Merge pull request #1197 from effigies/type/filehandling

TYPE: Annotate file-handling modules

Author: effigies

nibabel/dataobj_images.py

@@ -15,11 +15,14 @@
 
 from .arrayproxy import ArrayLike
 from .deprecated import deprecate_with_version
-from .filebasedimages import FileBasedHeader, FileBasedImage, FileMap, FileSpec
+from .filebasedimages import FileBasedHeader, FileBasedImage
+from .fileholders import FileMap
 
 if ty.TYPE_CHECKING:  # pragma: no cover
     import numpy.typing as npt
 
+    from .filename_parser import FileSpec
+
 ArrayImgT = ty.TypeVar('ArrayImgT', bound='DataobjImage')
 
 

nibabel/filebasedimages.py

@@ -10,18 +10,18 @@
 from __future__ import annotations
 
 import io
-import os
 import typing as ty
 from copy import deepcopy
 from typing import Type
 from urllib import request
 
-from .fileholders import FileHolder
-from .filename_parser import TypesFilenamesError, splitext_addext, types_filenames
+from .fileholders import FileHolder, FileMap
+from .filename_parser import TypesFilenamesError, _stringify_path, splitext_addext, types_filenames
 from .openers import ImageOpener
 
-FileSpec = ty.Union[str, os.PathLike]
-FileMap = ty.Mapping[str, FileHolder]
+if ty.TYPE_CHECKING:  # pragma: no cover
+    from .filename_parser import ExtensionSpec, FileSpec
+
 FileSniff = ty.Tuple[bytes, str]
 
 ImgT = ty.TypeVar('ImgT', bound='FileBasedImage')
@@ -160,7 +160,7 @@ class FileBasedImage:
     header_class: Type[FileBasedHeader] = FileBasedHeader
     _header: FileBasedHeader
     _meta_sniff_len: int = 0
-    files_types: tuple[tuple[str, str | None], ...] = (('image', None),)
+    files_types: tuple[ExtensionSpec, ...] = (('image', None),)
     valid_exts: tuple[str, ...] = ()
     _compressed_suffixes: tuple[str, ...] = ()
 
@@ -411,7 +411,7 @@ def _sniff_meta_for(
         t_fnames = types_filenames(
             filename, klass.files_types, trailing_suffixes=klass._compressed_suffixes
         )
-        meta_fname = t_fnames.get('header', filename)
+        meta_fname = t_fnames.get('header', _stringify_path(filename))
 
         # Do not re-sniff if it would be from the same file
         if sniff is not None and sniff[1] == meta_fname:

nibabel/fileholders.py

@@ -7,6 +7,10 @@
 #
 ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ##
 """Fileholder class"""
+from __future__ import annotations
+
+import io
+import typing as ty
 from copy import copy
 
 from .openers import ImageOpener
@@ -19,7 +23,12 @@ class FileHolderError(Exception):
 class FileHolder:
     """class to contain filename, fileobj and file position"""
 
-    def __init__(self, filename=None, fileobj=None, pos=0):
+    def __init__(
+        self,
+        filename: str | None = None,
+        fileobj: io.IOBase | None = None,
+        pos: int = 0,
+    ):
         """Initialize FileHolder instance
 
         Parameters
@@ -37,7 +46,7 @@ def __init__(self, filename=None, fileobj=None, pos=0):
         self.fileobj = fileobj
         self.pos = pos
 
-    def get_prepare_fileobj(self, *args, **kwargs):
+    def get_prepare_fileobj(self, *args, **kwargs) -> ImageOpener:
         """Return fileobj if present, or return fileobj from filename
 
         Set position to that given in self.pos
@@ -69,7 +78,7 @@ def get_prepare_fileobj(self, *args, **kwargs):
             raise FileHolderError('No filename or fileobj present')
         return obj
 
-    def same_file_as(self, other):
+    def same_file_as(self, other: FileHolder) -> bool:
         """Test if `self` refers to same files / fileobj as `other`
 
         Parameters
@@ -86,12 +95,15 @@ def same_file_as(self, other):
         return (self.filename == other.filename) and (self.fileobj == other.fileobj)
 
     @property
-    def file_like(self):
+    def file_like(self) -> str | io.IOBase | None:
         """Return ``self.fileobj`` if not None, otherwise ``self.filename``"""
         return self.fileobj if self.fileobj is not None else self.filename
 
 
-def copy_file_map(file_map):
+FileMap = ty.Mapping[str, FileHolder]
+
+
+def copy_file_map(file_map: FileMap) -> FileMap:
     r"""Copy mapping of fileholders given by `file_map`
 
     Parameters
@@ -105,7 +117,4 @@ def copy_file_map(file_map):
        Copy of `file_map`, using shallow copy of ``FileHolder``\s
 
     """
-    fm_copy = {}
-    for key, fh in file_map.items():
-        fm_copy[key] = copy(fh)
-    return fm_copy
+    return {key: copy(fh) for key, fh in file_map.items()}

nibabel/filename_parser.py

@@ -7,15 +7,21 @@
 #
 ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ##
 """Create filename pairs, triplets etc, with expected extensions"""
+from __future__ import annotations
+
 import os
-import pathlib
+import typing as ty
+
+if ty.TYPE_CHECKING:  # pragma: no cover
+    FileSpec = str | os.PathLike[str]
+    ExtensionSpec = tuple[str, str | None]
 
 
 class TypesFilenamesError(Exception):
     pass
 
 
-def _stringify_path(filepath_or_buffer):
+def _stringify_path(filepath_or_buffer: FileSpec) -> str:
     """Attempt to convert a path-like object to a string.
 
     Parameters
@@ -28,30 +34,21 @@ def _stringify_path(filepath_or_buffer):
 
     Notes
     -----
-    Objects supporting the fspath protocol (python 3.6+) are coerced
-    according to its __fspath__ method.
-    For backwards compatibility with older pythons, pathlib.Path objects
-    are specially coerced.
-    Any other object is passed through unchanged, which includes bytes,
-    strings, buffers, or anything else that's not even path-like.
-
-    Copied from:
-    https://github.com/pandas-dev/pandas/blob/325dd686de1589c17731cf93b649ed5ccb5a99b4/pandas/io/common.py#L131-L160
+    Adapted from:
+    https://github.com/pandas-dev/pandas/blob/325dd68/pandas/io/common.py#L131-L160
     """
-    if hasattr(filepath_or_buffer, '__fspath__'):
+    if isinstance(filepath_or_buffer, os.PathLike):
         return filepath_or_buffer.__fspath__()
-    elif isinstance(filepath_or_buffer, pathlib.Path):
-        return str(filepath_or_buffer)
     return filepath_or_buffer
 
 
 def types_filenames(
-    template_fname,
-    types_exts,
-    trailing_suffixes=('.gz', '.bz2'),
-    enforce_extensions=True,
-    match_case=False,
-):
+    template_fname: FileSpec,
+    types_exts: ty.Sequence[ExtensionSpec],
+    trailing_suffixes: ty.Sequence[str] = ('.gz', '.bz2'),
+    enforce_extensions: bool = True,
+    match_case: bool = False,
+) -> dict[str, str]:
     """Return filenames with standard extensions from template name
 
     The typical case is returning image and header filenames for an
@@ -152,12 +149,12 @@ def types_filenames(
     # we've found .IMG as the extension, we want .HDR as the matching
     # one.  Let's only do this when the extension is all upper or all
     # lower case.
-    proc_ext = lambda s: s
+    proc_ext: ty.Callable[[str], str] = lambda s: s
     if found_ext:
         if found_ext == found_ext.upper():
-            proc_ext = lambda s: s.upper()
+            proc_ext = str.upper
         elif found_ext == found_ext.lower():
-            proc_ext = lambda s: s.lower()
+            proc_ext = str.lower
     for name, ext in types_exts:
         if name == direct_set_name:
             tfns[name] = template_fname
@@ -171,7 +168,12 @@ def types_filenames(
     return tfns
 
 
-def parse_filename(filename, types_exts, trailing_suffixes, match_case=False):
+def parse_filename(
+    filename: FileSpec,
+    types_exts: ty.Sequence[ExtensionSpec],
+    trailing_suffixes: ty.Sequence[str],
+    match_case: bool = False,
+) -> tuple[str, str, str | None, str | None]:
     """Split filename into fileroot, extension, trailing suffix; guess type.
 
     Parameters
@@ -230,9 +232,9 @@ def parse_filename(filename, types_exts, trailing_suffixes, match_case=False):
             break
     guessed_name = None
     found_ext = None
-    for name, ext in types_exts:
-        if ext and endswith(filename, ext):
-            extpos = -len(ext)
+    for name, type_ext in types_exts:
+        if type_ext and endswith(filename, type_ext):
+            extpos = -len(type_ext)
             found_ext = filename[extpos:]
             filename = filename[:extpos]
             guessed_name = name
@@ -242,15 +244,19 @@ def parse_filename(filename, types_exts, trailing_suffixes, match_case=False):
     return (filename, found_ext, ignored, guessed_name)
 
 
-def _endswith(whole, end):
+def _endswith(whole: str, end: str) -> bool:
     return whole.endswith(end)
 
 
-def _iendswith(whole, end):
+def _iendswith(whole: str, end: str) -> bool:
     return whole.lower().endswith(end.lower())
 
 
-def splitext_addext(filename, addexts=('.gz', '.bz2', '.zst'), match_case=False):
+def splitext_addext(
+    filename: FileSpec,
+    addexts: ty.Sequence[str] = ('.gz', '.bz2', '.zst'),
+    match_case: bool = False,
+) -> tuple[str, str, str]:
     """Split ``/pth/fname.ext.gz`` into ``/pth/fname, .ext, .gz``
 
     where ``.gz`` may be any of passed `addext` trailing suffixes.