MongoDB-JAVA-Driver 3.2版本常用代码全整理(3) - 聚合

MongoDB的3.x版本Java驱动相对2.x做了全新的设计，类库和使用方法上有很大区别。例如用Document替换BasicDBObject、通过Builders类构建Bson替代直接输入$命令等，本文整理了基于3.2版本的常用增删改查操作的使用方法。为了避免冗长的篇幅，分为增删改、查询、聚合、地理索引等几部分。

聚合用于统计文档个数、求和、最大最小值、求平均值等，功能和函数名称和SQL中的count、distinct、group等关键字非常类似，此外，还可以通过javascript编写MapReduce实现复杂的计算(性能损耗也会非常严重)。

首先来看3.x驱动中的聚合方法的声明：

AggregateIterable<TDocument> aggregate(List<? extends Bson> pipeline)

参数类型是一个Bson的列表，而参数名称是pipeline，其构建方式正如其名，是以多个Bson建立起一条管道，前一个Bson的输出将作为后一个Bson的输入，例如：

mc.aggregate(Arrays.asList(match(eq("owner", "tom")), group("$author", sum("totalWords", "$words"))));

首先用$match查找出owner=tom的文档，并将结果集传递给$group并对字数求和。

下面来看更多命令用法，用于演示的类的基本代码如下

import static com.mongodb.client.model.Accumulators.*;
import static com.mongodb.client.model.Aggregates.*;
import static com.mongodb.client.model.Filters.eq;

import java.text.ParseException;
import java.util.Arrays;

import org.bson.Document;

import com.mongodb.Block;
import com.mongodb.MongoClient;
import com.mongodb.client.AggregateIterable;
import com.mongodb.client.MongoCollection;
import com.mongodb.client.MongoDatabase;

public class AggregatesExamples {

	public static void main(String[] args) throws ParseException {
		//根据实际环境修改ip和端口
		MongoClient mongoClient = new MongoClient("localhost", 27017);
		MongoDatabase database = mongoClient.getDatabase("lesson");

		AggregatesExamples client = new AggregatesExamples(database);
		client.show();
		mongoClient.close();
	}

	private MongoDatabase database;
	public AggregatesExamples(MongoDatabase database) {
		this.database = database;
	}

	public void show() {
		MongoCollection<Document> mc = database.getCollection("blog");
		//每次执行前清空集合以方便重复运行
		mc.drop();

		//插入用于测试的文档
		Document doc1 = new Document("title", "good day").append("owner", "tom").append("words", 300)
				.append("comments", Arrays.asList(new Document("author", "joe").append("score", 3).append("comment", "good"), new Document("author", "white").append("score", 1).append("comment", "oh no")));
		Document doc2 = new Document("title", "good").append("owner", "john").append("words", 400)
				.append("comments", Arrays.asList(new Document("author", "william").append("score", 4).append("comment", "good"), new Document("author", "white").append("score", 6).append("comment", "very good")));
		Document doc3 = new Document("title", "good night").append("owner", "mike").append("words", 200)
				.append("tag", Arrays.asList(1, 2, 3, 4));
		Document doc4 = new Document("title", "happiness").append("owner", "tom").append("words", 1480)
				.append("tag", Arrays.asList(2, 3, 4));
		Document doc5 = new Document("title", "a good thing").append("owner", "tom").append("words", 180)
				.append("tag", Arrays.asList(1, 2, 3, 4, 5));
		mc.insertMany(Arrays.asList(doc1, doc2, doc3, doc4, doc5));

		AggregateIterable<Document> iterable = mc.aggregate(Arrays.asList(match(eq("owner", "tom")),
                group("$author", sum("totalWords", "$words"))));
		printResult("", iterable);

		//TODO: 将在这里填充更多聚合示例
	}

	//打印聚合结果
	public void printResult(String doing, AggregateIterable<Document> iterable) {
		System.out.println(doing);
        iterable.forEach(new Block<Document>() {
            public void apply(final Document document) {
                System.out.println(document);
            }
        });
        System.out.println("------------------------------------------------------");
        System.out.println();
	}
}

如上面代码所示，将把所有的聚合操作集中在show()方法中演示，并且在执行后打印结果集以观察执行结果。下面用常用的聚合代码填充show()方法

		// $match 确定复合条件的文档, 可组合多个条件
		iterable = mc.aggregate(Arrays.asList(match(and(eq("owner", "tom"), gt("words", 300)))));
		printResult("$match only", iterable);

		// $sum求和 $avg平均值 $max最大值 $min最小值
		iterable = mc.aggregate(Arrays.asList(
				match(in("owner", "tom", "john", "mike")),
				group("$owner", sum("totalWords", "$words"),
						avg("averageWords", "$words"),
						max("maxWords", "$words"), min("minWords", "$words"))));
		printResult("$sum $avg $max $min", iterable);

		// $out 把聚合结果输出到集合
		mc.aggregate(Arrays.asList(
				match(in("owner", "tom", "john", "mike")),
				group("$owner", sum("totalWords", "$words"),
						avg("averageWords", "$words"),
						max("maxWords", "$words"), min("minWords", "$words")),
				out("wordsCount")));
		iterable = database.getCollection("wordsCount").aggregate(
				Arrays.asList(sample(3)));
		printResult("$out", iterable);

		// 随机取3个文档, 仅返回title和owner字段
		iterable = mc.aggregate(Arrays.asList(sample(3),
				project(fields(include("title", "owner"), excludeId()))));
		printResult("sample(3)", iterable);

		// 从第2个文档开始取2个文档, 仅返回title和owner字段
		iterable = mc.aggregate(Arrays.asList(skip(1), limit(2),
				project(fields(include("title", "owner"), excludeId()))));
		printResult("skip(1), limit(2)", iterable);

		// $lookup 和另一个集合关联
		database.getCollection("scores").drop();
		database.getCollection("scores").insertMany(
				Arrays.asList(
						new Document("writer", "tom").append("score", 100),
						new Document("writer", "joe").append("score", 95),
						new Document("writer", "john").append("score", 80)));
		iterable = mc.aggregate(Arrays.asList(lookup("scores", "owner",
				"writer", "joinedOutput")));
		printResult("lookup", iterable);

		// 拆分comments为单个文档
		iterable = mc.aggregate(Arrays.asList(match(size("comments", 2)),
				project(fields(include("comments"), excludeId())),
				unwind("$comments")));
		printResult("unwind comments", iterable);

		System.out.println("distinct");
		DistinctIterable<String> di = mc.distinct("owner", String.class);
		di.forEach(new Block<String>() {
			public void apply(final String str) {
				System.out.println(str);
			}
		});
		System.out.println("------------------------------------------------------");
		System.out.println();

		System.out.println("count");
		long count = mc.count(Filters.eq("owner", "tom"));
		System.out.println("count=" + count);
		System.out.println("------------------------------------------------------");
		System.out.println();

		System.out.println("mapreduce");
		String map = "function() { var category; "
				+ "if ( this.words >= 280 ) category = 'Long blogs'; "
				+ "else category = 'Short blogs'; "
				+ "emit(category, {title: this.title});}";

		String reduce = "function(key, values) { var cnt = 0; "
				+ "values.forEach(function(doc) { cnt += 1; }); "
				+ "return {count: cnt};} ";
		MapReduceIterable<Document> mi = mc.mapReduce(map, reduce);
		mi.forEach(new Block<Document>() {
			public void apply(final Document str) {
				System.out.println(str);
			}
		});
		System.out.println("------------------------------------------------------");
		System.out.println();

(完)