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

Re: Packaging SMALT for Debian



Hi Andreas,

Fine with me if you would like to forward our conversation to the list.

The attached patch should fix the glitch with the dependencies of bam_cigar_test.py you correctly identified. Perhaps I should try and package smalt myself but I have no experience with this and currently not much time.

Kind regards,
/Hannes

On 09/04/14 09:28, Andreas Tille wrote:
[Metaquestion: Usually we do all discussion in public on our
  mailing list.  Would you give me permission to forward our
  conversation to the list?]

Hi again,

I made some step forward with the tests (see below)

On Tue, Apr 08, 2014 at 07:38:54PM +0200, Andreas Tille wrote:
ERROR when mapping: returned with code 1
FAIL: bam_cigar_test.py
=================================
1 of 7 tests failed
Please report to hp3@sanger.ac.uk
=================================
According to the logfile this is related to the fact that bambamc
library is missing.  While I plan to package bambamc library as well
(just contacted the author[1]) for now I builded the package with the
option --with-bambamc=no.  Unfortunately this did not helped excluding
the test and I think the problem is in the test/Makefile.am which says:

if WITH_BAMBAMC
TESTS += ioform_test.py xali_test.py
endif

if WITH_SAMTOOLS_EXEC
TESTS += bam_cigar_test.py
endif


So bam_cigar_test.py is not excluded from the tests if bambamc is
missing.  I tried some patch to Makefile.am (attached) but this fails
unfortunately.  I have to admit I had to fight through strange automake
warnings telling me I should enable

    AUTOMAKE_OPTIONS = subdir-objects

and to set

    AM_PROG_AR

in the main Makefile.am.  Perhaps these are some "features" of later
automake versions.  Unfortunately I ended up with a Makefile which
contains a line

    ... am__expect_failure=no;; esac;  /usr/bin/python /bin/bash ../test-driver --test-name "$f"

which is totally stupid since python tries to call /bin/bash which for
sure ends up in a failure.  I have no clue why automake might create
this nonsense and I think I'll leave it to you how to drop the
bam_cigar_test.py apropriately from the tests if bambamc is missing.

Kind regards

          Andreas.


[1] https://lists.debian.org/debian-med/2014/04/msg00023.html



--
Hannes Ponstingl, Ph.D.
Sequence Assembly & Analysis Group     Tel: +44 (0)1223 834244  x7345
The Wellcome Trust Sanger Institute    Fax: +44 (0)1223 494919
Hinxton, Cambridge CB10 1SA, UK.       E-mail: hp3@sanger.ac.uk




--
The Wellcome Trust Sanger Institute is operated by Genome Research Limited, a charity registered in England with number 1021457 and a company registered in England with number 2742969, whose registered office is 215 Euston Road, London, NW1 2BE.
--- a/test/Makefile.am
+++ b/test/Makefile.am
@@ -89,11 +89,10 @@ TESTS += ${TEST_DRIVERS}
 
 if WITH_BAMBAMC
 TESTS += ioform_test.py xali_test.py
-endif
-
 if WITH_SAMTOOLS_EXEC
 TESTS += bam_cigar_test.py
 endif
+endif
 
 else
 TESTS =

Reply to: