Tao Li created BEAM-11930:
-----------------------------
Summary: Bug with "over window" query from Beam SQL
Key: BEAM-11930
URL: https://issues.apache.org/jira/browse/BEAM-11930
Project: Beam
Issue Type: Bug
Components: sdk-java-core
Reporter: Tao Li
The problem is that I don’t see the alias column “agg” in the query result's
schema.
This is my code (with Beam 2.28):
{noformat}
Schema appSchema = Schema
.builder()
.addInt32Field("foo")
.addInt32Field("bar")
.build();
Row rowOne = Row.withSchema(appSchema).addValues(1, 1).build();
Row rowTwo = Row.withSchema(appSchema).addValues(1, 2).build();
PCollection<Row> inputRows = pipeline
.apply(Create.of(rowOne, rowTwo))
.setRowSchema(appSchema);
String sql = "SELECT foo, bar, RANK() over (PARTITION BY foo ORDER BY
bar) AS agg FROM PCOLLECTION";
PCollection<Row> result = inputRows.apply("sql",
SqlTransform.query(sql));
{noformat}
The Beam schema of the query result is below. Row name is "w0$o0" instead of my
specified name "agg".
Field{name=foo, description=, type=FieldType{typeName=INT32, nullable=false,
logicalType=null, collectionElementType=null, mapKeyType=null,
mapValueType=null, rowSchema=null, metadata={}}, options={{}}}
Field{name=bar, description=, type=FieldType{typeName=INT32, nullable=false,
logicalType=null, collectionElementType=null, mapKeyType=null,
mapValueType=null, rowSchema=null, metadata={}}, options={{}}}
Field{name=w0$o0, description=, type=FieldType{typeName=INT64, nullable=false,
logicalType=null, collectionElementType=null, mapKeyType=null,
mapValueType=null, rowSchema=null, metadata={}}, options={{}}}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
