers and soures are referenced:
>
> https://paste.debian.net/1154499/

In general, ncbi-blast+ comes from NCBI's C++ Toolkit (home page
https://ncbi.github.io/cxx-toolkit/, sources from
https://ftp.ncbi.nih.gov/toolbox/ncbi_tools++/CURRENT), which builds
libraries for public consumption but is a behemoth with its own release
schedule.  Moreover, mecat gives the NCBI code it bundles completely
different paths (with, e.g., cmdline_args and str_util both containing
subsets of the C++ Toolkit's {include,src}/corelib), though I suppose
that's more of a formality.

That said, this duplication is of a sufficiently small portion of
ncbi-blast+ (let alone the entire C++ Toolkit) that I'm OK with letting
it slide, which is certainly fine from a legal perspective (NCBI's code
all being public domain).

Thanks for asking!

-- 
Aaron M. Ucko, KB1CJC (amu at alum.mit.edu, ucko at debian.org)
http://www.mit.edu/~amu/ | http://stuff.mit.edu/cgi/finger/?a...@monk.mit.edu

[RFS] elph

[Nilesh Patra]

Hi,
I've added autopkgtests for elph- builds with passing tests.
I've pushed my changes to the team-repo[1]
Please:

gbp clone --pristine-tar https://salsa.debian.org/med-team/elph

OR

Grant DM access: PGP key fingerprint:
3E99A526F5DCC0CBBF1CEEA600BAE74B343369F1

Thanks and regards
Nilesh

Re: [RFS] elph

[Steffen Möller]

Hi, granted!

If this package is regularly updated then, to avoid too much extra load
on https://snapshot.debian.org/, I suggest to wait for the next upstream
release, though, you decide.

Best,

Steffen

On 30.06.20 17:20, Nilesh Patra wrote:

Hi,
I've added autopkgtests for elph- builds with passing tests.
I've pushed my changes to the team-repo[1]
Please:

gbp clone --pristine-tar https://salsa.debian.org/med-team/elph


OR

Grant DM access: PGP key fingerprint:
3E99A526F5DCC0CBBF1CEEA600BAE74B343369F1

Thanks and regards
Nilesh

Re: [RFS] elph

[Nilesh Patra]

On Tue, 30 Jun 2020, 20:57 Steffen Möller,  wrote:

> Hi, granted!


Thank you :)
Just uploaded.


> If this package is regularly updated then, to avoid too much extra load
> on https://snapshot.debian.org/, I suggest to wait for the next upstream
> release, though, you decide.
>

As per it's release page, the last update was many, many years ago -
probably not as  frequently updated.

There have been bugs filed on this though, which resulted in more (Debian)
releases.

Regards,
Nilesh

>

Re: RFS: concurrentqueue feels ready to upload

[Andreas Tille]

On Tue, Jun 30, 2020 at 02:19:05PM +0200, Steffen Möller wrote:
> Upstream offers a series of additional tests. Not all are feasible to
> perform at build time, anyway, so having removed some source trees that
> go with a reduced set of tests seem fine to me.

Reviewed and uploaded, Andreas. 

-- 
http://fam-tille.de

Re: Need some help with bbmap

[Andreas Tille]

Hi Pranav,

I admit its even worse for me. :-(

On Tue, Jun 30, 2020 at 12:03:22PM +0530, Pranav Ballaney wrote:
> Hi,
> I was trying to add autopkgtests to bbmap [1] but I'm having some trouble
> getting it to work.
> Here is the command I'm trying to run along with the output (and I've
> attached the input file with this email).
> 
> > bbduk.sh in=SRR1753664_1.min.fastq out=clean1.fq

Please somehow commit your input files.  I simply tried:

  bbduk.sh in=something out=anything

> /usr/bin/bbduk.sh: line 348: /usr/share/bbmap/calcmem: No such file or
> directory
> /usr/bin/bbduk.sh: line 349: setEnvironment: command not found
> /usr/bin/bbduk.sh: line 350: parseXmx: command not found
> /usr/bin/bbduk.sh: line 354: freeRam: command not found
> java -Xmxm -Xmsm -cp /usr/share/java/bbmap.jar jgi.BBDuk
> in=SRR1753664_1.min.fastq out=clean1.fq
> Invalid maximum heap size: -Xmxm
> Error: Could not create the Java Virtual Machine.
> Error: A fatal exception has occurred. Program will exit.
>
> It says that it couldn't find calcmem, which is probably because in
> d/rules, line 30 is commented out.
> 
> # cp -a calcmem.sh $(PKGSHAREDIR)/calcmem # we don't have any name
> space collision when keeping .sh extension
> 
> After uncommenting this and building, this error goes away,

Yep.  I commited this.  Always feel free to push your local changes if
you find some enhancement.

> but I get the
> following error:
> 
> > bbduk.sh in=SRR1753664_1.min.fastq out=clean1.fq
> java -ea -Xmx1110m -Xms1110m -cp /usr/share/java/bbmap.jar jgi.BBDuk
> in=SRR1753664_1.min.fastq out=clean1.fq
> Executing jgi.BBDuk [in=SRR1753664_1.min.fastq, out=clean1.fq]
> Version 38.84
> 
> Warning!  Cannot find primes.txt.gz
> /home/pranav/code/debian-med/bbmap/SRR1753664/file:/usr/share/java/bbmap.jar!/primes.txt.gz
> java.lang.Exception
> at dna.Data.findPath(Data.java:1246)
> at dna.Data.findPath(Data.java:1193)
> at shared.Primes.fetchPrimes(Primes.java:167)
> at shared.Primes.(Primes.java:177)
> at kmer.ScheduleMaker.(ScheduleMaker.java:155)
> at jgi.BBDuk.(BBDuk.java:984)
> at jgi.BBDuk.main(BBDuk.java:77)
> Exception in thread "main" java.lang.ExceptionInInitializerError
> at kmer.ScheduleMaker.(ScheduleMaker.java:155)
> at jgi.BBDuk.(BBDuk.java:984)
> at jgi.BBDuk.main(BBDuk.java:77)
> Caused by: java.lang.NullPointerException
> at fileIO.ByteFile.(ByteFile.java:43)
> at fileIO.ByteFile1.(ByteFile1.java:98)
> at fileIO.ByteFile1.(ByteFile1.java:94)
> at shared.Primes.fetchPrimes(Primes.java:169)
> at shared.Primes.(Primes.java:177)
> ... 3 more
> 
> 
> I tried running the same command using the upstream package, and it works
> fine, so it's certainly some issue with the packaging here.
>
> It appears to have something to do with line 33 in d/rules, where the path
> is being replaced from ".DIR""current/" to the jar file.
> I've never worked with .jar files or Java in general, so I don't know what
> the exclamation mark means here: bbmap.jar!/primes.txt.gz
> Is it trying to find primes.txt.gz inside the jar file? If yes, looking at
> d/rules, it appears that nothing from resources/ in the original package is
> being installed.
> If this is indeed the case, please take a look and let me know how I can
> install these files and update the relevant paths.

I think your analysis is correct.  There is no "proper" build procedure
for this package.  Thus I had to use jh_build to get the JAR assembled.
Obviously the files in resources/ dir are missing but needed to let the
tools work properly.  I checked the jh_build manpage which says that it
is possible to add extra files via JH_JAR_EXTRA environment variable
which I tried in my last commit.  Unfortunately it either does not work
or I did it wrong.  I'd be super happy if you could ask for better advise
at debian-j...@lists.debian.org how to really get those files included.

Thanks a lot for spotting this.  It shows again how important all your
work is for Debian Med users.

Kind regards

  Andreas.

> [1] https://salsa.debian.org/med-team/bbmap

-- 
http://fam-tille.de

Re: RFS: concurrentqueue feels ready to upload

[Steffen Möller]

On 30.06.20 21:42, Andreas Tille wrote:

On Tue, Jun 30, 2020 at 02:19:05PM +0200, Steffen Möller wrote:

Upstream offers a series of additional tests. Not all are feasible to
perform at build time, anyway, so having removed some source trees that
go with a reduced set of tests seem fine to me.

Reviewed and uploaded, Andreas.

Thank you!

RFS: readerwriterqueue

[Steffen Möller]

Many thanks!

Steffen