2727F32_MIN = float (np .finfo (np .float32 ).min )
2828F32_MAX = float (np .finfo (np .float32 ).max )
2929
30- def dominant_timezone_per_column (schemas : list [pa .Schema ]) -> dict [str , tuple [str | None , str | None ]]:
30+
31+ def dominant_timezone_per_column (
32+ schemas : list [pa .Schema ],
33+ ) -> dict [str , tuple [str | None , str | None ]]:
3134 """
3235 For each timestamp column (by name) across all schemas, detect the most frequent timezone (including None).
3336 If None and a timezone are tied, prefer the timezone.
@@ -64,7 +67,10 @@ def dominant_timezone_per_column(schemas: list[pa.Schema]) -> dict[str, tuple[st
6467 dominant [name ] = (units [name ], tz )
6568 return dominant
6669
67- def standardize_schema_timezones_by_majority (schemas :list [pa .Schema ]) -> list [pa .Schema ]:
70+
71+ def standardize_schema_timezones_by_majority (
72+ schemas : list [pa .Schema ],
73+ ) -> list [pa .Schema ]:
6874 """
6975 For each timestamp column (by name) across all schemas, set the timezone to the most frequent (with tie-breaking).
7076 Returns a new list of schemas with updated timestamp timezones.
@@ -89,7 +95,10 @@ def standardize_schema_timezones_by_majority(schemas:list[pa.Schema]) -> list[pa
8995 new_schemas .append (pa .schema (fields , schema .metadata ))
9096 return new_schemas
9197
92- def standardize_schema_timezones (schemas :list [pa .Schema ], timezone :str | None = None ) -> list [pa .Schema ]:
98+
99+ def standardize_schema_timezones (
100+ schemas : list [pa .Schema ], timezone : str | None = None
101+ ) -> list [pa .Schema ]:
93102 """
94103 Standardize timezone info for all timestamp columns in a list of PyArrow schemas.
95104
@@ -125,7 +134,10 @@ def standardize_schema_timezones(schemas:list[pa.Schema], timezone:str|None=None
125134
126135
127136def unify_schemas (
128- schemas : list [pa .Schema ], use_large_dtypes : bool = False , timezone : str | None = None , standardize_timezones : bool = True
137+ schemas : list [pa .Schema ],
138+ use_large_dtypes : bool = False ,
139+ timezone : str | None = None ,
140+ standardize_timezones : bool = True ,
129141) -> pa .Schema :
130142 """
131143 Unify a list of PyArrow schemas into a single schema.
0 commit comments