Github user paul-rogers commented on a diff in the pull request:
https://github.com/apache/drill/pull/1121#discussion_r168675063
--- Diff:
exec/java-exec/src/main/java/org/apache/drill/exec/physical/impl/protocol/OperatorDriver.java
---
@@ -0,0 +1,183 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.drill.exec.physical.impl.protocol;
+
+import org.apache.drill.common.exceptions.UserException;
+import org.apache.drill.exec.ops.OperatorContext;
+import org.apache.drill.exec.record.RecordBatch.IterOutcome;
+
+/**
+ * State machine that drives the operator executable. Converts
+ * between the iterator protocol and the operator executable protocol.
+ * Implemented as a separate class in anticipation of eventually
+ * changing the record batch (iterator) protocol.
+ */
+
+public class OperatorDriver {
+ public enum State { START, SCHEMA, RUN, END, FAILED, CLOSED }
--- End diff --
Next, the question about the `END` and `FAILED` states. `END` means we
reached EOF and have no data left to deliver. Calling `next()` in the `END`
state simply returns `DONE` (there is *still* no more data.)
On the other hand, `FAILED` indicates that an error occurred. This seemed
like a useful thing to know. But, as I review this version of the code, I see
we are not actually using this state. So, rather than add a `CANCELLED` state,
maybe we can collapse `FAILED` into `END`: which is just a signal to `next()`
to return `DONE`.
Cleanup should be no different in the three cases: we must release all
resources regardless of the reason that `close()` is called.
Thoughts?
---
