[PEAK] access violation with py24/msvc compiler

alexander smishlajev alex at ank-sia.com
Thu Jan 27 14:20:14 EST 2005


Phillip J. Eby wrote, at 27.01.2005 20:43:
> 
>>> it is kjGraph.  the following command also raises access violation:
>>>    python -c "import kjbuckets; g=kjbuckets.kjGraph(); print g"
>>
>> well, i have found that the method function called here is 
>> WrapperPrint, and that the failing statement is in line 2002: 
>> 'fprintf(fp, "kjGraph(");'.
> 
> Interesting indeed.  It's probably a runtime library compatibility 
> error.  I suppose we should really get rid of kjbuckets; this isn't the 
> first time it's shown its age with a weird error.  It was originally 
> written for Python 1.3, so it's possible it's not properly up-to-date 
> with the Python C API.

i may be misunderstanding something, but i have checked the signature 
(printfunc) of tp_print member of PyTypeObject, and it seems to be 
correct except that kjbuckets declare return type as long while it is 
int in python headers.  MS says that msvc toolkit 2003 includes the same 
compiler and linker that ship with visual studio .net 2003.  as far as i 
know, these are the same tools (see for example 
http://blogs.msdn.com/arich/archive/2004/07/09/178503.aspx) that was 
used to build python distribution (python says it was built with visual 
studio 7.1).  so i am out of ideas.

best wishes,
alex.




More information about the PEAK mailing list