Merge pull request #49 from mdekstrand/tweak/pyright

Use Pyright for typechecking

.copier-answers.yml

@@ -1,9 +1,10 @@
 # Changes here will be overwritten by Copier
-_commit: ccc299c
+_commit: 8d75d6c
 _src_path: https://github.com/lenskit/lk-project-template
 package_name: binpickle
 project_descr: Optimized format for pickling binary data.
 project_name: binpickle
 project_title: BinPickle
 require_lint: true
 start_year: 2020
+typecheck: true

.github/workflows/check-sources.yml

@@ -11,64 +11,72 @@ concurrency:
 
 jobs:
   lint:
-    name: Check Source Style
+    name: Check Source Code
     runs-on: ubuntu-latest
 
     steps:
-    - name: 📥 Check out source code
-      uses: actions/checkout@v2
-      with:
-        fetch-depth: 0
+      - name: 📥 Check out source code
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
 
-    - name: 🐍 Set up Python
-      uses: actions/setup-python@v4
-      with:
-        python-version: "3.11"
-        cache: 'pip'
+      - name: 🐍 Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: "3.11"
+          cache: "pip"
 
-    - name: 🛠️ Install tools
-      run: |
-        pip install ruff
+      - name: 🛠️ Install development tools and dependencies
+        run: |
+          pip install -e .[dev]
 
-    - name: 🪮 Check source code formatting
-      id: format
-      run: |
-        if pipx run ruff format --diff $PKG_DIR; then
-          echo passed=yes >>"$GITHUB_OUTPUT"
-        else
-          echo passed=no >>"$GITHUB_OUTPUT"
-          echo "::error::source code not formatted"
-        fi
-      env:
-        PKG_DIR: binpickle
+      - name: 🪮 Check source code formatting
+        id: format
+        run: |
+          if ruff format --diff $PKG_DIR; then
+            echo passed=yes >>"$GITHUB_OUTPUT"
+          else
+            echo passed=no >>"$GITHUB_OUTPUT"
+            echo "::error::source code not formatted"
+          fi
+        env:
+          PKG_DIR: binpickle
 
-    - name: 🐜 Check source code lint rules
-      id: lint
-      run: |
-        if pipx run ruff check --output-format=github $PKG_DIR; then
-          echo passed=yes >>"$GITHUB_OUTPUT"
-        else
-          echo passed=no >>"$GITHUB_OUTPUT"
-          echo "::error::source code lint check failed"
-        fi
-      env:
-        PKG_DIR: binpickle
+      - name: 🐜 Check source code lint rules
+        id: lint
+        run: |
+          if ruff check --output-format=github $PKG_DIR; then
+            echo passed=yes >>"$GITHUB_OUTPUT"
+          else
+            echo passed=no >>"$GITHUB_OUTPUT"
+            echo "::error::source code lint check failed"
+          fi
+        env:
+          PKG_DIR: binpickle
 
-    - name: 🧾 Checking results
-      run: |
-        if [ "$FMT_PASSED" = no ]; then
-            echo "::error::format failed, failing build"
-            exit 1
-        fi
-        if [ "$LINT_PASSED" = no ]; then
-            if [ "$LINT_REQUIRED" = true ]; then
-                echo "::error::lint failed, failing build"
-                exit 2
-            else
-                echo "::error::lint failed but non-mandatory"
-            fi
-        fi
-      env:
-        FMT_PASSED: ${{ steps.fmt.outputs.passed }}
-        LINT_PASSED: ${{ steps.lint.outputs.passed }}
-        LINT_REQUIRED: True
+      - name: 🧾 Checking lint and format results
+        run: |
+          if [ "$FMT_PASSED" = no ]; then
+              echo "::error::format failed, failing build"
+              exit 1
+          fi
+          if [ "$LINT_PASSED" = no ]; then
+              if [ "$LINT_REQUIRED" = true ]; then
+                  echo "::error::lint failed, failing build"
+                  exit 2
+              else
+                  echo "::error::lint failed but non-mandatory"
+              fi
+          fi
+        env:
+          FMT_PASSED: ${{ steps.fmt.outputs.passed }}
+          LINT_PASSED: ${{ steps.lint.outputs.passed }}
+          LINT_REQUIRED: True
+
+
+      - name: 📐 Check types
+        id: typecheck
+        uses: jakebailey/pyright-action@v1
+        with:
+          extra-args: binpickle
+

.vscode/settings.json

@@ -1,7 +1,20 @@
 {
-  "mypy-type-checker.reportingScope": "workspace",
   "[python]": {
     "editor.defaultFormatter": "charliermarsh.ruff",
+    "editor.codeActionsOnSave": {
+      "source.organizeImports": true
+    },
+    "editor.formatOnSave": true,
+  },
+  "[jsonc]": {
+    "editor.formatOnSave": true,
+  },
+  "[yaml]": {
+    "editor.defaultFormatter": "redhat.vscode-yaml",
+    "editor.formatOnSave": true,
+  },
+  "[toml]": {
+    "editor.defaultFormatter": "tamasfe.even-better-toml",
     "editor.formatOnSave": true,
   },
 }

binpickle/__init__.py

@@ -2,10 +2,10 @@
 Optimized format for pickling binary data.
 """
 
-from importlib.metadata import version, PackageNotFoundError
+from importlib.metadata import PackageNotFoundError, version
 
-from .write import dump, BinPickler
-from .read import load, BinPickleFile, file_info
+from .read import BinPickleFile, file_info, load
+from .write import BinPickler, dump
 
 try:
     __version__ = version("binpickle")

binpickle/_util.py

@@ -2,8 +2,10 @@
 Internal utility functions for Binpickle.
 """
 from __future__ import annotations
-from typing import Optional, Any
+
 import hashlib
+from typing import Any, Optional
+
 from typing_extensions import Buffer
 
 naturalsize: Optional[Any]

binpickle/encode.py

@@ -3,11 +3,12 @@
 """
 
 from __future__ import annotations
-from typing import Optional, TypeAlias, Callable, overload
-from typing_extensions import Buffer
+
+from typing import Callable, Optional, TypeAlias, overload
 
 from numcodecs.abc import Codec
 from numcodecs.registry import get_codec
+from typing_extensions import Buffer
 
 from binpickle.format import CodecSpec
 

binpickle/format.py

@@ -2,10 +2,13 @@
 Constants and functions defining the binpickle format.
 """
 
-from dataclasses import dataclass, field, fields
-import struct
-from typing import TypeAlias
+from __future__ import annotations
+
 import enum
+import io
+import struct
+from dataclasses import dataclass, field, fields
+from typing import Any, TypeAlias
 
 from binpickle.errors import FormatError
 
@@ -71,7 +74,7 @@ def encode(self):
         return HEADER_FORMAT.pack(MAGIC, self.version, self.flags._value_, self.length)
 
     @classmethod
-    def decode(cls, buf: bytes, *, verify=True):
+    def decode(cls, buf: bytes | bytearray | memoryview, *, verify: bool = True) -> FileHeader:
         "Decode a file header from bytes."
         if len(buf) != HEADER_FORMAT.size:
             raise FormatError("incorrect header length")
@@ -88,7 +91,7 @@ def decode(cls, buf: bytes, *, verify=True):
         return cls(v, flags, off)
 
     @classmethod
-    def read(cls, file, **kwargs):
+    def read(cls, file: io.FileIO | io.BufferedReader, **kwargs: bool) -> FileHeader:
         buf = file.read(HEADER_FORMAT.size)
         return cls.decode(buf, **kwargs)
 
@@ -125,7 +128,7 @@ def encode(self):
         return TRAILER_FORMAT.pack(self.offset, self.length, self.hash)
 
     @classmethod
-    def decode(cls, buf, *, verify=True):
+    def decode(cls, buf: bytes | bytearray | memoryview, *, verify: bool = True) -> FileTrailer:
         "Decode a file trailer from bytes."
         off, len, ck = TRAILER_FORMAT.unpack(buf)
         return cls(off, len, ck)
@@ -155,7 +158,7 @@ def to_repr(self):
         return dict((f.name, getattr(self, f.name)) for f in fields(self))
 
     @classmethod
-    def from_repr(cls, repr):
+    def from_repr(cls, repr: dict[str, Any]):
         "Convert an index entry from its MsgPack-compatible representation"
         if not isinstance(repr, dict):
             raise TypeError("IndexEntry representation must be a dict")

binpickle/read.py

@@ -1,22 +1,23 @@
-import sys
-from dataclasses import dataclass
-from enum import Enum
 import hashlib
-import mmap
-import logging
 import io
-from os import PathLike
-from typing import Optional
-from typing_extensions import Buffer
+import logging
+import mmap
 import pickle
+import sys
 import warnings
+from dataclasses import dataclass
+from enum import Enum
+from os import PathLike
+from typing import Any, Optional
+
 import msgpack
+from typing_extensions import Buffer
 
 from binpickle.encode import resolve_codec
 from binpickle.errors import BinPickleError, FormatError, FormatWarning, IntegrityError
 
-from .format import FileHeader, Flags, IndexEntry, FileTrailer
 from ._util import hash_buffer
+from .format import FileHeader, FileTrailer, Flags, IndexEntry
 
 _log = logging.getLogger(__name__)
 
@@ -54,7 +55,7 @@ class BinPickleFile:
             If ``True`` (the default), verify file checksums while reading.
     """
 
-    filename: str | PathLike
+    filename: str | PathLike[str]
     direct: bool | str
     verify: bool
     header: FileHeader
@@ -64,7 +65,9 @@ class BinPickleFile:
     _index_buf: Optional[memoryview]
     entries: list[IndexEntry]
 
-    def __init__(self, filename, *, direct: bool | str = False, verify: bool = True):
+    def __init__(
+        self, filename: str | PathLike[str], *, direct: bool | str = False, verify: bool = True
+    ):
         self.filename = filename
         self.direct = direct
         self.verify = verify
@@ -77,7 +80,7 @@ def __init__(self, filename, *, direct: bool | str = False, verify: bool = True)
     def __enter__(self):
         return self
 
-    def __exit__(self, *args):
+    def __exit__(self, *args: Any):
         self.close()
         return False
 
@@ -110,7 +113,7 @@ def find_errors(self) -> list[str]:
         invalid msgpack formats in the index won't be detected here.  This method checks
         buffer hashes, offset overlaps, and such.
         """
-        errors = []
+        errors: list[str] = []
         assert self._index_buf is not None, "file not loaded"
 
         i_sum = hashlib.sha256(self._index_buf).digest()
@@ -173,7 +176,7 @@ def _read_index(self) -> None:
             self._index_buf = None
             raise e
 
-        self.entries = [IndexEntry.from_repr(e) for e in msgpack.unpackb(self._index_buf)]
+        self.entries = [IndexEntry.from_repr(e) for e in msgpack.unpackb(self._index_buf)]  # type: ignore
         _log.debug("read %d entries from file", len(self.entries))
 
     def _read_buffer(
@@ -219,7 +222,7 @@ def _verify_buffer(self, buf: memoryview, hash: bytes, msg: str = "buffer"):
                 raise IntegrityError(f"{msg} has incorrect hash, corrupt file?")
 
 
-def load(file: str | PathLike) -> object:
+def load(file: str | PathLike[str]) -> object:
     """
     Load an object from a BinPickle file.
 
@@ -231,7 +234,7 @@ def load(file: str | PathLike) -> object:
         return bpf.load()
 
 
-def file_info(file: str | PathLike) -> BPKInfo:
+def file_info(file: str | PathLike[str]) -> BPKInfo:
     """
     Test whether a file is a BinPickle file, and if so, return basic information
     about it.