Add type-checking using mypy

.github/workflows/checks.yml

@@ -26,7 +26,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        check: ['style', 'spellcheck']
+        check: ['style', 'spellcheck', 'type']
 
     steps:
       - uses: actions/checkout@v5

pyproject.toml

@@ -22,6 +22,7 @@ classifiers = [
 ]
 requires-python = ">=3.10"
 dependencies = [
+    "acres >=0.5",
     "attrs >=24.1",
     "bidsschematools >=1.1",
     "orjson>=3.11.3",
@@ -152,3 +153,28 @@ test = [
   "datalad >=1.1",
   "cattrs>=24.1.3",
 ]
+types = [
+    "mypy>=1.18.2",
+]
+
+[tool.mypy]
+strict = true
+exclude = [
+  "^tests/data",
+]
+
+[[tool.mypy.overrides]]
+module = 'bids_validator._version'
+ignore_errors = true
+
+[[tool.mypy.overrides]]
+module = 'datalad.*'
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = 'fsspec'
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = 'upath'
+ignore_missing_imports = true

src/bids_validator/__main__.py

@@ -12,8 +12,8 @@
 from typing import Annotated
 
 from bidsschematools.schema import load_schema
-from bidsschematools.types import Namespace
 from bidsschematools.types.context import Subject
+from bidsschematools.types.namespace import Namespace
 
 from bids_validator import BIDSValidator
 from bids_validator.context import Context, Dataset, Sessions
@@ -22,11 +22,11 @@
 app = typer.Typer()
 
 
-def is_subject_dir(tree):
+def is_subject_dir(tree: FileTree) -> bool:
     return tree.name.startswith('sub-')
 
 
-def walk(tree: FileTree, dataset: Dataset, subject: Subject = None) -> Iterator[Context]:
+def walk(tree: FileTree, dataset: Dataset, subject: Subject | None = None) -> Iterator[Context]:
     """Iterate over children of a FileTree and check if they are a directory or file.
 
     If it's a directory then run again recursively, if it's a file file check the file name is
@@ -52,7 +52,7 @@ def walk(tree: FileTree, dataset: Dataset, subject: Subject = None) -> Iterator[
             yield Context(child, dataset, subject)
 
 
-def validate(tree: FileTree, schema: Namespace):
+def validate(tree: FileTree, schema: Namespace) -> None:
     """Check if the file path is BIDS compliant.
 
     Parameters
@@ -71,14 +71,14 @@ def validate(tree: FileTree, schema: Namespace):
             print(f'{file.path} is not a valid bids filename')
 
 
-def show_version():
+def show_version() -> None:
     """Show bids-validator version."""
     from . import __version__
 
     print(f'bids-validator {__version__} (Python {sys.version.split()[0]})')
 
 
-def version_callback(value: bool):
+def version_callback(value: bool) -> None:
     """Run the callback for CLI version flag.
 
     Parameters

src/bids_validator/bids_validator.py

@@ -12,6 +12,8 @@
 import bidsschematools.utils
 import bidsschematools.validator
 
+from .types import _typings as t
+
 
 class LoggingContext:
     # From logging cookbook (CC0):
@@ -35,20 +37,31 @@ class LoggingContext:
 
     """
 
-    def __init__(self, logger, level=None, handler=None, close=True):
+    def __init__(
+        self,
+        logger: logging.Logger,
+        level: int | None = None,
+        handler: logging.Handler | None = None,
+        close: bool = True,
+    ) -> None:
         self.logger = logger
         self.level = level
         self.handler = handler
         self.close = close
 
-    def __enter__(self):
+    def __enter__(self) -> None:
         if self.level is not None:
             self.old_level = self.logger.level
             self.logger.setLevel(self.level)
         if self.handler:
             self.logger.addHandler(self.handler)
 
-    def __exit__(self, et, ev, tb):
+    def __exit__(
+        self,
+        et: type[BaseException] | None,
+        ev: BaseException | None,
+        tb: t.TracebackType,
+    ) -> None:
         if self.level is not None:
             self.logger.setLevel(self.old_level)
         if self.handler:
@@ -66,7 +79,7 @@ class BIDSValidator:
 
     regexes = None
 
-    def __init__(self, index_associated=True):
+    def __init__(self, index_associated: bool = True) -> None:
         """Initialize BIDSValidator object.
 
         Parameters
@@ -81,9 +94,9 @@ def __init__(self, index_associated=True):
         self.index_associated = index_associated
 
     @classmethod
-    def _init_regexes(cls):
+    def _init_regexes(cls) -> None:
         if cls.regexes is None:
-            with LoggingContext(bst.utils.get_logger(), level=logging.WARNING):
+            with LoggingContext(bst.utils.get_logger(), level=logging.WARNING):  # type: ignore[no-untyped-call]
                 schema = bst.schema.load_schema()
 
             all_rules = chain.from_iterable(
@@ -93,7 +106,7 @@ def _init_regexes(cls):
             cls.regexes = [rule['regex'] for rule in all_rules]
 
     @classmethod
-    def parse(cls, path):
+    def parse(cls, path: str) -> dict[str, str]:
         """Parse a file path into a dictionary of BIDS entities.
 
         Parameters
@@ -136,6 +149,7 @@ def parse(cls, path):
         """
         if cls.regexes is None:
             cls._init_regexes()
+            assert cls.regexes is not None  # noqa: S101
 
         if path.startswith(os.sep):
             path = path.replace(os.sep, '/')
@@ -155,7 +169,7 @@ def parse(cls, path):
 
     @classmethod
     @lru_cache
-    def is_bids(cls, path):
+    def is_bids(cls, path: str) -> bool:
         """Check if file path adheres to BIDS.
 
         Main method of the validator. Uses other class methods for checking
@@ -199,14 +213,14 @@ def is_bids(cls, path):
             return False
 
     @classmethod
-    def is_top_level(cls, path):
+    def is_top_level(cls, path: str) -> bool:
         """Check if the file has appropriate name for a top-level file."""
         parts = cls.parse(path)
         if not parts:
             return False
         return parts.get('subject') is None
 
-    def is_associated_data(self, path):
+    def is_associated_data(self, path: str) -> bool:
         """Check if file is appropriate associated data."""
         if not self.index_associated:
             return False
@@ -217,31 +231,31 @@ def is_associated_data(self, path):
         return parts.get('path') in ('code', 'derivatives', 'stimuli', 'sourcedata')
 
     @classmethod
-    def is_session_level(cls, path):
+    def is_session_level(cls, path: str) -> bool:
         """Check if the file has appropriate name for a session level."""
         parts = cls.parse(path)
         if not parts:
             return False
         return parts.get('datatype') is None and parts.get('suffix') != 'sessions'
 
     @classmethod
-    def is_subject_level(cls, path):
+    def is_subject_level(cls, path: str) -> bool:
         """Check if the file has appropriate name for a subject level."""
         parts = cls.parse(path)
         if not parts:
             return False
         return parts.get('suffix') == 'sessions'
 
     @classmethod
-    def is_phenotypic(cls, path):
+    def is_phenotypic(cls, path: str) -> bool:
         """Check if file is phenotypic data."""
         parts = cls.parse(path)
         if not parts:
             return False
         return parts.get('datatype') == 'phenotype'
 
     @classmethod
-    def is_file(cls, path):
+    def is_file(cls, path: str) -> bool:
         """Check if file is a data file or non-inherited metadata file."""
         parts = cls.parse(path)
         if not parts:

src/bids_validator/bidsignore.py

@@ -84,7 +84,7 @@ def from_file(cls, pathlike: os.PathLike):
 
     def match(self, relpath: str) -> bool:
         """Match a relative path against a collection of ignore patterns."""
-        if any(compile_pat(pattern).match(relpath) for pattern in self.patterns if pattern):
+        if any(regex.match(relpath) for pat in self.patterns if (regex := compile_pat(pat))):
             self.history.append(relpath)
             return True
         return False