TYP: improve type annotations and remove unnecessary type ignores (#62315)

Author: simonjayhawkins

pandas/_version.py

@@ -640,7 +640,7 @@ def render(pieces, style):
     }
 
 
-def get_versions():
+def get_versions() -> dict:
     """Get version information or return default if unable to do so."""
     # I am in _version.py, which lives at ROOT/VERSIONFILE_SOURCE. If we have
     # __file__, we can work backwards from there to the root. Some

pandas/core/algorithms.py

@@ -11,7 +11,9 @@
 from typing import (
     TYPE_CHECKING,
     Literal,
+    TypeVar,
     cast,
+    overload,
 )
 import warnings
 
@@ -104,6 +106,8 @@
         ExtensionArray,
     )
 
+    T = TypeVar("T", bound=Index | Categorical | ExtensionArray)
+
 
 # --------------- #
 # dtype access    #
@@ -314,6 +318,12 @@ def _check_object_for_strings(values: np.ndarray) -> str:
 # --------------- #
 
 
+@overload
+def unique(values: T) -> T: ...
+@overload
+def unique(values: np.ndarray | Series) -> np.ndarray: ...
+
+
 def unique(values):
     """
     Return unique values based on a hash table.

pandas/core/base.py

@@ -1102,7 +1102,7 @@ def unique(self):
             # i.e. ExtensionArray
             result = values.unique()
         else:
-            result = algorithms.unique1d(values)
+            result = algorithms.unique1d(values)  # type: ignore[assignment]
         return result
 
     @final

pandas/core/groupby/categorical.py

@@ -46,7 +46,7 @@ def recode_for_groupby(c: Categorical, sort: bool, observed: bool) -> Categorica
         # In cases with c.ordered, this is equivalent to
         #  return c.remove_unused_categories(), c
 
-        take_codes = unique1d(c.codes[c.codes != -1])  # type: ignore[no-untyped-call]
+        take_codes = unique1d(c.codes[c.codes != -1])
 
         if sort:
             take_codes = np.sort(take_codes)
@@ -68,7 +68,7 @@ def recode_for_groupby(c: Categorical, sort: bool, observed: bool) -> Categorica
 
     # GH:46909: Re-ordering codes faster than using (set|add|reorder)_categories
     # GH 38140: exclude nan from indexer for categories
-    unique_notnan_codes = unique1d(c.codes[c.codes != -1])  # type: ignore[no-untyped-call]
+    unique_notnan_codes = unique1d(c.codes[c.codes != -1])
     if sort:
         unique_notnan_codes = np.sort(unique_notnan_codes)
     if (num_cat := len(c.categories)) > len(unique_notnan_codes):

pandas/core/indexes/base.py

@@ -3304,7 +3304,7 @@ def _intersection(self, other: Index, sort: bool = False):
                     res = algos.unique1d(res_indexer)
                 else:
                     result = self.take(indexer)
-                    res = result.drop_duplicates()
+                    res = result.drop_duplicates()  # type: ignore[assignment]
                 return ensure_wrapped_if_datetimelike(res)
 
         res_values = self._intersection_via_get_indexer(other, sort=sort)

pandas/util/_print_versions.py

@@ -32,7 +32,7 @@ def _get_commit_hash() -> str | None:
     except ImportError:
         from pandas._version import get_versions
 
-        versions = get_versions()  # type: ignore[no-untyped-call]
+        versions = get_versions()
         return versions["full-revisionid"]