Files
test2/source/blender/blenkernel/intern/geometry_set.cc

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

751 lines
23 KiB
C++
Raw Normal View History

/* SPDX-License-Identifier: GPL-2.0-or-later */
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
#include "BLI_bounds.hh"
Geometry Nodes: Make instances real on-demand This commit makes the geometry output of the collection info usable. The output is the geometry of a collection instance, but this commit adds a utility to convert the instances to real geometry, used in the background whenever it is needed, like copy on write. The recursive nature of the "realize instances" code is essential, because collection instances in the `InstancesComponent`, might have no geometry sets of their own containing even more collection instances, which might then contain object instances, etc. Another consideration is that currently, every single instance contains a reference to its data. This is inefficient since most of the time there are many locations and only a few sets of unique data. So this commit adds a `GeometryInstanceGroup` to support this future optimization. The API for instances returns a vector of `GeometryInstanceGroup`. This may be less efficient when there are many instances, but it makes more complicated operations like point distribution that need to iterate over input geometry multiple times much simpler. Any code that needs to change data, like most of the attribute nodes, can simply call `geometry_set_realize_instances(geometry_set)`, which will move any geometry in the `InstancesComponent` to new "real" geometry components. Many nodes can support read-only access to instances in order to avoid making them real, this will be addressed where needed in the near future. Instances from the existing "dupli" system are not supported yet. Differential Revision: https://developer.blender.org/D10327
2021-02-12 11:58:15 -06:00
#include "BLI_map.hh"
#include "BLI_task.hh"
Geometry Nodes: Make instances real on-demand This commit makes the geometry output of the collection info usable. The output is the geometry of a collection instance, but this commit adds a utility to convert the instances to real geometry, used in the background whenever it is needed, like copy on write. The recursive nature of the "realize instances" code is essential, because collection instances in the `InstancesComponent`, might have no geometry sets of their own containing even more collection instances, which might then contain object instances, etc. Another consideration is that currently, every single instance contains a reference to its data. This is inefficient since most of the time there are many locations and only a few sets of unique data. So this commit adds a `GeometryInstanceGroup` to support this future optimization. The API for instances returns a vector of `GeometryInstanceGroup`. This may be less efficient when there are many instances, but it makes more complicated operations like point distribution that need to iterate over input geometry multiple times much simpler. Any code that needs to change data, like most of the attribute nodes, can simply call `geometry_set_realize_instances(geometry_set)`, which will move any geometry in the `InstancesComponent` to new "real" geometry components. Many nodes can support read-only access to instances in order to avoid making them real, this will be addressed where needed in the near future. Instances from the existing "dupli" system are not supported yet. Differential Revision: https://developer.blender.org/D10327
2021-02-12 11:58:15 -06:00
#include "BLT_translation.h"
Geometry Nodes: Make instances real on-demand This commit makes the geometry output of the collection info usable. The output is the geometry of a collection instance, but this commit adds a utility to convert the instances to real geometry, used in the background whenever it is needed, like copy on write. The recursive nature of the "realize instances" code is essential, because collection instances in the `InstancesComponent`, might have no geometry sets of their own containing even more collection instances, which might then contain object instances, etc. Another consideration is that currently, every single instance contains a reference to its data. This is inefficient since most of the time there are many locations and only a few sets of unique data. So this commit adds a `GeometryInstanceGroup` to support this future optimization. The API for instances returns a vector of `GeometryInstanceGroup`. This may be less efficient when there are many instances, but it makes more complicated operations like point distribution that need to iterate over input geometry multiple times much simpler. Any code that needs to change data, like most of the attribute nodes, can simply call `geometry_set_realize_instances(geometry_set)`, which will move any geometry in the `InstancesComponent` to new "real" geometry components. Many nodes can support read-only access to instances in order to avoid making them real, this will be addressed where needed in the near future. Instances from the existing "dupli" system are not supported yet. Differential Revision: https://developer.blender.org/D10327
2021-02-12 11:58:15 -06:00
#include "BKE_attribute.h"
#include "BKE_curves.hh"
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
#include "BKE_geometry_set.hh"
#include "BKE_instances.hh"
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
#include "BKE_lib_id.h"
#include "BKE_mesh.hh"
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
#include "BKE_mesh_wrapper.h"
Geometry Nodes: Make instances real on-demand This commit makes the geometry output of the collection info usable. The output is the geometry of a collection instance, but this commit adds a utility to convert the instances to real geometry, used in the background whenever it is needed, like copy on write. The recursive nature of the "realize instances" code is essential, because collection instances in the `InstancesComponent`, might have no geometry sets of their own containing even more collection instances, which might then contain object instances, etc. Another consideration is that currently, every single instance contains a reference to its data. This is inefficient since most of the time there are many locations and only a few sets of unique data. So this commit adds a `GeometryInstanceGroup` to support this future optimization. The API for instances returns a vector of `GeometryInstanceGroup`. This may be less efficient when there are many instances, but it makes more complicated operations like point distribution that need to iterate over input geometry multiple times much simpler. Any code that needs to change data, like most of the attribute nodes, can simply call `geometry_set_realize_instances(geometry_set)`, which will move any geometry in the `InstancesComponent` to new "real" geometry components. Many nodes can support read-only access to instances in order to avoid making them real, this will be addressed where needed in the near future. Instances from the existing "dupli" system are not supported yet. Differential Revision: https://developer.blender.org/D10327
2021-02-12 11:58:15 -06:00
#include "BKE_modifier.h"
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
#include "BKE_pointcloud.h"
#include "BKE_volume.h"
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
Geometry Nodes: Make instances real on-demand This commit makes the geometry output of the collection info usable. The output is the geometry of a collection instance, but this commit adds a utility to convert the instances to real geometry, used in the background whenever it is needed, like copy on write. The recursive nature of the "realize instances" code is essential, because collection instances in the `InstancesComponent`, might have no geometry sets of their own containing even more collection instances, which might then contain object instances, etc. Another consideration is that currently, every single instance contains a reference to its data. This is inefficient since most of the time there are many locations and only a few sets of unique data. So this commit adds a `GeometryInstanceGroup` to support this future optimization. The API for instances returns a vector of `GeometryInstanceGroup`. This may be less efficient when there are many instances, but it makes more complicated operations like point distribution that need to iterate over input geometry multiple times much simpler. Any code that needs to change data, like most of the attribute nodes, can simply call `geometry_set_realize_instances(geometry_set)`, which will move any geometry in the `InstancesComponent` to new "real" geometry components. Many nodes can support read-only access to instances in order to avoid making them real, this will be addressed where needed in the near future. Instances from the existing "dupli" system are not supported yet. Differential Revision: https://developer.blender.org/D10327
2021-02-12 11:58:15 -06:00
#include "DNA_collection_types.h"
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
#include "DNA_object_types.h"
#include "DNA_pointcloud_types.h"
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
#include "BLI_rand.hh"
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
#include "MEM_guardedalloc.h"
using blender::float3;
using blender::float4x4;
Geometry Nodes: Make instances real on-demand This commit makes the geometry output of the collection info usable. The output is the geometry of a collection instance, but this commit adds a utility to convert the instances to real geometry, used in the background whenever it is needed, like copy on write. The recursive nature of the "realize instances" code is essential, because collection instances in the `InstancesComponent`, might have no geometry sets of their own containing even more collection instances, which might then contain object instances, etc. Another consideration is that currently, every single instance contains a reference to its data. This is inefficient since most of the time there are many locations and only a few sets of unique data. So this commit adds a `GeometryInstanceGroup` to support this future optimization. The API for instances returns a vector of `GeometryInstanceGroup`. This may be less efficient when there are many instances, but it makes more complicated operations like point distribution that need to iterate over input geometry multiple times much simpler. Any code that needs to change data, like most of the attribute nodes, can simply call `geometry_set_realize_instances(geometry_set)`, which will move any geometry in the `InstancesComponent` to new "real" geometry components. Many nodes can support read-only access to instances in order to avoid making them real, this will be addressed where needed in the near future. Instances from the existing "dupli" system are not supported yet. Differential Revision: https://developer.blender.org/D10327
2021-02-12 11:58:15 -06:00
using blender::Map;
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
using blender::MutableSpan;
using blender::Span;
using blender::StringRef;
using blender::Vector;
using blender::bke::InstanceReference;
using blender::bke::Instances;
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
/* -------------------------------------------------------------------- */
/** \name Geometry Component
* \{ */
GeometryComponent::GeometryComponent(GeometryComponentType type) : type_(type)
{
}
GeometryComponent *GeometryComponent::create(GeometryComponentType component_type)
{
switch (component_type) {
case GEO_COMPONENT_TYPE_MESH:
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
return new MeshComponent();
case GEO_COMPONENT_TYPE_POINT_CLOUD:
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
return new PointCloudComponent();
case GEO_COMPONENT_TYPE_INSTANCES:
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
return new InstancesComponent();
case GEO_COMPONENT_TYPE_VOLUME:
return new VolumeComponent();
case GEO_COMPONENT_TYPE_CURVE:
return new CurveComponent();
case GEO_COMPONENT_TYPE_EDIT:
return new GeometryComponentEditData();
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
BLI_assert_unreachable();
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
return nullptr;
}
Geometry Nodes: new geometry attribute API Currently, there are two attribute API. The first, defined in `BKE_attribute.h` is accessible from RNA and C code. The second is implemented with `GeometryComponent` and is only accessible in C++ code. The second is widely used, but only being accessible through the `GeometrySet` API makes it awkward to use, and even impossible for types that don't correspond directly to a geometry component like `CurvesGeometry`. This patch adds a new attribute API, designed to replace the `GeometryComponent` attribute API now, and to eventually replace or be the basis of the other one. The basic idea is that there is an `AttributeAccessor` class that allows code to interact with a set of attributes owned by some geometry. The accessor itself has no ownership. `AttributeAccessor` is a simple type that can be passed around by value. That makes it easy to return it from functions and to store it in containers. For const-correctness, there is also a `MutableAttributeAccessor` that allows changing individual and can add or remove attributes. Currently, `AttributeAccessor` is composed of two pointers. The first is a pointer to the owner of the attribute data. The second is a pointer to a struct with function pointers, that is similar to a virtual function table. The functions know how to access attributes on the owner. The actual attribute access for geometries is still implemented with the `AttributeProvider` pattern, which makes it easy to support different sources of attributes on a geometry and simplifies dealing with built-in attributes. There are different ways to get an attribute accessor for a geometry: * `GeometryComponent.attributes()` * `CurvesGeometry.attributes()` * `bke::mesh_attributes(const Mesh &)` * `bke::pointcloud_attributes(const PointCloud &)` All of these also have a `_for_write` variant that returns a `MutabelAttributeAccessor`. Differential Revision: https://developer.blender.org/D15280
2022-07-08 16:16:56 +02:00
int GeometryComponent::attribute_domain_size(const eAttrDomain domain) const
{
if (this->is_empty()) {
return 0;
}
const std::optional<blender::bke::AttributeAccessor> attributes = this->attributes();
if (attributes.has_value()) {
return attributes->domain_size(domain);
}
return 0;
}
std::optional<blender::bke::AttributeAccessor> GeometryComponent::attributes() const
{
return std::nullopt;
};
std::optional<blender::bke::MutableAttributeAccessor> GeometryComponent::attributes_for_write()
{
return std::nullopt;
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
void GeometryComponent::user_add() const
{
users_.fetch_add(1);
}
void GeometryComponent::user_remove() const
{
const int new_users = users_.fetch_sub(1) - 1;
if (new_users == 0) {
delete this;
}
}
bool GeometryComponent::is_mutable() const
{
/* If the item is shared, it is read-only. */
/* The user count can be 0, when this is called from the destructor. */
return users_ <= 1;
}
GeometryComponentType GeometryComponent::type() const
{
return type_;
}
bool GeometryComponent::is_empty() const
{
return false;
}
/** \} */
/* -------------------------------------------------------------------- */
/** \name Geometry Set
* \{ */
GeometrySet::GeometrySet() = default;
GeometrySet::GeometrySet(const GeometrySet &other) = default;
GeometrySet::GeometrySet(GeometrySet &&other) = default;
GeometrySet::~GeometrySet() = default;
GeometrySet &GeometrySet::operator=(const GeometrySet &other) = default;
GeometrySet &GeometrySet::operator=(GeometrySet &&other) = default;
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
GeometryComponent &GeometrySet::get_component_for_write(GeometryComponentType component_type)
{
GeometryComponentPtr &component_ptr = components_[component_type];
if (!component_ptr) {
/* If the component did not exist before, create a new one. */
component_ptr = GeometryComponent::create(component_type);
return *component_ptr;
}
if (component_ptr->is_mutable()) {
/* If the referenced component is already mutable, return it directly. */
return *component_ptr;
}
/* If the referenced component is shared, make a copy. The copy is not shared and is
* therefore mutable. */
component_ptr = component_ptr->copy();
return *component_ptr;
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
GeometryComponent *GeometrySet::get_component_ptr(GeometryComponentType type)
{
if (this->has(type)) {
return &this->get_component_for_write(type);
}
return nullptr;
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
const GeometryComponent *GeometrySet::get_component_for_read(
GeometryComponentType component_type) const
{
return components_[component_type].get();
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
bool GeometrySet::has(const GeometryComponentType component_type) const
{
const GeometryComponentPtr &component = components_[component_type];
return component.has_value() && !component->is_empty();
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
void GeometrySet::remove(const GeometryComponentType component_type)
{
components_[component_type].reset();
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
void GeometrySet::keep_only(const blender::Span<GeometryComponentType> component_types)
{
for (GeometryComponentPtr &component_ptr : components_) {
if (component_ptr) {
if (!component_types.contains(component_ptr->type())) {
component_ptr.reset();
}
}
}
}
void GeometrySet::keep_only_during_modify(
const blender::Span<GeometryComponentType> component_types)
{
Vector<GeometryComponentType> extended_types = component_types;
extended_types.append_non_duplicates(GEO_COMPONENT_TYPE_INSTANCES);
extended_types.append_non_duplicates(GEO_COMPONENT_TYPE_EDIT);
this->keep_only(extended_types);
}
void GeometrySet::remove_geometry_during_modify()
{
this->keep_only_during_modify({});
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
void GeometrySet::add(const GeometryComponent &component)
{
BLI_assert(!components_[component.type()]);
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
component.user_add();
components_[component.type()] = const_cast<GeometryComponent *>(&component);
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
Vector<const GeometryComponent *> GeometrySet::get_components_for_read() const
{
Vector<const GeometryComponent *> components;
for (const GeometryComponentPtr &component_ptr : components_) {
if (component_ptr) {
components.append(component_ptr.get());
}
}
return components;
}
bool GeometrySet::compute_boundbox_without_instances(float3 *r_min, float3 *r_max) const
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
{
using namespace blender;
bool have_minmax = false;
if (const PointCloud *pointcloud = this->get_pointcloud_for_read()) {
Geometry: Cache bounds min and max, share between data-blocks Bounding box calculation can be a large in some situations, especially instancing. This patch caches the min and max of the bounding box in runtime data of meshes, point clouds, and curves, implementing part of T96968. Bounds are now calculated lazily-- only after they are tagged dirty. Also, cached bounds are also shared when copying geometry data-blocks that have equivalent data. When bounds are calculated on an evaluated data-block, they are also accessible on the original, and the next evaluated ID will also share them. A geometry will stop sharing bounds as soon as its positions (or radii) are changed. Just caching the bounds gave a 2-3x speedup with thousands of mesh geometry instances in the viewport. Sharing the bounds can eliminate recalculations entirely in cases like copying meshes in geometry nodes or the selection paint brush in curves sculpt mode, which causes a reevaluation but doesn't change the positions. **Implementation** The sharing is achieved with a `shared_ptr` that points to a cache mutex (from D16419) and the cached bounds data. When geometries are copied, the bounds are shared by default, and only "un-shared" when the bounds are tagged dirty. Point clouds have a new runtime struct to store this data. Functions for tagging the data dirty are improved for added for point clouds and improved for curves. A missing tag has also been fixed for mesh sculpt mode. **Future** There are further improvements which can be worked on next - Apply changes to volume objects and other types where it makes sense - Continue cleanup changes described in T96968 - Apply shared cache design to more expensive data like triangulation or normals Differential Revision: https://developer.blender.org/D16204
2022-11-15 13:46:55 -06:00
have_minmax |= pointcloud->bounds_min_max(*r_min, *r_max);
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
if (const Mesh *mesh = this->get_mesh_for_read()) {
have_minmax |= BKE_mesh_wrapper_minmax(mesh, *r_min, *r_max);
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
if (const Volume *volume = this->get_volume_for_read()) {
have_minmax |= BKE_volume_min_max(volume, *r_min, *r_max);
}
if (const Curves *curves_id = this->get_curves_for_read()) {
const bke::CurvesGeometry &curves = curves_id->geometry.wrap();
Geometry: Cache bounds min and max, share between data-blocks Bounding box calculation can be a large in some situations, especially instancing. This patch caches the min and max of the bounding box in runtime data of meshes, point clouds, and curves, implementing part of T96968. Bounds are now calculated lazily-- only after they are tagged dirty. Also, cached bounds are also shared when copying geometry data-blocks that have equivalent data. When bounds are calculated on an evaluated data-block, they are also accessible on the original, and the next evaluated ID will also share them. A geometry will stop sharing bounds as soon as its positions (or radii) are changed. Just caching the bounds gave a 2-3x speedup with thousands of mesh geometry instances in the viewport. Sharing the bounds can eliminate recalculations entirely in cases like copying meshes in geometry nodes or the selection paint brush in curves sculpt mode, which causes a reevaluation but doesn't change the positions. **Implementation** The sharing is achieved with a `shared_ptr` that points to a cache mutex (from D16419) and the cached bounds data. When geometries are copied, the bounds are shared by default, and only "un-shared" when the bounds are tagged dirty. Point clouds have a new runtime struct to store this data. Functions for tagging the data dirty are improved for added for point clouds and improved for curves. A missing tag has also been fixed for mesh sculpt mode. **Future** There are further improvements which can be worked on next - Apply changes to volume objects and other types where it makes sense - Continue cleanup changes described in T96968 - Apply shared cache design to more expensive data like triangulation or normals Differential Revision: https://developer.blender.org/D16204
2022-11-15 13:46:55 -06:00
have_minmax |= curves.bounds_min_max(*r_min, *r_max);
}
return have_minmax;
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
std::ostream &operator<<(std::ostream &stream, const GeometrySet &geometry_set)
{
Vector<std::string> parts;
if (const Mesh *mesh = geometry_set.get_mesh_for_read()) {
parts.append(std::to_string(mesh->totvert) + " verts");
parts.append(std::to_string(mesh->totedge) + " edges");
parts.append(std::to_string(mesh->totpoly) + " polys");
parts.append(std::to_string(mesh->totloop) + " corners");
}
if (const Curves *curves = geometry_set.get_curves_for_read()) {
parts.append(std::to_string(curves->geometry.point_num) + " control points");
parts.append(std::to_string(curves->geometry.curve_num) + " curves");
}
if (const PointCloud *point_cloud = geometry_set.get_pointcloud_for_read()) {
parts.append(std::to_string(point_cloud->totpoint) + " points");
}
if (const Volume *volume = geometry_set.get_volume_for_read()) {
parts.append(std::to_string(BKE_volume_num_grids(volume)) + " volume grids");
}
if (geometry_set.has_instances()) {
parts.append(std::to_string(geometry_set.get_instances_for_read()->instances_num()) +
" instances");
}
if (geometry_set.get_curve_edit_hints_for_read()) {
parts.append("curve edit hints");
}
stream << "<GeometrySet: ";
for (const int i : parts.index_range()) {
stream << parts[i];
if (i < parts.size() - 1) {
stream << ", ";
}
}
stream << ">";
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
return stream;
}
void GeometrySet::clear()
{
for (GeometryComponentPtr &component_ptr : components_) {
component_ptr.reset();
}
}
void GeometrySet::ensure_owns_direct_data()
{
for (GeometryComponentPtr &component_ptr : components_) {
if (!component_ptr) {
continue;
}
if (component_ptr->owns_direct_data()) {
continue;
}
GeometryComponent &component_for_write = this->get_component_for_write(component_ptr->type());
component_for_write.ensure_owns_direct_data();
}
}
Geometry Nodes: support for geometry instancing Previously, the Point Instance node in geometry nodes could only instance existing objects or collections. The reason was that large parts of Blender worked under the assumption that objects are the main unit of instancing. Now we also want to instance geometry within an object, so a slightly larger refactor was necessary. This should not affect files that do not use the new kind of instances. The main change is a redefinition of what "instanced data" is. Now, an instances is a cow-object + object-data (the geometry). This can be nicely seen in `struct DupliObject`. This allows the same object to generate multiple geometries of different types which can be instanced individually. A nice side effect of this refactor is that having multiple geometry components is not a special case in the depsgraph object iterator anymore, because those components are integrated with the `DupliObject` system. Unfortunately, different systems that work with instances in Blender (e.g. render engines and exporters) often work under the assumption that objects are the main unit of instancing. So those have to be updated as well to be able to handle the new instances. This patch updates Cycles, EEVEE and other viewport engines. Exporters have not been updated yet. Some minimal (not master-ready) changes to update the obj and alembic exporters can be found in P2336 and P2335. Different file formats may want to handle these new instances in different ways. For users, the only thing that changed is that the Point Instance node now has a geometry mode. This also fixes T88454. Differential Revision: https://developer.blender.org/D11841
2021-09-06 18:22:24 +02:00
bool GeometrySet::owns_direct_data() const
{
for (const GeometryComponentPtr &component_ptr : components_) {
if (component_ptr) {
if (!component_ptr->owns_direct_data()) {
return false;
}
Geometry Nodes: support for geometry instancing Previously, the Point Instance node in geometry nodes could only instance existing objects or collections. The reason was that large parts of Blender worked under the assumption that objects are the main unit of instancing. Now we also want to instance geometry within an object, so a slightly larger refactor was necessary. This should not affect files that do not use the new kind of instances. The main change is a redefinition of what "instanced data" is. Now, an instances is a cow-object + object-data (the geometry). This can be nicely seen in `struct DupliObject`. This allows the same object to generate multiple geometries of different types which can be instanced individually. A nice side effect of this refactor is that having multiple geometry components is not a special case in the depsgraph object iterator anymore, because those components are integrated with the `DupliObject` system. Unfortunately, different systems that work with instances in Blender (e.g. render engines and exporters) often work under the assumption that objects are the main unit of instancing. So those have to be updated as well to be able to handle the new instances. This patch updates Cycles, EEVEE and other viewport engines. Exporters have not been updated yet. Some minimal (not master-ready) changes to update the obj and alembic exporters can be found in P2336 and P2335. Different file formats may want to handle these new instances in different ways. For users, the only thing that changed is that the Point Instance node now has a geometry mode. This also fixes T88454. Differential Revision: https://developer.blender.org/D11841
2021-09-06 18:22:24 +02:00
}
}
return true;
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
const Mesh *GeometrySet::get_mesh_for_read() const
{
const MeshComponent *component = this->get_component_for_read<MeshComponent>();
return (component == nullptr) ? nullptr : component->get_for_read();
}
bool GeometrySet::has_mesh() const
{
const MeshComponent *component = this->get_component_for_read<MeshComponent>();
return component != nullptr && component->has_mesh();
}
const PointCloud *GeometrySet::get_pointcloud_for_read() const
{
const PointCloudComponent *component = this->get_component_for_read<PointCloudComponent>();
return (component == nullptr) ? nullptr : component->get_for_read();
}
const Volume *GeometrySet::get_volume_for_read() const
{
const VolumeComponent *component = this->get_component_for_read<VolumeComponent>();
return (component == nullptr) ? nullptr : component->get_for_read();
}
const Curves *GeometrySet::get_curves_for_read() const
{
const CurveComponent *component = this->get_component_for_read<CurveComponent>();
return (component == nullptr) ? nullptr : component->get_for_read();
}
const Instances *GeometrySet::get_instances_for_read() const
{
const InstancesComponent *component = this->get_component_for_read<InstancesComponent>();
return (component == nullptr) ? nullptr : component->get_for_read();
}
const blender::bke::CurvesEditHints *GeometrySet::get_curve_edit_hints_for_read() const
{
const GeometryComponentEditData *component =
this->get_component_for_read<GeometryComponentEditData>();
return (component == nullptr) ? nullptr : component->curves_edit_hints_.get();
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
bool GeometrySet::has_pointcloud() const
{
const PointCloudComponent *component = this->get_component_for_read<PointCloudComponent>();
return component != nullptr && component->has_pointcloud();
}
bool GeometrySet::has_instances() const
{
const InstancesComponent *component = this->get_component_for_read<InstancesComponent>();
return component != nullptr && component->get_for_read() != nullptr &&
component->get_for_read()->instances_num() >= 1;
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
bool GeometrySet::has_volume() const
{
const VolumeComponent *component = this->get_component_for_read<VolumeComponent>();
return component != nullptr && component->has_volume();
}
bool GeometrySet::has_curves() const
{
const CurveComponent *component = this->get_component_for_read<CurveComponent>();
return component != nullptr && component->has_curves();
}
bool GeometrySet::has_realized_data() const
{
for (const GeometryComponentPtr &component_ptr : components_) {
if (component_ptr) {
if (component_ptr->type() != GEO_COMPONENT_TYPE_INSTANCES) {
return true;
}
}
}
return false;
}
bool GeometrySet::is_empty() const
{
return !(this->has_mesh() || this->has_curves() || this->has_pointcloud() ||
this->has_volume() || this->has_instances());
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
GeometrySet GeometrySet::create_with_mesh(Mesh *mesh, GeometryOwnershipType ownership)
{
GeometrySet geometry_set;
if (mesh != nullptr) {
MeshComponent &component = geometry_set.get_component_for_write<MeshComponent>();
component.replace(mesh, ownership);
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
return geometry_set;
}
GeometrySet GeometrySet::create_with_volume(Volume *volume, GeometryOwnershipType ownership)
{
GeometrySet geometry_set;
if (volume != nullptr) {
VolumeComponent &component = geometry_set.get_component_for_write<VolumeComponent>();
component.replace(volume, ownership);
}
return geometry_set;
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
GeometrySet GeometrySet::create_with_pointcloud(PointCloud *pointcloud,
GeometryOwnershipType ownership)
{
GeometrySet geometry_set;
if (pointcloud != nullptr) {
PointCloudComponent &component = geometry_set.get_component_for_write<PointCloudComponent>();
component.replace(pointcloud, ownership);
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
return geometry_set;
}
GeometrySet GeometrySet::create_with_curves(Curves *curves, GeometryOwnershipType ownership)
{
GeometrySet geometry_set;
if (curves != nullptr) {
CurveComponent &component = geometry_set.get_component_for_write<CurveComponent>();
component.replace(curves, ownership);
}
return geometry_set;
}
GeometrySet GeometrySet::create_with_instances(Instances *instances,
GeometryOwnershipType ownership)
{
GeometrySet geometry_set;
geometry_set.replace_instances(instances, ownership);
return geometry_set;
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
void GeometrySet::replace_mesh(Mesh *mesh, GeometryOwnershipType ownership)
{
if (mesh == nullptr) {
this->remove<MeshComponent>();
return;
}
if (mesh == this->get_mesh_for_read()) {
return;
}
this->remove<MeshComponent>();
MeshComponent &component = this->get_component_for_write<MeshComponent>();
component.replace(mesh, ownership);
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
void GeometrySet::replace_curves(Curves *curves, GeometryOwnershipType ownership)
{
if (curves == nullptr) {
this->remove<CurveComponent>();
return;
}
if (curves == this->get_curves_for_read()) {
return;
}
this->remove<CurveComponent>();
CurveComponent &component = this->get_component_for_write<CurveComponent>();
component.replace(curves, ownership);
}
void GeometrySet::replace_instances(Instances *instances, GeometryOwnershipType ownership)
{
if (instances == nullptr) {
this->remove<InstancesComponent>();
return;
}
if (instances == this->get_instances_for_read()) {
return;
}
this->remove<InstancesComponent>();
InstancesComponent &component = this->get_component_for_write<InstancesComponent>();
component.replace(instances, ownership);
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
void GeometrySet::replace_pointcloud(PointCloud *pointcloud, GeometryOwnershipType ownership)
{
if (pointcloud == nullptr) {
this->remove<PointCloudComponent>();
return;
}
if (pointcloud == this->get_pointcloud_for_read()) {
return;
}
this->remove<PointCloudComponent>();
PointCloudComponent &component = this->get_component_for_write<PointCloudComponent>();
component.replace(pointcloud, ownership);
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
void GeometrySet::replace_volume(Volume *volume, GeometryOwnershipType ownership)
{
if (volume == nullptr) {
this->remove<VolumeComponent>();
return;
}
if (volume == this->get_volume_for_read()) {
return;
}
this->remove<VolumeComponent>();
VolumeComponent &component = this->get_component_for_write<VolumeComponent>();
component.replace(volume, ownership);
}
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
Mesh *GeometrySet::get_mesh_for_write()
{
MeshComponent *component = this->get_component_ptr<MeshComponent>();
return component == nullptr ? nullptr : component->get_for_write();
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
PointCloud *GeometrySet::get_pointcloud_for_write()
{
PointCloudComponent *component = this->get_component_ptr<PointCloudComponent>();
return component == nullptr ? nullptr : component->get_for_write();
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
Volume *GeometrySet::get_volume_for_write()
{
VolumeComponent *component = this->get_component_ptr<VolumeComponent>();
return component == nullptr ? nullptr : component->get_for_write();
}
Curves *GeometrySet::get_curves_for_write()
{
CurveComponent *component = this->get_component_ptr<CurveComponent>();
return component == nullptr ? nullptr : component->get_for_write();
}
Instances *GeometrySet::get_instances_for_write()
{
InstancesComponent *component = this->get_component_ptr<InstancesComponent>();
return component == nullptr ? nullptr : component->get_for_write();
}
blender::bke::CurvesEditHints *GeometrySet::get_curve_edit_hints_for_write()
{
if (!this->has<GeometryComponentEditData>()) {
return nullptr;
}
GeometryComponentEditData &component =
this->get_component_for_write<GeometryComponentEditData>();
return component.curves_edit_hints_.get();
}
void GeometrySet::attribute_foreach(const Span<GeometryComponentType> component_types,
const bool include_instances,
const AttributeForeachCallback callback) const
{
using namespace blender;
using namespace blender::bke;
for (const GeometryComponentType component_type : component_types) {
if (!this->has(component_type)) {
continue;
}
const GeometryComponent &component = *this->get_component_for_read(component_type);
Geometry Nodes: new geometry attribute API Currently, there are two attribute API. The first, defined in `BKE_attribute.h` is accessible from RNA and C code. The second is implemented with `GeometryComponent` and is only accessible in C++ code. The second is widely used, but only being accessible through the `GeometrySet` API makes it awkward to use, and even impossible for types that don't correspond directly to a geometry component like `CurvesGeometry`. This patch adds a new attribute API, designed to replace the `GeometryComponent` attribute API now, and to eventually replace or be the basis of the other one. The basic idea is that there is an `AttributeAccessor` class that allows code to interact with a set of attributes owned by some geometry. The accessor itself has no ownership. `AttributeAccessor` is a simple type that can be passed around by value. That makes it easy to return it from functions and to store it in containers. For const-correctness, there is also a `MutableAttributeAccessor` that allows changing individual and can add or remove attributes. Currently, `AttributeAccessor` is composed of two pointers. The first is a pointer to the owner of the attribute data. The second is a pointer to a struct with function pointers, that is similar to a virtual function table. The functions know how to access attributes on the owner. The actual attribute access for geometries is still implemented with the `AttributeProvider` pattern, which makes it easy to support different sources of attributes on a geometry and simplifies dealing with built-in attributes. There are different ways to get an attribute accessor for a geometry: * `GeometryComponent.attributes()` * `CurvesGeometry.attributes()` * `bke::mesh_attributes(const Mesh &)` * `bke::pointcloud_attributes(const PointCloud &)` All of these also have a `_for_write` variant that returns a `MutabelAttributeAccessor`. Differential Revision: https://developer.blender.org/D15280
2022-07-08 16:16:56 +02:00
const std::optional<AttributeAccessor> attributes = component.attributes();
if (attributes.has_value()) {
attributes->for_all(
[&](const AttributeIDRef &attribute_id, const AttributeMetaData &meta_data) {
callback(attribute_id, meta_data, component);
return true;
});
}
}
if (include_instances && this->has_instances()) {
const Instances &instances = *this->get_instances_for_read();
instances.foreach_referenced_geometry([&](const GeometrySet &instance_geometry_set) {
instance_geometry_set.attribute_foreach(component_types, include_instances, callback);
});
}
}
void GeometrySet::gather_attributes_for_propagation(
const Span<GeometryComponentType> component_types,
const GeometryComponentType dst_component_type,
bool include_instances,
const blender::bke::AnonymousAttributePropagationInfo &propagation_info,
Geometry Nodes: new geometry attribute API Currently, there are two attribute API. The first, defined in `BKE_attribute.h` is accessible from RNA and C code. The second is implemented with `GeometryComponent` and is only accessible in C++ code. The second is widely used, but only being accessible through the `GeometrySet` API makes it awkward to use, and even impossible for types that don't correspond directly to a geometry component like `CurvesGeometry`. This patch adds a new attribute API, designed to replace the `GeometryComponent` attribute API now, and to eventually replace or be the basis of the other one. The basic idea is that there is an `AttributeAccessor` class that allows code to interact with a set of attributes owned by some geometry. The accessor itself has no ownership. `AttributeAccessor` is a simple type that can be passed around by value. That makes it easy to return it from functions and to store it in containers. For const-correctness, there is also a `MutableAttributeAccessor` that allows changing individual and can add or remove attributes. Currently, `AttributeAccessor` is composed of two pointers. The first is a pointer to the owner of the attribute data. The second is a pointer to a struct with function pointers, that is similar to a virtual function table. The functions know how to access attributes on the owner. The actual attribute access for geometries is still implemented with the `AttributeProvider` pattern, which makes it easy to support different sources of attributes on a geometry and simplifies dealing with built-in attributes. There are different ways to get an attribute accessor for a geometry: * `GeometryComponent.attributes()` * `CurvesGeometry.attributes()` * `bke::mesh_attributes(const Mesh &)` * `bke::pointcloud_attributes(const PointCloud &)` All of these also have a `_for_write` variant that returns a `MutabelAttributeAccessor`. Differential Revision: https://developer.blender.org/D15280
2022-07-08 16:16:56 +02:00
blender::Map<blender::bke::AttributeIDRef, blender::bke::AttributeKind> &r_attributes) const
{
using namespace blender;
using namespace blender::bke;
/* Only needed right now to check if an attribute is built-in on this component type.
* TODO: Get rid of the dummy component. */
const GeometryComponent *dummy_component = GeometryComponent::create(dst_component_type);
this->attribute_foreach(
component_types,
include_instances,
[&](const AttributeIDRef &attribute_id,
const AttributeMetaData &meta_data,
const GeometryComponent &component) {
Geometry Nodes: new geometry attribute API Currently, there are two attribute API. The first, defined in `BKE_attribute.h` is accessible from RNA and C code. The second is implemented with `GeometryComponent` and is only accessible in C++ code. The second is widely used, but only being accessible through the `GeometrySet` API makes it awkward to use, and even impossible for types that don't correspond directly to a geometry component like `CurvesGeometry`. This patch adds a new attribute API, designed to replace the `GeometryComponent` attribute API now, and to eventually replace or be the basis of the other one. The basic idea is that there is an `AttributeAccessor` class that allows code to interact with a set of attributes owned by some geometry. The accessor itself has no ownership. `AttributeAccessor` is a simple type that can be passed around by value. That makes it easy to return it from functions and to store it in containers. For const-correctness, there is also a `MutableAttributeAccessor` that allows changing individual and can add or remove attributes. Currently, `AttributeAccessor` is composed of two pointers. The first is a pointer to the owner of the attribute data. The second is a pointer to a struct with function pointers, that is similar to a virtual function table. The functions know how to access attributes on the owner. The actual attribute access for geometries is still implemented with the `AttributeProvider` pattern, which makes it easy to support different sources of attributes on a geometry and simplifies dealing with built-in attributes. There are different ways to get an attribute accessor for a geometry: * `GeometryComponent.attributes()` * `CurvesGeometry.attributes()` * `bke::mesh_attributes(const Mesh &)` * `bke::pointcloud_attributes(const PointCloud &)` All of these also have a `_for_write` variant that returns a `MutabelAttributeAccessor`. Differential Revision: https://developer.blender.org/D15280
2022-07-08 16:16:56 +02:00
if (component.attributes()->is_builtin(attribute_id)) {
if (!dummy_component->attributes()->is_builtin(attribute_id)) {
/* Don't propagate built-in attributes that are not built-in on the destination
* component. */
return;
}
}
if (meta_data.data_type == CD_PROP_STRING) {
/* Propagating string attributes is not supported yet. */
return;
}
if (attribute_id.is_anonymous() &&
!propagation_info.propagate(attribute_id.anonymous_id())) {
return;
}
eAttrDomain domain = meta_data.domain;
Geometry Nodes: support instance attributes when realizing instances This patch refactors the instance-realization code and adds new functionality. * Named and anonymous attributes are propagated from instances to the realized geometry. If the same attribute exists on the geometry and on an instance, the attribute on the geometry has precedence. * The id attribute has special handling to avoid creating the same id on many output points. This is necessary to make e.g. the Random Value node work as expected afterwards. Realizing instance attributes has an effect on existing files, especially due to the id attribute. To avoid breaking existing files, the Realize Instances node now has a legacy option that is enabled for all already existing Realize Instances nodes. Removing this legacy behavior does affect some existing files (although not many). We can decide whether it's worth to remove the old behavior as a separate step. This refactor also improves performance when realizing instances. That is mainly due to multi-threading. See D13446 to get the file used for benchmarking. The curve code is not as optimized as it could be yet. That's mainly because the storage for these attributes might change soonish and it wasn't worth optimizing for the current storage format right now. ``` 1,000,000 x mesh vertex: 530 ms -> 130 ms 1,000,000 x simple cube: 1290 ms -> 190 ms 1,000,000 x point: 1000 ms -> 150 ms 1,000,000 x curve spiral: 1740 ms -> 330 ms 1,000,000 x curve line: 1110 ms -> 210 ms 10,000 x subdivided cylinder: 170 ms -> 40 ms 10 x subdivided spiral: 180 ms -> 180 ms ``` Differential Revision: https://developer.blender.org/D13446
2021-12-14 15:57:58 +01:00
if (dst_component_type != GEO_COMPONENT_TYPE_INSTANCES && domain == ATTR_DOMAIN_INSTANCE) {
domain = ATTR_DOMAIN_POINT;
}
auto add_info = [&](AttributeKind *attribute_kind) {
Geometry Nodes: support instance attributes when realizing instances This patch refactors the instance-realization code and adds new functionality. * Named and anonymous attributes are propagated from instances to the realized geometry. If the same attribute exists on the geometry and on an instance, the attribute on the geometry has precedence. * The id attribute has special handling to avoid creating the same id on many output points. This is necessary to make e.g. the Random Value node work as expected afterwards. Realizing instance attributes has an effect on existing files, especially due to the id attribute. To avoid breaking existing files, the Realize Instances node now has a legacy option that is enabled for all already existing Realize Instances nodes. Removing this legacy behavior does affect some existing files (although not many). We can decide whether it's worth to remove the old behavior as a separate step. This refactor also improves performance when realizing instances. That is mainly due to multi-threading. See D13446 to get the file used for benchmarking. The curve code is not as optimized as it could be yet. That's mainly because the storage for these attributes might change soonish and it wasn't worth optimizing for the current storage format right now. ``` 1,000,000 x mesh vertex: 530 ms -> 130 ms 1,000,000 x simple cube: 1290 ms -> 190 ms 1,000,000 x point: 1000 ms -> 150 ms 1,000,000 x curve spiral: 1740 ms -> 330 ms 1,000,000 x curve line: 1110 ms -> 210 ms 10,000 x subdivided cylinder: 170 ms -> 40 ms 10 x subdivided spiral: 180 ms -> 180 ms ``` Differential Revision: https://developer.blender.org/D13446
2021-12-14 15:57:58 +01:00
attribute_kind->domain = domain;
attribute_kind->data_type = meta_data.data_type;
};
auto modify_info = [&](AttributeKind *attribute_kind) {
attribute_kind->domain = bke::attribute_domain_highest_priority(
Geometry Nodes: support instance attributes when realizing instances This patch refactors the instance-realization code and adds new functionality. * Named and anonymous attributes are propagated from instances to the realized geometry. If the same attribute exists on the geometry and on an instance, the attribute on the geometry has precedence. * The id attribute has special handling to avoid creating the same id on many output points. This is necessary to make e.g. the Random Value node work as expected afterwards. Realizing instance attributes has an effect on existing files, especially due to the id attribute. To avoid breaking existing files, the Realize Instances node now has a legacy option that is enabled for all already existing Realize Instances nodes. Removing this legacy behavior does affect some existing files (although not many). We can decide whether it's worth to remove the old behavior as a separate step. This refactor also improves performance when realizing instances. That is mainly due to multi-threading. See D13446 to get the file used for benchmarking. The curve code is not as optimized as it could be yet. That's mainly because the storage for these attributes might change soonish and it wasn't worth optimizing for the current storage format right now. ``` 1,000,000 x mesh vertex: 530 ms -> 130 ms 1,000,000 x simple cube: 1290 ms -> 190 ms 1,000,000 x point: 1000 ms -> 150 ms 1,000,000 x curve spiral: 1740 ms -> 330 ms 1,000,000 x curve line: 1110 ms -> 210 ms 10,000 x subdivided cylinder: 170 ms -> 40 ms 10 x subdivided spiral: 180 ms -> 180 ms ``` Differential Revision: https://developer.blender.org/D13446
2021-12-14 15:57:58 +01:00
{attribute_kind->domain, domain});
attribute_kind->data_type = bke::attribute_data_type_highest_complexity(
{attribute_kind->data_type, meta_data.data_type});
};
r_attributes.add_or_modify(attribute_id, add_info, modify_info);
});
delete dummy_component;
}
static void gather_component_types_recursive(const GeometrySet &geometry_set,
const bool include_instances,
const bool ignore_empty,
Vector<GeometryComponentType> &r_types)
{
for (const GeometryComponent *component : geometry_set.get_components_for_read()) {
if (ignore_empty) {
if (component->is_empty()) {
continue;
}
}
r_types.append_non_duplicates(component->type());
}
if (!include_instances) {
return;
}
const blender::bke::Instances *instances = geometry_set.get_instances_for_read();
if (instances == nullptr) {
return;
}
instances->foreach_referenced_geometry([&](const GeometrySet &instance_geometry_set) {
gather_component_types_recursive(
instance_geometry_set, include_instances, ignore_empty, r_types);
});
}
blender::Vector<GeometryComponentType> GeometrySet::gather_component_types(
const bool include_instances, bool ignore_empty) const
{
Vector<GeometryComponentType> types;
gather_component_types_recursive(*this, include_instances, ignore_empty, types);
return types;
}
static void gather_mutable_geometry_sets(GeometrySet &geometry_set,
Vector<GeometrySet *> &r_geometry_sets)
{
r_geometry_sets.append(&geometry_set);
if (!geometry_set.has_instances()) {
return;
}
/* In the future this can be improved by deduplicating instance references across different
* instances. */
Instances &instances = *geometry_set.get_instances_for_write();
instances.ensure_geometry_instances();
for (const int handle : instances.references().index_range()) {
if (instances.references()[handle].type() == InstanceReference::Type::GeometrySet) {
GeometrySet &instance_geometry = instances.geometry_set_from_reference(handle);
gather_mutable_geometry_sets(instance_geometry, r_geometry_sets);
}
}
}
void GeometrySet::modify_geometry_sets(ForeachSubGeometryCallback callback)
{
Vector<GeometrySet *> geometry_sets;
gather_mutable_geometry_sets(*this, geometry_sets);
if (geometry_sets.size() == 1) {
/* Avoid possible overhead and a large call stack when multithreading is pointless. */
callback(*geometry_sets.first());
}
else {
blender::threading::parallel_for_each(
geometry_sets, [&](GeometrySet *geometry_set) { callback(*geometry_set); });
}
}
/** \} */
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
/* -------------------------------------------------------------------- */
/** \name C API
* \{ */
void BKE_geometry_set_free(GeometrySet *geometry_set)
{
delete geometry_set;
}
Geometry Nodes: support for geometry instancing Previously, the Point Instance node in geometry nodes could only instance existing objects or collections. The reason was that large parts of Blender worked under the assumption that objects are the main unit of instancing. Now we also want to instance geometry within an object, so a slightly larger refactor was necessary. This should not affect files that do not use the new kind of instances. The main change is a redefinition of what "instanced data" is. Now, an instances is a cow-object + object-data (the geometry). This can be nicely seen in `struct DupliObject`. This allows the same object to generate multiple geometries of different types which can be instanced individually. A nice side effect of this refactor is that having multiple geometry components is not a special case in the depsgraph object iterator anymore, because those components are integrated with the `DupliObject` system. Unfortunately, different systems that work with instances in Blender (e.g. render engines and exporters) often work under the assumption that objects are the main unit of instancing. So those have to be updated as well to be able to handle the new instances. This patch updates Cycles, EEVEE and other viewport engines. Exporters have not been updated yet. Some minimal (not master-ready) changes to update the obj and alembic exporters can be found in P2336 and P2335. Different file formats may want to handle these new instances in different ways. For users, the only thing that changed is that the Point Instance node now has a geometry mode. This also fixes T88454. Differential Revision: https://developer.blender.org/D11841
2021-09-06 18:22:24 +02:00
bool BKE_object_has_geometry_set_instances(const Object *ob)
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
{
Geometry Nodes: support for geometry instancing Previously, the Point Instance node in geometry nodes could only instance existing objects or collections. The reason was that large parts of Blender worked under the assumption that objects are the main unit of instancing. Now we also want to instance geometry within an object, so a slightly larger refactor was necessary. This should not affect files that do not use the new kind of instances. The main change is a redefinition of what "instanced data" is. Now, an instances is a cow-object + object-data (the geometry). This can be nicely seen in `struct DupliObject`. This allows the same object to generate multiple geometries of different types which can be instanced individually. A nice side effect of this refactor is that having multiple geometry components is not a special case in the depsgraph object iterator anymore, because those components are integrated with the `DupliObject` system. Unfortunately, different systems that work with instances in Blender (e.g. render engines and exporters) often work under the assumption that objects are the main unit of instancing. So those have to be updated as well to be able to handle the new instances. This patch updates Cycles, EEVEE and other viewport engines. Exporters have not been updated yet. Some minimal (not master-ready) changes to update the obj and alembic exporters can be found in P2336 and P2335. Different file formats may want to handle these new instances in different ways. For users, the only thing that changed is that the Point Instance node now has a geometry mode. This also fixes T88454. Differential Revision: https://developer.blender.org/D11841
2021-09-06 18:22:24 +02:00
const GeometrySet *geometry_set = ob->runtime.geometry_set_eval;
if (geometry_set == nullptr) {
return false;
}
for (const GeometryComponent *component : geometry_set->get_components_for_read()) {
if (component->is_empty()) {
continue;
}
const GeometryComponentType type = component->type();
bool is_instance = false;
switch (type) {
case GEO_COMPONENT_TYPE_MESH:
is_instance = ob->type != OB_MESH;
break;
case GEO_COMPONENT_TYPE_POINT_CLOUD:
is_instance = ob->type != OB_POINTCLOUD;
break;
case GEO_COMPONENT_TYPE_INSTANCES:
is_instance = true;
break;
case GEO_COMPONENT_TYPE_VOLUME:
is_instance = ob->type != OB_VOLUME;
break;
case GEO_COMPONENT_TYPE_CURVE:
is_instance = !ELEM(ob->type, OB_CURVES_LEGACY, OB_FONT);
break;
case GEO_COMPONENT_TYPE_EDIT:
break;
}
if (is_instance) {
return true;
}
Geometry Nodes: support for geometry instancing Previously, the Point Instance node in geometry nodes could only instance existing objects or collections. The reason was that large parts of Blender worked under the assumption that objects are the main unit of instancing. Now we also want to instance geometry within an object, so a slightly larger refactor was necessary. This should not affect files that do not use the new kind of instances. The main change is a redefinition of what "instanced data" is. Now, an instances is a cow-object + object-data (the geometry). This can be nicely seen in `struct DupliObject`. This allows the same object to generate multiple geometries of different types which can be instanced individually. A nice side effect of this refactor is that having multiple geometry components is not a special case in the depsgraph object iterator anymore, because those components are integrated with the `DupliObject` system. Unfortunately, different systems that work with instances in Blender (e.g. render engines and exporters) often work under the assumption that objects are the main unit of instancing. So those have to be updated as well to be able to handle the new instances. This patch updates Cycles, EEVEE and other viewport engines. Exporters have not been updated yet. Some minimal (not master-ready) changes to update the obj and alembic exporters can be found in P2336 and P2335. Different file formats may want to handle these new instances in different ways. For users, the only thing that changed is that the Point Instance node now has a geometry mode. This also fixes T88454. Differential Revision: https://developer.blender.org/D11841
2021-09-06 18:22:24 +02:00
}
return false;
Geometry Nodes: initial scattering and geometry nodes This is the initial merge from the geometry-nodes branch. Nodes: * Attribute Math * Boolean * Edge Split * Float Compare * Object Info * Point Distribute * Point Instance * Random Attribute * Random Float * Subdivision Surface * Transform * Triangulate It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier. Notes on the Generic attribute access API The API adds an indirection for attribute access. That has the following benefits: * Most code does not have to care about how an attribute is stored internally. This is mainly necessary, because we have to deal with "legacy" attributes such as vertex weights and attributes that are embedded into other structs such as vertex positions. * When reading from an attribute, we generally don't care what domain the attribute is stored on. So we want to abstract away the interpolation that that adapts attributes from one domain to another domain (this is not actually implemented yet). Other possible improvements for later iterations include: * Actually implement interpolation between domains. * Don't use inheritance for the different attribute types. A single class for read access and one for write access might be enough, because we know all the ways in which attributes are stored internally. We don't want more different internal structures in the future. On the contrary, ideally we can consolidate the different storage formats in the future to reduce the need for this indirection. * Remove the need for heap allocations when creating attribute accessors. It includes commits from: * Dalai Felinto * Hans Goudey * Jacques Lucke * Léo Depoix
2020-12-02 13:25:25 +01:00
}
/** \} */