refactor: Store Enum categories as list instead of pl.Series (#150)

Author: AndreasAlbertQC

dataframely/columns/_base.py

@@ -381,15 +381,6 @@ def _attributes_match(
         if name == "check":
             return _compare_checks(lhs, rhs, column_expr)
 
-        lhs_is_series = isinstance(lhs, pl.Series)
-        rhs_is_series = isinstance(rhs, pl.Series)
-
-        if lhs_is_series != rhs_is_series:
-            return False
-
-        if lhs_is_series and rhs_is_series:
-            return _compare_series(lhs, rhs)
-
         return lhs == rhs
 
     # -------------------------------- DUNDER METHODS -------------------------------- #
@@ -413,10 +404,6 @@ def __str__(self) -> str:
         return self.__class__.__name__.lower()
 
 
-def _compare_series(lhs: pl.Series, rhs: pl.Series) -> bool:
-    return (len(lhs) == len(rhs)) and lhs.equals(rhs)
-
-
 def _compare_checks(lhs: Check | None, rhs: Check | None, expr: pl.Expr) -> bool:
     match (lhs, rhs):
         case (None, None):

dataframely/columns/enum.py

@@ -67,12 +67,8 @@ def __init__(
             metadata=metadata,
         )
         if isclass(categories) and issubclass(categories, enum.Enum):
-            categories = pl.Series(
-                values=[getattr(v, "value", v) for v in categories.__members__.values()]
-            )
-        elif not isinstance(categories, pl.Series):
-            categories = pl.Series(values=categories)
-        self.categories = categories
+            categories = (item.value for item in categories)
+        self.categories = list(categories)
 
     @property
     def dtype(self) -> pl.DataType:
@@ -81,7 +77,7 @@ def dtype(self) -> pl.DataType:
     def validate_dtype(self, dtype: PolarsDataType) -> bool:
         if not isinstance(dtype, pl.Enum):
             return False
-        return self.categories.equals(dtype.categories)
+        return self.categories == dtype.categories.to_list()
 
     def sqlalchemy_dtype(self, dialect: sa.Dialect) -> sa_TypeEngine:
         category_lengths = [len(c) for c in self.categories]
@@ -102,6 +98,6 @@ def pyarrow_dtype(self) -> pa.DataType:
     def _sample_unchecked(self, generator: Generator, n: int) -> pl.Series:
         return generator.sample_choice(
             n,
-            choices=self.categories.to_list(),
+            choices=self.categories,
             null_probability=self._null_probability,
         ).cast(self.dtype)