[Python-Dev] RE: [Python-checkins] CVS: python/dist/src/Modules stropmodule.c,2.76,2.77

[Tim Peters]

[Fredrik Lundh]
> any reason why the
>
> #ifdef MALLOC_ZERO_RETURNS_NULL
>
> macro (in pyport.h) isn't set / doesn't take care of this?

The code uses PyMem_MALLOC, which after a chain of umpteen #defines ends up
being plain malloc.  As Michael noted in the bug report, it could have used
PyMem_Malloc() instead and avoided the problem.  But I chose not to do that,
since special-casing a result of 0 was more efficient for reasons other than
malloc.  However:

> (and is it just me, or does the strop.replace function allocate
> a buffer, copy the result to that buffer, only to copy it into a
> string and throw the buffer away?

Yes.  And I'm returning something now that musn't be free()'ed when the
result length is 0.  Will fix.

> no wonder u"".replace() is 30% faster than "".replace() ;-)

For a given number of characters or bytes <wink>?