CGAL users discussion list

[marcosscriven <>]

>BTW, what are you trying to do, a model running in the browser?

Precisely - I guess you've seen my myriad other posts on here :)

I have it working in a single worker thread, but I'd like to delegate CSG
operations on Nef_polyhedron_3 objects to further worker threads. Trouble is
of course worker threads don't share memory - I can  only pass a message.

So that message is a list of polys, plus an operation, and the message back
is a single resulting poly of that operation.

I asked the question here:

https://groups.google.com/forum/?fromgroups=#!topic/emscripten-discuss/JeLvBMn9X60

And the lead guy there suggested the possilbilty of a 'bump' allocator.
After further reading I understand this to mean, for instance, setting a
custom frame allocator, creating the instances of the polys one requires
with that allocator, and then saying bytes[] = allocatorInstance.getBytes[].

Then on the other side somehow being able to offset the pointers internal to
the Nef_polyhedron_3 object?

I *sounds* like this might work, but I'm just not sure on the details at the
moment.

Mracos



--
View this message in context: 
http://cgal-discuss.949826.n4.nabble.com/What-s-the-fastest-way-to-serialise-Nef-3-poly-tp4657024p4657038.html
Sent from the cgal-discuss mailing list archive at Nabble.com.