[Biopython-dev] Compiling Biopython on 64 bit Windows, was: Volunteer buildslave machines?

[Manlio Calvi]

Works as expected because I have Numpy modules running in other
scripts (dumps the path to __init__.py)

In C:\Python34\Lib\site-packages\Bio the non-py files are two:
cpairwise2.pyd
trie.pyd

The same are in the folder I used to build them, and by the extension
should be callable/importable.
I tried to import cpairwise2 but (don't) work the same way, same error
raised as before.


On Mon, Apr 21, 2014 at 6:12 PM, Peter Cock <p.j.a.cock at googlemail.com> wrote:
> On Mon, Apr 21, 2014 at 4:45 PM, Manlio Calvi <manlio.calvi at gmail.com> wrote:
>>> test_Cluster ... skipping. If you want to use Bio.Cluster, install NumPy first and then reinstall Biopython
>>
>> And that's strange, Numpy IS installed in my case and usable by other
>> scripts... I tried to reinstall and rebuild, all seems the same as
>> before.
>
> The error message is just triggered if "from Bio import Cluster"
> fails with an ImportError - much like all the other C code modules'
> imports are failing on your Python 3.4 setup.
>
>> Apart from Numpy, seems the .pyd file is there but isn't importable by
>> python as module as it raise an "ImportError" kind of error.
>>
>> This is the dump:
>> ####################
>> Traceback (most recent call last):
>>   File "<string>", line 1, in <module>
>> ImportError: cannot import name 'trie'
>> ####################
>>
>> Now, seems strange to me that File<string> thing
>>
>> Manlio
>
> What does this do?
>
> python3.4 -c "import numpy; print(numpy.__file__)"
>
> If it works, can you see any compiled modules in that
> folder, and how they are named?
>
> Peter