ARROW-16651 : [Python] Casting Table to new schema ignores nullabilit…

…y of fields (apache#14048)

```python
table = pa.table({'a': [None, 1], 'b': [None, True]})
new_schema = pa.schema([pa.field("a", "int64", nullable=True), pa.field("b", "bool", nullable=False)])
casted = table.cast(new_schema)

```

Now leads to
```
RuntimeError: Casting field 'b' with null values to non-nullable
```

Authored-by: kshitij12345 <kshitijkalambarkar@gmail.com>
Signed-off-by: Joris Van den Bossche <jorisvandenbossche@gmail.com>

python/pyarrow/table.pxi

@@ -3401,6 +3401,9 @@ cdef class Table(_PandasConvertible):
                              .format(self.schema.names, target_schema.names))
 
         for column, field in zip(self.itercolumns(), target_schema):
+            if not field.nullable and column.null_count > 0:
+                raise ValueError("Casting field {!r} with null values to non-nullable"
+                                 .format(field.name))
             casted = column.cast(field.type, safe=safe, options=options)
             newcols.append(casted)
 

python/pyarrow/tests/test_table.py

@@ -2192,3 +2192,15 @@ def test_table_join_many_columns():
         "col6": ["A", "B", None, "Z"],
         "col7": ["A", "B", None, "Z"],
     })
+
+
+def test_table_cast_invalid():
+    # Casting a nullable field to non-nullable should be invalid!
+    table = pa.table({'a': [None, 1], 'b': [None, True]})
+    new_schema = pa.schema([pa.field("a", "int64", nullable=True),
+                            pa.field("b", "bool", nullable=False)])
+    with pytest.raises(ValueError):
+        table.cast(new_schema)
+
+    table = pa.table({'a': [None, 1], 'b': [False, True]})
+    assert table.cast(new_schema).schema == new_schema