bump zarr conventions; migrate type checking to pyright

.github/workflows/ci.yml

@@ -10,19 +10,25 @@ permissions:
   contents: read
 
 jobs:
-  pre-commit:
+  lint:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@df4cb1c069e1874edd31b4311f1884172cec0e10 # v6.0.3
-      with:
-        persist-credentials: false
-    - uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6
-      with:
-        python-version: '3.12'
-    - name: Install pre-commit
-      run: pip install pre-commit
-    - name: Run pre-commit
-      run: pre-commit run --all-files
+      - uses: actions/checkout@df4cb1c069e1874edd31b4311f1884172cec0e10 # v6.0.3
+        with:
+          persist-credentials: false
+      - name: Set up Python
+        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
+        with:
+          python-version: "3.12"
+      - name: Install uv
+        uses: astral-sh/setup-uv@fac544c07dec837d0ccb6301d7b5580bf5edae39 # v8.2.0
+        with:
+          enable-cache: true
+      - name: Install dependencies
+        # The pyright pre-commit hook runs `uv run --frozen pyright`, so the
+        # project environment must be present.
+        run: uv sync --group dev --group test
+      - uses: j178/prek-action@bdca6f102f98e2b4c7029491a53dfd366469e33d # v2.0.4
 
   test:
     runs-on: ${{ matrix.os }}

.pre-commit-config.yaml

@@ -11,13 +11,12 @@ repos:
         args: ["--fix", "--show-fixes"]
       - id: ruff-format
 
-  - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.17.0
+  - repo: local
     hooks:
-      - id: mypy
-        language_version: python
-        exclude: tests/.*
-        additional_dependencies:
-        - types-attrs
-        - typing-extensions>=4.15.0
-        - pydantic>=2.12
+      - id: pyright
+        name: pyright
+        language: system
+        entry: uv run --frozen pyright
+        pass_filenames: false
+        always_run: true
+        types_or: [python, pyi]

pyproject.toml

@@ -36,7 +36,7 @@ dependencies = [
     "rioxarray>=0.13.0",
     "cf-xarray>=0.8.0",
     "typing-extensions>=4.15.0",
-    "zarr-cm>=0.2.0",
+    "zarr-cm>=0.4.1",
     "aiohttp>=3.14.0",
     "s3fs>=2024.6.0",
     "boto3>=1.34.0",
@@ -48,7 +48,7 @@ dependencies = [
 dev = [
     "pytest>=7.0.0",
     "pytest-cov>=4.0.0",
-    "mypy>=1.0.0",
+    "pyright>=1.1.390",
     "pre-commit>=3.0.0",
     "bandit[toml]>=1.7.0",
 ]
@@ -149,40 +149,19 @@ ignore = [
     "TRY003",  # Long exception messages outside class - common pattern
 ]
 
-[tool.mypy]
-python_version = "3.12"
-warn_return_any = true
-warn_unused_configs = true
-disallow_untyped_defs = true
-disallow_incomplete_defs = true
-check_untyped_defs = true
-disallow_untyped_decorators = true
-no_implicit_optional = true
-warn_redundant_casts = true
-warn_unused_ignores = true
-warn_no_return = true
-warn_unreachable = true
-strict_equality = true
-plugins = ["pydantic.mypy"]
-
-[tool.pydantic-mypy]
-init_forbid_extra = true
-init_typed = true
-warn_required_dynamic_aliases = true
-warn_untyped_fields = true
-
-[[tool.mypy.overrides]]
-module = ["zarr.*", "xarray.*", "rioxarray.*", "cf_xarray.*", "dask.*"]
-ignore_missing_imports = true
-
-[[tool.mypy.overrides]]
-module = [
-    "eopf_geozarr.data_api.s1",
-    "eopf_geozarr.data_api.s2",
-    "eopf_geozarr.data_api.geozarr.v2",
-    "eopf_geozarr.data_api.geozarr.store",
-]
-disable_error_code = ["valid-type"]
+[tool.pyright]
+include = ["src", "tests"]
+pythonVersion = "3.12"
+typeCheckingMode = "standard"
+# Several runtime deps ship no type stubs; we can't fix their types here, so
+# don't report missing stubs/sources for them. (Imports still resolve because
+# the packages are installed in the environment.)
+reportMissingTypeStubs = false
+reportMissingModuleSource = false
+# Match the strictness we relied on under mypy.
+reportUnnecessaryTypeIgnoreComment = true
+reportReturnType = "error"
+reportUnnecessaryCast = "error"
 
 [tool.pytest.ini_options]
 minversion = "7.0"

src/eopf_geozarr/cli.py

@@ -354,7 +354,7 @@ def format_data_vars(data_vars: dict[str, xr.DataArray]) -> str:
             # Fallback to simple format if xarray HTML fails
             vars_html = []
             for name, var in data_vars.items():
-                dims_str = format_dimensions(dict(zip(var.dims, var.shape, strict=True)))
+                dims_str = format_dimensions(dict(zip(map(str, var.dims), var.shape, strict=True)))
                 dtype_str = str(var.dtype)
                 vars_html.append(
                     f"""
@@ -450,7 +450,7 @@ def render_node(node: xr.DataTree, path: str = "", level: int = 0) -> str:
                 <div class="tree-section">
                     <h4 class="section-title">Variables</h4>
                     <div class="tree-variables">
-                        {format_data_vars(node.ds.data_vars)}
+                        {format_data_vars({str(k): v for k, v in node.ds.data_vars.items()})}
                     </div>
                 </div>
             """

src/eopf_geozarr/conversion/fs_utils.py

@@ -3,7 +3,7 @@
 import json
 import os
 from collections.abc import Mapping
-from typing import TYPE_CHECKING, Any
+from typing import TYPE_CHECKING, Any, Final, Literal, cast
 from urllib.parse import urlparse
 
 import s3fs
@@ -16,6 +16,8 @@
 if TYPE_CHECKING:
     import xarray as xr
 
+ZarrOpenMode = Literal["r", "r+", "a", "w", "w-"]
+
 _MISSING = object()  # sentinel for missing optional attrs
 
 
@@ -52,18 +54,44 @@ def replace_json_invalid_floats(obj: object) -> object:
     return obj
 
 
+def _sanitize_attrs(attrs: Mapping[str, object]) -> dict[str, object]:
+    """Sanitize an attributes mapping, returning a typed ``dict``.
+
+    Wraps :func:`replace_json_invalid_floats` (which is typed ``object -> object``)
+    and verifies the dict-in/dict-out invariant at runtime instead of casting.
+    """
+    sanitized = replace_json_invalid_floats(dict(attrs))
+    if not isinstance(sanitized, dict):  # pragma: no cover - invariant guard
+        raise TypeError(f"expected a dict after sanitizing attrs, got {type(sanitized).__name__}")
+    return sanitized
+
+
+_ZARR_MODES: Final = ("r", "r+", "a", "w", "w-")
+
+
+def _zarr_mode(mode: str) -> ZarrOpenMode:
+    """Validate *mode* against zarr's accepted access modes and narrow its type.
+
+    Checks the value at runtime instead of casting a bare ``str`` to the
+    ``Literal`` zarr expects.
+    """
+    if mode not in _ZARR_MODES:
+        raise ValueError(f"Invalid zarr access mode {mode!r}; expected one of {_ZARR_MODES}")
+    return mode
+
+
 class NanCompatibleJSONEncoder(json.JSONEncoder):
     """
     Custom JSON encoder that converts NaN, Inf, -Inf values to JSON-safe equivalents
     to ensure valid JSON output.
     """
 
-    def encode(self, obj: Any) -> str:
+    def encode(self, o: Any) -> str:
         """
         Encode object to JSON string, converting NaN values to "NaN".
         """
 
-        converted_obj = replace_json_invalid_floats(obj)
+        converted_obj = replace_json_invalid_floats(o)
         return super().encode(converted_obj)
 
 
@@ -87,22 +115,22 @@ def sanitize_dataset_attributes(ds: "xr.Dataset") -> "xr.Dataset":
     ds_clean = ds.copy()
 
     # Sanitize dataset attributes
-    ds_clean.attrs = replace_json_invalid_floats(ds_clean.attrs)
+    ds_clean.attrs = _sanitize_attrs(ds_clean.attrs)
 
     # Sanitize variable attributes
     for var_name in ds_clean.data_vars:
         var = ds_clean[var_name]
         # Preserve _FillValue as-is — xarray encodes it via FillValueCoder on write;
         # converting np.nan to the string "NaN" would break FillValueCoder.encode.
         fill_value = var.attrs.get("_FillValue", _MISSING)
-        var.attrs = replace_json_invalid_floats(var.attrs)
+        var.attrs = _sanitize_attrs(var.attrs)
         if fill_value is not _MISSING:
             var.attrs["_FillValue"] = fill_value
 
     # Sanitize coordinate attributes
     for coord_name in ds_clean.coords:
         coord = ds_clean[coord_name]
-        coord.attrs = replace_json_invalid_floats(coord.attrs)
+        coord.attrs = _sanitize_attrs(coord.attrs)
 
     return ds_clean
 
@@ -405,7 +433,12 @@ def open_s3_zarr_group(s3_path: str, mode: str = "r", **s3_kwargs: Any) -> zarr.
         Zarr group
     """
     storage_options = get_s3_storage_options(s3_path, **s3_kwargs)
-    return zarr.open_group(s3_path, mode=mode, zarr_format=3, storage_options=storage_options)
+    return zarr.open_group(
+        s3_path,
+        mode=_zarr_mode(mode),
+        zarr_format=3,
+        storage_options=cast("dict[str, object]", storage_options),
+    )
 
 
 def get_s3_credentials_info() -> S3Credentials:
@@ -589,4 +622,9 @@ def open_zarr_group(path: str, mode: str = "r", **kwargs: Any) -> zarr.Group:
         Zarr group
     """
     storage_options = get_storage_options(path, **kwargs)
-    return zarr.open_group(path, mode=mode, zarr_format=3, storage_options=storage_options)
+    return zarr.open_group(
+        path,
+        mode=_zarr_mode(mode),
+        zarr_format=3,
+        storage_options=cast("dict[str, object] | None", storage_options),
+    )