test_to_records fails on BE boxes: (0L, 'a') != (0L, 'a')

[yarikoptic]

see e.g. https://buildd.debian.org/status/fetch.php?pkg=pandas&arch=sparc&ver=0.15.1%2Bgit125-ge463818-1&stamp=1417703234&file=log

======================================================================
FAIL: test_to_records (pandas.tests.test_categorical.TestCategoricalAsBlock)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/«BUILDDIR»/pandas-0.15.1+git125-ge463818/debian/tmp/usr/lib/python2.7/dist-packages/pandas/tests/test_categorical.py", line 2393, in test_to_records
    tm.assert_almost_equal(result,expected)
  File "das/src/testing.pyx", line 58, in pandas._testing.assert_almost_equal (pandas/src/testing.c:2745)
  File "das/src/testing.pyx", line 93, in pandas._testing.assert_almost_equal (pandas/src/testing.c:1830)
  File "das/src/testing.pyx", line 143, in pandas._testing.assert_almost_equal (pandas/src/testing.c:2618)
AssertionError: (0L, 'a') != (0L, 'a')

[jreback]

odd, those look the same?

[yarikoptic]

On Thu, 04 Dec 2014, jreback wrote:

odd, those look the same?

that is the fun part indeed ;)

Yaroslav O. Halchenko, Ph.D.
http://neuro.debian.net http://www.pymvpa.org http://www.fail2ban.org
Research Scientist, Psychological and Brain Sciences Dept.
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834 Fax: +1 (603) 646-1419
WWW: http://www.linkedin.com/in/yarik

[yarikoptic]

feels like a numpy buy somewhere

(Pdb) print result
[(0L, 'a') (1L, 'b') (2L, 'c')]
(Pdb) print expected
[(0L, 'a') (1L, 'b') (2L, 'c')]
*(Pdb) print result==expected
False
*(Pdb) print result[0]==expected[0]
False
*(Pdb) print result[0][0]==expected[0][0]
True
*(Pdb) print result[0][1]==expected[0][1]
True
*(Pdb) print type(result[0]), type(expected[0])  
<class 'numpy.core.records.record'> <class 'numpy.core.records.record'>
(Pdb) print result[0].dtype
[('index', '>i8'), ('0', 'O')]
*(Pdb) print result[1].dtype
[('index', '>i8'), ('0', 'O')]

[jreback]

@yarikoptic originally gave this the wrong name, but please give #9015 a test

[yarikoptic]

my name was just the failure message, bit sure his it can be wrong.
anyways, look at my pdb session above, it it's not just a matter of array comparison

[jreback]

dupe of #10425 (and closed by #10438)

[jreback]

@yarikoptic the fix was merged in #10438 should fix the big-endian issues. If there are most pls post.

[yarikoptic]

On Fri, 26 Jun 2015, jreback wrote:

@yarikoptic the fix was merged in #10438 should fix the big-endian issues.
If there are most pls post.

ok... I will then attempt a fresh build across

Cheers!

Yaroslav O. Halchenko, Ph.D.
http://neuro.debian.net http://www.pymvpa.org http://www.fail2ban.org
Research Scientist, Psychological and Brain Sciences Dept.
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834 Fax: +1 (603) 646-1419
WWW: http://www.linkedin.com/in/yarik