Made PropertyDefinition::m_name private.
Moved code around for a cleaner diff in MaterialTypeSourceData.h.
Added API comments.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
I removed support for property rename version updates in MaterialTypeSourceData (i.e. ApplyPropertyRenames) because MaterialSourceData serialization no longer loads material property definitions from the .materialtype file, per a recent change on the development branch. The unit tests were broken and it wasn't worth updating them since we don't need this functionality anymore. Material property renames and other version updates are now exclusively applied by the MaterialAsset class.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
There were lots of material system conflicts that had to be resolved. I expect the build is broken at this commit, and I'll fix it in followup commits.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
Here the class has been generalized for a list of group names and a final property name, rather than assuming a single group containing the property. This included removing the unused GetPropertyName and GetGroupName functions. All that's really need from this class is conversion to a full property ID string.
Testing:
New unit test.
Reprocessed all core material types and StandardPBR test materials used in Atom Sample Viewer's material screenshot test.
Atom Sample Viewer material screenshot test script.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
These changes have the added benefit of simplifying some of the serialization code. MaterialSourceDataSerializer is no longer needed, as its main purpose was to pass the MaterialTypeSourceData down to the MaterialPropertyValueSerializer.
Before, the JSON serialization system gave a lot of data flexibility because it did best-effort conversions, like allowing a float to be loaded as an int for example. But now the material serialization code doesn't know target data type, so it has to assume the data type based on what's in the .material file, and then the MaterialAsset will convert the data to the appropriate type later when Finalize() is called.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
I found a mistake where MaterialAssetCreator needs to clear the raw data when configured to finalize the material asset.
Since MaterialSourceData no longer relies on the material type source file at all, I was able to change MaterialSourceDataTest to avoid saving the source data to disk.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
I made MaterialAsset::Finalize private so I could add some parameters specifically for MaterialAssetCreator to use. Now MaterialAssetCreator::Begin has an option to finalize the material or not.
Moved MaterialAssetCreatorCommon::ValidateDataType to MaterialPropertyDescriptor as "ValidateMaterialPropertyDataType" so that MaterialAsset::Finalize could use it too
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
Before, the material builder was loading the MaterialTypeAsset and doing some processing with it, but was avoiding declaring job dependencies that would cause reprocessing lots of assets when a shader or .materialtype file changes. Reading the asset data isn't safe when not declaring a job dependency (or when declaring a weak job dependency like OrderOnce which is the case here). This caused to several known bugs.
The main change here is it no longer loads the MaterialTypeAsset at all; all other changes flow from there.
The biggest changes (when deferred material processing is enabled) are ...
1) MaterialSourceData no longer loads MaterialTypeAsset. All it really needs is to determine whether a string is an image file reference or an enum value, which is easy to do by just looking for the "." for the extension.
2) MaterialAssetCreator no longer produces a finalized material asset. It no longer uses MaterialAssetCreatorCommon because that only produces a non-finalized MaterialAsset, which has very different needs for the SetPropertyValue function. (We could consider merging MaterialAssetCreatorCommon into MaterialTypeAssetCreator since that's the only subclass at this point). And it doesn't do any validation against the properties layout since that can be done at runtime.
3) Moved processing of enum property values from MaterialSourceData to MaterialAsset::Finalize (this was the only thing being done in the builder that actually needed to read the material type asset data).
Also...
- Updated the MaterialAsset class mostly to clarify and formalize the two different modes it can be in: whether it is finalized or not.
- Merged the separate "IncludeMaterialPropertyNames" registry settings from MaterialConverterSystemComponent and MaterialBuilder into one "FinalizeMaterialAssets" setting used for both.
- Removed MaterialSourceData::ApplyVersionUpdates. Now the flow of data is the same regardless of whether the materials are finalized by the AP or at runtime. Version updates are always applied on the MaterialAsset.
- Added a validation check to MaterialTypeAssetCreator ensuring that once a property is renamed, the old name can never be used again for a new property. This assumption was already made previously, but not formalized, in that Material::FindPropertyIndex does not expect every caller to provide a version number for the material property name, also the material asset's list of raw property names was never versioned. The only way for this to be a safe assumption is to prevent reuse of old names.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
I actually tried this approach before and it didn't seem to work the way we needed, but I realized that's because PropertyAssetCtrl wasn't handling missing assets properly. I fixed a few issues there including showing the error button when the asset can't be found, and fixing a broken reference to the error icon file.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
The main reason for this is to give consistent results between the AP and Material Editor, where a placholder texture can be used if a texture is missing. Otherwise, you could get a placeholder texture in Material Editor and stale data in the runtime; this inconsistency would be confusing.
As a consequence, it is possible for example that the user could mess up the name of a property in a .material file and not notice the problem because it is now a warning instead of an error. If warnings-as-errors is desirable, you can enable the new "/O3DE/Atom/RPI/MaterialBuilder/WarningsAsErrors" registry setting.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
Material Editor also warns the user when saving a material that is populated with fallback image references.
Factored out the path strings for the default images to ImateSystemInterface.h.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
Made it a warning instead of an error when MaterialAssetCreator can't find a texture.
Updated the Material Editor's MaterialDocument class to not elevate warnings to errors.
Material Editor uses new features in TraceRecorder to show a message dialog when warnings are detected so the user is notified of the missing texture.
Next I will work on making MaterialAssetCreator put a "missing" texture in place of the requested one.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
Changed material document to load a source data are for the parent material as well. It was also a previously loading the parent material products asset which would be out of date compared to the source data.
Changed material document to track source file dependency changes instead of product asset changes.
Fixed a bug or copy paste error in the document manager that was using the same container to track documents the modified externally and from other dependency changes.
Returning source data dependencies when creating a material asset from source.
Signed-off-by: Guthrie Adams <guthadam@amazon.com>
This RenderStates is used to override the values in the final draw packet, if the values are valid; it's supposed to be initialized to invalid values, but it wasn't. So the depth compare function was getting set to Less instead of GreaterEqual. This wasn't a problem when using serialized assets from disk, because the deserialization uses the default constructor which did initialize m_renderStatesOverlay. No all Item constructors initialize m_renderStatesOverlay.
Signed-off-by: santorac <55155825+santorac@users.noreply.github.com>
Signed-off-by: Guthrie Adams <guthadam@amazon.com>
santorac
puvvadar
Guthrie Adams