[Bf-committers] CVS commit: blender/intern/python/freeze freeze_ac.py

Ton Roosendaal bf-committers@blender.org
Wed, 1 Jan 2003 12:47:23 +0100


Hi,

Yep, Strubi wrote it under big deadline pressure, and still attempted  
to make it an ambitious improvement for Blender.

Unless it works smoothly, and compiles out of the box for all platforms  
now, I would propose to get rid of it, even if that means no vrml2  
input. I prefer to wait for a motivated and experienced (group of)  
Python coder(s) to pick it up again.

Anyone feels comfortable with what and how to #ifdef or comment it out?

-Ton-


On Wednesday, Jan 1, 2003, at 04:33 Europe/Amsterdam, Carsten Wartmann  
wrote:

> LarstiQ writes:
>> On Wed, Jan 01, 2003 at 03:07:47AM +0100, Unprivileged User wrote:
>>> nobody      2003/01/01 03:07:47 CET
>>>
>>>   Modified files:
>>>     blender/intern/python/freeze freeze_ac.py
>>>
>>>   Log:
>>>   A kludge to use the right compiler while freezing
>>>   using gcc under Irix -- without this it uses only
>>>   'cc -n32' (since this is what the sgi employee
>>>   who compiled the python stuff used).
>>
>> Which brings up the issue of how the freeze currently works, I don't
>> know why exactly it tries to pull build information from the python
>> installation, we can hack around this, or provide a more structural
>> solution that actually looks at build choices the user made, but is
>> there anyone that knows how the freeze works and why? This isn't the
>> only goofy thing going on.
>
> I talked with Strubi 2 days ago and he said something about people
> beeing angry about this hack etc. ;-) So if I am right it is only for
> the VRML2 import hack and should probably be just an (default off)
> compiletime option?
>
> Carsten.
> _______________________________________________
> Bf-committers mailing list
> Bf-committers@blender.org
> http://www.blender.org/mailman/listinfo/bf-committers
>
>
------------------------------------------------------------------------ 
--
Ton Roosendaal  Blender Foundation ton@blender.org