[Bf-committers] Klocwork source analysis

[Joe Eagar]

Stephen Swaney wrote:
> On Fri, Dec 01, 2006 at 09:01:21PM -0700, Joe Eagar wrote:
>
>   
>> Hi.  I looked at the ID property stuff in the report, and the report 
>> seems completely useless on it.  It makes no sense, the line numbers 
>> don't match, and frankly I have no clue what's going on there :S
>>     
>
> The report was done before you made some changes so the line 
> numbers don't match.
>
> In BPy_IDProperty_getattr(), you are not checking the pointer
> returned from PyObject_New() before using it.  It can be NULL.
>   
Well I don't see the point of doing a NULL check there; it seems as 
pointless as checking for NULL after alloc functions.

As for the Py_XINCREF, the ref count is being incremented because the 
wrapped object is stored in ->data_wrap.  Seems fine to me, though I'm 
not an expert of refcounting. :S

Joe