Hi,

On 15 June 2011 21:11, Tom Bachmann <[email protected]> wrote:

> On 15.06.2011 20:08, Aaron Meurer wrote:
>
>> The int cache test tests the integer cache, which is not the same
>> after a second execution (it tests that something is *not* in the
>> cache).  See http://code.google.com/p/sympy/issues/detail?id=1946.
>>
>> The mpmath test fails because test_lambdify.py sets mpmath.mp.dps,
>> which changes the precision globally.  Is there a way to use a
>> different precision in that test without changing it globally?
>>
>>
> Isn't the common idiom to do
>
> oldprec = mpmath.dps
> mpmath.dps = ...
> ...
> mpmath.dps = oldprec


The best way is to do:

dps = mpmath.dps
mpmath.dps = ...

try:
    ...
finally:
    mpmath.dps = dps

(until we will be able to use with statement in the library).


>
>
> ?
>
>  Aaron Meurer
>>
>> On Wed, Jun 15, 2011 at 12:04 PM, Aaron Meurer<[email protected]>
>>  wrote:
>>
>>> Regarding the last two, it seems that testing warnings is not as easy
>>> as I though.  To quote
>>> http://docs.python.org/library/warnings.html?highlight=warnings:
>>>
>>> "One thing to be aware of is that if a warning has already been raised
>>> because of a once/default rule, then no matter what filters are set
>>> the warning will not be seen again unless the warnings registry
>>> related to the warning has been cleared."
>>>
>>> I can't figure out how to actually reset this registry
>>> (warnings.resetwarnings() does not do it).  So I will XFAIL these
>>> tests and create an issue for this.
>>>
>>> Aaron Meurer
>>>
>>> On Wed, Jun 15, 2011 at 11:49 AM, Aaron Meurer<[email protected]>
>>>  wrote:
>>>
>>>> Hi.
>>>>
>>>> Thanks for reporting this.  I think I can fix the bugs you give below,
>>>> but I don't think I can fix the Windows bugs without some help, as I
>>>> don't have Windows and can't reproduce those on my Mac.  Do you think
>>>> you could help me debug them?  If possible, pull in the latest 0.7.0
>>>> branch on Windows, because I'll be pushing in fixes for the below
>>>> errors there soon.
>>>>
>>>> Aaron Meurer
>>>>
>>>> On Wed, Jun 15, 2011 at 2:22 AM, Renato Coutinho
>>>> <[email protected]>  wrote:
>>>>
>>>>> Hello,
>>>>>
>>>>> I just tested the new rc. In linux 64-bits using tox, all tests pass
>>>>> with python and gmpy ground types.
>>>>>
>>>>> In Windows XP, all tests pass with python 2.5-2.7. With python2.4 I
>>>>> get a lot of errors, and the test suite doesn't even finish. I
>>>>> uploaded the output to https://gist.github.com/1026644. I'm not sure
>>>>> if it's worthwhile to pursue all those errors though.
>>>>>
>>>>> I also noticed that caching introduces some errors. In python2.7 in
>>>>> linux, if I run sympy.test() twice in the same session, I get the
>>>>> errors below. Fortunately, none of the failures look too dangerous,
>>>>> but it shows caching does have some colateral effects.
>>>>>
>>>>> Renato
>>>>>
>>>>>
>>>>> _____________ sympy/core/tests/test_numbers.py:test_integers_cache
>>>>> _____________
>>>>>  File
>>>>> "/home/renato/down/sympy-0.7.0.rc2/sympy/core/tests/test_numbers.py",
>>>>> line 14, in test_integers_cache
>>>>>    value += 1
>>>>> UnboundLocalError: local variable 'value' referenced before assignment
>>>>>
>>>>>
>>>>> ________________________________________________________________________________
>>>>> _________ sympy/assumptions/tests/test_query.py:test_key_extensibility
>>>>> _________
>>>>>  File
>>>>> "/home/renato/down/sympy-0.7.0.rc2/sympy/assumptions/tests/test_query.py",
>>>>> line 976, in test_key_extensibility
>>>>>    raises(AttributeError, "ask(Q.my_key(x))")
>>>>>  File "sympy/utilities/pytest.py", line 49, in raises
>>>>>    raise AssertionError("DID NOT RAISE")
>>>>> AssertionError: DID NOT RAISE
>>>>>
>>>>> ________________________________________________________________________________
>>>>> __________ sympy/core/tests/test_numbers.py:test_conversion_to_mpmath
>>>>> __________
>>>>>  File
>>>>> "/home/renato/down/sympy-0.7.0.rc2/sympy/core/tests/test_numbers.py",
>>>>> line 703, in test_conversion_to_mpmath
>>>>>    assert mpmath.mpmathify(Float('1.23')) == mpmath.mpf('1.23')
>>>>> AssertionError
>>>>>
>>>>> ________________________________________________________________________________
>>>>> _________________ sympy/core/tests/test_symbol.py:test_symbols
>>>>> _________________
>>>>>  File
>>>>> "/home/renato/down/sympy-0.7.0.rc2/sympy/core/tests/test_symbol.py",
>>>>> line 123, in test_symbols
>>>>>    raises(DeprecationWarning, "symbols('xyz', each_char=True)")
>>>>>  File "sympy/utilities/pytest.py", line 49, in raises
>>>>>    raise AssertionError("DID NOT RAISE")
>>>>> AssertionError: DID NOT RAISE
>>>>>
>>>>> ________________________________________________________________________________
>>>>> ______________ sympy/geometry/tests/test_geometry.py:test_polygon
>>>>> ______________
>>>>>  File
>>>>> "/home/renato/down/sympy-0.7.0.rc2/sympy/geometry/tests/test_geometry.py",
>>>>> line 593, in test_polygon
>>>>>    raises(UserWarning, "p1.distance(p2)")
>>>>>  File "sympy/utilities/pytest.py", line 49, in raises
>>>>>    raise AssertionError("DID NOT RAISE")
>>>>> AssertionError: DID NOT RAISE
>>>>>
>>>>> --
>>>>> You received this message because you are subscribed to the Google
>>>>> Groups "sympy" group.
>>>>> To post to this group, send email to [email protected].
>>>>> To unsubscribe from this group, send email to
>>>>> [email protected].
>>>>> For more options, visit this group at
>>>>> http://groups.google.com/group/sympy?hl=en.
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>
> --
> You received this message because you are subscribed to the Google Groups
> "sympy" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to
> [email protected].
> For more options, visit this group at
> http://groups.google.com/group/sympy?hl=en.
>
>
Mateusz

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/sympy?hl=en.

Reply via email to