Files
test2/scripts/startup/bl_operators/geometry_nodes.py

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

245 lines
8.5 KiB
Python
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
import bpy
from bpy.types import Operator
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
from bpy.app.translations import pgettext_data as data_
2020-12-16 18:02:40 +11:00
def build_default_empty_geometry_node_group(name):
group = bpy.data.node_groups.new(name, 'GeometryNodeTree')
group.inputs.new('NodeSocketGeometry', data_("Geometry"))
group.outputs.new('NodeSocketGeometry', data_("Geometry"))
input_node = group.nodes.new('NodeGroupInput')
output_node = group.nodes.new('NodeGroupOutput')
output_node.is_active_output = True
input_node.select = False
output_node.select = False
input_node.location.x = -200 - input_node.width
output_node.location.x = 200
return group
def geometry_node_group_empty_new():
group = build_default_empty_geometry_node_group(data_("Geometry Nodes"))
group.links.new(group.nodes[data_("Group Input")].outputs[0], group.nodes[data_("Group Output")].inputs[0])
return group
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
2020-12-16 18:02:40 +11:00
def geometry_modifier_poll(context):
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
ob = context.object
# Test object support for geometry node modifier
if not ob or ob.type not in {'MESH', 'POINTCLOUD', 'VOLUME', 'CURVE', 'FONT', 'CURVES'}:
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 False
return True
2020-12-16 18:02:40 +11:00
def get_context_modifier(context):
if context.area.type == 'PROPERTIES':
modifier = context.modifier
else:
modifier = context.object.modifiers.active
if modifier is None or modifier.type != 'NODES':
return None
return modifier
def edit_geometry_nodes_modifier_poll(context):
return get_context_modifier(context) is not None
def socket_idname_to_attribute_type(idname):
if idname.startswith("NodeSocketInt"):
return "INT"
elif idname.startswith("NodeSocketColor"):
return "FLOAT_COLOR"
elif idname.startswith("NodeSocketVector"):
return "FLOAT_VECTOR"
elif idname.startswith("NodeSocketBool"):
return "BOOLEAN"
elif idname.startswith("NodeSocketFloat"):
return "FLOAT"
raise ValueError("Unsupported socket type")
return ""
def modifier_attribute_name_get(modifier, identifier):
try:
return modifier[identifier + "_attribute_name"]
except KeyError:
return None
def modifier_input_use_attribute(modifier, identifier):
try:
return modifier[identifier + "_use_attribute"] != 0
except KeyError:
return False
def get_socket_with_identifier(sockets, identifier):
for socket in sockets:
if socket.identifier == identifier:
return socket
return None
def get_enabled_socket_with_name(sockets, name):
for socket in sockets:
if socket.name == name and socket.enabled:
return socket
return None
class MoveModifierToNodes(Operator):
"""Move inputs and outputs from in the modifier to a new node group"""
bl_idname = "object.geometry_nodes_move_to_nodes"
bl_label = "Move to Nodes"
bl_options = {'REGISTER', 'UNDO'}
@classmethod
def poll(cls, context):
return edit_geometry_nodes_modifier_poll(context)
def execute(self, context):
modifier = get_context_modifier(context)
if not modifier:
return {'CANCELLED'}
old_group = modifier.node_group
if not old_group:
return {'CANCELLED'}
wrapper_name = old_group.name + ".wrapper"
group = build_default_empty_geometry_node_group(wrapper_name)
group_node = group.nodes.new("GeometryNodeGroup")
group_node.node_tree = old_group
group_node.update()
group_input_node = group.nodes[data_("Group Input")]
group_output_node = group.nodes[data_("Group Output")]
# Copy default values for inputs and create named attribute input nodes.
input_nodes = []
first_geometry_input = None
for input_socket in old_group.inputs:
identifier = input_socket.identifier
group_node_input = get_socket_with_identifier(group_node.inputs, identifier)
if modifier_input_use_attribute(modifier, identifier):
input_node = group.nodes.new("GeometryNodeInputNamedAttribute")
input_nodes.append(input_node)
input_node.data_type = socket_idname_to_attribute_type(input_socket.bl_socket_idname)
attribute_name = modifier_attribute_name_get(modifier, identifier)
input_node.inputs["Name"].default_value = attribute_name
output_socket = get_enabled_socket_with_name(input_node.outputs, "Attribute")
group.links.new(output_socket, group_node_input)
elif hasattr(input_socket, "default_value"):
group_node_input.default_value = modifier[identifier]
elif input_socket.bl_socket_idname == 'NodeSocketGeometry':
if not first_geometry_input:
first_geometry_input = group_node_input
group.links.new(group_input_node.outputs[0], first_geometry_input)
# Adjust locations of named attribute input nodes and group input node to make some space.
if input_nodes:
for i, node in enumerate(input_nodes):
node.location.x = -175
node.location.y = i * -50
group_input_node.location.x = -350
# Connect outputs to store named attribute nodes to replace modifier attribute outputs.
store_nodes = []
first_geometry_output = None
for output_socket in old_group.outputs:
identifier = output_socket.identifier
group_node_output = get_socket_with_identifier(group_node.outputs, identifier)
attribute_name = modifier_attribute_name_get(modifier, identifier)
if attribute_name:
store_node = group.nodes.new("GeometryNodeStoreNamedAttribute")
store_nodes.append(store_node)
store_node.data_type = socket_idname_to_attribute_type(output_socket.bl_socket_idname)
store_node.domain = output_socket.attribute_domain
store_node.inputs["Name"].default_value = attribute_name
input_socket = get_enabled_socket_with_name(store_node.inputs, "Value")
group.links.new(group_node_output, input_socket)
elif output_socket.bl_socket_idname == 'NodeSocketGeometry':
if not first_geometry_output:
first_geometry_output = group_node_output
# Adjust locations of store named attribute nodes and move group output.
# Note that the node group has its sockets names translated, while the built-in nodes don't.
if store_nodes:
for i, node in enumerate(store_nodes):
node.location.x = (i + 1) * 175
node.location.y = 0
group_output_node.location.x = (len(store_nodes) + 1) * 175
group.links.new(first_geometry_output, store_nodes[0].inputs["Geometry"])
for i in range(len(store_nodes) - 1):
group.links.new(store_nodes[i].outputs["Geometry"], store_nodes[i + 1].inputs["Geometry"])
group.links.new(store_nodes[-1].outputs["Geometry"], group_output_node.inputs[data_("Geometry")])
else:
group.links.new(first_geometry_output, group_output_node.inputs[data_("Geometry")])
modifier.node_group = group
return {'FINISHED'}
class NewGeometryNodesModifier(Operator):
"""Create a new modifier with a new geometry node group"""
bl_idname = "node.new_geometry_nodes_modifier"
bl_label = "New Geometry Node Modifier"
bl_options = {'REGISTER', 'UNDO'}
@classmethod
def poll(cls, context):
return geometry_modifier_poll(context)
def execute(self, context):
modifier = context.object.modifiers.new(data_("GeometryNodes"), "NODES")
if not modifier:
return {'CANCELLED'}
group = geometry_node_group_empty_new()
modifier.node_group = group
return {'FINISHED'}
class NewGeometryNodeTreeAssign(Operator):
2021-01-04 15:42:15 -06:00
"""Create a new geometry node group and assign it to the active modifier"""
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
bl_idname = "node.new_geometry_node_group_assign"
bl_label = "Assign New Geometry Node Group"
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
bl_options = {'REGISTER', 'UNDO'}
@classmethod
def poll(cls, context):
return geometry_modifier_poll(context)
def execute(self, context):
modifier = get_context_modifier(context)
if not modifier:
return {'CANCELLED'}
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
2021-03-06 18:21:17 +11:00
group = geometry_node_group_empty_new()
modifier.node_group = group
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 {'FINISHED'}
classes = (
NewGeometryNodesModifier,
NewGeometryNodeTreeAssign,
MoveModifierToNodes,
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
)