CHANGELOG.md

Changelog / release notes

WebGPU and wgpu-native are still changing fast, and with that we do to. We do not yet attempt to make things backwards compatible. Instead we try to be precise about tracking changes to the public API.

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

Possible sections in each release:

• Added: for new features.
• Changed: for changes in existing functionality.
• Deprecated: for soon-to-be removed features.
• Removed: for now removed features.
• Fixed: for any bug fixes.
• Security: in case of vulnerabilities.

[v0.19.3] - 10-12-2024

Fixed:

• Ensure that wgpu is compatible both with imgui 1.6.0 and older by @hmaarrfk in #649
• Clean up async code and add proper trio / rendercanvas support by @fyellin in #631
• Remove timeout in awaitable by @almarklein in #651
• Allow rendercanvas.get_context() before a backend is selected by @almarklein in #652

[v0.19.2] - 25-11-2024

Changed:

• Update to Imgui 1.6+ by @panxinmiao in #645

[v0.19.1] - 19-11-2024

Some internal refactoring, fix the doc theme, and compatibility with rendercanvas.

Changed:

• Do not confound STDERR and STDOUT by @hmaarrfk in #638
• More consolidation by @fyellin in #641
• Use rtd theme for docs by @almarklein in #639
• Refactor present-method mechanic by @almarklein in #642

[v0.19.0] - 04-11-2024

Added:

• Overrideable constants by @fyellin in #579
• Implemention for multi-draw features by @fyellin in #583
• Statistics query by @fyellin in #605
• Examples for asyncio and trio by @almarklein in #608
• Add example that uses PySide6 with asyncio compat by @almarklein in #612
• Add pre-commit hooks for ruff by @claydugo in #629
• Include pre-commit in optional dependencies by @claydugo in #630

Changed:

• Async API by @almarklein in #598
  • Renamed request_adapter() to request_adapter_sync() (old method still works with warning).
  • Renamed enumerate_adapters() to enumerate_adapters_sync() (old method still works with warning).
  • Renamed request_device() to request_device_sync() (old method still works with warning).
  • Renamed buffer.map() to buffer.map_sync() (old method still works with warning).
• Renamed request_device_trace -> request_device by @fyellin in #589
• Make depth_stencil_attachment follow the spec by @fyellin in #611
• If data in create_buffer_with_data isn't a multiple of 4, just round up. by @fyellin in #626
• Drop support for Python 3.8.

Fixed:

• Fix snake case for 1d/2d/3d suffix by @almarklein in #617

Docs:

• Add note on examples by @almarklein in #597
• Tweaks to examples by @almarklein in #610
• Add extra version info on dev installs by @almarklein in #623
• Fix small typo in README intro by @otterbotter in #625

Internal changes:

• Remove _release code that's duplicated 19 times. by @fyellin in #590
• Fix some small errors by @fyellin in #591
• Refactor build system by @almarklein in #596
• Use ruff instead of black and flake8 by @almarklein in #599
• Fixes to testing by @almarklein in #604
• Remove stuff that's been deprecated for a while by @almarklein in #607
• Improve generated type hints and defaults by @almarklein in #606
• Add better async support for wgpu-native by @almarklein in #609
• Nitpicky changes flagged by PyCharm by @fyellin in #613
• Python 3.13. Remove ruff flakiness by @fyellin in #614
• Structure checking by @fyellin in #615
• Eliminate warning in test suite by @fyellin in #621
• Fix install of mesa drivers by @almarklein in #622
• Only keep alive those objects that are necessary. by @fyellin in #628
• Refactor canvas context to allow presenting as image by @almarklein in #586
• Prevent extra errors when canvascontext creation fails by @almarklein in #635

[v0.18.1] - 17-09-2024

Added:

• Add support for push constants (a wgpu native extension).

Fixed:

• Fixed jupyter gui backend.

[v0.18.0] - 16-09-2024

Added:

• Now also building wheels for Windows aarch64.

Changed:

• Updated to latest wgpu-native (v22.1.0.5).

Fixed:

• Various typos in the docs.
• Fixed the wheels for 32bit Windows.

[v0.17.3] - 10-09-2024

Added:

• Support for occlusion queries.
• Support for render bundles.
• Support for IMGUI, via wgpu.utils.imgui.
• Wx is now a fully supported GUI backend.
• A BaseEnum class was added to wgpu.utils, so it can be used in downstream libs like pygfx.
• The WGPUCanvas.add_event_handler() method now has an order arg.

Changed:

• The flags and enums are implemented using a new enum class, enabling better static code analysis (i.e. autocompletion in IDE's).
• Native (desktop) features must now be specified in the same way as normal (WebGPU) features: lowercase and with hyphens between the words.
• Bindings can omit offset and size (the full size will be used). This makes our API follow WebGPU better.
• Support omitting fields from BindGroupLayoutEntry, BufferBindingLayout, SamplerBindingLayout, StorageTextureBindingLayout, VertexState. See #534 for details.
• In cases where a view_dimension is given, it must be provided as a string (e.g. '2d'). Ints are no longer allowed, because e.g. 2 does not mean '2d', which can be a source of confusion.

(Due to problems with the CD process, we had to bump the version a few times.)

[v0.16.0] - 13-06-2024

Changed:

• Updated to wgpu-native 0.19.4.1. Does not incur API changes.
• Updated to latest IDL.
• Replaced the adapter.request_adapter_info() method with the adapter.info property.

Changed:

• The Adapter.request_adapter_info() method is replaced by the .info property.

[v0.15.3] - 12-06-2024

Added:

• Implement canvas.set_title() by @almarklein in #508
• Add experimental support for get_mapped_range() by @almarklein in #522

Changed:

• Allow create_render_pipeline() to use AutoLayout by @fyellin in #500
• Support omitting the topology field in the primitive dict passed to create_render_pipeline() by @fyellin in #503
• Support omitting last arguments in set_bind_group() by @panxinmiao in #511
• Try to optimize proxy_func by @almarklein in #507

Fixed:

• Fix wx examples by @panxinmiao in #520
• Support timestamp queryset for begin_render_pass() by @almarklein in #505
• Implement destroy() the right way by @almarklein in #519

[v0.15.2] - 10-05-2024

Added:

• New example to demonstrate manual GLFW integration by @Korijn in #480
• More details in docs of enumerate_adapters by @almarklein in #489
• Add GPU cloud compute info to docs by @kushalkolar in #495

Changed:

• Make modifiers and buttons tuples for immutability by @hmaarrfk in #492

Fixed:

• Fix WxWgpuWindow by not calling 'add_event_handler' by @cansik in #486

[v0.15.1] - 28-03-2024

Added:

• Users can set the WGPU_GUI_BACKEND environment variable to prefer a specific backend.
• Added wgpu.gpu.enumerate_adapters().
• Added adapter.summary property, to easily get a one-line description of each adapter.

Changed:

• Avoid present-related warning messages when an exception is raised from the draw function.
• The adapter.request_adapter_info() now also contains vendor_id and device_id.

Fixed:

• For GUI backend selection:
  • Honor %gui in IPython and Jupyter.
  • Don't enter Qt event loop when Qt is active by the env (e.g. IPython).
  • Prefer glfw when we detect a running asyncio loop.
  • Make the glfw backend work in IPython.
  • The run() function can be called multiple times (similar to plt.plot()).

[v0.15.0] - 07-03-2024

Changed:

• Updated to wgpu-native 0.19.3.1. Does not incur API changes.
• Updated to latest IDL. Does not incur API changes.

Fixed:

• Wayland is finally properly supported, albeit via XWayland.

Added:

• Tests for detecting nan and inf values in shaders.

[v0.14.1] - 15-02-2024

Added:

• The wgpu.DiagnosticsBase is now a public base class that can be subclassed by downstream libs (like pygfx) to provide diagnostics.

Changed:

• Diagnostics renders bools as ✓ or -, and large ints get scientific notation.

[v0.14.0] - 13-02-2024

Added:

• The feature "float32-filterable" is now available natively.
• Add list of projects using wgpu-py to README. (#456)

Changed:

• Updated to wgpu-native 0.19.1.1. (#458)
• Canvas.get_context() now raises an error if no backend is selected yet (instead of returning a worthless base context object).
• Omitting the call to context.get_current_texture() results in a warning instead of an exception.

Removed:

• Shadertoy util is removed from the wgpu-py. It is now available as a separate package: wgpu-shadertoy. (#455)

Fixed:

• Devices no longer leak memory.

[v0.13.2] - 21-12-2023

Added:

• Implement support for timestamp QuerySet.
• Add texture input and iFrameRate builtin to Shadertoy util #453

[v0.13.1] - 08-12-2023

Fixed:

• Prevent access violation errors with GLFW on Windows.
• Prevent a segfault when deleting a GPUPipelineLayout (observed in a very specific use-case on LavaPipe).
• Fix triangle_glsl.py example.
• Fix that when logger is set to debug, errors are produced when Python exits.

Added:

• Support for linux-aarch64 (binary wheels available)! This includes Raspberry Pi's with a 64-bit OS, and adds support for building linux docker images on Apple Silicon devices without having to emulate x86 (no need for --platform linux/amd64).

[v0.13.0] - 24-11-2023

Added:

• Add iDate builtin to Shadertoy utility.
• Allow "auto" layout args for create_compute_pipeline().
• Official support for Python 3.12 and pypy.

Changed:

• Update to wgpu-native 0.18.1.2.
• CanvasContext.get_current_texture() now returns a GPUTexture instead of a GPUTextureView.
• OffscreenCanvasBase.present() now receives a GPUTexture instead of a GPUTextureView, and this is a new texture on each draw (no re-use).
• Renamed ``wgpu.gui.WgpuOffscreenCanvastoWgpuOffscreenCanvasBase`.
• The wgpu.base submodule that defined the GPU classes is renamed to be a private module. The new wgpu.classes namespace contains all GPU classes (and nothing else).
• The __repr__ of the GPU classes shows a shorter canonical class name.
• Flags and Enums have a more useful __repr__.

Fixed:

• Dragging a window between windows with different scale factor (with Qt on Windows) no longer puts the window in an invalid state. A warning is still produced though.
• GPUCommandEncoder.begin_render_pass() binds the lifetime of passed texture views to the returned render pass object to prevent premature destruction when no reference to a texture view is kept.

[v0.12.0] - 15-11-2023

This is a big release that contains many improvements, but also multiple API changes.

Most backward incompatible changes are due to two things: the backend system has been refactored, making it simpler and future-proof. And we have revised the buffer mapping API, making it more similar to the WebGPU spec, and providing more flexible and performant ways to set buffer data.

A summary to help you update your code:

# X import wgpu.backends.rs
import wgpu


# X wgpu.request_adapter(canvas=None, power_preference="high-performance")
wgpu.gpu.request_adapter(power_preference="high-performance")

# X buffer.map_read()
buffer.map("READ")
buffer.read_mapped(...)
buffer.read_mapped(...)
buffer.unmap()

# X buffer.map_write()
buffer.map("WRITE")
buffer.write_mapped(data1, ...)
buffer.write_mapped(data2, ...)
buffer.unmap()

Added:

• The wgpu.gpu object, which represents the API entrypoint. This makes the API more clear and more similar to the WebGPU API.
• A convenience auto backend, and a stub js_webgpu backend.
• New function enumerate_adapters() in the wgpu_native backend.
• Warning about pip when wgpu-native binary is missing on Linux
• The GPUBuffer has new methods map(), map_async(), unmap(). These have been part of the WebGPU spec for a long time, but we had an alternative API, until now.
• The GPUBuffer has new methods read_mapped() and write_mapped(). These are not present in the WebGPU spec; they are the Pythonic alternative to getMappedRange().
• Flags can now be passed as strings, and can even be combined using "MAP_READ|COPY_DIST".
• GUI events have an extra "timestamp" field, and wheel events an additional "buttons" field.
• A diagnostics subsystem that amongst other things counts GPU objects. Try e.g. wgpu.diagnostics.print_report().
• Several improvements to the shadertoy util: offscreen support and a snapshot method.

Changed:

• Can create a buffer that is initially mapped: device.create_buffer(..., mapped_at_creation=True) is enabled again.
• The wgpu.request_adapter() function is moved to wgpu.gpu.request_adapter(). Same for the async version.
• The canvas argument of the request_adapter() function is now optional.
• The rs backend is renamed to wgpu_native.
• It is no longer necessary to explicitly import the backend.
• The GPUDevice.request_device_tracing() method is now a function in the wgpu_native backend.
• We no longer force using Vulkan on Windows. For now wgpu-native still prefers Vulkan over D3D12.
• The wgpu.utils subpackage is imported by default, but most submodules are not. This means that compute_with_buffers must be explicitly imported from wgpu.utils.compute.

Deprecated:

• wgpu.request_adapter() and its async version. Use wgpu.gpu.request_adapter() instead.
• The GPUBuffer methods map_read()and map_write() are deprecated, in favor of map(), unmap(), read_mapped() and write_mapped().

To be clear, these are not changed:

• The convenient device.create_buffer_with_data() (not part of the WebGPU spec) is still available.
• The GPUQueue.read_buffer() and GPUQueue.write_buffer() methods are unchanged.

Fixed:

• The shaderutil now re-uses the default device, avoiding memoryleaks when running multiple consecutively.
• The GUI backend selection takes into account whether a backend module is already imported.
• The offscreen GUI backend no longer uses asyncio (it does not need an event loop).
• Prevent a few classes of memoryleaks. Mind that creating many GPUDevice objects still leaks.

[v0.11.0] - 11-10-2023

Changed:

• Update to wgpu-native 0.17.2.1. No changes are needed in downstream code.

[v0.10.0] - 09-10-2023

In this release the API is aligned with the latest webgpu.idl, and we updated to wgpu-native (v0.17.0.2).

Added:

• New wgpu.wgsl_language_features property, which for now always returns an empty set.
• The GPUShaderModule.compilation_info property (and its async version) are replaced with a get_compilation_info() method.
• The WebGPU features "bgra8unorm-storage" and "float32-filterable" are now available.

Changed:

• The binary wheels are now based on manylinux 2.28, and the 32bit Linux wheels are no longer built.
• In WGSL: toplevel constants must be defined using const, using let will now fail.
• In WGSL: it is no longer possible to re-declare an existing variable name.
• Error messages may look a bit different, since wgpu-native now produces nice messages replacing our custom ones.
• Errors produced by a call into a wgpu-native function now produce a Python exception (no more async logging of errors).

[v0.9.5] - 02-10-2023

Fixed:

• Fixed setting the dpi awareness in the Qt backend, by correctly looking up the Qt version.

Changed:

• Links to readthedocs now point to stable instead of latest, so that people reading the docs see these that reflect the latest release.
• Don't enable any features by default (previously WGPUNativeFeature_TEXTURE_ADAPTER_SPECIFIC_FORMAT_FEATURES was enabled).

[v0.9.4] - 23-02-2023

Fixed:

• Fixed issue related to winid (native widgets) on embedded Qt widgets on Windows (#348).
• Fixed our example screenshot tests.

[v0.9.3] - 20-02-2023

Changed:

• The offscreen WgpuCanvas.draw() method now returns a memoryview instead of a numpy array.
• The shadertoy util changed internally from using numpy to using a memoryview.

[v0.9.2] - 17-02-2023

Fixed:

• Fixed that get_preferred_format() could crash (in wgpuSurfaceGetSupportedFormats) due to an upstream bug in wgpu-native (#342)

Added:

• The shadertoy util now supports GLSL, so code from the shadertoy website can be directly copied and run with wgpu (#343)

[v0.9.1] - 13-02-2023

Changed:

• Improved documentation.

Added:

• Added print_report() to get a report on the internals of wgpu.
• Added command_encoder.clear_buffer()
• Added support for GLSL.

[v0.9.0] - 25-01-2023

In this release the API is aligned with the latest webgpu.idl, and we updated to the latest release of wgpu-native (v0.14.2.3).

Changed:

• To use the default min_binding_size in create_bind_group_layout, it should be None instead of zero.
• If the depth-stencil texture has not room for stencil data, the stencil_read_mask and stencil_write_mask fields in the DepthStencilState struct passed to create_render_pipeline() must be set to 0.
• In WGSL, @stage(compute) must now be @compute. Same for vertex and fragment.
• In WGSL, the list of reserved words has been extended, including e.g. mod, matrix and ref.
• In WGSL, smoothStep is now smoothstep.

Added:

• New IDL: texture has new props weight, height, depth_or_array_layers.
• New IDL: Buffer has new prop map_state.

[v0.8.4] - 10-01-2023

Fixed:

• The offscreen canvas's mainloop prevents leaking callbacks better (#322)
• Prevent error messages when Qt examples/apps are closed (#326)

[v0.8.3] - 06-01-2023

Fixed:

• Prevent Qt warning about setting dpi awareness (#320)
• Make canvases close when they get deleted (#319)
• Fix qt canvas in ipython (#315)
• Make offscreen canvas closable (#309)
• Fix that the offscreen canvas had it size hardcoded, ignoring the given size (#317)
• Fixed renaming of queue in docs (#308)
• Fix using .draw_frame on qt canvas (#304)
• Add missing dev dependencies (#295)

Added:

• A shadertoy utility, plus examples (#312)

Changed:

• Improve the error prompt when wgsl code is multi line error (#311, #316)
• Tests: execute examples in the test process (#310)
• Package only the release binary (not the debug build) (#299)
• Codegen: uses in-memory file system during code generation (#303)
• Improve readme (#290, #302, #314)

[v0.8.2] - 06-10-2022

Fixed:

• Fixed imports for PyQt6.
• Keyboard events work again for Qt 6.3.
• Fixed that overloading handle_event() did not work for a canvas based on a Qt or wx main widget/window.

Added:

• Can now add a wildcard ("*") to add_event_handler to handle all events.
• Shader error messages show more context, making shader debugging much easier.
• VSync can now be turned off to raise the frame rate when needed. Note that FPS measurements are still a poor performance benchmark!

Changed:

• GLFW canvas does not draw when minimized anymore.
• The offscreen and Jupyter canvas now use the srgb format for consistency with normal canvases.
• The examples have been adjusted for srgb colors.

[v0.8.1] - 29-04-2022

Fixed:

• Fixed regression that canvas.handle_event() could no longer be overloaded to handle move and wheel events.

Changed:

• Added a note in the docs to explain that the version of the examples must match the runtime version.

[v0.8.0] - 20-04-2022

Changed:

• Now targeting wgpu-native 0.12.0.1.
• Updated API to the latest WebGPU spec.
• Better error logging using the new callbacks in wgpu-native.
• All destructors (drop methods) are now working as they should.

To update, you need to adjust to the following API changes:

• The encoder's end_pass() are renamed to end().
• The compute encoder's dispatch() is renamed dispatch_workgroups, and dispatch_indirect to dispatch_workgroups_indirect.
• The load_value is replaced with clear_value and load_op.
• Same for depth_load_value and stencil_load_value.
• The device.create_sampler() method for mipmap filtering now uses the MipmapFilterMode enum instead of the FilterMode enum. Since the fields of these enums are the same, you probably don't need to change anything.

To update, your shaders need the following changes:

• The decorators have changed from [[...]] to @....
  • E.g. change [[location(0)]] to @location(0).
  • E.g. change [[group(0), binding(0)]] to @group(0) @binding(0).
• Structs now use , to separate fields instead of ;.
• The elseif keyword is now else if.
• Buffers bound as arrays don't need to be defined via a struct anymore.

[v0.7.7] - 12-04-2022

Fixed:

• Fixed that event handlers could not be added while in an event handler.
• Prevent swap chain errors when minimizing a window.

Added:

• The QWgpuWidget now also supports the autogui events.
• Our CI now tests the examples (including image comparisons).

[v0.7.6] - 28-03-2022

Changed:

• Pointer move and wheel events are now rate-limited, leading to better performance if e.g. picking is done at each event.

Added:

• Added wgpu.gui.base.log_exception, a context-manager to catch and log exceptions, e.g. in event callbacks.

[v0.7.5] - 17-03-2022

Fixed:

• Mouse down events were not emitted during double clicks in the Qt canvas.
• Mouse move events were not emitted no button is pressed in the Qt canvas.

[v0.7.4] - 04-02-2022

Fixed:

• Position of glfw pointer events on MacOS.

[v0.7.3] - 25-01-2022

Added:

• Expanded the auto gui backend, which can now also select qt framework if available.
• The qt gui backend (like the glfw gui backend) supports user events in the same manner as the jupyter gui backend.
• Expanded the auto gui backend to also support an offscreen canvas intended for automated tests.

Fixed:

• Size of glfw windows on MacOS.

[v0.7.2] - 24-12-2021

Fixed:

• Exceptions in user-interaction callbacks don't break the glfw loop anymore.
• Pointer events in glfw have the correct key modifiers now.

[v0.7.1] - 22-12-2021

Added:

• #224 - Added add_event_handler and remove_event_handler to GLFW and Jupyter GUI canvases.

[v0.7.0] - 21-12-2021

Changed:

• Now targeting wgpu-native v0.11.0.1, containing many upstream fixes and improvements.
• The [[block]] syntax in shaders has been dropped.
• Renamed ProgrammableStage.clamp_depth -> unclipped_depth.

[v0.6.0] - 16-12-2021

Added:

• Official support for Windows 3.10.
• The max_fps argument can be provided to a canvas.
• The glfw gui backend supports user events in the same manner as the jupyter gui backend, using the jupyter_rfb event specification.
• Introduce the auto gui backend, which selects either glfw or jupyter.

Fixed:

• The wx gui backend is now fully functional.

Changed:

• The qt and wx gui backend now contain WgpuCanvas for a toplevel window, and WgpuWidget for an embeddable widget.
• All gui backends (can) now limit the FPS.
• No changes to the wgpu API.

[v0.5.9] - 11-10-2021

Fixed:

• Include the correct binaries in macOS arm64 wheels
• Options for arch argument of download-wgpu-native.py script

[v0.5.8] - 09-10-2021

Added:

• Experimental support for macos_arm64 (M1).

Changed:

• The Qt examples use PySide6 instead of PyQt5.

[v0.5.7] - 07-10-2021

Changed:

• Update to the latest wgpu-native (including latest Naga).
• The power-preference is actually taken into account.
• The adapter actually reports its limits.
• The limits in request_device are actually used.
• The Adapter.is_software property is renamed to Adapter.is_fallback_adapter.

[v0.5.6] - 30-08-2021

Added:

• An offscreen canvas to take snapshots without needing a window.

Changed:

• On Windows, the Vulkan backend is now forced unless WGPU_BACKEND_TYPE is set.

Fixed:

• Better support for multiple canvases by fixing a specific Qt issue.
• Fixed that canvas was not passed to low level function of request_adapter.
• Support calling get_current_texture() multiple times during a draw.

[v0.5.5] - 09-08-2021

Added:

• The wgpu backend can be forced using the WGPU_BACKEND_TYPE env variable. Values can be e.g. "D3D12", "Metal", "Vulkan".
• Initial support for off-screen canvases.
• Adds adapter.is_software property.

Changed:

• The GPUPresentationContext class has been renamed to GPUCanvasContext.
• The functionality of the swap-chain has moved to the GPUCanvasContext.
• The now removed GPUSwapChain was used as a context manager. Instead, the frame is presented (ala GL swapbuffers) automatically at the end of a draw.
• The canvas.configure_swap_chain() method has been removed. Instead, canvas.get_context() should be used, to obtain a present/canvas context.
• The adapter.request_device() method has its arguments non_guaranteed_features and non_guaranteed_limits replaced with required_features and required_limits.
• The enum field StoreOp.clear is now StoreOp.discard.
• The flag field TextureUsage.SAMPLED is now TextureUsage.TEXTURE_BINDING .
• The flag field TextureUsage.STORAGE is now TextureUsage.STORAGE_BINDING .
• The enum InputStepMode is now VertexStepMode.
• WGSL: arrays must be declared as var (not let) in order to allow dynamic indexing.
• WGSL: storage classes are written differently.

[v0.5.4] - 11-06-2021

Changed:

• The backend selection is automatic by default. To force a backend, the WGPU_BACKEND_TYPE environment variable can be set to e.g. "Vulkan". It could be good to do this on Windows to prevent selection of DX12 for now.

[v0.5.3] - 04-06-2021

Added:

• adapter.properties now has actual values, allowing inspection of the selected GPU and backend.
• Added back support for filtering float32 textures by enabling a certain wgpu feature by default.

Fixed:

• An error in the docs of create_render_pipeline.
• Vulkan backend is now forced to prevent DX12 being select and causing probems because it's less mature.

[v0.5.2] - 23-05-2021

This release uses a new version of wgpu-native which has changed quite a bit internally. There is more validation (thus more restrictions). There are only a few changes to the API. However, one big change is that shaders can now be provided as both SpirV and WGSL. Due to the strict validation, most shaders compiled by PyShader are not usable anymore. We recommend using WGSL instead.

Added:

• Added GPUAdapter.properties (the amount of information it contains will increase in the future).
• Added proper support for WGSL.

Changed:

• Renamed renderpass.set_blend_color -> set_blend_constant.
• Stricter validation of SpirV shaders.
• Float32 texture formats must now use a non-filtering sampler and texture-sample-type.
• Integer texture formats can no longer use a texture (use textureLoad instead).
• ... and more tighter restrictions.

Removed:

• The API concerning debug markers and groups is temporarily removed.
• Adapter and device features is temporarily removed.
• Adapter and device limits is temporarily removed.

[v0.4] - 21-05-2021

This release represents about half a year of progress on the WebGPU API, so the API has changed quite a bit. The wgpu-py API more closely reflects the webgpu API - wgpu-native does not affect the API except for a few additional features.

Added:

• Added GPUQueue.read_buffer as extra API (next to write_buffer which is original WebGPU API).
• Added GPUQueue.read_texture as extra API.

y Removed:

• Removed GPUBuffer.read_data(). Use device.queue.read_buffer() instead. Note that usage MAP_READ should be replaced with COPY_SRC.
• Removed GPUBuffer.write_data(). Use device.queue.write_buffer() instead. Note that usage MAP_WRITE should be replaced with COPY_DST.

Changed:

• GPUCanvasContext.get_swap_chain_preferred_format(): now takes an adapter instead of a device.
• GPUAdapter.extensions: is now called features.
• GPUAdapter.request_device(): the extensions and limit args are now non_guaranteed_features and non_guaranteed_limits.
• GPUDevice.default_queue: is now called queue.
• GPUDevice.create_compute_pipeline(): the compute_stage arg is now called compute.
• GPUDevice.create_bind_group_layout() has changed the required structure of the layout entry dicts.
• GPUDevice.create_render_pipeline() has changed a lot in terms of shape of input dicts. See new docs.
• GPUTexture.create_view(): args mip_level_count and array_layer_count are default None instead of 0.
• GPUCommandEncoder.begin_render_pass(): the color_attachments and depth_stencil_attachment arguments have their attachment field renamed to view.
• GPURenderEncoderBase.set_index_buffer() has an extra argument (after the buffer) to specify the format. The index format is no longer specified in device.create_render_pipeline().
• Flag TextureUsage has field OUTPUT_ATTACHMENT renamed to RENDER_ATTACHMENT.
• Enum BindingType is split up in different enums for buffer, sampler, sampled texture and storage texture.
• Enum BlendFactor has some of its field names changed.
• Enum VertexFormat has its field names changed, e.g. ushort2 -> uint16x2.
• The API is more restrictive in the use of buffer/texture usage combinations.
• The API is more restrictive in formats for storage buffers/textures.
• When copying from/to textures, the bytes_per_row must now be a multiple of 256.

[v0.3.0] - 2020-07-05

With this update we're using a later release of wgpu-native, and follow changes is the WebGPU spec. Further, we've removed the need for ctypes to communicate data arrays. Instead, wgpu-py can consume any object that supports the buffer protocol, and it returns memoryview objects.

Added:

• The texture object has more properties to query the parameters that it was created with.
• The texture view object has a texture property.
• The render and compute pipeline objects have a property layout and a method get_bind_group_layout().
• The shader object got a compilation_info method, but this does not do anything yet.
• The create_shader_module() has a source_map attribute, but this is yet unused.
• Log messages from wgpu-native (Rust) are now injected into Python's logger.
• The queue object got two new methods write_buffer and write_texture.
• The buffer has read_data() and write_data() methods. Note: the latter may be removed later.
• The device create_buffer_with_data is added as a convenience function. This will likely stay.

Changed:

• Targets wgpu-native v.0.5.2. The first release build from the wgpu-native repo itself.
• The array_layer in copy operations involving a texture is removed.
• The utils.compute_with_buffers function now accepts any data dat supports the buffer protocol (not just ctypes arrays). The outputs are memoryview objects, which shape and format can be specified. When a ctypes array type is specified, the output will be an instance of that type. This means that these changes are fully backwards compatible.

Removed:

• The buffer (for now) no longer exposes a data mapping API. Instead use read_data() and write_data().
• The device create_buffer_mapped method is similarly removed. Use create_buffer_with_data instead.

[v0.2.0] - 2020-04-16

Added:

• The canvas now has a request_draw method.
• More and better docs.
• The canvas can be passed to request_adapter so that the created surface can be selected on it.
  • Support for debug markers.

Changed:

• Targets wgpu-native v0.5.1. This is the last release when wgpu-native was still part of wgpu-core.
• The bindings in bind groups and bind group layouts are now called entries.
• There is no more generic storage texture, only a readonly and a writeonly one.
• The set_index_buffer and set_vertex_buffer methods got a size argument.
• The row_pitch and image_height args in copy operations involving a texture are renamed to bytes_per_row and rows_per_image.
• Rendering is now done under the swap_chain's context: with swap_chain as current_texture_view

[v0.1.6] - 2020-04-01

This release is the first moderately mature version of wgpu-py.