bug#22738: Build failure in vigra

bug#22738: Build failure in vigra

[Danny Milosavljevic]

/tmp/guix-build-vigra-1.10.0.drv-0/build
...
======================================================================
ERROR: test_arraytypes.testMethods
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/gnu/store/m71dy36xl5jcp1qgw9dkyab3yv9gpibx-python2-nose-1.3.7/lib/python2.7/site-packages/nose-1.3.7-py2.7.egg/nose/case.py", line 197, in runTest
    self.test(*self.arg)
  File "/tmp/guix-build-vigra-1.10.0.drv-0/build/vigranumpy/test/test_arraytypes.py", line 1331, in testMethods
    nz = n.nonzero()
  File "/tmp/guix-build-vigra-1.10.0.drv-0/build/vigranumpy/vigra/arraytypes.py", line 1285, in nonzero
    res[k].axistags = AxisTags(AxisInfo(self.axistags[k]))
AttributeError: 'numpy.ndarray' object has no attribute 'axistags'

----------------------------------------------------------------------

bug#22738: Build failure in vigra

[Leo Famulari]

On Fri, Feb 19, 2016 at 02:02:32AM +0100, Danny Milosavljevic wrote:
> /tmp/guix-build-vigra-1.10.0.drv-0/build
> ...
> ======================================================================
> ERROR: test_arraytypes.testMethods
> ----------------------------------------------------------------------
> Traceback (most recent call last):
>   File "/gnu/store/m71dy36xl5jcp1qgw9dkyab3yv9gpibx-python2-nose-1.3.7/lib/python2.7/site-packages/nose-1.3.7-py2.7.egg/nose/case.py", line 197, in runTest
>     self.test(*self.arg)
>   File "/tmp/guix-build-vigra-1.10.0.drv-0/build/vigranumpy/test/test_arraytypes.py", line 1331, in testMethods
>     nz = n.nonzero()
>   File "/tmp/guix-build-vigra-1.10.0.drv-0/build/vigranumpy/vigra/arraytypes.py", line 1285, in nonzero
>     res[k].axistags = AxisTags(AxisInfo(self.axistags[k]))
> AttributeError: 'numpy.ndarray' object has no attribute 'axistags'

It looks like we need this module 'vigranumpy':
https://ukoethe.github.io/vigra/doc-release/vigranumpy/index.html

bug#22738: Build failure in vigra

[Andreas Enge]

On Thu, Feb 18, 2016 at 10:59:11PM -0500, Leo Famulari wrote:
> It looks like we need this module 'vigranumpy':
> https://ukoethe.github.io/vigra/doc-release/vigranumpy/index.html

This is part of the vigra source code. There should soon be a new vigra
release:
   https://mailhost.informatik.uni-hamburg.de/pipermail/vigra/2016-February/001317.html
but it also failed its tests for me (in another place).

I made a bug report on the vigra mailing list:
   https://mailhost.informatik.uni-hamburg.de/pipermail/vigra/2016-February/001319.html

Andreas

bug#22738: Build failure in vigra

[Ricardo Wurmus]

Andreas Enge <andreas@enge.fr> writes:

> On Thu, Feb 18, 2016 at 10:59:11PM -0500, Leo Famulari wrote:
>> It looks like we need this module 'vigranumpy':
>> https://ukoethe.github.io/vigra/doc-release/vigranumpy/index.html
>
> This is part of the vigra source code. There should soon be a new vigra
> release:
>    https://mailhost.informatik.uni-hamburg.de/pipermail/vigra/2016-February/001317.html
> but it also failed its tests for me (in another place).
>
> I made a bug report on the vigra mailing list:
>    https://mailhost.informatik.uni-hamburg.de/pipermail/vigra/2016-February/001319.html

I’m told[1] that this is solved on branch “version-1-11-rc”, but I
haven’t been able to test this yet.  Would anyone else like to try?

~~ Ricardo

[1]: https://github.com/ukoethe/vigra/issues/326#issuecomment-187669279

bug#22738: Build failure in vigra

[Andreas Enge]

On Wed, Feb 24, 2016 at 12:30:38PM +0100, Ricardo Wurmus wrote:
> I’m told[1] that this is solved on branch “version-1-11-rc”, but I
> haven’t been able to test this yet.  Would anyone else like to try?
> [1]: https://github.com/ukoethe/vigra/issues/326#issuecomment-187669279

I have, but there is another problem there (see the second half
of my bug report).

Sorry I did not remember that you had opened an issue.

Andreas

bug#22738: Build failure in vigra

[Efraim Flashner]

[-- Attachment #1: Type: text/plain, Size: 1215 bytes --]

On Wed, 24 Feb 2016 11:32:52 +0100
Andreas Enge <andreas@enge.fr> wrote:

> On Thu, Feb 18, 2016 at 10:59:11PM -0500, Leo Famulari wrote:
>  [...]  
> 
> This is part of the vigra source code. There should soon be a new vigra
> release:
>    https://mailhost.informatik.uni-hamburg.de/pipermail/vigra/2016-February/001317.html
> but it also failed its tests for me (in another place).
> 
> I made a bug report on the vigra mailing list:
>    https://mailhost.informatik.uni-hamburg.de/pipermail/vigra/2016-February/001319.html
> 
> Andreas

As a side note, once this builds libreoffice should build again. Currently
our version of libreoffice is vulnerable to cve-2016-{0794,0795} and should
be upgraded to 5.1.1. I didn't want to push the upgrade until I was sure it
would build correctly, but based on hydra's listed build times I'd be looking
at ~20 hours, so I'm not a good one to push that update when it comes.
https://download.documentfoundation.org/libreoffice/src/5.1.1/

-- 
Efraim Flashner   <efraim@flashner.co.il>   אפרים פלשנר
GPG key = A28B F40C 3E55 1372 662D  14F7 41AA E7DC CA3D 8351
Confidentiality cannot be guaranteed on emails sent or received unencrypted

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

bug#22738: Build failure in vigra

[Christopher Allan Webber]

Andreas Enge writes:

> On Thu, Feb 18, 2016 at 10:59:11PM -0500, Leo Famulari wrote:
>> It looks like we need this module 'vigranumpy':
>> https://ukoethe.github.io/vigra/doc-release/vigranumpy/index.html
>
> This is part of the vigra source code. There should soon be a new vigra
> release:
>    https://mailhost.informatik.uni-hamburg.de/pipermail/vigra/2016-February/001317.html
> but it also failed its tests for me (in another place).
>
> I made a bug report on the vigra mailing list:
>    https://mailhost.informatik.uni-hamburg.de/pipermail/vigra/2016-February/001319.html
>
> Andreas

It looks like this is fixed:
  https://mailhost.informatik.uni-hamburg.de/pipermail/vigra/2016-February/001325.html

Is it possible to get an updated version of this package in, then?

bug#22738: Build failure in vigra

[Ludovic Courtès]

Hi,

Did the update in 019b38758c3895d80a27610d6488ae59455465d6 (March 17)
fix it?

(There are currently no substitutes for this on x86_64.)

Ludo’.

bug#22738: Build failure in vigra

[Leo Famulari]

On Fri, Mar 25, 2016 at 09:44:27AM +0100, Ludovic Courtès wrote:
> Hi,
> 
> Did the update in 019b38758c3895d80a27610d6488ae59455465d6 (March 17)
> fix it?
> 
> (There are currently no substitutes for this on x86_64.)

From the 'fix-letsencrypt' branch on Savannah (which includes that
commit), it works for me on x86_64, but when cross-building to i686, the
test suite fails with:

---
Entering test suite GraphAlgorithmTestSuite

Failure in GraphAlgorithmTest::testEdgeWeightComputation()
Assertion failed: Sequences differ at index 5 [4.24264 != 4.24264]
(/tmp/guix-build-vigra-1.11.0.drv-0/vigra-1.11.0/test/graph_algorithm/test.cxx:543)

1 of 6 tests failed in test suite GraphAlgorithmTestSuite
---

So, 4.24264 != 4.24264?

bug#22653: bug#22738: Build failure in vigra

[Ludovic Courtès]

Vigra builds fine now, presumably since commit
019b38758c3895d80a27610d6488ae59455465d6 by Andreas, so closing this
bug.

Ludo’.