CHANGELOG.md

Changelog

All notable changes to Xentica project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

Unreleased

0.2.0 - 2019-06-29

Changed

• Core.
  • Disable nvcc compiler warnings.
• Exceptions.
  • Raise exception when assigning to DeferredExpression.
  • Raise exception if Variable has no initial value.
• QA.
  • Disable faulty pylint issues.
  • Raise max locals limit for pylint.
  • Remove PyPy3 from supported versions.
• Refactoring.
  • Change Constant API to rid of eval.
  • Move DeferredExpression to separate module.
  • Move code generation to CellularAutomaton.__init__
  • Change Constant defining mechanism.

Added

• Core.
  • Add TotalisticRuleProperty.
  • Add RandomProperty.
  • Add FloatVariable.
  • Add long int to supported types.
  • Add BscaDetectorMixin to Property.
  • Add Parameter class.
  • Integrate parameters to core engine.
  • Add meta-params population from Experiment class.
  • Implement interactive parameters.
  • Add validation for field's dimensionality.
• Seeds.
  • Add ChainedPattern functionality.
  • Add support for RandInt in mixed expressions.
  • Add option to RandInt making it constant.
• Tools.
  • Add xentica.tools package.
  • Add color conversion helpers module (tools.color).
    • Basic HSV to RGB conversion.
    • Genome coloring: positional and modular methods.
  • Add math functions wrappers (tools.xmath)
    • Min / max functions over a number of arguments.
    • Float / int type casting.
    • Counting a number of non-zero bits in integer (popc).
  • Add CA rules helpers (tools.rules).
    • Convert Golly nonation to integer.
    • Convert integer to Golly notation.
  • Add genome manipulation helpers (tools.genetics).
    • Crossover several genomes in stochastic way.
    • Mutation during genomes' crossover.
• Examples.
  • Add EvoLife example model.
  • Add NoiseTV example model.
• Tests.
  • Implement full test suite with tox.
  • Add test for broad bit width.
  • Add test for RendererPlain class.
  • Add test for illegal assign to DeferredExpression.
  • Add test for Variable without init value.
  • Add tests for Lifelike rules helpers.
  • Add test for code property.
  • Add tests for core.variables coverage.
  • Add tests for tools.xmath module.
• Documentation.
  • Add separate section for testing.
  • Add core.expressions to docs.
• Miscellaneous.
  • Add NoiseTV and EvoLife to benchmark.

Fixed

• Major fixes.
  • Fix augmented assigns to properties and variables.
  • Fix direct assignments to variables.
  • Fix variables names in declarations.
  • Fix variable's fallback name.
  • Fix BscaDetectorMixin.
  • Fix ColorEffect.
  • Fix dtype/ctype usage in CellularAutomaton class.
• Minor fixes.
  • Remove unnecessary conditions.
  • Fix default field size.
  • Fix minor codestyle issues.
  • Fix typos and grammar in docs.
  • Fix imports in examples.

0.1.1 - 2018-07-24

Changed

• Optimization.
  • Optimize BigBang/PrimordialSoup seed generation.
• Examples.
  • Add Shifting Sands example.
• Refactoring.
  • Make some private properties public.
  • Rename LocalRandom streams.
  • Add properties getter to ContainerProperty.
  • Move GPU stuff to separate classes.
  • Decompose BSCA class creation into smaller methods.
  • Move core.topology.mixins to core.mixins.

Added

• QA.
  • Add pylint config.
  • Add pylint run script.
• Tests.
  • Add test for non-uniform buffer interactions.
  • Add test for unset viewport.

Fixed

• Major fixes.
  • Fix direct buffer assignments.
  • Fix property assignments.
  • Fix direct Variable assignments.
• Minor fixes.
  • Fix Lattice.index_to_coord()
  • Fix Experiment.seed.generate() arguments.
  • Fix RandInt to generate NumPy arrays.
• Refactoring.
  • Fix most of pylint issues through the code.
• Tests.
  • Fix BigBang/PrimordialSoup tests.
  • Fix checksums in tests.
• Miscellaneous.
  • Fix typos in docstrings.

0.1.0 - 2018-05-05

Added

• Base functionality.
  • Translate Python functions describing CA logic into CUDA C.
  • Evolve CA step by step, using resulting GPU kernels.
  • Render the current CA state into 2D NumPy array for subsequent visualization.
  • Save and load current CA state.
• Multi-dimensional CA topologies.
  • N-dimensional orthogonal lattice.
  • N-dimensional cases of Moore and Von Neumann neighborhoods.
  • Wrap the board into N-torus.
  • Surround the board by static cells.
• CA inner logic.
  • Multiple integer 'properties' per each cell state.
  • Buffered interaction between neighbors.
  • Lazy read access and deferred write access to GPU memory.
  • Packing and unpacking cell's state into raw bit sequence.
  • Mixed values in expressions with variables/properties.
• Rendering.
  • Computed colors per each cell.
  • 'Moving average' color effect, for smoother visualization.
  • Projection into 2D plain from higher dimensions.
• Experiments.
  • Experiment classes describing initial state and meta-parameters.
  • Independent deterministic RNG stream per each CA.
  • 'Big Bang' and 'Primordial Soup' initial random patterns.
• Bridge with Moire GUI.
• Tests.
  • 100% code coverage by unittests.
  • Performance check script.
• Documentation.
  • Throughout API documentation via docstrings.
  • Sphinx configuration to automatically build and publish at Read The Docs.
  • Project overview.
  • Brief installation instructions.
  • Placeholder for tutorial.
• Miscellaneous.
  • MIT License.
  • Readme.
  • Contributing Guidelines.
  • Code of Conduct
  • Changelog.