The modifier was called "Empty", which reflects the fact that the modifier
doesn't do anything when you add it. However, that is not true after you
change the modifier's node group, and the fact that the modifier uses
nodes is the one thing that distinguishes it at this point. This commit
changes the name to "Geometry Nodes", which is simple and consistent
with the "Geometry Node Editor" and the naming of the project so far.
The handling of muted nodes is handled at the derived node tree
level now. This is also where expanding node groups is handled.
Muted nodes are relinked and removed from the derived tree
during construction. The geometry node evaluation code does
not have to know about muted nodes this way.
Note: This also changes the Shader socket color, to match "Shading" in the Outliner.
Theme update for shader nodes will be committed separately.
Related to T82689
Before, attributes could only be accessed via setter and getter functions.
While this works well, it can be cumbersome and slow.
It would be better if most code could just operate on the underlying
attribute array directly. The issue is that some attributes are currently
not stored in plain arrays (e.g. vertex positions, vertex groups, ...).
This patch implements a solution where these "special" attributes are
converted into a temporary array for easy access and then copied
back into their correct form after all modifications are done.
Attribute accessors now have a `get_span` and `apply_span` method,
which can be used exactly for that purpose. In the case of attributes
that are plain arrays internally, only small constant overhead is added.
Material indices and smooth states are not destroyed anymore.
If the two incoming meshes use different materials, the materials of
the first input will be used. This is obviously not ideal, because the
second input might change its material.
In the future we want to handle that differently and keep both materials,
but for now this is consistent with the boolean node.
This commit makes the way the selector in the middle of the node editor
header interacts with the active modifier much more intuitive.
- With no active modifier, or when the active modifier is a non-node
modifier, the "new" operator adds a new nodes modifier.
- With an active node modifier, the drop-down affects the modifier's
selected node group.
- With no active object, there is no drop-down in the header.
- The node editor's node group always updates, even when there is no
active modifier.
- Pinning disables the drop-down and keeps the same node group
selected, regardless of the context.
This is designed to work similarly to materials and material slots,
which should be familiar to any Blender user. One workflow not allowed
by this design is editing any geometry node group, regardless of the
context, but if that turns out to be an issue it should be solved in a
similar way for all node editors.
This adds "modifier" as a new context variable in the property editor.
It is used instead of getting the active modifier directly. This way it
displays in the breadcrumbs panel, and the variable can be set only
in the modifier tab, meaning it won't work from the shortcut even
when the modifiers aren't visible.
While we can support different domains already in this node, there is not
really anything you could do with that data yet. It can be added back when
there is a use case for it.
Currently, implementing a node that supports math operations requires
you to write a long switch statement. The goal of this abstraction
is to reduce the number of these switch statements.
Note, this commit adds new such long switch cases. However, in the upcoming
commits existing math nodes are updated to use this abstraction, resulting in
less code overall.
Based on discussion with the team, summarized in T82691, it makes sense
to simplify the shortcut behavior and remove the way the shortcuts
currently activate on hover if there is no active modifier. This commit
also removes the behavior of clearing the active modifier when you
click on blank space in the modifier tab.
This commit includes these improvements to operators and how they
relate to the active modifier:
- Shortcuts performed while there is an active modifier will affect
only that modifier.
- Clicking anywhere on the empty space in a modifier's panel will make
it active.
- Clicking anywhere in the empty properties space will clear the
active.
These changes require some refactoring of object modifier code. First
is splitting up the modifier property invoke callback, which now needs
to be able to get the active modifier separately from the hovered
modifier for the different operators. Second is a change to removing
modifiers, where there is now a separate function to remove a modifier
from an object's list, in order to handle changing the active. Finally
the panel handler neeeds a small tweak so that this "click in panel"
event can be handled afterwards.
This provide a simple implementation for basic math operations. Add,
subtract, multiple, and divide are implemented so far. Domain
interpolation is not implemented yet either.
This commit also adds two "Type" options to the node for choosing
explicitly whether to use an attribute or a float value as one of the
inputs. It is not possible to use two floats as inputs, because that
would be a regular math node.
Some cleanup to avoid some code duplication will come in a future
commit.
This node creates a new geometry based on two incoming geometries.
In the future the node ui can be extended to support more than two inputs.
The output geometry will have all the attributes of the inputs.
If both inputs have non-matching set of attributes, missing data is either
interpolated, type cast or zeroed.
Ref T82786.
* It now knows its own component type.
* Add virtual method to check if the component is empty.
* Add virtual methods to find all attribute names.
* Support adding existing components to a geometry set.
This changes the meaning of the order of the two geometry input sockets
to match the existing boolean modifier. The first socket is now the base
mesh and the second socket is the "cutter" geometry.
Similar to how the node editor context works for materials, this commit
makes the node group displayed in the node editor depend on the active
object and its active modifier. To keep the node group from changing,
just pin the node group in the header.
There are still missing updates when switching the modifier's node group.
This exposes the operator to set the active modifier as the type icon
in the header of the modifier panels. Tweaks to the widget drawing
code were necessary to use the red alert for non-emboss operator
buttons. Then, the panel for the active modifier gets a border around it
(which currently uses the property search theme color), requiring
an "active property" field in the PanelType struct.
Previously, it was always outputting an empty geometry in that case.
This was confusing because e.g. the output of a union operation
with a single non-empty mesh is also non-empty.
This will probably not stay exactly like this, because it looks a bit ugly.
Adding this now, because it increases flexibility a lot and allows me
to test the attribute api a bit more.
Originally, there were technical reasons for why two different domains
were used. Those reasons have been lost unfortunately, and there
don't seem to be any issues with combining the domains now.
This reduces confusion when users have to select the domain type
in enums.
Attributes can now be accessed through a GeometryComponent.
MeshComponent and PointCloudComponent implement the specific
attribute access mechanisms using virtual methods.
This simplifies code that should work on different geometry components
quite a bit, because it only has to be written once without using templates.
The Random Attribute node benefits from that a lot.
Although disabling this option is equivalent to setting the edge angle
to 180 degrees, there is no great reason to change the settings from what
is exposed in the modifier.
Differential Revision: https://developer.blender.org/D9564
The random attribute node currently doesn't support every attribute
domain or data type. With a bit of boilerplate code we can remove these
unsupported enum items from the UI for now, and also get a system to
remove unsupported items from the UI in the future. If all attribute
creation nodes end up always supporting all attibute types and domains,
this can be removed.
There will be an operator to access this property instead, which provides
a better name and tooltip to expose in the UI, and makes it more clear
that the property is dependent on the other modifiers.
This commit adds functions to set and get the object's active
modifier, which is stored as a flag in the ModifierData struct,
similar to constraints. This will be used to set the context in
the node editor. There are no visible changes in this commit.
This adds some ui boilerplate code for the Random Attribute node and
provides an initial implementation.
Note, while the implementation can already randomize attributes, it might
not behave as expected under all circumstances yet. It's still work in progress.
I will have to do a couple more iterations on this api in the upcoming weeks,
but for now it should be good enough.
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).
Accessing attributes through this indirection does have a performance penalty.
In later iterations of this API I want to reduce this penalty and extend the API
so that performance critical code does not have to go through the indirection
for every attribute access.
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.
Until there is a icon made specially for this, the nodetree icon is up
for grabs. Using it in the nodegroup + modifier + editor helps the users
to make a connection on where to edit those modifiers.
The point cloud object is the only one that will support instancing at
first. So we can expose it as a regular object.
It is limited since it has no edit mode. But this is not different than
the volume object.
As part of preparing this work for master, with an initial subset of
working functionalities the Geometry Nodes Editor can be exposed for
everyone.
(it also helps people that are testing the branch)
Previously, GeometrySets and GeometryComponents has reference
counters and could be shared. This commit changes it so that
only GeometryComponents are shared. A GeometrySet is a fairly
small type that is cheap to copy.
A lot of code simplifies when we can assume that GeometrySet
is cheap to copy.
Brecht and I decided to move the GeometrySet classes
to the global namespace for now. This way we can use
the same type in C and C++ and don't have to use
reinterpret_cast as often.
More than output for the modifier's node group is fine, because the
node group may be used in other situations, but the modifier will
not work if there is no geometry output socket.
This diff remove "Combine Strings" and "Group Instance ID" nodes from the UI (but keep them in the source code to be able to bring them back if needed).
This is part of "Cleanup 1st Sprint Nodes" (https://developer.blender.org/T82370).
Differential Revision: https://developer.blender.org/D9558
This introduces a regression. Users need to re-create their modifiers
and set their values again. The NodeGroups are intact so all they need
is to create new "Empty" modifiers and set the node group.
---
Original problem: Eventually most modifiers will be node-based, so naming
this new modifier "Nodes" will get outdated pretty quickly.
Calling it "Empty" it's fairly descriptive, since the node tree simply connects
the modifier Input to an Output, without any effect.
There is a potential problem, that it could be associated with the
Empty object type. Like some sort of Hook.
This is the task T82700, and the design task T82537.
---
Leave the internal DNA data struct as "Nodes", and only rename the
modifier and user interface to Empty.
This implements a workaround for the issue that (for historical reasons)
the names of vertex groups are stored on the object while the actual vertex
group data is stored on the mesh.
The solution is to copy the vertex group names from the object into
the `MeshComponent` so that the information is not lost, when the
object cannot be accessed.
Previously, the execution function of a geometry node has three parameters.
Now it has only one. This makes it easier to pass more information to the
execution function, that might only be used by a few nodes, because we
don't have to add more parameters that are unused in most cases.
* Support vertex weights to control density.
* O(n) performance instead of O(n^2).
* More stable when density weights are changed.
The vertex group has to be specified using an index for now.
This is a technical limitation that will resolved a bit later.
Objects can evaluate to a geometry set instead of a single ID (only point cloud
objects for now). In the depsgraph object iterator, those geometry components
are expanded into temporary objects.
It's important to note that instanced objects can also contain geometry
components. Therefore, they have to be split up into multiple objects
as well in some cases.
At a high level the iterator works like so:
```
for object in depsgraph:
for component in object:
yield object_from_component(component)
for dupli in make_duplis_list(object):
for component in dupli:
yield object_from_component(component)
```
DEG_iterator_objects_next has been cleaned up, to make this structure
a bit more apparent.
This changes the signature of the modifyPointCloud function.
I'm doing that instead of making a new callback, because it requires
changes to significantly fewer files. Eventually it would be good
combine modifyMesh, modifyHair, modifyPointCloud and modifyVolume
into one modifyGeometrySet.
I temporarily disabled the displacement only modifiers for point clouds.
Support can be added back a bit later. I assume those have not been
used anywhere anyway.
The output of point cloud modifiers can not only be another point cloud,
but also a mesh and/or some instances. I added a new geometry_set_eval
field to Object_Runtime. For point cloud objects, the final geometry is
now referenced by that pointer instead of data_eval. The data_eval field
is still initialized after modifier evaluation to make some other code happy.
The evaluated geometry set is not yet passed to the renderer, so a point
cloud is currently rendered empty.
This node takes an object as input and outputs its location, rotation,
scale and geometry. Right now the loc/rot/scale are extracted from the
objects `obmat`. The geometry is just the mesh in the local space of the
source object.
We will likely need some more control over space transformations
using enums in the node, but those can be added a bit later.
The fundamental difference between object sockets and the
other existing data sockets is that an object is an ID data block.
Changing the value of an object socket also changes the depsgraph.
The modifier has to analyse the node tree to figure out which other
objects it depends on. Currently, this is done very simply by just
looping over all sockets and collecting the objects. In the future
this can be improved by also figuring out what components of
an object are needed.
Instead of passing object pointers around in the node tree, we actually
use a handle. This handle is just a number internally that identifies
a specific object. The conversion between handles and object pointers
is done using a map that is provided by the modifier.
This approach has a couple of benefits. It protects us a bit from passing
around pointers that are not known to the modifier and therefore are
not in the depsgraph. Furthermore, the object pointer can change
while the handle stays the same. This is not important right now, but
is not unlikely to become useful in the future.
The API for how nodes access object pointers is not ideal yet and
will be improved in the future.
This commits implements multiple changes:
* Adds a simple C API.
* Improves the ownership handling by introducing GeometryOwnershipType.
* Adds an InstancesComponent that stores positions and Object pointers.
This should not change any functionality.
After talking to Brecht, we agreed that it might be good
not to have a class called Geometry for now. In the future
we might want to use a Geometry class as base class for
meshes, curves, etc.
This commit renames the Geometry class to GeometrySet,
because it is essentially a container that can contain
multiple geometries of different types.
This node is similar to the subdivision surface modifier. It subdivides
the mesh component of a geometry. The options have the same meaning
as in the modifier.
Differential Revision: https://developer.blender.org/D9364
Naming based on supported geometry components only works in some cases.
For nodes that work with multiple data types, it's better to categorize
them based on their functionality.
Theoretically the modifier's properties could be changed or removed by
Python, so it may be useful to have errors printed and added to the
modifier in that case.
I kept the check as a separate step from the `compute_geometry`
pass because the object is needed to set the modifier error message.
But it could easily be moved there in the future.
If we add the "name" label manually with uiItemR, we can use the socket
identifier instead of the name for the IDProperty name. This will also
allow us more flexibility in how to draw the settings in the future, and
removes the empty _RNA_UI proprety that was drawn before.
The system for exposing property settings like min, max, default, subtype,
etc. for ID properties is quite convoluted currently, so I won't give a
full description here, but this commit creates the tree of ID properties
needed to store that information. This means that property subtypes like
"angle" or "XYZ" will affect the display in the modifier.
Limitations:
- The _RNA_UI property is displayed in the modifier. This may require a
modification to uiDefAutoButsRNA to fix.
- IDProperties must have unique names, but node sockets don't have
that limitation. This can be solved by adding a "UI name" field to
IDProperties.
We can't use that for now, because the original mesh might be freed
before the new copied mesh. When the original mesh is freed, the shared
layers will be freed as well.
This can probably be improved in the future, so that we can actually
share attribute arrays between meshes. An approach similar to how
geometries and components are shared should work as well.
A geometry now contains zero or more geometry components.
A geometry component is a subclass off GeometryComponent.
Currently, there is a MeshComponent and PointCloudComponent.
A geometry contains at most one component of each type.
Individual components can be shared between multiple geometries
to avoid unnecessary copies. For that, each component has a user
count that determines when the component will be freed and whether
it is mutable.
Code working with geometries can either work with the components
directly, or use the utility functions on the geometry that cover the
most common operations.
In the future, additional component types can be added. For example,
we'll probably need components for curves and volumes.
Furthermore, something like an InstancesComponent can be added,
which contains points with attributes and references a geometry/object/collection
that is instanced on all the points.
The execute callback of a geometry node gets more domain specific
types as parameters now: GeoNodeInputs and GeoNodeOutputs.
Those types are also aware of what node is being executed and can
provide better error messages when they are used incorrectly.
The first is a "Point Distribute" node, which takes a mesh and outputs a
pointcloud, scattering points randomly on the surface based on the input
density. The distribution algorithm is extremely basic at this point, and
doesn't take into account an attribute weight at each vertex.
The second node is a "Point Instance" node, which is mainly a placeholder
until the engineering design for how to work with instancing in geometry
nodes is decided. For now it just runs the same code that the convert
pointcloud to mesh operator uses.
This adds another set of the same mesh functions but for pointclouds.
There are probably better ways to generalize this functionality, but that
may have to be rethought in the future anyway if we want to store
multiple of each type. And anyway it's handy to have a specific set of
"pointcloud" functions available in the node implementations.
This uses the code from the rewritten boolean modifier from 2.91 as a node.
The implementation is about as minimal as it can get, since for now the use
case of this node is mostly to test multiple geometry outputs and inputs.
The boolean code requires a BMesh, so the node converts the input meshes
to BMesh, so the bmesh is added as a dependency to the nodes module.
The settings can only be set via Python currently. The matching between properties
and group inputs is based on the socket identifier (which is e.g. `Input_5`).
Maybe we'll have to use a different matching strategy in the future, will see.
The properties are not used yet, but can already be accessed with Python
using `modifier.settings['setting name']`.
The plan is to use id properties to store the parameters that the modifier
passes into the geometry node group.
Group inputs are not yet exposed in the modifier. For now I just added
a simple float setting that will be passed to every float input of the group.
Inputs of other types have some default value.
We might not want to have all those nodes in a final version.
Some of them have been added with particle nodes in mind.
However, to test the evaluation system it is useful to have a
couple of nodes available.
Those nodes should "just" work, because their implementation
is reused from the particle nodes project.
This adds support for nodes that have a multi-function implementation.
That includes various function nodes like Math, Combine Vector, ...
Furthermore, there is support for implicit conversions now. So it should
work when one connects e.g. a float to an integer and vice versa.
These two functions "snode_notify" and "ED_node_tag_update_id" appear to
be mostly duplicates. However, there is already a case for each type of
built-in node tree, so it makes sense to add one for the geometry node
tree as well. This doesn't solve the update issues for changing number
in buttons, that must be handled somewhere else.
This adds a layer of abstraction between the code calling callbacks
on sockets and the implementation of those callbacks.
This abstraction layer allows some sockets to not implement all
callbacks when those can be derived from some other callback.
This is still very basic and does quite a few unnecessary computations.
Also the error handling is quite weak currently, so when invalid things are
connected, it will probably just crash.
Also the interface that individual nodes have to implement will have to change,
but the current solution is a good starting point.
Only the triangulate node is implemented for now.
Adjust the threshold for switching from the base case to trace > 0,
based on very similar example code from www.euclideanspace.com to
avoid float precision issues when trace is close to -1.
Also, remove conversions to and from double, because using double
here doesn't really have benefit, especially with the new threshold.
Finally, add quaternion-matrix-quaternion round trip tests with
full coverage for all 4 branches.
Differential Revision: https://developer.blender.org/D9675
Extract initialisation code of box selection into separate functions.
No functional changes.
Reviewed By: zeddb, sybren
Differential Revision: https://developer.blender.org/D9196
Fix various problems in the Action Group rearranging code. All fixes are
necessary to resolve the bug.
- Before groups are rearranged, the channels are moved into their
respective groups (so no longer referenced by `action->channels`). A
temporary group is made for ungrouped channels. The code made
assumptions about the channels being in the same order as the groups;
that assumption has been removed.
- Looping over channels in an Action Group should stop when reaching the
last channel, and not until `NULL`.
- After all the reshuffling is done, the `action->channels` linked list
wasn't terminated properly. Now `first.prev` and `last.next` are set
to `NULL` to avoid infinite loops.
Previously image accessor was sharing array pointer for tracks access.
Now it is possible to pass a temporary array valid only during the
initialization process.
Should be no functional changes.
This is something not-so-trivial to see from just reading code, which
shown an important of proper comments and clear naming.
Main source of confusion was that it is not immediately clear that
AutoTrack context is to see all tracks, but tracking to only operate on
selected ones.
The windows build leaves a convenience helper script
in the build folder called, rebuild.cmd. This change
passes any parameters you give rebuild.cmd to ninja
so you can easily pass it additional parameters without
having to edit the batch file manually.
This operators reset the vertex color information of the strokes.
If nothing is selected, all strokes are reset. If any is selected, only selected strokes are reset.
Also added a new menu Paint in Vertex Color mode.
Differential Revision: https://developer.blender.org/D9647
The event queue can contain events from before pointer warping, ignore those
now. This is an old issue, but became more common now that we disabled event
coalescing and started using the event mouse location rather than the current
mouse location.
Thanks to Yevgeny Makarov and Nicholas Rishel for helping solve this.
Ref D9662
Blacklist a bunch of tests on the GPU, which are known to currently have
differences with CPU. These can be enabled as they are fixed or the test
is modified to give compatible results when there are known limitations.
OSL tests were also moved to their own directory since those are not
supported on the GPU.
Ref T82193
CYCLES_TEST_DEVICES is a list of devices (CPU, CUDA, OPTIX, OPENCL). It is set
to CPU only by default.
Test output is now writen to build/tests/cycles/<device>, and the HTML report
has separate report pages for the different devices, with option to compare
between CPU and GPU renders.
Various GPU tests are still failing due to CPU/GPU differences, these are to be
fixed or blacklisted still.
Ref T82193
Allows to use mutex, scoped_lock, and conditional_variable from within
the libmv namespace.
Implementation is coming from C++11. Other configurations are easy to
implement, but currently C++11 is the way to go.
During revision of {D8952} one of the comments was to make a function that converts the render percentage to a factor. This to avoid code duplication. However the duplicated code was already all over the compositor code. So in order to avoid this code duplication for {D8952} I propose to first cleanup the duplicated code and build patch {D8952} based on this clean up.
The method that converts the render percentage to a factor is put in the CompositorContext. Why? The CompositorContext keeps DNA information like the renderdata. DNA, and thus the CompositorContext, keeps the size of the render resolution in percentage (user oriented). The compositor needs the size of the render resolution as a factor. So the CompositorContext seems like the obvious place to have this conversion method.
Why not in de NodeBase? The method could've been added to the nodebase, but I wanted to keep the nodebase as clean as possible and not put simple "conversion" methods into this base class. Also I didn't really like the call flow: you'd always have to get the renderdata size from the context and then convert.
Putting it in the CompositorContext avoids this extra invoke of a call.
Why not in the Converter? See nodebase. And the Converter seems more like a class for "structural" and complex node tree conversions. Not the simple conversions.
Reviewed By: Sergey Sharybin
Differential Revision: https://developer.blender.org/D9566
This commit replaces while loops and for loops with the equivalent
macro. This results in much more readable code in some places,
and it's now more apparent when the situation is more complicated
than just iterating through a linked list.
As proposed in T74432 and already implemented in several commits,
"region" is the preferred name for `ARegion` variables, rather than
any variant of "ar". This commit changes a few "ar" variables that have
popped up over time and also adjusted names of variants like "arnew".
This separates out PugiXML that was previously
bundled by OIIO.
As this linux/mac libs are not available
this commit only contains the builder and windows
changes, and the option to enable pugixml is
guarded by a platform if, this can be removed
once all platforms have committed the svn libs.
For details see D8628
By checking if a cell has already been processed in the finding patch
component code, an enormous speedup happens. This only will be
noticeable if there are lots of patches, and some cells with a
large number of patches.
In case where there are coplanar instersections where
each part has a lot of triangles, the finding-cells algorithm was
very inefficient. This uses a Set instead of a Vector to keep track
of a cell's patches, avoids going through all patch x patch combinations,
avoids going through all patches to renumber after a merge, and
merges smaller patch-sixe cells into larger ones.
All this reduces the time to find cells in the cited case by a factor of 10.
Ensure consistent order of pose bones. Now it should always match the
one from bones in armature obdata (as exposed by e.g. RNA, i.e.
children-first).
Previously when some pose bones would need to be added or removed from a
pose due to changes in the bone armature, or if bones in armature were
re-ordered, the bones order in pose would not match anymore the one from
armature, and could even become different between e.g. a proxy and its
linked source.
This was not really nice, but not a big issue before either. But with
diffing process of override, consistent order of items between reference
linked collection and local override one is crucial.
Reviewed By: #animation_rigging, sybren
Maniphest Tasks: T82758
Differential Revision: https://developer.blender.org/D9646
The change was intentional so that the orientation matrices match
(`rv3d->viewinv` becomes equal to the orientation matrix).
But, although in a projection matrix the Z axis is negative, this
should not be so from the user's point of view.
So the solution here is to negate the Z axis when the View orientation
is chosen.
This affects all modes, but is only evident for rotation.
---
Another change here is to use the final rotation value (`values_final`)
for the gizmo drawing since this value can be changed by the mode.
Avoid the evaluation of constraints when computing the parent-inverse
matrix.
Constraints are meant to be evaluated last; object transforms are
computed this order:
1. `parent->obmat` (the parent object's world matrix)
2. `ob->parentinv` (the object's parent-inverse matrix)
3. Object's loc/rot/scale
4. Object's constraint evaluation
When the constraints are used to compute the parent-inverse matrix,
their effect is moved from step 4 to step 2 in this list, potentially
rotating or scaling the object's local transform. This causes unwanted
movement as reported in T82156.
Reviewed By: looch
Differential Revision: https://developer.blender.org/D9413
Move the logic for determining if the item at a given x position is an
icon into the function. This is used for determining selection over an
icon, and will be used in a later commit for checking for hover over an
icon. No functional changes.
This adds support for incremental updates in the sculpt transform
code. Now tools can define if they need the displacement applied
for the original coordinates or incrementally.
This is needed for features like elastic transform or cloth deformation
target in the transform tool.
No functional changes.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D9547
This was reported in the form of the eyedropper of the 'Parent' property
creating a custom property 'parent' if self was picked.
Problem arises when certain checks for setting rna pointer properties
failed (for example: the PROP_ID_SELF_CHECK check) and then a different
code path was entered (which was only meant for IDProperties).
Problem was introduced in rBa7b3047cefcb.
To solve, now first enter the branch for rna-based pointer properties,
then perform the sanity-checks (and if these fail: dont enter the other
unrelated codepath but instead do nothing)
Maniphest Tasks: T83055
Differential Revision: https://developer.blender.org/D9652
dynamic paint modifiers
Looks like prior to the introduction of mantaflow, the former
SmokeModifierData always had a domain initialized (even if its type was
set to None). Since mataflow, the FluidModifierData type needs to be set
to MOD_FLUID_TYPE_DOMAIN (otherwise domain is invalid)
Maniphest Tasks: T82729
Differential Revision: https://developer.blender.org/D9644
Code was actually not applying any override operation over linked data.
Reasonn behind that was that if library file is saved with latest
override applied then this is not needed, since data saved for the
override in the lib file is already up to date.
But this is actually fully breaking in case someone update the lib file
of the lib file, without re-saving the libfile itself.
So now we alwaya apply overrides also on linked data.
Note that this will not fix the case where a resync is needed.
The panel title text intersected any buttons in the header because
the label offset retrieved from the layout code was not scaled by
the block's zoom level. Error in rB0d93bd8d63980.
The transform modifiers are confirmed by releasing any button.
Thus, the operation can be falsely confirmed if the button that launched
the operation is released after the modifier has been activated.
Previously the events that confirmed the modifiers were hardcoded.
An option to fix this would be to add custom confirmation keyitens for
specific modifiers. But this can be a bit confusing and would make the
modal keymap even bigger.
So the solution here is to skip the button that launched the operation
when confirming the modifier.
Orthogonalize the constraint target's matrix before decomposing it into
Euler angles. This removes sheer, and is actually a requirement for
correct decomposition.
It's conceivable that someone has used the incorrect behaviour in a rig.
As the shear caused unpredictable flipping of the constrained object,
this is unlikely.
Reviewed By: angavrilov, sybren
Differential Revision: https://developer.blender.org/D8915
OpenImageIO's ustring makes assumptions about the interals of
std::string which aren't true on macOS/ARM. A patch for OpenImageIO
addresses this for macOS/ARM builds only, at the expense of potentially
duplicated allocations of strings.
The build script for macOS dependencies expects to find bison provided
by Homebrew under /usr/local. Since the default install location for
Homebrew changes on macOS/ARM, the script now looks at the host
architecture and switches to /opt/homebrew on Macs with Apple Silicon.
This update introduces two improvements from the Mantaflow repository:
(1) Improved particle sampling:
- Liquid and secondary particles are sampled more predictably. With all parameters being equal, baked particles will be computed at the exact same position during every bake.
- Before, this was not guaranteed.
(2) Sparse grid caching:
- While saving grid data to disk, grids will from now on be saved in a sparse structure whenever possible (e.g. density, flame but not levelsets).
- With the sparse optimization grid cells with a value under the 'Empty Space' value (already present in domain settings) will not be cached.
- The main benefits of this optimization are: Smaller cache sizes and faster playback of simulation data in the viewport.
- This optimization works 'out-of-the-box'. There is no option in the UI to enable it.
- For now, only smoke simulation grids will take advantage of this optimization.
Previously, the alternate row color in the Video Sequence Editor was
just a shaded version of the editor's background color. This makes it
theme-able just like in the file browser and outliner, although the
default color is very slightly different.
Differential Revision: https://developer.blender.org/D9634
If the cursor is not over the mesh, the operator was still using the
last Face Set ID updated by the drawing cursor code when the cursor was
over the mesh.
This now cancels the operator instead of modifying a Face Set that will
look random to the user.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D9545
The original name was chosen because symmetry was a property of
the Paint struct. Now symmetry is a property of meshes, which
also makes the name of the property easier to understand.
Reviewed By: sergey, dbystedt
Differential Revision: https://developer.blender.org/D9618
Since D6133 fluid particle code uses thread local storage to collect
springs created during a time step before adding them to the actual
spring array.
Prior to the switch to TBB there was a single finalize callback which
was called on the main thread, so it could use psys_sph_flush_springs
and insert the new entries into the final buffer. However in D7394 it
was replaced with a reduce callback, which is supposed to be thread
safe and have no side effects. This means that the only thing it can
safely do is copy entries to the other temporary buffer.
In addition, careful checking reveals that the 'classical' solver
doesn't actually add springs, so reduce isn't needed there.
Differential Revision: https://developer.blender.org/D9632
The roll value may not be as expected when a matrix is not orthogonal
or has a negative determinant.
This can lead to confusion as seen in T82930.
Therefore, make it clear that this is a limitation and that a value for
the roll is somewhat indeterminable in these cases.
This fixes T82930
In `calc_curve_deform` a factor is calculated without checking if
the divisior is zero or close to zero. This patch adds the missing
checks and sets the factor to zero if the division shouldn't be
computed.
Reviewed By: mont29
Differential Revision: https://developer.blender.org/D9645
Blender uses 64bit atomics to manipulate SessionUUID, and these atomics
were not defined on any of 32bit platforms.
While official support is limited to 64bit platforms only, the code
should not make assumptions about bitness or endianess, in terms that
there should be codepaths and fallback (or provision of them) for 32bit
platforms.
This change makes 64bit atomic functions defined for all platforms.
The atomic_test was compiled and successfully tested on i686 and armv7l
platforms. The rest of compilation process of Blender will be very
tedious, so that was not done.
This change is essential, but not necessarily enough to make Blender
compilable on i686 (ability to compile Blender on 32bit platforms was
lost during the 2.91 development).
This is a functional part of original fix done by Brecht in D9577.
There is a special defines block needed for ARM on Linux. Move it from
public header to an implementation file.
No functional changes.
This is a non-functional part of original fix done by Brecht in D9577.
The logic was duplicated.
Should be no functional changes. The modified function is expected
to give same exact results for all inputs.
On the "caching last-used track" topic. The code was using last_marker
to allow faster lookup of marker closest to the frame. With this
change it is still the case since the BKE_tracking_marker_get() does
cache last used marker.
Selecting an object by clicking on its instances only worked,
when the object itself is visible. However, it is possible to hide
the object and still keep the instances visible.
The solution is to give every object the correct `select_id` in the
depsgraph object iterator right before rendering.
Reviewers: fclem, brecht
Differential Revision: https://developer.blender.org/D9640
This change will use the image engine to draw the backdrop of the compositor. With this patch the alpha blending will be done in Linear Scene Reference space and shows pure emissive colors.
See differential for an example image.
**Technical changes**
As only the backdrop drawing is done using the draw manager there are some technical changes.
1. The overlay buffer is partly drawn outside the draw manager. When drawing the backdrop image the overlay buffer needs to be masked to simulate premultiplied alpha under.
2. The backdrop of the node editor is done in region pixel space. A `DRWView` is constructed with this space.
3. UDIM textures uses world position to generate the UV coordinates. This has been implemented more strict by the `IMAGE_DRAW_FLAG_USE_WORLD_POS`. When the flag isn't used the local coordinates are used to generate the UV coordinates what is image space.
4. The draw manager now checks the actual `eSpaceType` of the space data to use different code paths. In the future the movie clip editor will be added.
NOTE: The preview images in nodes are drawn in display space and cannot show pure emissive colors. As preview images are used on more locations it is best to fix this in a separate patch.
Reviewed By: Clément Foucault
Differential Revision: https://developer.blender.org/D9451
The spikes were caused by non-initialized tangent matrix used during
smoothing process. The reason tangent matrix was not initialized was
because wrong usage of API: n-gons should pass corner of 0 to the
matrix construction function.
Corrected usage of the API and added assert() to help catching such
kind of non-initialized issues easier.
When clicking on an already-selected keyframe, mark it as active if the
click caused the previously-active keyframe to become deselected.
When clicking on a key in the graph editor, it is selected and all other
keys are deselected. If that key was already selected before the click,
it would not become the active keyframe. This is now fixed.
Reviewed by: Severin
Differential Revision: https://developer.blender.org/D9639
Regression introduced by {rB042143440d76}. The deform group lookup was
performed on the wrong object. Before the lookup was performed on
the lattice object. This patch changes this back.
Reviewed By: Bastien Montagne
Differential Revision: https://developer.blender.org/D9638
According to feedback the outliner to properties editor tab switching
was annoying when it always changed tabs on selection, especially for
selecting individual objects. This limits the tab switching behavior to
only when the icons in the outliner are selected.
Adds smooth scrolling with the trackpad for popovers. Also fixes the position of the scroll arrows on high-DPI.
Differential Revision: https://developer.blender.org/D9533
Reviewed by Brecht Van Lommel
Layout block safety sizes were not scaled correctly with interface scale.
Differential Revision: https://developer.blender.org/D9569
Reviewed by Brecht Van Lommel
The existing hash function didn't work well with Set's method of
masking to the lower bits, because many verts have zeros in the
lower bits.
Also, replaced VectorSet with Set for Vert deduping.
Recent changes introduced `acc` parameter into the texture read
functions. When nanovdb isn't enabled this leads to compilation errors
as the `acc` variable wasn't defined. OpenCL only compiles needed
features what made it more prominent.
Reviewed By: Patrick Mours
Differential Revision: https://developer.blender.org/D9629
ROCm 3.9 already defined `NULL`. This patch will first check if it was
already defined to remove compilation warnings.
NOTE: This doesn't add official support for ROCm as it still fails to
render correctly (crashes with default cube).
Reviewed By: Brecht van Lommel
Differential Revision: https://developer.blender.org/D9610
Split some of the code of `graph_edit.c` into:
* `graph_view.c`: preview range, view all, view selected etc.
* `graph_slider_ops.c`: the decimate modal operator code.
The latter file will be extended later with more slider-based operators.
Maniphest Tasks: T81785
Reviewed By: sybren
Differential Revision: https://developer.blender.org/D9312
This new discontinuity filter performs actions on the entire Euler
rotation, rather than only on the individual X/Y/Z channels. This makes
it fix a wider range of discontinuities, for example those in T52744.
The filter now runs twice on the selected channels, in this order:
- New: Convert X+Y+Z rotation to matrix, then back to Euler angles.
- Old: Add/remove factors of 360° to minimize jumps.
The messaging is streamlined; it now reports how many channels were
filtered, and only warns (instead of errors) when there was an actual
problem with the selected channels (like selecting three or more
channels, but without X/Y/Z triplet).
A new kernel function `BKE_fcurve_keyframe_move_value_with_handles()` is
introduced, to make it possible to move a keyframe's value and move its
handles at the same time.
Manifest Task: T52744
Reviewed By: looch
Differential Revision: https://developer.blender.org/D9602
The caller is still responsible for allocating list of markers,
but the track allocation and initialization can now be reused.
Currently no functional changes, preparing for an upcoming
development.
Having up to nine levels of indentation make this function hard to
follow. Instead of indenting the rest of the loop for a simple special
case, just continue.
By using floating point filters, the speed improves by a factor of 2 to 10.
This will help speed up some cases of the Exact Boolean modifier.
Changed the interface of mpq2::isect_seg_seg to not return mu, as it was
not needed and not calculating it saved 15% time.
When lowering the wireframe opacity with sculpt overlays enabled, the
wireframe overlay was creating white artifacts along the edges.
Reviewed By: fclem
Differential Revision: https://developer.blender.org/D9607
This adds an invert toggle for the outliner object state filters.
There are some cases where we want a filter for invertable states (Selected,
Unselected) and having a single toggle to invert the filter reduces the
number of separate filter types needed. This removes the "Hidden" filter
which can now be replicated with an inverted "Visible" filter.
Differential Revision: https://developer.blender.org/D9598
In case we do not use names, code adding new insert operations in
collections was broken.
Not a proble in practice so far, since this case was not yet in use, but
will be soon with NLA overrides.
Since add-ons now unregister on exit
(as of fa566157a5)
clearing functions in `bpy.app.handlers` caused an error on exit.
Resolve by restoring handlers before exiting.
When an projecting onto a plane that is orthogonal to the views Z axis,
project onto a view aligned axis then map it back to the original plane.
see: ED_view3d_win_to_3d_on_plane_with_fallback
Since the depth can't be properly visualized in 3D, display a 2D so
it's possible to to tell the depth from the cursor motion.
When using surface orientation and nothing is under the mouse-cursor,
pick the axis that's closest to the view's Z axis.
Now by default default, drawing into empty spaces wont use a plane
orthogonal to the view.
Generating the thumbnail left the view matrices set to values
that couldn't be used for cursor drawing.
Backup/restore the matrices for off-screen drawing.
The Normal Map node was falling back to (0, 0, 0) when it was missing
the required attributes to calculate a new normal.
(0, 0, 0) is not a valid normal and can lead to NaNs when it is
normalized later in the shader. Instead, we now return sd->N,
the unperturbed surface normal.
The issue was that the same point cache was read by multiple
threads at the same time (the same object was evaluated for
render and for the viewport).
Both threads incremented PTCacheMem->cur which lead to the crash.
The fix is to remove the PTCacheMem->cur and store it on the
stack instead. This way every thread has its own cur.
Reviewers: brecht
Differential Revision: https://developer.blender.org/D9606
Cover all atomic functions with unit tests.
The tests are quite simple, nothing special so far. The goal is to:
- Make sure implementation exists.
- Implementation behaves the same way on all platforms
(We had issue when MSVC and GCC were behaving differently in the
past).
- Proper bitness is used for implementation and non-fixed-size
function implementation.
The tests can be extended further to make sure, for example, that
CAS operations do not cast arguments to a more narrow type for
comparison. Considering it a possible further improvement, as it is
better be done being focused on that specific task.
There is an annoying ifdef around 64bit implementation, which uses
same internal ifdef as the header does. This check is aimed to be
removed, so is easier to simply accept such duplication for now.
The tests seems somewhat duplicate for signed/unsigned variants and
things like this. The reason for that is to keep test code as simple
as possible: attempting to do something smart/tricky in the test code
often causes the test code to be a subject of being covered with its
own unit tests.
Differential Revision: https://developer.blender.org/D9590
Previously the lock-free tests were actually testing guarded allocator
because the main entry point of tests was switching allocator to the
guarded one.
There seems to be no allocations happening between the initialization
sequence and the fixture's SetUp(), so easiest seems to be just to
switch to lockfree implementation in the fixture's SetUp().
The test are passing locally, so the "should work" has high chance
of actually being truth :)
Differential Revision: https://developer.blender.org/D9584
Previously the only way to use lockfree implementation was to start
executable and never switch to guarded allocator.
Surely, it is not possible to switch implementation once any allocation
did happen, but some tests are desired to test lock-free implementation
of the allocator. Those tests did not operate properly because the main
entry point of tests are forcing guarded allocator to help catching
bugs.
This change makes it possible for those tests to ensure they do operate
on lock-free implementation.
There is no functional changes here, preparing boilerplate for an
upcoming work on the allocator tests themselves.
Don't refresh the list of sockets, so that when the .blend file is restored the
links remain valid. Also display such nodes in red to indicate an error, same
as when the node type info is missing.
Some DNA headers already did this, most did not. Even though many of them would
be included in C++ files and thus compiled as C++. This would be confusing and
developers may think they have to add `extern "C"` too a whole lot of
(indirect) includes to be able to use a C header in C++.
However, this is a misconception.
`extern "C"` does not cause code to be compiled with C rather than C++! It only
causes the linker to not use C++ function name mangling. See
https://stackoverflow.com/a/1041880.
Because extern DNA headers don't have function declarations, using `extern "C"`
actually should not have any effect. On the other hand, adding it causes no
harm and avoids confusion. So let's just have it consistently in C header
files.
Differential Revision: https://developer.blender.org/D9578
Reviewed by: Bastien Montagne, Sybren Stüvel
Don't allocate StripElem for movieclip, scene and mask strips. This
struct is not handled in seq_dupli function. This caused field to be
uninitialized in COW datablock.
StripElem is not allocated when adding strip with operator and it is
not needed for these strip types.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D9600
Split up the Euler filter function into two more functions. The Euler
filter operator works in two stages (find channels that define the X/Y/Z
Euler rotations, and perform filtering on those channels), and each
stage now has its own function. This makes it clearer which data are
used in which part of the code, and makes future improvements easier.
No functional changes.
This is actually a user interface issue, introduced on 0688309988.
but->tip is checked for NULL in other parts, and rightly so, since here
it crashes Blender.
On a side note I don't know how to reliably reproduce the original bug
whose fix introduced this issue. That got on the way of seeing whether
this is the correct fix, if it happens outside greasepencil operators.
Maybe it is an operator that is missing a tooltip? Impossible to tell
without more information from said commit. That said since this happened
during the end of bcon3 and we are not in bcon4 I'm going ahead with
this NULL check commit.
Patch reviewed by Philipp Oeser and Sergey Sharybin.
Code was rewritten to work on per-sequence basis returning reference to
created strip.
There should be no functional changes.
Selection logic is left as is for now. I could simplify it, but it
belongs to operator, not split logic.
Reviewed By: sergey
Differential Revision: developer.blender.org/D9592
My last cleanup commit for this function missed this case. It likely
happens because the panel's block size doesn't update properly somewhere.
Short of investigating that right now, it makes sense to return early
in this case anyway.
This option joins any stroke with an end near the actual stroke. Now it is not limited to the last stroke, any stroke in the same layer for the actual frame can be joined. The join can join two strokes drawing a third stroke.
If the end and the start of the result stroke are very small, the stroke is changed to be cyclic automatically.
There is a limit distance to join the stroke, if the distance is greater than this value, the strokes are not joined. Actually, a constant, threshold distance is used, but we could expose
as a parameter in the UI in the future.
The tool can be used with freehand drawing or with primitives.
Note: Great part of the patch is just a refactor of the old code to make it accessible and to keep code organized.
Reviewed By: mendio
Maniphest Tasks: T82377
Differential Revision: https://developer.blender.org/D9440
The existing panel drawing function was a bit convoluted with dependent
conditions in different scopes, redundant and unecessary computations,
and un-helpful naming.
This commit separates the function into two parts, the backdrop and the
widgets. It also improves naming and uses const where possible, and in
general cleans up the code.
There are some slight visual changes, mostly with the placement of the
drag icon, which moves a bit downward to be centered with the triangle
icon. The black rectangle displayed while dragging is also removed.
The logic for separator-spacers (used here for right-alignment) didn't take
region scaling into account. Usually that's not an issue because they are
otherwise only used in headers which can't zoom.
The logic for separator-spacers (used here for right-alignment) didn't take
region scaling into account. Usually that's not an issue because they are
otherwise only used in headers which can't zoom.
During some operators like rotate in grease pencil edit mode the
hierarchy lines in the outliner would draw twice as thick. Set the width
before drawing the lines in the outliner.
Differential Revision: https://developer.blender.org/D9589
This is an addendum to previous boolean fix, where the object
transformation was "cleaned". Now the operand one is too.
This fixes the issue shown in the video in T82301 when you move
a column around the XY plane with the top and bottom faces
supposedly coplanar with a cube. The transformation matrix when
you do that has a tiny offset in the z component.
Scaling of forces needs more work. Before making changes to them it would be nice to have a setup, that works physically correct across multiple modifiers (cloth, rigid bodies, fluid).
This will be a to do for 2.92.
Active tile could be NULL when it was on the second tile before
switching back and forth between the Image/UDIM.
In the future we might also check that the active_tile_index is always
valid.
The root cause of this bug is that the function that updates the PBVH
normals is drw_sculpt_generate_calls. As now both the overlays and
mesh can be drawn without using pbvh drawing, the normals were not
updating. This patch forces a normals updates also in the no PBVH
drawing code path of the overlays. This was affecting both shading and
sculpt surface sampling in both flat and smooth shading modes.
Having the sculpt normals being updated by the drawing code is a wrong
design which also causes other issues like:
Brushes that sample the surface and do multiple stroke steps between
redraws will sample invalid normals, creating artifacts during the
stroke clearly visible in some brushes.
Brushes that do not need to sample the surface update the normals on
each redraw. This affects performance a lot as in some cases, updating the
normals takes more time than doing the brush deformation. If flat shading
is being used, this is only necessary to do once after the stroke ends.
Reviewed By: fclem
Differential Revision: https://developer.blender.org/D9535
Pie menu had inconsistent behavior for dragging & releasing the key
compared to clicking on the button.
This was caused by the `onfree` argument being set to true,
preventing the button from running callbacks such as setting
up undo data & auto-keyframe
This argument should only be used when freeing the button,
set this to false as is done for regular menus.
This brush needs to be disabled for dyntopo as it stores its custom data
and deforms from original coordiantes.
Reviewed By: sergey
Maniphest Tasks: T82542
Differential Revision: https://developer.blender.org/D9516
Panels for active uiBlocks always have a type, because the process that
makes them uses the types. Add an assert just to make it clear that the
assumption is purposeful.
Regression introduced by {b17cca6966}. When centralizing the gpu texture
premultiplication setting it was assumed that generated images
(`IMA_TYPE_UV_TEST`) were stored as premultiplied. That assumption was
totally wrong as the alpha association is determined by the existing of
the float/byte buffer.
NOTE: This change will render generated images with pure emissive
colors (show colors when alpha=0.0) what might add more reports. Any
reports could be merged in the next report {T82790}.
Reviewed By: Clément Foucault, Philipp Oeser
Differential Revision: https://developer.blender.org/D9585
After recent changes to the context panel layout (rB187cc5e26d28b1a8),
there has been an error printed when running propery search:
> Error: separator_spacer() not supported in popups.
The layout code thinks it's drawing in a menu because region->visible
isn't properly set for the other tab searches. This patch sets that field
for the temporary searching region, but it also disables searching in the
context breadcrumbs panel, because at best this will just give results
for the names of the active object, etc. This isn't helpful since
those labels are mostly in every tab anyway.
Differential Revision: https://developer.blender.org/D9425
Logic was incorrect, mistake in f3b8792b96.
Updated comment to make intent more clear.
Same as fd78f8699e, but needed conflict resolution as the panel-type
flags where renamed in master.
Edges with 3 or more connected UV's caused UV pack to fail.
Instead of using functions from uvedit_parametrizer.c which are intended
specifically for ABF/LSCM unwrapping, use a simpler method for packing
which stores arrays of BMesh faces.
The clone tool in the image editor can show a second texture on top
of the image. This wasn't ported and now results into alpha and depth
issues. This fix adds the clone tool drawing to the overlay engine.
Reviewed By: Clément Foucault
Differential Revision: https://developer.blender.org/D9352
Remove redundant call to `ofstream::close()` from `~PSStrokeRenderer`
and `~TextStrokeRenderer`. ofstream will be destructed automatically.
- For `~Depsgraph`, `std::function`'s constructor can throw.
- Passing throwing statements in the lambda will not be detected by
clang-tidy.
Fix these issues by using lambda as function argument.
Reviewed By: sergey, sybren
Differential Revision: https://developer.blender.org/D9497
In 8d1978a8e0 bmain entry is updated when prefetching is started,
but this must be done before seq_prefetch_update_context(). Otherwise
created cache keys will be incorrect.
This happens immediately after prefetch is started. When it's refreshed
problem goes away.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D9581
Some RNA structs, like operators or keymaps, are not allowed to have ID
pointer properties. now this check will ignore those, and report an
error message in the console.
Related to T82597.
Notes: While a bit more involved than rBf39fbb3e6046, this commit
remains fairly localized and non-intrusive.
It relies on some rather obscure and weird behaviors of our RNA code
though, a cleaner solution could be e.g. to add a tye to
`StructOrFunctionRNA`, so that we could properly 'rebuild' (re-cast) the
pointer to either `StructRNA` or `FunctionRNA` when needed in internal
code...
When offsetting strips to the left, to make space when syncing strip
length, move the preceding NLA strips instead of the succeeding strips.
The bug seems to be from a copy/pasting typo.
Reviewed By: sybren
Differential Revision: https://developer.blender.org/D9583
Select only editable keys on selected curves when "Only Show Selected
Curves" is enabled in the Graph Editor's View menu.
This prevents selecting invisible keys with box, circle, and lasso
select in the Graph Editor.
Reviewed By: looch
Differential Revision: https://developer.blender.org/D9580
Crash is related to the definition of the GL_MAX_TEXTURE_SIZE. OpenGL does not clearly
defined `GL_MAX_TEXTURE_SIZE` exactly means. Both on AMD and NVIDIA we have issues with
huge textures that they don't get created even if they are smaller. (See {D9530} for
research).
This patch will try to create the texture in a smaller size when the texture creation
failed.
Final implementation by: Clément Foucault
We should create a solution that doesn't need downscaling. For this specific case ARB_sparse_texture might help to create cleaner code, but you still have to commit the whole image what introduces several draw calls. Other improvement is to optimize the scaling; current implementation isn't optimized for performance.
Reviewed By: Clément Foucault
Differential Revision: https://developer.blender.org/D9524
This is was caused by incorrectly set preview_render_size in VSE
rendering context. Value was set to SEQ_PROXY_RENDER_SIZE_FULL, but
it should be SEQ_PROXY_RENDER_SIZE_SCENE as scene render size is
being used.
This is same fix as 0d7036b40e, but I did not checked openGL
render pipeline.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D9562
Use either scene render size or fixed preview scale factor.
Previously scene render size was used as baseline value for text size
correction. This is incorrect.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D9563
The icons originally chosen for the collection colors were selected
during development and had a few issues with contrast in the light
theme, and the gray color was not a good choice against the default gray
backgrounds.
The new colors are more readable in both default Blender themes. Gray
was replaced with pink.
Differential Revision: https://developer.blender.org/D9504
Coalescing on macOS overwrites a singular unprocessed mouse event. To
receive all mouse and tablet events coalescing is disabled.
Disabling coalescing for macOS disables coalescing for trackpad
gestures. Repeat trackpad events are unnecessary and found to
negatively impact performance thus are re-coalesced in Window Manager.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D9574
This was introduced by rBdb7d8281c5a2.
The color needs to be premultiplied as there is no blend mode and
the output color is replacing the framebuffer color & alpha.
The OpenVDB data structure can store voxel data in leaf nodes or tiles
when all the nodes in a given region have a constant value. However,
Cycles is using the leaf nodes to generate the acceleration structure
for computing volume intersections which did not include constant tiles.
To fix this, we simply voxelize all the active tiles prior to generating
the volume bounding mesh. As we are using a MaskGrid, this will not
allocate actual voxel buffers for each leaf, so the memory usage will be
kept low.
Differential Revision: https://developer.blender.org/D9557
Reviewed by: brecht, JacquesLucke
X axis panel dragging traces back to Blender versions before 2.5,
where panels could be aligned horizontally. But for many years now
panels have been vertically aligned. Considering this, keeping the
X axis dragging around is a bit odd. It makes interaction confusing,
or at least more complicated. It also looks bad, since any part of
the panel outside the region is cropped.
Differential Revision: https://developer.blender.org/D9549
This reverts commit f39fbb3e60.
Code is not valid, `DefRNA.laststruct` does not always point to the
proper struct when defined from Python, need to be done differently.
The original code for viewlayer collection flag syncing across moves
from D9158 didn't consider the case where the collection could no longer
be found in its original view layer (moving a collections betwen scenes).
The fix is to just check if the collection starts in the same scene as
it will be moved to before trying to do the flag syncing. I thought about
this for a while and tried a couple other solutions, but I couldn't come
up with a proper way to support syncing the layer collection flags across
scenes without making too many changes.
Differential Revision: https://developer.blender.org/D9568
Root of the issue is that point caches are added to the object's list
on-demand, which often ends up with them being added only during
depsgraph evaluation, i.e. on COW objects.
This could result in having 'orig' data caches with invalid/unset stack
index at some points (e.g. when reading a file and applying
liboverrides), leading to discarding valid existing disk cache files.
Fact that one of those index is signed, and the other not, does not
help...
While this is very weak, fixing broken PointCache code is out of the
scope of a bug fix, so this patch merely:
* Simplifies and factorizes the code generating the 'extension' part of
caches filenames;
* Ensures `BKE_object_insert_ptcache` is called when needed so that we
always have a valid stack index to generate that filename extension.
This is only a bandaid, but it is simple and should be safe enough for
now.
Related to T82503.
Some RNA structs, like operators or keymaps, are not allowed to have ID
pointer properties. now this check will ignore those, and report an
error message in the console.
Related to T82597.
Remove `return` from for-loop which blocked the recalculation of driven
values when it found the first driver.
Reviewed By: sybren, sergey
Differential Revision: https://developer.blender.org/D9515
Previously the return value of `ufopen` wasn't checked and if it failed,
`NULL` was passed into `fclose()` which resulted in a crash. This patch
avoids this by returning from `BLI_gzopen` when the file cannot be created.
Reviewed By: sebbas, iss
Differential Revision: https://developer.blender.org/D9576
is present
Caused by rB4212b6528afb.
'updateGLSLCurveMapping()' compares cacheIDs and in certain scenarios,
these are the same when they should not.
- whenever we had multiple viewports that are colormanaged with
curvemappings this worked right (cacheIDs were different)
- for example, this also worked right when the ImageEditor displays a
Render Result or a Compositor Viewer
- but it worked wrong when the Image Editor displays any other Image (or
no Image at all)
- it also worked right if there were multiple Image Editors [and one of
them displays a Render Result e.g]
Now why is this so?
For comparison, the curve mapping's pointer/address is used.
- update_glsl_display_processor frees the curve_mapping, see
BKE_curvemapping_free(global_glsl_state.curve_mapping)
- similar, update_glsl_display_processor creates a new curvemapping, see
BKE_curvemapping_copy(view_settings->curve_mapping)
- now for the situation that a viewport with curvemapping and a viewport
without curvemapping is present and you make changes to the curvemapping
the following happens:
-- curve_mapping_settings->cache_id is set once [to the memory address
of curvemapping before change]
-- change happens
-- viewport 1 frees curvemapping
-- viewport 2 duplicates using BKE_curvemapping_copy, but this one gets
the same address like before the change
-- this means we have different data on the same address with the same
cacheID...
Solution: to really make the cache ID unique we can combine the pointer
with its 'changed_timestamp' [which increases on every change].
Reviewers: jbakker
Maniphest Tasks: T82460
Differential Revision: https://developer.blender.org/D9559
The issues was that Blender was trying to refresh the
splash screen region. However, opening the file browser
closed the splash screen and freed the region.
The fix is to simply not refresh the region.
Has been approved in T81817.
First detect the other selected object, then check it can be written to.
Otherwise the target object could be the first one found when
looping over objects which is random from the user perspective.
Move the type check to the operator, which also checks the data isn't
library data which was being ignored.
Move most functions into separate files.
File sequencer.c is reserved for alloc, free and duplicating functions.
There should be no functional changes.
Two problems were fixed. One, the code for dissolving vertices
left a face around if dissolving a vertex would leave less than
three vertices. Instead, the face should be deleted.
Two, with transformations like "rotate 180 degrees", this should
be no problem with exact, but the current transformation matrix
has very small non-zero entries where it shouldn't. Cleaning the
transformation matrix makes it more likely that user expectations
about coplanar faces will be fulfilled.
It should not be possible to set the scene collection's color tag
through rna. Also adds a missing notifier for setting the collection
color tag from python.
After rB452a1c7b3838 there were still a few cases where the old
collection icon was used in the interface. Replace these with the new
filled collection icon.
Differential Revision: https://developer.blender.org/D8660
This patch is the result of the GSoC 2020 "Editing Grease Pencil Strokes
Using Curves" project. It adds a submode to greasepencil edit mode that
allows for the transformation of greasepencil strokes using bezier
curves. More information about the project can be found
here: https://wiki.blender.org/wiki/User:Filedescriptor/GSoC_2020.
The overlap with the `Panel` flags that start with "PNL" was quite
confusing because wasn't clear which enum a flag was from. The
new names are a bit longer, but the clarity is worth it.
Instead to use the ID of the object, now the parameter is an Enum with Selected object or New.
If use selected mode, the first grease pencil object selected is used. If none of the selected objects is a grease pencil object, a new object is created.
Small cleanup changes to the original patch.
Differential Revision: https://developer.blender.org/D9529
LibOverrides only support a small sub-set of PointCache features for now
(one cannot add new caches, baking in memory is not supported...).
Part of first step of T82503: support disk cache in liboverrides.
This was done as some sort of safety, but should not actually be needed,
and including tags like `ID_RECALC_POINT_CACHE` e.g. makes usage of
point caches impossible with liboverrides (since it would systematically
invalidate all cache on file load).
In theory we should not have to tag anything here in fact, RNA accessors
are supposed to take care of it, but for now we keep the
`ID_RECALC_COPY_ON_WRITE` one.
Part of first step of T82503: support disk cache in liboverrides.
Note that due to convoluted layout of point caches in RNA (active one
also storing list of all available ones), we'll often have the
pointcache overrides rules twice. Should not be a huge problem,
practically speaking.
Part of first step of T82503: support disk cache in liboverrides.
We already were using one of earlier RC of the library, so there is no
expected big changes. Just making the update official, using official
version and stating it in the readme file.
Becomes rather annoying to duplicate them across C/C++ GCC/Clang sets,
almost as if the test should test both C and C++, and to do it for all
compilers.
Solves strict warning in the upstream of Ceres library.
This reverts commit 9d172f007e.
Got a second thought and remembered why it was not done in the first place.
The issue here is that the server needs to communicate codesign result back
and that must happen within the new protocol. So if the client talks old
protocol it is possible to receieve data from it, but is not possible to
communicate result back to it.
Image format code checked the file type against an enum except for
zero which is used when the format can't be detected.
Also add doc-strings to some of the image file type callbacks.
the function vec_roll_to_mat3_normalized() has a bug as described in T82455. This Differential is only for refactoring the code such that it becomes more clear what the function does and how the bug can be fixed. This differential is supposed to not introduce any functional changes.
Reviewed By: sybren
Differential Revision: https://developer.blender.org/D9410
HDRI preview should have resolution dependent on dpi, viewport scale and HDRI gizmo size.
This patch uses a LOD to render a more round sphere.
Reviewed By: Jeroen Bakker
Differential Revision: https://developer.blender.org/D9382
This would allow python script to access `lineWidth` uniform when drawing lines
without using `glLineWidth`.
Reviewed By: Jeroen Bakker
Differential Revision: https://developer.blender.org/D9518
Standard deviation formula wasn't being applied correctly when selecting
R G B cases. Issue is there since Blender 2.64 as it was incorrectly
ported over from the previous compositor.
Reviewed By: Sergey Sharybin, Jeroen Bakker
Differential Revision: https://developer.blender.org/D9384
This patch adds an opacity slider to the wireframe overlay. The previous
wireframe in dense geometry scenes could be too dark and sometimes the
user just wants an impression of the geometry during modelling.
Reviewed By: Jeroen Bakker
Differential Revision: https://developer.blender.org/D7622
This commit moves some of the logic around so that the logic in
panel_activate_state is clearly separated by the state being activated.
There are fewer nested and redundant checks, and it's easier to see
the progression of interaction with the panel handler.
This callback made some sense before moving the file-type information
from a bit-flag to an enum: e142ae77ca
Since then, we can compare the type value directly.
Also replace loops over file types with IMB_file_type_from_{ibuf/ftype}.
This commit makes the property editor switch to the physics tab instead
of the modifier tab when selecting physics modifiers. Since the modifier
isn't visible then, it's confusing to change the expansion, so this commit
also disables the modifier expansion for these modifiers.
Differential Revision: https://developer.blender.org/D9544
Now that the Blender sync mechanism deletes nodes from the scene, we need to
ensure scene update is stopped before we do this.
Also add some more early out in scene geometry update to ensure we do not
continue working on incomplete geometry data, though that was not the cause of
this crash.
Two issues:
* Automatic deduplication of OpenVDB grid data was failing when Cycles had
already cleared the OpenVDB grid, causing an empty grid. Instead rely on
Blender return the same OpenVDB grid pointer when deduplication is possible.
* The volume bounds mesh was not properly cleared when the OpenVDB grid was
empty, causing a mismatch between mesh and voxel data.
The function vec_roll_to_mat3_normalized() basically has to handle 3 scenarios:
- When a bone is oriented along the negative Y axis
- When a bone is very close to the negative Y axis
- All other cases
The tests in the Differential make sure that all 3 situations are covered.
Reviewed By: sybren, mont29
Differential Revision: https://developer.blender.org/D9525
Only a single DEG operation node `POSE_SPLINE_IK_SOLVER` should
be added in this case [ see `build_splineik_pose`, same is already done
for overlapping IK in `build_ik_pose`]
ref T82347.
Reviewers: sybren
Maniphest Tasks: T82347
Differential Revision: https://developer.blender.org/D9471
This caused warnings when e.g. building the lite profile because NanoVDB was not disabled, but
OpenVDB was. This Fixes this by setting the "WITH_NANOVDB" flag too.
Add a new property `co_ui` to Keyframes, the modification of which will
apply to the keyframe itself as well as its Bézier handles.
Dragging the "Keyframe" slider in the properties panel now maintains the
deltas between the keyframe and its handles, just like moving the key in
the graph editor would.
Reviewed by @sybren in T81813.
This is more in line to other eyedropper usages throughout blender.
Affected operators:
- Sample Dyntopo detail
- Extract Face Set (as reported in T82615)
ref T82615
Maniphest Tasks: T82615
Differential Revision: https://developer.blender.org/D9531
Pass codesign errors (if any) from codesign buildbot server to the
buildbot worker, so that the latter one can abort build process if
the error happens. This solves issues when non-properly-notarized
DMG package gets uploaded to the buildbot website.
When creating an armature from the skin modifier, resulting bones would
always be flagged BONE_CONNECTED.
Those bones cannot be transformed (just rotated).
Now only flag bones that really have a parent BONE_CONNECTED.
Maniphest Tasks: T82624
Differential Revision: https://developer.blender.org/D9534
Issue exposed by rB4c7b1766a7f1.
Main idea is that non-memfile first undo step should check into previous
memfile and tag the ID it is editing as `future_changed`.
That way, when we go back and undo to the memfile, said IDs are properly
detected as changed and re-read from the memfile.
Otherwise, undo system sees them as unchanged, and just re-use the
current data instead.
Note that currently only Sculpt mode seems affected (probably because it
is storing the mode switch itself as a Sculpt undo step instead of a
memfile one), but similar action might be needed in some other cases
too.
Maniphest Tasks: T82388
Differential Revision: https://developer.blender.org/D9510
Pass codesign errors (if any) from codesign buildbot server to the
buildbot worker, so that the latter one can abort build process if
the error happens. This solves issues when non-properly-notarized
DMG package gets uploaded to the buildbot website.
When painting in the image editor on data images (Non-color, Raw) the
color mismatched between the sampled color and the actual effect that
the painting has on the image. The root cause is that the sampling is
color managed, but the painting still uses a fixed color management
pipeline with a lot of assumptions. Due to recent changes the drawing
of the image editor is color managed, but the painting isn't what made
these changes show up.
This patch is a work-a-round so that the sampled colors and the effect
the paint has on the texture matches. This isn't the correct solution
as that would be to migrate all the painting tools to use proper color
management.
Reviewed By: Pablo Dobarro
Differential Revision: https://developer.blender.org/D9411
There are two implementations of the Sample Color operation. One is used
by the paint texture and one by the image editor. The image editor
variant sampled from the ibuf directly, but the paint texture variant
was sampling from the screen front buffer. This can lead into incorrect
samples due to color pipeline.
This patch will use the image editor variant when sampling a color for
2d texture painting
Reviewed By: Pablo Dobarro
Differential Revision: https://developer.blender.org/D9408
Previously the header was a fixed size and assumed to be zeroed.
Now read in bytes up to `HEADER_SIZE`, pass the number or bytes
read to the callback which must not read past those bytes.
We haven't supported 32bit mac builds for a while so this should be safe to remove.
Reviewed By: #platform_macos, brecht
Differential Revision: https://developer.blender.org/D9489
Needed a better normal to for plane to offset into when there are
non in-plane edges between two beveled edges. It was using the vertex
normal, which is just wrong.
Differential Revision: https://developer.blender.org/D9508
These are generally only one or two word phrases and are not sentences.
This change slightly improves readability.
Note, the check when display labels:
```
Tip Label (only for buttons not already showing the label).
```
Could be improved here because there are a lot of false positives.
See https://developer.blender.org/D9499.
It's odd to include a C++ header (".hh") in C code, we should avoid that. All
of the Outliner code should be moved to C++, I don't expect this C header to
stay for long.
See https://developer.blender.org/D9499.
"View" leads to weird names like `TreeViewViewLayer` and after all they are
specific to what we call a "display mode", so "display" is more appropriate.
Also add, update and correct comments.
See https://developer.blender.org/D9499.
* Use C++17 nested namespaces.
* Use `_` suffix rather than prefix for private member variables.
Also: Simplify code visually in `tree_view.cc` with `using namespace`.
See https://developer.blender.org/D9499.
* Turn functions into member functions (makes API for a type more obvious &
local, allows implicitly sharing data through member variables, enables order
independend definition of functions, allows more natural language for
function names because of the obvious context).
* Prefer references over pointers for passing by reference (makes clear that
NULL is not a valid value and that the current scope is not the owner).
* Reduce indentation levels, use `continue` in loops to ensure preconditions
are met.
* Add asserts for sanity checks.
See https://developer.blender.org/D9499.
Also:
* Add `space_outliner/tree/common.cc` for functions shared between display
modes.
* I had to add a cast to `ListBaseWrapper` to make it work with ID lists.
* Cleanup: Remove internal `Tree` alias for `ListBase`. That was more confusing
than helpful.
See https://developer.blender.org/D9499.
* Add comments to explain the design ideas better.
* Follow code style guide for class layout.
* Avoid uninitialized value after construction (general good practice).
See https://developer.blender.org/D9499.
* Use `blender::Map` over `GHash`
* Use `blender::Vector` over allocated `ListBase *`
Benefits:
* Significantly reduces the amount of heap allocations in large trees (e.g.
from O(n) to O(log(n)), where n is number of objects).
* Higher type safety (no `void *`, virtually no casts).
* More optimized (e.g. small buffer optimization).
* More practicable, const-correct APIs with well-defined exception behavior.
Code generally becomes more readable (less lines of code, less boilerplate,
more logic-focused APIs because of greater language flexibility).
See https://developer.blender.org/D9499.
* Turn functions into member functions (makes API for a type more obvious &
local, allows implicitly sharing data through member variables, enables order
independend definition of functions, allows more natural language for
function names because of the obvious context).
* Move important variables to classes rather than passing around all the time
(shorter, more task-focused code, localizes important data names).
* Add helper class for adding object children sub-trees (smaller, more focused
units are easier to reason about, have higher coherence, better testability,
can manage own resources easily with RAII).
* Use C++ iterators over C-macros (arguably more readable, less macros is
generally preferred)
* Add doxygen groups (visually emphasizes the coherence of code sections,
provide place for higher level comments on sections).
* Prefer references over pointers for passing by reference (makes clear that
NULL is not a valid value and that the current scope is not the owner).
This introduces a new C++ abstraction "tree-display" (in this commit named
tree-view, renamed in a followup) to help constructing and managing the tree
for the different display types (View Layer, Scene, Blender file, etc.).
See https://developer.blender.org/D9499 for more context. Other developers
approved this rather significantly different design approach there.
----
Motivation
General problems with current design:
* The Outliner tree building code is messy and hard to follow.
* Hard-coded display mode checks are scattered over many places.
* Data is passed around in rather unsafe ways (e.g. lots of `void *`).
* There are no individually testable units.
* Data-structure use is inefficient.
The current Outliner code needs quite some untangling, the tree building seems
like a good place to start. This and the followup commits tackle that.
----
Design Idea
Idea is to have an abstract base class (`AbstractTreeDisplay`), and then
sub-classes with the implementation for each display type (e.g.
`TreeDisplayViewLayer`, `TreeDisplayDataAPI`, etc). The tree-display is kept
alive until tree-rebuild as runtime data of the space, so that further queries
based on the display type can be executed (e.g. "does the display support
selection syncing?", "does it support restriction toggle columns?", etc.).
New files are in a new `space_outliner/tree` sub-directory.
With the new design, display modes become proper units, making them more
maintainable, safer and testable. It should also be easier now to add new
display modes.
2020-11-11 18:51:57 +01:00
957 changed files with 34685 additions and 16722 deletions
// Assign the values of the computed covariance using the
// inverse permutation used in the QR factorization.
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
