chore: bump (#93)

- bump deps (incl. torchcodec)
- pre-commit -> prek

Author: egorchakov

.github/workflows/ci.yaml

@@ -17,19 +17,19 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: setup ssh
-        uses: webfactory/ssh-agent@v0.9.0
+        uses: webfactory/ssh-agent@v0.9.1
         with:
           ssh-private-key: ${{ secrets.YAAK_IDL_REPO_SSH_KEY }}
 
       - name: checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           submodules: recursive
           persist-credentials: false
           lfs: true
 
       - name: setup just
-        uses: extractions/setup-just@v2
+        uses: extractions/setup-just@v3
 
       - name: setup uv
         uses: astral-sh/setup-uv@v7
@@ -50,6 +50,6 @@ jobs:
           only: ytt
 
       - name: install ffmpeg
-        run: sudo apt update && sudo apt install -y ffmpeg
+        run: sudo apt install -y ffmpeg
 
       - run: just test

.github/workflows/release.yaml

@@ -21,12 +21,12 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: setup ssh
-        uses: webfactory/ssh-agent@v0.9.0
+        uses: webfactory/ssh-agent@v0.9.1
         with:
           ssh-private-key: ${{ secrets.YAAK_IDL_REPO_SSH_KEY }}
 
       - name: checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           submodules: recursive
           persist-credentials: false
@@ -44,7 +44,7 @@ jobs:
     needs: build
 
     steps:
-      - uses: actions/download-artifact@v4
+      - uses: actions/download-artifact@v6
         with:
           name: Packages
           path: dist

.pre-commit-config.yaml

@@ -1,5 +1,10 @@
 ---
 repos:
+  - repo: builtin
+    hooks:
+      - id: trailing-whitespace
+      - id: check-added-large-files
+
   - repo: https://github.com/abravalheri/validate-pyproject
     rev: v0.24.1
     hooks:
@@ -11,24 +16,14 @@ repos:
       - id: pyupgrade
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.14.5
+    rev: v0.14.7
     hooks:
       - id: ruff-check
         args: [--fix]
       - id: ruff-format
 
-  - repo: local
-    # TODO: replace once available
-    hooks:
-      - id: ty
-        name: ty check
-        entry: uv run ty check
-        language: system
-        always_run: true
-        pass_filenames: false
-
   - repo: https://github.com/tombi-toml/tombi-pre-commit
-    rev: v0.6.52
+    rev: v0.7.0
     hooks:
       - id: tombi-format
       - id: tombi-lint
@@ -38,8 +33,28 @@ repos:
     hooks:
       - id: nbstripout
 
+  - repo: https://github.com/codespell-project/codespell
+    rev: v2.4.1
+    hooks:
+      - id: codespell
+
+  - repo: https://github.com/hukkin/mdformat
+    rev: 1.0.0
+    hooks:
+      - id: mdformat
+        additional_dependencies:
+          - mdformat-gfm
+
   - repo: local
     hooks:
+      # TODO: replace once available
+      - id: ty
+        name: ty check
+        entry: uv run ty check
+        language: system
+        always_run: true
+        pass_filenames: false
+
       - id: just-format
         name: just-format
         language: system

LICENSE.txt

@@ -71,4 +71,3 @@ distributed under the License is distributed on an "AS IS" BASIS,
 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License.
-

flake.lock

@@ -35,9 +35,7 @@ class BuildYaakIdlProtosHook(BuildHookInterface):
     @override
     def clean(self, versions: list[str]) -> None:
         for path in self._get_yaak_idl_proto_paths():
-            if path.exists():
-                logger.warning("removing %s", path)
-                path.unlink()
+            path.unlink(missing_ok=True)
 
     @override
     def initialize(self, version: str, build_data: dict[str, Any]) -> None:
@@ -78,4 +76,4 @@ def _build_yaak_idl_protos(cls) -> None:
     @staticmethod
     def _overwrite_callback(file: Path, text: str) -> None:
         logger.warning("overwriting %s", file)
-        _ = file.write_text(text, encoding="utf-8")
+        file.write_text(text, encoding="utf-8")

hatch_build.py

@@ -17,7 +17,7 @@ sync:
 setup: sync
     git submodule update --init --recursive --force --remote
     git lfs pull
-    uvx --with=pre-commit-uv pre-commit install --install-hooks
+    uvx prek install
 
 build:
     uv build
@@ -27,8 +27,8 @@ check:
     uv run ruff check
     uv run ty check
 
-pre-commit *ARGS: build
-    uvx --with=pre-commit-uv pre-commit run --all-files --color=always {{ ARGS }}
+prek *ARGS: build
+    uvx prek --all-files {{ ARGS }}
 
 generate-config:
     ytt --file {{ justfile_directory() }}/config/_templates \

justfile

@@ -1,6 +1,6 @@
 [project]
 name = "rbyte"
-version = "0.34.1"
+version = "0.34.2"
 description = "Multimodal PyTorch dataset library"
 readme = "README.md"
 requires-python = ">=3.12,<3.13"
@@ -24,10 +24,10 @@ dependencies = [
   "makefun>=1.16.0",
   "more-itertools>=10.8.0",
   "numpy",
-  "optree>=0.17.0",
-  "pipefunc[autodoc]>=0.87.1",
+  "optree>=0.18.0",
+  "pipefunc[autodoc]>=0.89.0",
   "polars==1.34.0",  # https://github.com/pola-rs/polars/issues/25567
-  "pydantic>=2.12.4",
+  "pydantic>=2.12.5",
   "structlog>=25.4.0",
   "tensordict-nightly==2025.9.24",  # https://github.com/pytorch/tensordict/issues/1440
   "torch",
@@ -45,12 +45,12 @@ rbyte-visualize = "rbyte.scripts.visualize:main"
 
 [project.optional-dependencies]
 build = ["hatchling>=1.27.0"]
-geo = ["polars-st==0.4.0"]
+geo = ["polars-st==0.4.3"]
 hdf5 = ["h5py>=3.14.0"]
-jpeg = ["simplejpeg>=1.8.2"]
+jpeg = ["simplejpeg>=1.9.0"]
 mcap = ["mcap>=1.3.0", "mcap-protobuf-support>=0.5.3", "protobuf"]
 protos = ["grpcio-tools==1.71.0", "protoletariat>=3.3.10"]
-video = ["torchcodec>=0.7.0"]
+video = ["torchcodec>=0.8.1"]
 visualize = [
   "einops>=0.8.1",
   "rerun-sdk[notebook]>=0.27.2",
@@ -72,12 +72,12 @@ dev = [
   "wat-inspector>=0.4.3",
 ]
 check = [
-  "ruff>=0.14.5",
-  "ty==0.0.1a21",
+  "ruff>=0.14.6",
+  "ty==0.0.1a31",
 ]
 test = [
   "dill>=0.4.0",
-  "pytest>=8.4.1",
+  "pytest>=9.0.1",
   "pytest-lazy-fixtures>=1.4.0",
   "testbook>=0.4.2",
 ]

pyproject.toml

@@ -28,7 +28,7 @@ class MapAndCollate[T]:
     def __init__(
         self,
         dataset: InstanceOf[BatchIndexableDataset],
-        collate_fn: Callable[[list[T]], Any],
+        collate_fn: Callable[[...], Any],
     ) -> None:
         self._dataset = dataset
         self._collate_fn = collate_fn

src/rbyte/dataloader.py

@@ -66,7 +66,7 @@ def __init__(
         if streams is not None and (
             missing_stream_indexes := (
                 {stream_config.index for stream_config in streams.values()}
-                - (data_keys := data.keys())
+                - (data_keys := set(data.keys()))
             )
         ):
             logger.error(
@@ -119,7 +119,7 @@ def streams(self) -> StreamsConfig | None:
 
     @override
     def __getitem__(self, index: int) -> Batch:
-        return self.get_batch([index])[0]
+        return self.get_batch([index])[0]  # ty: ignore[invalid-return-type]
 
     def __getitems__(self, index: Sequence[int]) -> Batch:  # noqa: PLW3201
         return self.get_batch(index)
@@ -134,25 +134,25 @@ def get_batch(
         include_streams: bool | None = None,
         include_meta: bool = True,
     ) -> Batch:
-        data = self.data[index]
+        data = self.data[index]  # ty: ignore[invalid-argument-type]
         meta = self.meta[index]
 
         match include_streams, self.streams:
             case None | True, dict():
-                stream_data = {stream_id: [] for stream_id in self.streams}
+                stream_data = {stream_id: [] for stream_id in self.streams}  # ty: ignore[not-iterable]
 
                 for sample, input_id in zip(data, meta["input_id"], strict=True):
-                    for stream_id, stream_config in self.streams.items():
+                    for stream_id, stream_config in self.streams.items():  # ty: ignore[possibly-missing-attribute]
                         stream_index = sample[stream_config.index].tolist()
                         source = self._get_source(stream_id, input_id)
                         stream_data[stream_id].append(source[stream_index])
 
                 stream_data = {k: torch.stack(v) for k, v in stream_data.items()}
 
-                if data.is_locked:
-                    data = data.clone(recurse=True)
+                if data.is_locked:  # ty: ignore[possibly-missing-attribute]
+                    data = data.clone(recurse=True)  # ty: ignore[unknown-argument]
 
-                data = data.update(stream_data, inplace=False)
+                data = data.update(stream_data, inplace=False)  # ty: ignore[possibly-missing-attribute]
 
             case True, None:
                 msg = "`include_streams` is True but no streams specified"
@@ -173,6 +173,10 @@ def get_batch(
 
     @cachedmethod(lambda self: self._stream_source_cache)
     def _get_source(self, stream_id: str, input_id: str) -> TensorSource:
+        if self.streams is None:
+            msg = "streams not specified"
+            raise RuntimeError(msg)
+
         return self.streams[stream_id].sources[input_id].instantiate()
 
     @classmethod
@@ -194,7 +198,7 @@ def _build_samples(
                 )
 
         output_name = pipeline.unique_leaf_node.output_name
-        results = pipeline.map(  # ty: ignore[missing-argument]
+        results = pipeline.map(
             executor=executor, **samples.model_dump(exclude={"pipeline", "executor"})
         )