[Bf-committers] [patch] RNA functions multiple returns, new features and cleanup
ideasman42 at gmail.com
Sat Jan 23 22:12:09 CET 2010
Hi, thanks for working on this.
Went over the patch, looks good to me. (+1)
Python may need to own memory at some point, have a patch on my system
for this but would like to avoid this if possible so glad you managed
On Sat, Jan 23, 2010 at 6:24 PM, Elia Sarti <vekoon at gmail.com> wrote:
> this is actually more specifically directed to Brecht and Campbell but
> of course others are free to check as well, especially Joshua, because I
> added in the Action.get_frame_range() function which was commented out
> because there was no support yet for returning arrays (which this patch
> enables). I tested the function of course and seems to work but if you
> want to edit/remove it, feel free.
> Anyway, I've made a patch because this ended up touching and modifying
> slightly more parts that I initially expected. Although I've tested both
> UI and rigify and everything seems to work properly. Also, I'm using the
> mailing list after unsuccessfully trying to get in touch in IRC.
> The patch: http://vekoon.googlecode.com/files/rna_func_dynamic_fix.patch
> This code fixes and completes support for returning multiple values.
> This includes adding support for returning arrays, both fixed and
> dynamically sized. The way this is achieved is by storing an additional
> int value next to the dynamic parameter in the ParameterList stack which
> gets passed to the C function as an additional parameter. In the case of
> return parameters it is duty of the C function to set this int to the
> correct length value for the dynamic parameter (which makes sense).
> On the BPY side I was worried about the additional allocated memory for
> dynamically sized returns but actually it turned out BPY duplicates the
> result into a Py object and so original memory gets freed properly. I
> don't think this is a big issue as of yet (memory duplication), so can
> be left as is.
> Also, I renamed PROP_RETURN as PROP_OUTPUT, which represents better the
> reality now that there are multiple returns.
> If the patch looks fine I might as well just commit it.
> Bf-committers mailing list
> Bf-committers at blender.org
More information about the Bf-committers