The issue was that the propagation of referenced anonymous attributes treated
geometry outputs of the foreach zone as "normal". That means that every
anonymous attributes referenced by the input socket would also be referenced by
the output socket.
However, just like in the repeat zone, this so called "propagate relation" needs
some special behavior, because anonymous attributes references created inside a
zone have to remain inside that zone. Instead, the output node creates a new
anonymous attribute reference that is used outside of the zone.
Note, this is the same as #132560 but also implements the right-to-left pass,
whereas before only the left-to-right pass was implemented.
Pull Request: https://projects.blender.org/blender/blender/pulls/132607
The issue was that the propagation of referenced anonymous attributes treated
geometry outputs of the foreach zone as "normal". That means that every
anonymous attributes referenced by the input socket would also be referenced by
the output socket.
However, just like in the repeat zone, this so called "propagate relation" needs
some special behavior, because anonymous attributes references created inside a
zone have to remain inside that zone. Instead, the output node creates a new
anonymous attribute reference that is used outside of the zone.
Pull Request: https://projects.blender.org/blender/blender/pulls/132560
Instead of storing an optional in the runtime data of each node,
move the field status for the socket to an array in the node tree's
runtime data. This wasn't possible before because selecting nodes
reordered nodes which caused too much cache recomputation.
The benefit is having less geometry-nodes specific data in the
node socket runtime struct, and a slight move to a more data-
oriented storage format.
Pull Request: https://projects.blender.org/blender/blender/pulls/131822
This refactors the lifetime analysis of anonymous attributes in geometry nodes.
The refactor has a couple of goals:
* Use a better and simpler abstraction that can be used when building the
lazy-function graph. We currently have a bunch of duplicate code to handle
"field source" and "caller propagation" attributes. This is now unified so
that one only has to worry about one kind of "reference sets".
* Make the abstraction compatible with handling bundles and closures in case we
want to support them in the future. Both types can contain geometries and
fields so they need to be taken into account when determining lifetimes.
* Make more parts independent of the concept of "anonymous attributes". In
theory, there could be more kinds of referenced data whose lifetimes need to
be managed. I don't have any concrete plans for adding any though.
At its core, deterministic anonymous attributes still work the same they have
been since they became deterministic [0]. Even the generated lazy-function graph
is still pretty much or even exactly the same as before.
The patch renames `AnonymousAttributeSet` to the more general
`GeometryNodesReferenceSet` which is more. This also makes more places
independent of the concept of anonymous attributes. Functionally, this still the
same though. It's only used in the internals of geometry nodes nowadays. Most
code just gets an `AttributeFilter` that is based on it.
[0]: https://archive.blender.org/developer/D16858
Pull Request: https://projects.blender.org/blender/blender/pulls/128667
