HBASE-26688 Threads shared EMPTY_RESULT may lead to unexpected client job down. ...
[hbase.git] / hbase-client / src / main / java / org / apache / hadoop / hbase / client / AdvancedScanResultConsumer.java
blob10933abf3cf2202994fc0517c5314130e18639b0
1 /**
2 * Licensed to the Apache Software Foundation (ASF) under one
3 * or more contributor license agreements. See the NOTICE file
4 * distributed with this work for additional information
5 * regarding copyright ownership. The ASF licenses this file
6 * to you under the Apache License, Version 2.0 (the
7 * "License"); you may not use this file except in compliance
8 * with the License. You may obtain a copy of the License at
10 * http://www.apache.org/licenses/LICENSE-2.0
12 * Unless required by applicable law or agreed to in writing, software
13 * distributed under the License is distributed on an "AS IS" BASIS,
14 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
15 * See the License for the specific language governing permissions and
16 * limitations under the License.
18 package org.apache.hadoop.hbase.client;
20 import java.util.Optional;
22 import org.apache.yetus.audience.InterfaceAudience;
24 /**
25 * This is the low level API for asynchronous scan.
26 * <p>
27 * All results that match the given scan object will be passed to this class by calling
28 * {@link #onNext(Result[], ScanController)}. {@link #onComplete()} means the scan is finished, and
29 * {@link #onError(Throwable)} means we hit an unrecoverable error and the scan is terminated.
30 * {@link #onHeartbeat(ScanController)} means the RS is still working but we can not get a valid
31 * result to call {@link #onNext(Result[], ScanController)}. This is usually because the matched
32 * results are too sparse, for example, a filter which almost filters out everything is specified.
33 * <p>
34 * Notice that, all the methods here will be called directly in the thread which we send request to
35 * HBase service. So if you want the asynchronous scanner fetch data from HBase in background while
36 * you process the returned data, you need to move the processing work to another thread to make the
37 * {@link #onNext(Result[], ScanController)} call return immediately. And please do NOT do any time
38 * consuming tasks in these methods unless you know what you are doing.
39 * @since 2.0.0
41 @InterfaceAudience.Public
42 public interface AdvancedScanResultConsumer extends ScanResultConsumerBase {
44 /**
45 * Used to resume a scan.
47 @InterfaceAudience.Public
48 interface ScanResumer {
50 /**
51 * Resume the scan. You are free to call it multiple time but only the first call will take
52 * effect.
54 void resume();
57 /**
58 * Used to suspend or stop a scan, or get a scan cursor if available.
59 * <p>
60 * Notice that, you should only call the {@link #suspend()} or {@link #terminate()} inside onNext
61 * or onHeartbeat method. A IllegalStateException will be thrown if you call them at other places.
62 * <p>
63 * You can only call one of the {@link #suspend()} and {@link #terminate()} methods(of course you
64 * are free to not call them both), and the methods are not reentrant. An IllegalStateException
65 * will be thrown if you have already called one of the methods.
67 @InterfaceAudience.Public
68 interface ScanController {
70 /**
71 * Suspend the scan.
72 * <p>
73 * This means we will stop fetching data in background, i.e., will not call onNext any more
74 * before you resume the scan.
75 * @return A resumer used to resume the scan later.
77 ScanResumer suspend();
79 /**
80 * Terminate the scan.
81 * <p>
82 * This is useful when you have got enough results and want to stop the scan in onNext method,
83 * or you want to stop the scan in onHeartbeat method because it has spent too many time.
85 void terminate();
87 /**
88 * Get the scan cursor if available.
89 * @return The scan cursor.
91 Optional<Cursor> cursor();
94 /**
95 * Indicate that we have receive some data.
96 * @param results the data fetched from HBase service.
97 * @param controller used to suspend or terminate the scan. Notice that the {@code controller}
98 * instance is only valid within scope of onNext method. You can only call its method in
99 * onNext, do NOT store it and call it later outside onNext.
101 void onNext(Result[] results, ScanController controller);
104 * Indicate that there is a heartbeat message but we have not cumulated enough cells to call
105 * {@link #onNext(Result[], ScanController)}.
106 * <p>
107 * Note that this method will always be called when RS returns something to us but we do not have
108 * enough cells to call {@link #onNext(Result[], ScanController)}. Sometimes it may not be a
109 * 'heartbeat' message for RS, for example, we have a large row with many cells and size limit is
110 * exceeded before sending all the cells for this row. For RS it does send some data to us and the
111 * time limit has not been reached, but we can not return the data to client so here we call this
112 * method to tell client we have already received something.
113 * <p>
114 * This method give you a chance to terminate a slow scan operation.
115 * @param controller used to suspend or terminate the scan. Notice that the {@code controller}
116 * instance is only valid within the scope of onHeartbeat method. You can only call its
117 * method in onHeartbeat, do NOT store it and call it later outside onHeartbeat.
119 default void onHeartbeat(ScanController controller) {