PR feedback

ihmeuw · Nov 4, 2024 · b285eda · b285eda
1 parent d71bb7b
commit b285eda
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 8 deletions.
diff --git a/src/pseudopeople/configuration/noise_configuration.py b/src/pseudopeople/configuration/noise_configuration.py
@@ -1,4 +1,3 @@
-# mypy: disable-error-code="unused-ignore"
 from __future__ import annotations
 
 from typing import Any
@@ -24,8 +23,7 @@ def __init__(self, config: LayeredConfigTree):
         self._config = config
 
     def to_dict(self) -> dict:
-        # TODO: remove ignore when dropping support for Python 3.9
-        config_dict: dict = self._config.to_dict()  # type: ignore [assignment]
+        config_dict: dict = self._config.to_dict()
         return config_dict
 
     def get_value(
@@ -80,8 +78,7 @@ def get_value(
             )
         noise_value: int | float | LayeredConfigTree = parameter_tree.get(parameter_name)
         converted_noise_value: int | float | dict = (
-            # not sure how to tell mypy the types in this dict
-            noise_value.to_dict()  # type: ignore [assignment]
+            noise_value.to_dict()
             if isinstance(noise_value, LayeredConfigTree)
             else noise_value
         )

diff --git a/src/pseudopeople/utilities.py b/src/pseudopeople/utilities.py
@@ -220,7 +220,7 @@ def count_number_of_tokens_per_string(s1: pd.Series, s2: pd.Series) -> pd.Series
     s2 = s2.astype(str)
     strings = s2.unique()
     tokens_per_string = pd.Series(
-        list(sum(count_occurrences(s, str(token)) for token in s1) for s in strings),
+        [sum(count_occurrences(s, str(token)) for token in s1) for s in strings],
         index=strings,
     )
 

diff --git a/tests/unit/test_column_noise.py b/tests/unit/test_column_noise.py
@@ -318,8 +318,6 @@ def test_leave_blank(dataset: Dataset, fuzzy_checker: FuzzyChecker) -> None:
     noised_data: pd.Series = dataset.data[column_name]
     # Calculate newly missing data, ie data that didn't come in as already missing
     data = data.squeeze()
-    # mypy can't know that squeeze() will not produce a scalar value
-    noised_data = noised_data.squeeze()  # type: ignore [assignment]
     is_not_missing = (data.notna()) & (data != "")
     orig_non_missing_idx = is_not_missing[is_not_missing].index
     newly_missing_idx = noised_data.index[