Move user-facing README (installation, usage) into onnxruntime_ep_webgpu/ so it is bundled in the wheel and shown on PyPI. Add developer-facing README in plugin-ep-webgpu/python/ with build and test instructions.

The package has no CPython extension modules, only pre-built native libraries, so a single wheel works across all Python versions. Override bdist_wheel.get_tag() to produce py3-none-{platform} instead of cp3XX-cp3XX-{platform}.

…version stamp

- Build wheel in a temporary directory instead of mutating the source tree

- Copy only the files needed (pyproject.toml, setup.py, onnxruntime_ep_webgpu/) instead of using an exclude list

- Change version placeholder to VERSION_PLACEHOLDER and fail hard if not found

- Disable CPU EP fallback in test to ensure WebGPU EP runs the model

- Simplify docstring and README descriptions

- macOS/Windows: add setup-build-tools.yml to Python Package and Test jobs

- Linux: run Python packaging and testing inside Docker for manylinux compatibility and auditwheel support

- Windows: skip Python package/test jobs for arm64 (cross-compiled, can't run on x64 agents)

- Linux: add gpu_machine_pool parameter for test job pool

…h executable

Print environment info (Python version, platform, ORT version, relevant env vars), package directory contents, library file size, device enumeration details, session providers, and full tracebacks on failure.

Apple Silicon requires all executable code to be signed. Without this, dlopen triggers a SIGBUS (bus error) when loading the unsigned dylib.

ESRP requires .zip or .dmg input. Zip the dylib before signing, then unzip the signed result and verify.

The Docker image does not have pip pre-installed. Use ensurepip to bootstrap it before installing wheel build dependencies.

Use python -u in all three platform pipelines so prints are flushed immediately, even if the process crashes during native DLL load. Add ort.set_default_logger_severity(0) in the test script for verbose ORT logging.

Remove reinterpret_cast of OrtKernelInfo* to internal OpKernelInfo* that breaks ABI across DLL boundaries (vtable mismatch between plugin EP and ORT core).

- KernelInfoCache: use Ort::ConstKernelInfo::GetEp() instead of casting to OpKernelInfo* and calling GetExecutionProvider()->GetOrtEp()

- GetAllocator: use C API KernelInfoGetAllocator + IAllocatorImplWrappingOrtAllocator instead of casting to OpKernelInfo*

- Remove #include core/framework/op_kernel_info.h (no longer needed)

- Add #include core/session/allocator_adapters.h for IAllocatorImplWrappingOrtAllocator

…ging_python_fix

…p_adapter_cast_issue

…ast_issue' into edgchen1/webgpu_packaging_python_fix

Materialize the glob generator to a list so the emptiness check works, and delete each raw wheel after auditwheel repair so only the manylinux wheel remains.

Add the missing set-nightly-build-option-variable-step.yml template to all three platform Python_Package jobs for consistency with the Build jobs.

Materialize the glob generator to a list so the emptiness check works, and delete each raw wheel after auditwheel repair so only the manylinux wheel remains.

Add the missing set-nightly-build-option-variable-step.yml template to all three platform Python_Package jobs for consistency with the Build jobs.

…ging_python_fix

…stage.

…ging_python

- build_wheel.py: materialize glob() generators with list() so empty-wheel checks in collect_wheels() and auditwheel_repair() actually trigger.

- plugin-linux-webgpu-stage.yml: remove duplicate setup-feeds-and-python-steps.yml include.

- test_webgpu_plugin_ep.py: have create_mul_model() write into a caller-provided directory and use TemporaryDirectory() for cleanup.

- Move the minimum onnxruntime version into plugin-ep-webgpu/MIN_ONNXRUNTIME_VERSION so future plugin EP packages can reuse it.

- Convert pyproject.toml into pyproject.toml.in with @var@ template markers.

- Add a small gen_file_from_template() helper in build_wheel.py (modeled on tools/ci_build/github/apple/package_assembly_utils.py) and use it to render pyproject.toml at build time.

- Replace ad-hoc print+sys.exit(1) error paths with raised exceptions for consistency.

- Make auditwheel_repair fail loudly if there is no wheel to repair or the repair produces no output.

- gen_file_from_template: docstring caveat that values are inserted verbatim and the caller is responsible for any required quoting/escaping.

- plugin-ep-webgpu/python/README.md: note that pip install / pip wheel against this directory is unsupported (source has pyproject.toml.in, not pyproject.toml).

- WebGPU Dockerfile: SwiftShader pin comment now points at Dawn's DEPS (third_party/swiftshader) and notes the SHA tracks the Dawn commit pinned in cmake/deps.txt, with a refresh URL template.

Add an epVersionFile pipeline variable in plugin-cuda-pipeline.yml (set to 'VERSION_NUMBER') and pass it as version_file through plugin-cuda-packaging-stage.yml down to plugin-linux-cuda-stage.yml and plugin-win-cuda-stage.yml, mirroring the pattern used by the WebGPU plugin pipeline.