Re: python-pybedtools: fixing the failing tests
Hi Liubov,
On Sun, Jul 15, 2018 at 02:12:12PM +0300, Liubov Chuprikova wrote:
> > I have started with python-pybedtools and pushed a patch that fixes the
> > tests that were failing. Apart from that, I wrote to the upstream about
> > this bug [1]. At the moment the package builds successfully, although there
> > is a couple of lintian errors and some warnings that I will try to fix.
> >
>
> I fixed some lintian warnings, please, take a look at the changes [2]! I am
> not sure about d/changelog. Will it be right if I just increment the
> release number to 0.7.10-3 leaving its distribution field as UNRELEASED?
We can sort out the changelog before the actual upload. Its a bit
unfortunate that it was removed since passing new is always a time delay
and creates work for others.
> What else needs to be done so that we could get the package back into
> distribution.
It does not build for me in a pbuilder chroot. I get:
======================================================================
FAIL: shuffle, i=gzip ibam=False, {'seed': 1, 'chrom': True, 'genome': 'hg19'}
----------------------------------------------------------------------
Traceback (most recent call last):
File "/usr/lib/python2.7/dist-packages/nose/case.py", line 197, in runTest
self.test(*self.arg)
File "/build/python-pybedtools-0.7.10/.pybuild/cpython2_2.7_pybedtools/build/pybedtools/test/test_iter.py", line 68, in run
assert res == expected
AssertionError:
-------------------- >> begin captured stdout << ---------------------
/build/python-pybedtools-0.7.10/.pybuild/cpython2_2.7_pybedtools/build/pybedtools/test/tmp/pybedtools.rH92eu.tmp
Method call:
BedTool.shuffle(seed=1, chrom=True, genome=hg19)
Got:
chr1 46341498 46341597 feature1 0 +
chr1 45615582 45615682 feature2 0 +
chr1 102762672 102763022 feature3 0 -
chr1 17293432 17293482 feature4 0 +
Expected:
chr1 59535036 59535135 feature1 0 +
chr1 99179023 99179123 feature2 0 +
chr1 186189051 186189401 feature3 0 -
chr1 219133189 219133239 feature4 0 +
Diff:
---
+++
@@ -1,4 +1,4 @@
-chr1 46341498 46341597 feature1 0 +
-chr1 45615582 45615682 feature2 0 +
-chr1 102762672 102763022 feature3 0 -
-chr1 17293432 17293482 feature4 0 +
+chr1 59535036 59535135 feature1 0 +
+chr1 99179023 99179123 feature2 0 +
+chr1 186189051 186189401 feature3 0 -
+chr1 219133189 219133239 feature4 0 +
Showing special characters:
Got:
chr1\t46341498\t46341597\tfeature1\t0\t+\n
chr1\t45615582\t45615682\tfeature2\t0\t+\n
chr1\t102762672\t102763022\tfeature3\t0\t-\n
chr1\t17293432\t17293482\tfeature4\t0\t+\n
Expected:
chr1\t59535036\t59535135\tfeature1\t0\t+\n
chr1\t99179023\t99179123\tfeature2\t0\t+\n
chr1\t186189051\t186189401\tfeature3\t0\t-\n
chr1\t219133189\t219133239\tfeature4\t0\t+\n
> Will it be sufficient if it builds and most lintians are
> fixed?
There is no need to get full lintian cleanlyness.
> Could generating documentation and writing autopkgtest be done
> later, as it is probably more urgent now to take care of other packages?
I agree that we can do tests later. I wonder if it might be
a good idea to implement some generic way to run the build
time test as autopkgtest. This could solve the issue for
several Python packages.
Kind regards
Andreas.
> [1] https://github.com/daler/pybedtools/issues/248
> >
> [2] https://salsa.debian.org/med-team/python-pybedtools/commits/master
--
http://fam-tille.de
Reply to: