[Biopython-dev] Versions of numpy/Numeric

[Bruce Southey]

Peter wrote:
>> Hi,
>> I just built and installed Python 2.6 with gcc version 4.3.2. I then
>> installed numpy 1.2 with it (so no Numeric).
>>     
>
> I thought you might be the first to try Biopython with python 2.6 was
> I knew it was out.
>
>   
>> I did a cvs update on biopython and installed with Python 2.5.2 and Python
>> 2.6.  In both cases I noticed many gcc warnings 'differ in signedness'
>>  (should I file a bug report?) in Bio/cstringfnsmodule.c and Bio/trie.c and
>> also Bio/triemodule.c has a couple of other warnings.
>>     
>
> I've noticed differ in signedness warnings from trie before with an
> older gcc - we propably should fix these so please file a (low
> priority) bug for that.
>   
Filed bug reports for these and I think that the Bio/cstringfnsmodule.c 
and Bio/trie.c are related to the declaration of 'unsigned char' . I 
change this to 'char' in   cstringfnsmodule.c and the warning goes away. 
However, that probably is not be the best thing to do without checking 
that reason for using 'unsigned char' in the code (may be essential to 
maintain sign).

There are some interesting comments on the usage of strlen() warnings 
such Linus Torvalds:
"..and my argument is that a warning which doesn't allow you to call
"strlen()" on a "unsigned char" array without triggering is a bogus
warning, and must be removed."

So perhaps these can be ignored.

Regards
Bruce