make ptest - best way to fix errors?

asked 2015-04-16 04:25:36 +0100

lfearnley gravatar image

Hi all,

I've installed a few extra modules and run make/make ptest. A few of the tests are failing (details at the end of the post):

sage -t --warn-long 28.0 src/sage/geometry/fan.py # 4 doctests failed sage -t --warn-long 28.0 src/sage/homology/examples.py # 2 doctests failed sage -t --warn-long 28.0 src/sage/homology/chain_complex.py # 2 doctests failed

What's the best way to go about fixing these problems/figuring out what's going wrong here? Is there a protocol beyond 'post the log and open a bug on the tracker'?

I'm compiling under 64-bit OSX 10.10, with current Xcode command line tools installed.

In Homology/Chain Complex:

**********************************************************************
File "src/sage/homology/chain_complex.py", line 1137, in sage.homology.chain_complex.ChainComplex_class.homology
Failed example:
D.homology()
Expected:
{0: 0, 1: 0, 4: 0, 5: 0}
Got:
{}
**********************************************************************
File "src/sage/homology/chain_complex.py", line 1403, in sage.homology.chain_complex.ChainComplex_class.torsion_list
Failed example:
C.torsion_list(5)
Expected:
[(2, [1]), (3, [3])]
Got:
[(2, []), (3, [])]
**********************************************************************

In SimplicalComplexExamples:

**********************************************************************
File "src/sage/homology/examples.py", line 1325, in sage.homology.examples.SimplicialComplexExamples.SumComplex
Failed example:
    S.homology(1)
Expected:
    C140183
Got:
    0
**********************************************************************
File "src/sage/homology/examples.py", line 1331, in sage.homology.examples.SimplicialComplexExamples.SumComplex
Failed example:
    S.homology(1)
Expected:
    C5670599
Got:
    0
**********************************************************************

In Geometry/Fan:

**********************************************************************
File "src/sage/geometry/fan.py", line 3282, in sage.geometry.fan.?.complex
Failed example:
    K_extended.homology()
Expected:
    {-1: 0, 0: 0, 1: 0, 2: 0, 3: 0}
Got:
    {}
**********************************************************************
File "src/sage/geometry/fan.py", line 3320, in sage.geometry.fan.?.complex
Failed example:
    fan.complex().homology()
Expected:
    {0: 0, 1: 0, 2: Z x Z, 3: 0}
Got:
    {2: Z x Z, 3: 0}
**********************************************************************
File "src/sage/geometry/fan.py", line 3324, in sage.geometry.fan.?.complex
Failed example:
    fan.complex().homology()
Expected:
    {0: 0, 1: 0, 2: Z, 3: 0}
Got:
    {2: Z, 3: 0}
**********************************************************************
File "src/sage/geometry/fan.py", line 3327, in sage.geometry.fan.?.complex
Failed example:
    fan.complex().homology()
Expected:
    {0: 0, 1: 0, 2: 0, 3: 0}
Got:
    {}
**********************************************************************
edit retag flag offensive close merge delete

Comments

Which extra packages - perhaps Chomp, which has caused errors in the past? Also, note that currently Xcode 6.3 apparently has a cpp compiler bug or something, though I think the symptoms were a failed compilation, not obscure homology errors!

kcrisman gravatar imagekcrisman ( 2015-04-16 05:13:32 +0100 )edit

I do have chomp installed - full list is:

4ti2, biopython, chomp, cryptominisat, dot2tex, gap_packages, gcc, git, gnuplotpy, guppy, nauty, nose, pyx, pyzmq, qhull, zeromq, PyVTK, gnuplot, graphviz, latte_int, and pexpect.

I've had a look around for methods to remove chomp for testing purposes - couldn't find anything much. Is there a list of installed files that need deleting somewhere?

lfearnley gravatar imagelfearnley ( 2015-04-17 02:27:09 +0100 )edit

I don't have 10.10 but I would be pretty surprised if this wasn't somehow related to 10.10+chomp+6.6 or something like that. I think reporting a Trac ticket would be better than trying to diagnose it here, since it probably will need a patch eventually anyway.

kcrisman gravatar imagekcrisman ( 2015-04-17 03:34:17 +0100 )edit
1

This might be related to http://trac.sagemath.org/ticket/16364.

John Palmieri gravatar imageJohn Palmieri ( 2015-04-19 01:55:36 +0100 )edit