4040if TYPE_CHECKING :
4141 import pathlib
4242
43- import pandas
44- import polars
45- import pyarrow
43+ import pandas as pd
44+ import polars as pl
45+ import pyarrow as pa
4646
4747 from datafusion .plan import ExecutionPlan , LogicalPlan
4848
@@ -537,7 +537,7 @@ def register_listing_table(
537537 path : str | pathlib .Path ,
538538 table_partition_cols : list [tuple [str , str ]] | None = None ,
539539 file_extension : str = ".parquet" ,
540- schema : pyarrow .Schema | None = None ,
540+ schema : pa .Schema | None = None ,
541541 file_sort_order : list [list [Expr | SortExpr ]] | None = None ,
542542 ) -> None :
543543 """Register multiple files as a single table.
@@ -606,14 +606,14 @@ def sql_with_options(self, query: str, options: SQLOptions) -> DataFrame:
606606
607607 def create_dataframe (
608608 self ,
609- partitions : list [list [pyarrow .RecordBatch ]],
609+ partitions : list [list [pa .RecordBatch ]],
610610 name : str | None = None ,
611- schema : pyarrow .Schema | None = None ,
611+ schema : pa .Schema | None = None ,
612612 ) -> DataFrame :
613613 """Create and return a dataframe using the provided partitions.
614614
615615 Args:
616- partitions: :py:class:`pyarrow .RecordBatch` partitions to register.
616+ partitions: :py:class:`pa .RecordBatch` partitions to register.
617617 name: Resultant dataframe name.
618618 schema: Schema for the partitions.
619619
@@ -684,16 +684,14 @@ def from_arrow(
684684 return DataFrame (self .ctx .from_arrow (data , name ))
685685
686686 @deprecated ("Use ``from_arrow`` instead." )
687- def from_arrow_table (
688- self , data : pyarrow .Table , name : str | None = None
689- ) -> DataFrame :
687+ def from_arrow_table (self , data : pa .Table , name : str | None = None ) -> DataFrame :
690688 """Create a :py:class:`~datafusion.dataframe.DataFrame` from an Arrow table.
691689
692690 This is an alias for :py:func:`from_arrow`.
693691 """
694692 return self .from_arrow (data , name )
695693
696- def from_pandas (self , data : pandas .DataFrame , name : str | None = None ) -> DataFrame :
694+ def from_pandas (self , data : pd .DataFrame , name : str | None = None ) -> DataFrame :
697695 """Create a :py:class:`~datafusion.dataframe.DataFrame` from a Pandas DataFrame.
698696
699697 Args:
@@ -705,7 +703,7 @@ def from_pandas(self, data: pandas.DataFrame, name: str | None = None) -> DataFr
705703 """
706704 return DataFrame (self .ctx .from_pandas (data , name ))
707705
708- def from_polars (self , data : polars .DataFrame , name : str | None = None ) -> DataFrame :
706+ def from_polars (self , data : pl .DataFrame , name : str | None = None ) -> DataFrame :
709707 """Create a :py:class:`~datafusion.dataframe.DataFrame` from a Polars DataFrame.
710708
711709 Args:
@@ -755,7 +753,7 @@ def register_table_provider(
755753 self .ctx .register_table_provider (name , provider )
756754
757755 def register_record_batches (
758- self , name : str , partitions : list [list [pyarrow .RecordBatch ]]
756+ self , name : str , partitions : list [list [pa .RecordBatch ]]
759757 ) -> None :
760758 """Register record batches as a table.
761759
@@ -776,7 +774,7 @@ def register_parquet(
776774 parquet_pruning : bool = True ,
777775 file_extension : str = ".parquet" ,
778776 skip_metadata : bool = True ,
779- schema : pyarrow .Schema | None = None ,
777+ schema : pa .Schema | None = None ,
780778 file_sort_order : list [list [SortExpr ]] | None = None ,
781779 ) -> None :
782780 """Register a Parquet file as a table.
@@ -817,7 +815,7 @@ def register_csv(
817815 self ,
818816 name : str ,
819817 path : str | pathlib .Path | list [str | pathlib .Path ],
820- schema : pyarrow .Schema | None = None ,
818+ schema : pa .Schema | None = None ,
821819 has_header : bool = True ,
822820 delimiter : str = "," ,
823821 schema_infer_max_records : int = 1000 ,
@@ -863,7 +861,7 @@ def register_json(
863861 self ,
864862 name : str ,
865863 path : str | pathlib .Path ,
866- schema : pyarrow .Schema | None = None ,
864+ schema : pa .Schema | None = None ,
867865 schema_infer_max_records : int = 1000 ,
868866 file_extension : str = ".json" ,
869867 table_partition_cols : list [tuple [str , str ]] | None = None ,
@@ -901,7 +899,7 @@ def register_avro(
901899 self ,
902900 name : str ,
903901 path : str | pathlib .Path ,
904- schema : pyarrow .Schema | None = None ,
902+ schema : pa .Schema | None = None ,
905903 file_extension : str = ".avro" ,
906904 table_partition_cols : list [tuple [str , str ]] | None = None ,
907905 ) -> None :
@@ -923,8 +921,8 @@ def register_avro(
923921 name , str (path ), schema , file_extension , table_partition_cols
924922 )
925923
926- def register_dataset (self , name : str , dataset : pyarrow .dataset .Dataset ) -> None :
927- """Register a :py:class:`pyarrow .dataset.Dataset` as a table.
924+ def register_dataset (self , name : str , dataset : pa .dataset .Dataset ) -> None :
925+ """Register a :py:class:`pa .dataset.Dataset` as a table.
928926
929927 Args:
930928 name: Name of the table to register.
@@ -975,7 +973,7 @@ def session_id(self) -> str:
975973 def read_json (
976974 self ,
977975 path : str | pathlib .Path ,
978- schema : pyarrow .Schema | None = None ,
976+ schema : pa .Schema | None = None ,
979977 schema_infer_max_records : int = 1000 ,
980978 file_extension : str = ".json" ,
981979 table_partition_cols : list [tuple [str , str ]] | None = None ,
@@ -1012,7 +1010,7 @@ def read_json(
10121010 def read_csv (
10131011 self ,
10141012 path : str | pathlib .Path | list [str ] | list [pathlib .Path ],
1015- schema : pyarrow .Schema | None = None ,
1013+ schema : pa .Schema | None = None ,
10161014 has_header : bool = True ,
10171015 delimiter : str = "," ,
10181016 schema_infer_max_records : int = 1000 ,
@@ -1065,7 +1063,7 @@ def read_parquet(
10651063 parquet_pruning : bool = True ,
10661064 file_extension : str = ".parquet" ,
10671065 skip_metadata : bool = True ,
1068- schema : pyarrow .Schema | None = None ,
1066+ schema : pa .Schema | None = None ,
10691067 file_sort_order : list [list [Expr | SortExpr ]] | None = None ,
10701068 ) -> DataFrame :
10711069 """Read a Parquet source into a :py:class:`~datafusion.dataframe.Dataframe`.
@@ -1110,7 +1108,7 @@ def read_parquet(
11101108 def read_avro (
11111109 self ,
11121110 path : str | pathlib .Path ,
1113- schema : pyarrow .Schema | None = None ,
1111+ schema : pa .Schema | None = None ,
11141112 file_partition_cols : list [tuple [str , str ]] | None = None ,
11151113 file_extension : str = ".avro" ,
11161114 ) -> DataFrame :
0 commit comments