Revert fastparquet nullable dtype support (#42954)

Author: lithomas1

doc/source/whatsnew/v1.3.2.rst

@@ -45,7 +45,6 @@ Bug fixes
 
 Other
 ~~~~~
-- :meth:`pandas.read_parquet` now supports reading nullable dtypes with ``fastparquet`` versions above 0.7.1.
 -
 
 .. ---------------------------------------------------------------------------

pandas/io/parquet.py

@@ -309,20 +309,16 @@ def write(
     def read(
         self, path, columns=None, storage_options: StorageOptions = None, **kwargs
     ):
-        parquet_kwargs = {}
+        parquet_kwargs: dict[str, Any] = {}
         use_nullable_dtypes = kwargs.pop("use_nullable_dtypes", False)
-        # Technically works with 0.7.0, but was incorrect
-        # so lets just require 0.7.1
         if Version(self.api.__version__) >= Version("0.7.1"):
-            # Need to set even for use_nullable_dtypes = False,
-            # since our defaults differ
-            parquet_kwargs["pandas_nulls"] = use_nullable_dtypes
-        else:
-            if use_nullable_dtypes:
-                raise ValueError(
-                    "The 'use_nullable_dtypes' argument is not supported for the "
-                    "fastparquet engine for fastparquet versions less than 0.7.1"
-                )
+            # We are disabling nullable dtypes for fastparquet pending discussion
+            parquet_kwargs["pandas_nulls"] = False
+        if use_nullable_dtypes:
+            raise ValueError(
+                "The 'use_nullable_dtypes' argument is not supported for the "
+                "fastparquet engine"
+            )
         path = stringify_path(path)
         handles = None
         if is_fsspec_url(path):
@@ -478,18 +474,15 @@ def read_parquet(
 
     use_nullable_dtypes : bool, default False
         If True, use dtypes that use ``pd.NA`` as missing value indicator
-        for the resulting DataFrame.
+        for the resulting DataFrame. (only applicable for the ``pyarrow``
+        engine)
         As new dtypes are added that support ``pd.NA`` in the future, the
         output with this option will change to use those dtypes.
         Note: this is an experimental option, and behaviour (e.g. additional
         support dtypes) may change without notice.
 
         .. versionadded:: 1.2.0
 
-        .. versionchanged:: 1.3.2
-            ``use_nullable_dtypes`` now works with the the ``fastparquet`` engine
-            if ``fastparquet`` is version 0.7.1 or higher.
-
     **kwargs
         Any additional kwargs are passed to the engine.
 

pandas/tests/io/test_parquet.py

@@ -600,18 +600,18 @@ def test_use_nullable_dtypes(self, engine):
         import pyarrow.parquet as pq
 
         if engine == "fastparquet":
-            pytest.importorskip(
-                "fastparquet",
-                "0.7.1",
-                reason="fastparquet must be 0.7.1 or higher for nullable dtype support",
-            )
+            # We are manually disabling fastparquet's
+            # nullable dtype support pending discussion
+            pytest.skip("Fastparquet nullable dtype support is disabled")
 
         table = pyarrow.table(
             {
                 "a": pyarrow.array([1, 2, 3, None], "int64"),
                 "b": pyarrow.array([1, 2, 3, None], "uint8"),
                 "c": pyarrow.array(["a", "b", "c", None]),
                 "d": pyarrow.array([True, False, True, None]),
+                # Test that nullable dtypes used even in absence of nulls
+                "e": pyarrow.array([1, 2, 3, 4], "int64"),
             }
         )
         with tm.ensure_clean() as path:
@@ -627,6 +627,7 @@ def test_use_nullable_dtypes(self, engine):
                 "b": pd.array([1, 2, 3, None], dtype="UInt8"),
                 "c": pd.array(["a", "b", "c", None], dtype="string"),
                 "d": pd.array([True, False, True, None], dtype="boolean"),
+                "e": pd.array([1, 2, 3, 4], dtype="Int64"),
             }
         )
         if engine == "fastparquet":