fix test cover

Author: chelsea-lin

db_dtypes/json.py

@@ -81,8 +81,7 @@ def _box_pa(
         cls, value, pa_type: pa.DataType | None = None
     ) -> pa.Array | pa.ChunkedArray | pa.Scalar:
         """Box value into a pyarrow Array, ChunkedArray or Scalar."""
-        if pa_type is not None and pa_type != pa.string():
-            raise ValueError(f"Unsupported type '{pa_type}' for JSONArray")
+        assert pa_type is None or pa_type == pa.string()
 
         if isinstance(value, pa.Scalar) or not (
             common.is_list_like(value) and not common.is_dict_like(value)
@@ -93,8 +92,6 @@ def _box_pa(
     @classmethod
     def _box_pa_scalar(cls, value) -> pa.Scalar:
         """Box value into a pyarrow Scalar."""
-        if isinstance(value, pa.Scalar):
-            pa_scalar = value
         if pd.isna(value):
             pa_scalar = pa.scalar(None, type=pa.string())
         else:
@@ -104,33 +101,21 @@ def _box_pa_scalar(cls, value) -> pa.Scalar:
         return pa_scalar
 
     @classmethod
-    def _box_pa_array(
-        cls, value, pa_type: pa.DataType | None = None, copy: bool = False
-    ) -> pa.Array | pa.ChunkedArray:
+    def _box_pa_array(cls, value, copy: bool = False) -> pa.Array | pa.ChunkedArray:
         """Box value into a pyarrow Array or ChunkedArray."""
         if isinstance(value, cls):
             pa_array = value._pa_array
-        elif isinstance(value, (pa.Array, pa.ChunkedArray)):
-            pa_array = value
         else:
-            try:
-                value = [JSONArray._serialize_json(x) for x in value]
-                pa_array = pa.array(value, type=pa_type, from_pandas=True)
-            except (pa.ArrowInvalid, pa.ArrowTypeError):
-                # https://github.com/pandas-dev/pandas/pull/50430:
-                # let pyarrow infer type, then cast
-                pa_array = pa.array(value, from_pandas=True)
-
-        if pa_type is not None and pa_array.type != pa_type:
-            pa_array = pa_array.cast(pa_type)
-
+            value = [JSONArray._serialize_json(x) for x in value]
+            pa_array = pa.array(value, type=pa.string(), from_pandas=True)
         return pa_array
 
     @classmethod
     def _from_sequence(cls, scalars, *, dtype=None, copy=False):
         """Construct a new ExtensionArray from a sequence of scalars."""
-        result = [JSONArray._serialize_json(scalar) for scalar in scalars]
-        return cls(pa.array(result, type=pa.string(), from_pandas=True))
+        pa_array = cls._box_pa(scalars)
+        arr = cls(pa_array)
+        return arr
 
     @classmethod
     def _concat_same_type(cls, to_concat) -> JSONArray:
@@ -139,11 +124,6 @@ def _concat_same_type(cls, to_concat) -> JSONArray:
         arr = pa.chunked_array(chunks, type=pa.string())
         return cls(arr)
 
-    @classmethod
-    def _from_factorized(cls, values, original):
-        """Reconstruct an ExtensionArray after factorization."""
-        return cls._from_sequence(values, dtype=original.dtype)
-
     @staticmethod
     def _serialize_json(value):
         """A static method that converts a JSON value into a string representation."""
@@ -202,19 +182,6 @@ def __getitem__(self, item):
                 r"only integers, slices (`:`), ellipsis (`...`), numpy.newaxis "
                 r"(`None`) and integer or boolean arrays are valid indices"
             )
-        # We are not an array indexer, so maybe e.g. a slice or integer
-        # indexer. We dispatch to pyarrow.
-        if isinstance(item, slice):
-            # Arrow bug https://github.com/apache/arrow/issues/38768
-            if item.start == item.stop:
-                pass
-            elif (
-                item.stop is not None
-                and item.stop < -len(self)
-                and item.step is not None
-                and item.step < 0
-            ):
-                item = slice(item.start, None, item.step)
 
         value = self._pa_array[item]
         if isinstance(value, pa.ChunkedArray):
@@ -229,7 +196,8 @@ def __getitem__(self, item):
     def __iter__(self):
         """Iterate over elements of the array."""
         for value in self._pa_array:
-            val = JSONArray._deserialize_json(value.as_py())
+            val = value.as_py()
+            # val = JSONArray._deserialize_json(value.as_py())
             if val is None:
                 yield self._dtype.na_value
             else:

tests/unit/test_json.py

@@ -15,8 +15,8 @@
 
 import json
 
+import numpy as np
 import pandas as pd
-import pandas.testing
 import pytest
 
 import db_dtypes
@@ -45,7 +45,12 @@
 }
 
 
-def test_get_items():
+def test_construct_w_unspported_types():
+    with pytest.raises(ValueError):
+        db_dtypes.JSONArray(100)
+
+
+def test_getitems_return_json_objects():
     data = db_dtypes.JSONArray._from_sequence(JSON_DATA.values())
     for id, key in enumerate(JSON_DATA.keys()):
         if key == "null":
@@ -54,7 +59,7 @@ def test_get_items():
             assert data[id] == JSON_DATA[key]
 
 
-def test_get_items_unbox_object():
+def test_getitems_w_unboxed_dict():
     data = db_dtypes.JSONArray._from_sequence([JSON_DATA["dict"]])
     assert len(data[0]) == 2
 
@@ -67,6 +72,20 @@ def test_get_items_unbox_object():
         data[0]["unknown"]
 
 
+def test_getitems_w_invalid_numpy_array():
+    data = db_dtypes.JSONArray._from_sequence(JSON_DATA.values())
+    idx = np.array(["str"])
+    with pytest.raises(IndexError):
+        data[idx]
+
+
+def test_getitems_when_iter_with_null():
+    data = db_dtypes.JSONArray._from_sequence([JSON_DATA["null"]])
+    s = pd.Series(data)
+    result = s[:1].item()
+    assert pd.isna(result)
+
+
 def test_to_numpy():
     s = pd.Series(db_dtypes.JSONArray._from_sequence(JSON_DATA.values()))
     data = s.to_numpy()