more static typing-related fixes

abydos/distance/_azzoo.py

@@ -19,7 +19,7 @@
 AZZOO similarity
 """
 
-from typing import Any, Collection, Counter as TCounter, Optional, Union
+from typing import Any, Collection, Counter as TCounter, Optional, Union, cast
 
 from ._token_distance import _TokenDistance
 from ..tokenizer import _Tokenizer
@@ -141,7 +141,7 @@ def sim_score(self, src: str, tar: str) -> float:
         a = self._intersection_card()
         d = self._total_complement_card()
 
-        return a + self.params['sigma'] * d
+        return cast(float, a + self.params['sigma'] * d)
 
     def sim(self, src: str, tar: str) -> float:
         """Return the AZZOO similarity of two strings.

abydos/distance/_hellinger.py

@@ -151,7 +151,7 @@ def dist(self, src: str, tar: str) -> float:
         norm = (
             2
             * sum(
-                max(self._src_tokens[tok], self._tar_tokens[tok]) ** 2  # type: ignore  # noqa: E501
+                max(self._src_tokens[tok], self._tar_tokens[tok]) ** 2
                 for tok in self._total().keys()
             )
         ) ** 0.5

abydos/distance/_meta_levenshtein.py

@@ -91,7 +91,7 @@ def __init__(
         """
         super(MetaLevenshtein, self).__init__(**kwargs)
         self._corpus = corpus
-        self._metric = cast(_Distance, metric)
+        self._metric = JaroWinkler() if metric is None else metric
         self._normalizer = normalizer
 
         qval = 2 if 'qval' not in self.params else self.params['qval']
@@ -103,9 +103,6 @@ def __init__(
             else QGrams(qval=qval, start_stop='$#', skip=0, scaler=None)
         )
 
-        if self._metric is None:
-            self._metric = JaroWinkler()
-
     def dist_abs(self, src: str, tar: str) -> float:
         """Return the Meta-Levenshtein distance of two strings.
 
@@ -195,7 +192,7 @@ def _dist(s_tok: str, t_tok: str) -> float:
                     + _dist(src_ordered[i], tar_ordered[j]),  # sub/==
                 )
 
-        return d_mat[len(src_ordered), len(tar_ordered)]
+        return cast(float, d_mat[len(src_ordered), len(tar_ordered)])
 
     def dist(self, src: str, tar: str) -> float:
         """Return the normalized Levenshtein distance between two strings.

abydos/distance/_minhash.py

@@ -20,7 +20,7 @@
 """
 
 from hashlib import sha512
-from typing import Any, Optional
+from typing import Any, Optional, cast
 
 import numpy as np
 
@@ -150,7 +150,7 @@ def sim(self, src: str, tar: str) -> float:
                 ),
             )
 
-        return (hashes_src == hashes_tar).sum() / k
+        return cast(float, (hashes_src == hashes_tar).sum() / k)
 
 
 if __name__ == '__main__':

abydos/distance/_minkowski.py

@@ -19,7 +19,7 @@
 Minkowski distance & similarity
 """
 
-from typing import Any, Collection, Counter as TCounter, Optional, Union
+from typing import Any, Collection, Counter as TCounter, Optional, Union, cast
 
 from ._token_distance import _TokenDistance
 from ..tokenizer import _Tokenizer
@@ -144,10 +144,11 @@ def dist_abs(self, src: str, tar: str, normalized: bool = False) -> float:
         if self.params['pval'] == 0:
             # This is the l_0 "norm" as developed by David Donoho
             return sum(_ != 0 for _ in diffs) / normalizer
-        return (
+        return cast(
+            float,
             sum(_ ** self.params['pval'] for _ in diffs)
             ** (1 / self.params['pval'])
-            / normalizer
+            / normalizer,
         )
 
     def dist(self, src: str, tar: str) -> float:

abydos/distance/_mutual_information.py

@@ -20,7 +20,7 @@
 """
 
 from math import log2
-from typing import Any, Collection, Counter as TCounter, Optional, Union
+from typing import Any, Collection, Counter as TCounter, Optional, Union, cast
 
 from ._token_distance import _TokenDistance
 from ..tokenizer import _Tokenizer
@@ -179,7 +179,7 @@ def sim(self, src: str, tar: str) -> float:
             if not norm:
                 norm = [1]
 
-            return (1.0 + score / max(norm)) / 2.0
+            return cast(float, (1.0 + score / max(norm)) / 2.0)
         return 0.0
 
 

abydos/distance/_quantitative_dice.py

@@ -108,7 +108,7 @@ def sim(self, src: str, tar: str) -> float:
         return (
             2
             * sum(
-                self._src_tokens[tok] * self._tar_tokens[tok]  # type: ignore
+                self._src_tokens[tok] * self._tar_tokens[tok]
                 for tok in alphabet
             )
             / (

abydos/distance/_rouge_w.py

@@ -62,7 +62,7 @@ def __init__(
         super(RougeW, self).__init__(**kwargs)
 
         if f_func is not None:
-            self._f_func = f_func # type: Callable[[float], float]
+            self._f_func = f_func  # type: Callable[[float], float]
         else:
             self._f_func = RougeW._square  # noqa: SF01
 

abydos/distance/_saps.py

@@ -19,7 +19,7 @@
 Syllable Alignment Pattern Searching tokenizer
 """
 
-from typing import Any, Callable, List, Optional, Tuple
+from typing import Any, Callable, List, Optional, Tuple, cast
 
 from numpy import int as np_int
 from numpy import zeros as np_zeros

abydos/distance/_sift4_extended.py

@@ -19,7 +19,7 @@
 Sift4 Extended approximate string distance
 """
 
-from typing import Any, Callable, Dict, List, Optional, Union, cast
+from typing import Any, Callable, Dict, List, Optional, Union
 
 from ._distance import _Distance
 from ._sift4 import Sift4

abydos/distance/_softtf_idf.py

@@ -21,7 +21,7 @@
 
 from collections import defaultdict
 from math import log1p
-from typing import Any, DefaultDict, Optional, Tuple, cast
+from typing import Any, DefaultDict, Optional, Tuple
 
 from ._distance import _Distance
 from ._jaro_winkler import JaroWinkler
@@ -103,11 +103,8 @@ def __init__(
         """
         super(SoftTFIDF, self).__init__(tokenizer=tokenizer, **kwargs)
         self._corpus = corpus
-        self._metric = cast(_Distance, metric)
         self._threshold = threshold
-
-        if self._metric is None:
-            self._metric = JaroWinkler()
+        self._metric = JaroWinkler() if metric is None else metric
 
     def sim(self, src: str, tar: str) -> float:
         """Return the SoftTF-IDF similarity of two strings.

abydos/distance/_token_distance.py

@@ -30,6 +30,7 @@
     Optional,
     Tuple,
     Union,
+    cast,
 )
 
 import numpy as np
@@ -286,7 +287,9 @@ def _norm_inverse(x: float, _squares: int, pop: float) -> float:
     def _norm_complement(x: float, _squares: int, pop: float) -> float:
         return pop - x
 
-    def _tokenize(self, src: Union[str, TCounter], tar: Union[str, TCounter]) -> '_TokenDistance':
+    def _tokenize(
+        self, src: Union[str, TCounter], tar: Union[str, TCounter]
+    ) -> '_TokenDistance':
         """Return the Q-Grams in src & tar.
 
         Parameters
@@ -595,12 +598,13 @@ def _soft_intersection(self) -> TCounter[str]:
 
         def _membership(src: str, tar: str) -> float:
             greater_length = max(len(src), len(tar))
-            return (
+            return cast(
+                float,
                 max(
                     greater_length - self.params['metric'].dist_abs(src, tar),
                     self._lcprefix.dist_abs(src, tar),
                 )
-                / greater_length
+                / greater_length,
             )
 
         def _token_src_tar_int(

abydos/distance/_tversky.py

@@ -19,7 +19,7 @@
 Tversky index
 """
 
-from typing import Any, Optional
+from typing import Any, Optional, cast
 
 from ._token_distance import _TokenDistance
 from ..tokenizer import _Tokenizer
@@ -175,21 +175,29 @@ def sim(self, src: str, tar: str) -> float:
             return 0.0
 
         if self.params['bias'] is None:
-            return q_intersection_mag / (
+            return cast(
+                float,
                 q_intersection_mag
-                + self.params['alpha'] * q_src_mag
-                + self.params['beta'] * q_tar_mag
+                / (
+                    q_intersection_mag
+                    + self.params['alpha'] * q_src_mag
+                    + self.params['beta'] * q_tar_mag
+                ),
             )
 
         a_val, b_val = sorted((q_src_mag, q_tar_mag))
         c_val = q_intersection_mag + self.params['bias']
-        return c_val / (
-            self.params['beta']
-            * (
-                self.params['alpha'] * a_val
-                + (1 - self.params['alpha']) * b_val
-            )
-            + c_val
+        return cast(
+            float,
+            c_val
+            / (
+                self.params['beta']
+                * (
+                    self.params['alpha'] * a_val
+                    + (1 - self.params['alpha']) * b_val
+                )
+                + c_val
+            ),
         )
 
 

abydos/distance/_typo.py

@@ -216,12 +216,8 @@ def dist_abs(self, src: str, tar: str) -> float:
         else:
             keyboard = self._keyboard[self._layout]
 
-        lowercase = {
-            item for sublist in keyboard[0] for item in sublist
-        }
-        uppercase = {
-            item for sublist in keyboard[1] for item in sublist
-        }
+        lowercase = {item for sublist in keyboard[0] for item in sublist}
+        uppercase = {item for sublist in keyboard[1] for item in sublist}
         keys = set(chain(*chain(*keyboard)))
 
         def _kb_array_for_char(char: str) -> Tuple[Tuple[str, ...], ...]: