fixed: formatted with black

dfschema/core/column.py

@@ -52,16 +52,16 @@ def _validate_column_presence(
             raise DataFrameValidationError(text)
 
 
-def _is_string(series: pd.Series, strict:bool=False) -> bool:
-    '''Check if series is string-like
+def _is_string(series: pd.Series, strict: bool = False) -> bool:
+    """Check if series is string-like
     NOTE: Pandas 2 does not accept object dtype as string;
-    THis is a workaround. 
+    THis is a workaround.
     TODO: explicitly check for object dtype and raise warning
-    '''
+    """
     result = pd.api.types.is_string_dtype(series) or pd.isnull(series).all()
     if strict:
         return result
-    
+
     object_like = pd.api.types.is_object_dtype(series)
 
     return result or object_like
@@ -227,7 +227,7 @@ class ColSchema(BaseModel):
         lt=1.0,
         description="limit of missing values. If set to true, will raise if all values are empty. If set to a number, will raise if more than given perecnt of values are empty (Nan)",
         alias="na_limit",
-        alias_priority=2
+        alias_priority=2,
     )
 
     value_limits: Optional[ValueLimits] = Field(
@@ -264,9 +264,6 @@ def _map_dtype(
                 warn(f"Unsupported dtype: {self.dtype}")
         return None
 
-
-
-
     # abstract dtypes with a corresponding checker
     _dtype_test_func = {
         "numeric": pd.api.types.is_numeric_dtype,

dfschema/core/core.py

@@ -282,7 +282,10 @@ def from_df(
         """
 
         schema = generate_schema_dict_from_df(df)
-        schema["metadata"] = {"protocol_version": 2.0, "version": datetime.now().isoformat()}
+        schema["metadata"] = {
+            "protocol_version": 2.0,
+            "version": datetime.now().isoformat(),
+        }
         subset_schemas = []
         if subset_predicates:
             for predicate in subset_predicates:
@@ -296,7 +299,7 @@ def from_df(
 
         if return_dict:
             return schema
-        
+
         return cls(**schema)
 
 

dfschema/core/dtype.py

@@ -3,6 +3,7 @@
 mostly based on pandas.core.dtypes.dtypes.
 https://pandas.pydata.org/docs/user_guide/basics.html#basics-dtypes
 """
+
 import sys
 
 if sys.version_info >= (3, 8):

dfschema/core/generate.py

@@ -1,5 +1,6 @@
 import pandas as pd
 
+
 def generate_schema_dict_from_df(df: pd.DataFrame) -> dict:
     """
     generate Schema object from given dataframe.

pyproject.toml

@@ -42,6 +42,7 @@ MarkupSafe = "2.0.1"
 jupyter = "^1.0.0"
 watermark = "^2.3.1"
 gitchangelog = "^3.0.4"
+black = "^24.8.0"
 
 [build-system]
 requires = ["poetry-core>=1.0.0"]

tests/test_core.py

@@ -2,7 +2,6 @@
 from pydantic.v1.error_wrappers import ValidationError
 
 
-
 def test_schema_objects(good_schema: dict):
     from dfschema.core.core import DfSchema
 

tests/test_str_patterns.py

@@ -11,7 +11,7 @@ def str_df() -> pd.DataFrame:
             "z": ["A-1", "A-10", "A-12", None],
         }
     )
-    
+
     for col in df.columns:
         df[col] = df[col].astype("string")
     return df

tests/test_validate.py

@@ -146,23 +146,25 @@ def test_validate_df3_raises(df3, schema):
         validate(df3, schema)
 
 
-
 prediction_good_schemas = [
     Path(__file__).parent / "test_schemas/v2/good/v2_predictions.json",
-    Path(__file__).parent / "test_schemas/v2/good/v2_predictions2.json"
+    Path(__file__).parent / "test_schemas/v2/good/v2_predictions2.json",
 ]
 
 
 @pytest.fixture
 def df4():
     import pandas as pd
-    df = pd.DataFrame({
-        "value": [100000, 200000, 300000, 500000],
-        "inferred_at": ["2020-01-01", "2020-01-02", "2020-01-03", "2020-01-03"],
-        "trained_at": ["2020-01-01", "2020-01-02", "2020-01-03", "2020-01-03"],
-        "model": ["main", "certainty_low", "certainty_hight", "certainty_median"],
-        "version": ["1.0.0", "1.0.1", "1.0.2", "1.0.3"],
-    })
+
+    df = pd.DataFrame(
+        {
+            "value": [100000, 200000, 300000, 500000],
+            "inferred_at": ["2020-01-01", "2020-01-02", "2020-01-03", "2020-01-03"],
+            "trained_at": ["2020-01-01", "2020-01-02", "2020-01-03", "2020-01-03"],
+            "model": ["main", "certainty_low", "certainty_hight", "certainty_median"],
+            "version": ["1.0.0", "1.0.1", "1.0.2", "1.0.3"],
+        }
+    )
     for col in ["inferred_at", "trained_at"]:
         df[col] = pd.to_datetime(df[col])
     return df
@@ -171,5 +173,6 @@ def df4():
 @pytest.mark.parametrize("schema", prediction_good_schemas)
 def test_validate_df4(df4, schema):
     from dfschema import DfSchema
+
     schema = DfSchema.from_file(schema)
-    schema.validate_df(df4)
+    schema.validate_df(df4)