BUG: read_json raising with table orient and NA values (#50332)

* BUG: read_json raising with table orient and NA values

* Add test

Author: phofl

doc/source/whatsnew/v2.0.0.rst

@@ -879,6 +879,7 @@ I/O
 - Bug when a pickling a subset PyArrow-backed data that would serialize the entire data instead of the subset (:issue:`42600`)
 - Bug in :func:`read_sql_query` ignoring ``dtype`` argument when ``chunksize`` is specified and result is empty (:issue:`50245`)
 - Bug in :func:`read_csv` for a single-line csv with fewer columns than ``names`` raised :class:`.errors.ParserError` with ``engine="c"`` (:issue:`47566`)
+- Bug in :func:`read_json` raising with ``orient="table"`` and ``NA`` value (:issue:`40255`)
 - Bug in displaying ``string`` dtypes not showing storage option (:issue:`50099`)
 - Bug in :meth:`DataFrame.to_string` with ``header=False`` that printed the index name on the same line as the first row of the data (:issue:`49230`)
 - Bug in :meth:`DataFrame.to_string` ignoring float formatter for extension arrays (:issue:`39336`)

pandas/io/json/_table_schema.py

@@ -196,11 +196,11 @@ def convert_json_field_to_pandas_type(field) -> str | CategoricalDtype:
     if typ == "string":
         return "object"
     elif typ == "integer":
-        return "int64"
+        return field.get("extDtype", "int64")
     elif typ == "number":
-        return "float64"
+        return field.get("extDtype", "float64")
     elif typ == "boolean":
-        return "bool"
+        return field.get("extDtype", "bool")
     elif typ == "duration":
         return "timedelta64"
     elif typ == "datetime":

pandas/tests/io/json/test_json_table_schema_ext_dtype.py

@@ -8,9 +8,13 @@
 import pytest
 
 from pandas import (
+    NA,
     DataFrame,
+    Index,
     array,
+    read_json,
 )
+import pandas._testing as tm
 from pandas.core.arrays.integer import Int64Dtype
 from pandas.core.arrays.string_ import StringDtype
 from pandas.core.series import Series
@@ -273,3 +277,43 @@ def test_to_json(self, df):
         expected = OrderedDict([("schema", schema), ("data", data)])
 
         assert result == expected
+
+    def test_json_ext_dtype_reading_roundtrip(self):
+        # GH#40255
+        df = DataFrame(
+            {
+                "a": Series([2, NA], dtype="Int64"),
+                "b": Series([1.5, NA], dtype="Float64"),
+                "c": Series([True, NA], dtype="boolean"),
+            },
+            index=Index([1, NA], dtype="Int64"),
+        )
+        expected = df.copy()
+        data_json = df.to_json(orient="table", indent=4)
+        result = read_json(data_json, orient="table")
+        tm.assert_frame_equal(result, expected)
+
+    def test_json_ext_dtype_reading(self):
+        # GH#40255
+        data_json = """{
+            "schema":{
+                "fields":[
+                    {
+                        "name":"a",
+                        "type":"integer",
+                        "extDtype":"Int64"
+                    }
+                ],
+            },
+            "data":[
+                {
+                    "a":2
+                },
+                {
+                    "a":null
+                }
+            ]
+        }"""
+        result = read_json(data_json, orient="table")
+        expected = DataFrame({"a": Series([2, NA], dtype="Int64")})
+        tm.assert_frame_equal(result, expected)