On Thu, Oct 16, 2008 at 5:00 AM, Dag Sverre Seljebotn
<[EMAIL PROTECTED]> wrote:
>
> Huh? Your patch doesn't solve this problem either? (I just
> tried...fixnoneleaks.diff does nothing about the
>
> cdef object foo1
>
> not being declared issue).

Ups! I wrote that at home last night, I believe it worked. But perhaps
I got confused.

> The two should be orthogonal issues. I really believe that my patch is how
> the first problem should be solved. Attaching now how I believe the second
> problem should be solved, for which there seems to be no patch yet anyway.

OK, I'll give a look right now.

> If you agree I can simply make a push of my patches BTW, but there's been
> holes in my patches before so I'm waiting. If you find a way you think is
> better then feel free.

You are far more experienced with Cython internals than me. So if you
said you way is the way, then I trust you. Just let me play a bit with
the corner cases.

>> BTW, perhaps if we could support the --clenaup option as a compiler
>> directive, then I would be able to write a rether good testcase.
>>
>> Do this request make sense, I mean, adding a 'cleanup' directive
>> enabling the highest level of cleanup? I ALWAY build my code with
>> cleanup enabled, so I would love to be hable to use in my preambles::
>> #cython: cleanup=True
>
> +1
>
> But one should block its use as a decorator and with statement. Perhaps a
> more descriptive name -- module_cleanup or similar?

Well, I would ask for 'modulecleanup' for consistency with the naming
of the other directives.  But I do not know yet how to disable
decorator and with statement. BTW, should this directive be an
integer, supporting various levels of cleanups, like the former
--cleanup option? Or it should mean a full cleanup?



-- 
Lisandro Dalcín
---------------
Centro Internacional de Métodos Computacionales en Ingeniería (CIMEC)
Instituto de Desarrollo Tecnológico para la Industria Química (INTEC)
Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
PTLC - Güemes 3450, (3000) Santa Fe, Argentina
Tel/Fax: +54-(0)342-451.1594
_______________________________________________
Cython-dev mailing list
[email protected]
http://codespeak.net/mailman/listinfo/cython-dev

Reply via email to