[PYTHON] Use isinstance() instead of type() == for type checks in pandas/types.py

Replace all `type(dt) == SomeType` comparisons with `isinstance(dt, SomeType)` in `to_arrow_type`, `_check_arrow_array_timestamps_localize`, and `_to_corrected_pandas_type` for correctness with subclasses.

Co-authored-by: Isaac

Author: zhengruifeng

python/pyspark/sql/pandas/types.py

@@ -116,38 +116,38 @@ def to_arrow_type(
     """
     import pyarrow as pa
 
-    if type(dt) == BooleanType:
+    if isinstance(dt, BooleanType):
         arrow_type = pa.bool_()
-    elif type(dt) == ByteType:
+    elif isinstance(dt, ByteType):
         arrow_type = pa.int8()
-    elif type(dt) == ShortType:
+    elif isinstance(dt, ShortType):
         arrow_type = pa.int16()
-    elif type(dt) == IntegerType:
+    elif isinstance(dt, IntegerType):
         arrow_type = pa.int32()
-    elif type(dt) == LongType:
+    elif isinstance(dt, LongType):
         arrow_type = pa.int64()
-    elif type(dt) == FloatType:
+    elif isinstance(dt, FloatType):
         arrow_type = pa.float32()
-    elif type(dt) == DoubleType:
+    elif isinstance(dt, DoubleType):
         arrow_type = pa.float64()
-    elif type(dt) == DecimalType:
+    elif isinstance(dt, DecimalType):
         arrow_type = pa.decimal128(dt.precision, dt.scale)
-    elif type(dt) == StringType:
+    elif isinstance(dt, StringType):
         arrow_type = pa.large_string() if prefers_large_types else pa.string()
-    elif type(dt) == BinaryType:
+    elif isinstance(dt, BinaryType):
         arrow_type = pa.large_binary() if prefers_large_types else pa.binary()
-    elif type(dt) == DateType:
+    elif isinstance(dt, DateType):
         arrow_type = pa.date32()
-    elif type(dt) == TimestampType:
+    elif isinstance(dt, TimestampType):
         assert timezone is not None
         arrow_type = pa.timestamp("us", tz=timezone)
-    elif type(dt) == TimestampNTZType:
+    elif isinstance(dt, TimestampNTZType):
         arrow_type = pa.timestamp("us", tz=None)
-    elif type(dt) == DayTimeIntervalType:
+    elif isinstance(dt, DayTimeIntervalType):
         arrow_type = pa.duration("us")
-    elif type(dt) == TimeType:
+    elif isinstance(dt, TimeType):
         arrow_type = pa.time64("ns")
-    elif type(dt) == ArrayType:
+    elif isinstance(dt, ArrayType):
         field = pa.field(
             "element",
             to_arrow_type(
@@ -159,7 +159,7 @@ def to_arrow_type(
             nullable=dt.containsNull,
         )
         arrow_type = pa.list_(field)
-    elif type(dt) == MapType:
+    elif isinstance(dt, MapType):
         key_field = pa.field(
             "key",
             to_arrow_type(
@@ -181,7 +181,7 @@ def to_arrow_type(
             nullable=dt.valueContainsNull,
         )
         arrow_type = pa.map_(key_field, value_field)
-    elif type(dt) == StructType:
+    elif isinstance(dt, StructType):
         field_names = dt.names
         if error_on_duplicated_field_names_in_struct and len(set(field_names)) != len(field_names):
             raise UnsupportedOperationException(
@@ -203,7 +203,7 @@ def to_arrow_type(
             for field in dt
         ]
         arrow_type = pa.struct(fields)
-    elif type(dt) == NullType:
+    elif isinstance(dt, NullType):
         arrow_type = pa.null()
     elif isinstance(dt, UserDefinedType):
         arrow_type = to_arrow_type(
@@ -212,15 +212,15 @@ def to_arrow_type(
             timezone=timezone,
             prefers_large_types=prefers_large_types,
         )
-    elif type(dt) == VariantType:
+    elif isinstance(dt, VariantType):
         fields = [
             pa.field("value", pa.binary(), nullable=False),
             # The metadata field is tagged so we can identify that the arrow struct actually
             # represents a variant.
             pa.field("metadata", pa.binary(), nullable=False, metadata={b"variant": b"true"}),
         ]
         arrow_type = pa.struct(fields)
-    elif type(dt) == GeometryType:
+    elif isinstance(dt, GeometryType):
         fields = [
             pa.field("srid", pa.int32(), nullable=False),
             pa.field(
@@ -231,7 +231,7 @@ def to_arrow_type(
             ),
         ]
         arrow_type = pa.struct(fields)
-    elif type(dt) == GeographyType:
+    elif isinstance(dt, GeographyType):
         fields = [
             pa.field("srid", pa.int32(), nullable=False),
             pa.field(
@@ -545,7 +545,7 @@ def _check_arrow_array_timestamps_localize(
     if types.is_timestamp(a.type) and truncate and a.type.unit == "ns":
         a = pc.floor_temporal(a, unit="microsecond")
 
-    if types.is_timestamp(a.type) and a.type.tz is None and type(dt) == TimestampType:
+    if types.is_timestamp(a.type) and a.type.tz is None and isinstance(dt, TimestampType):
         assert timezone is not None
 
         # Only localize timestamps that will become Spark TimestampType columns.
@@ -867,31 +867,31 @@ def _to_corrected_pandas_type(dt: DataType) -> Optional[Any]:
     import numpy as np
     import pandas as pd
 
-    if type(dt) == ByteType:
+    if isinstance(dt, ByteType):
         return np.int8
-    elif type(dt) == ShortType:
+    elif isinstance(dt, ShortType):
         return np.int16
-    elif type(dt) == IntegerType:
+    elif isinstance(dt, IntegerType):
         return np.int32
-    elif type(dt) == LongType:
+    elif isinstance(dt, LongType):
         return np.int64
-    elif type(dt) == FloatType:
+    elif isinstance(dt, FloatType):
         return np.float32
-    elif type(dt) == DoubleType:
+    elif isinstance(dt, DoubleType):
         return np.float64
-    elif type(dt) == BooleanType:
+    elif isinstance(dt, BooleanType):
         return bool
-    elif type(dt) == TimestampType:
+    elif isinstance(dt, TimestampType):
         if LooseVersion(pd.__version__) < "3.0.0":
             return np.dtype("datetime64[ns]")
         else:
             return np.dtype("datetime64[us]")
-    elif type(dt) == TimestampNTZType:
+    elif isinstance(dt, TimestampNTZType):
         if LooseVersion(pd.__version__) < "3.0.0":
             return np.dtype("datetime64[ns]")
         else:
             return np.dtype("datetime64[us]")
-    elif type(dt) == DayTimeIntervalType:
+    elif isinstance(dt, DayTimeIntervalType):
         if LooseVersion(pd.__version__) < "3.0.0":
             return np.dtype("timedelta64[ns]")
         else: