/
WordCount.java
187 lines (168 loc) · 9.26 KB
/
WordCount.java
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
/*
* 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.flink.streaming.examples.wordcount;
import org.apache.flink.api.common.eventtime.WatermarkStrategy;
import org.apache.flink.api.common.functions.FlatMapFunction;
import org.apache.flink.api.common.serialization.SimpleStringEncoder;
import org.apache.flink.api.java.tuple.Tuple2;
import org.apache.flink.configuration.MemorySize;
import org.apache.flink.connector.file.sink.FileSink;
import org.apache.flink.connector.file.src.FileSource;
import org.apache.flink.connector.file.src.reader.TextLineInputFormat;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.api.functions.sink.filesystem.rollingpolicies.DefaultRollingPolicy;
import org.apache.flink.streaming.examples.wordcount.util.CLI;
import org.apache.flink.streaming.examples.wordcount.util.WordCountData;
import org.apache.flink.util.Collector;
import java.time.Duration;
/**
* Implements the "WordCount" program that computes a simple word occurrence histogram over text
* files. This Job can be executed in both streaming and batch execution modes.
*
* <p>The input is a [list of] plain text file[s] with lines separated by a newline character.
*
* <p>Usage:
*
* <ul>
* <li><code>--input <path></code>A list of input files and / or directories to read. If no
* input is provided, the program is run with default data from {@link WordCountData}.
* <li><code>--discovery-interval <duration></code>Turns the file reader into a continuous
* source that will monitor the provided input directories every interval and read any new
* files.
* <li><code>--output <path></code>The output directory where the Job will write the
* results. If no output path is provided, the Job will print the results to <code>stdout
* </code>.
* <li><code>--execution-mode <mode></code>The execution mode (BATCH, STREAMING, or
* AUTOMATIC) of this pipeline.
* </ul>
*
* <p>This example shows how to:
*
* <ul>
* <li>Write a simple Flink DataStream program
* <li>Use tuple data types
* <li>Write and use a user-defined function
* </ul>
*/
public class WordCount {
// *************************************************************************
// PROGRAM
// *************************************************************************
public static void main(String[] args) throws Exception {
final CLI params = CLI.fromArgs(args);
// Create the execution environment. This is the main entrypoint
// to building a Flink application.
final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
// Apache Flink’s unified approach to stream and batch processing means that a DataStream
// application executed over bounded input will produce the same final results regardless
// of the configured execution mode. It is important to note what final means here: a job
// executing in STREAMING mode might produce incremental updates (think upserts in
// a database) while in BATCH mode, it would only produce one final result at the end. The
// final result will be the same if interpreted correctly, but getting there can be
// different.
//
// The “classic” execution behavior of the DataStream API is called STREAMING execution
// mode. Applications should use streaming execution for unbounded jobs that require
// continuous incremental processing and are expected to stay online indefinitely.
//
// By enabling BATCH execution, we allow Flink to apply additional optimizations that we
// can only do when we know that our input is bounded. For example, different
// join/aggregation strategies can be used, in addition to a different shuffle
// implementation that allows more efficient task scheduling and failure recovery behavior.
//
// By setting the runtime mode to AUTOMATIC, Flink will choose BATCH if all sources
// are bounded and otherwise STREAMING.
env.setRuntimeMode(params.getExecutionMode());
// This optional step makes the input parameters
// available in the Flink UI.
env.getConfig().setGlobalJobParameters(params);
DataStream<String> text;
if (params.getInputs().isPresent()) {
// Create a new file source that will read files from a given set of directories.
// Each file will be processed as plain text and split based on newlines.
FileSource.FileSourceBuilder<String> builder =
FileSource.forRecordStreamFormat(
new TextLineInputFormat(), params.getInputs().get());
// If a discovery interval is provided, the source will
// continuously watch the given directories for new files.
params.getDiscoveryInterval().ifPresent(builder::monitorContinuously);
text = env.fromSource(builder.build(), WatermarkStrategy.noWatermarks(), "file-input");
} else {
text = env.fromData(WordCountData.WORDS).name("in-memory-input");
}
DataStream<Tuple2<String, Integer>> counts =
// The text lines read from the source are split into words
// using a user-defined function. The tokenizer, implemented below,
// will output each word as a (2-tuple) containing (word, 1)
text.flatMap(new Tokenizer())
.name("tokenizer")
// keyBy groups tuples based on the "0" field, the word.
// Using a keyBy allows performing aggregations and other
// stateful transformations over data on a per-key basis.
// This is similar to a GROUP BY clause in a SQL query.
.keyBy(value -> value.f0)
// For each key, we perform a simple sum of the "1" field, the count.
// If the input data stream is bounded, sum will output a final count for
// each word. If it is unbounded, it will continuously output updates
// each time it sees a new instance of each word in the stream.
.sum(1)
.name("counter");
if (params.getOutput().isPresent()) {
// Given an output directory, Flink will write the results to a file
// using a simple string encoding. In a production environment, this might
// be something more structured like CSV, Avro, JSON, or Parquet.
counts.sinkTo(
FileSink.<Tuple2<String, Integer>>forRowFormat(
params.getOutput().get(), new SimpleStringEncoder<>())
.withRollingPolicy(
DefaultRollingPolicy.builder()
.withMaxPartSize(MemorySize.ofMebiBytes(1))
.withRolloverInterval(Duration.ofSeconds(10))
.build())
.build())
.name("file-sink");
} else {
counts.print().name("print-sink");
}
// Apache Flink applications are composed lazily. Calling execute
// submits the Job and begins processing.
env.execute("WordCount");
}
// *************************************************************************
// USER FUNCTIONS
// *************************************************************************
/**
* Implements the string tokenizer that splits sentences into words as a user-defined
* FlatMapFunction. The function takes a line (String) and splits it into multiple pairs in the
* form of "(word,1)" ({@code Tuple2<String, Integer>}).
*/
public static final class Tokenizer
implements FlatMapFunction<String, Tuple2<String, Integer>> {
@Override
public void flatMap(String value, Collector<Tuple2<String, Integer>> out) {
// normalize and split the line
String[] tokens = value.toLowerCase().split("\\W+");
// emit the pairs
for (String token : tokens) {
if (token.length() > 0) {
out.collect(new Tuple2<>(token, 1));
}
}
}
}
}