[SPARK-42666] Fix `createDataFrame` to work properly with rows and schema

• Type: Sub-task
• Status: Resolved
• Priority: Major
• Resolution: Duplicate
• Affects Version/s: 3.5.0
• Fix Version/s: 3.4.0
• Component/s: Connect
• Assignee: Unassigned
• Reporter: Haejoon Lee
• Created: 03/Mar/23 20:40
• Updated: 15/Mar/23 03:17
• Resolved: 15/Mar/23 03:17

Description

The code below is not working properly in Spark Connect:

>>> sdf = spark.range(10)
>>> spark.createDataFrame(sdf.tail(5), sdf.schema) 
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/.../spark/python/pyspark/sql/connect/dataframe.py", line 94, in __repr__
    return "DataFrame[%s]" % (", ".join("%s: %s" % c for c in self.dtypes))
  File "/.../spark/python/pyspark/sql/connect/dataframe.py", line 162, in dtypes
    return [(str(f.name), f.dataType.simpleString()) for f in self.schema.fields]
  File "/.../spark/python/pyspark/sql/connect/dataframe.py", line 1346, in schema
    self._schema = self._session.client.schema(query)
  File "/.../spark/python/pyspark/sql/connect/client.py", line 614, in schema
    proto_schema = self._analyze(method="schema", plan=plan).schema
  File "/.../spark/python/pyspark/sql/connect/client.py", line 755, in _analyze
    self._handle_error(rpc_error)
  File "/.../spark/python/pyspark/sql/connect/client.py", line 894, in _handle_error
    raise convert_exception(info, status.message) from None
pyspark.errors.exceptions.connect.AnalysisException: [NULLABLE_COLUMN_OR_FIELD] Column or field `id` is nullable while it's required to be non-nullable.

whereas working properly in regular PySpark:

>>> sdf = spark.range(10)
>>> spark.createDataFrame(sdf.tail(5), sdf.schema).show()
+---+
| id|
+---+
|  5|
|  6|
|  7|
|  8|
|  9|
+---+ 

Issue Links

• is duplicated by: SPARK-42679 createDataFrame doesn't work with non-nullable schema.