elasticSearch聚合sum查询

有时需要统计一段时间内，订单的总金额。类似于sql的sum，针对某一字段求和。这就涉及到es的聚合查询，来看看用spring-data-elasticSearch怎么写：

1. QueryBuilder queryBuilder = QueryBuilders.boolQuery()
2. .must(QueryBuilders.rangeQuery("orderTime").gte(from).lte(to));
3. // 聚合查询。goodsSales是要统计的字段，sum_sales是自定义的别名
4. SumAggregationBuilder sumBuilder = AggregationBuilders.sum("sum_sales").field("goodsSales");
6. SearchQuery searchQuery = new NativeSearchQueryBuilder()
7. .withQuery(queryBuilder)
8. .addAggregation(sumBuilder)
9. .build();
11. double saleAmount = elasticsearchTemplate.query(searchQuery, response -> {
12. InternalSum sum = (InternalSum)response.getAggregations().asList().get(0);
13. return sum.getValue();
14. });

对应es查询语句：

1. {
2. "query": {
3. "bool": {
4. "must": [
5. {
6. "range": {
7. "orderTime": {
8. "gte": 20180313,
9. "lte": 20180314
10. }
11. }
12. },
13. {
14. "match": {
15. "skuNo": "888"
16. }
17. }
18. ]
19. }
20. },
21. "aggs": {
22. "sum_value": {
23. "sum_sales": {
24. "field": "goodsSales"
25. }
26. }
27. }
28. }

以下还有一些常用聚合查询（参考http://blog.csdn.net/u010454030/article/details/63266035）：

1. （1）统计某个字段的数量
2. ValueCountBuilder vcb= AggregationBuilders.count("count_uid").field("uid");
3. （2）去重统计某个字段的数量（有少量误差）
4. CardinalityBuilder cb= AggregationBuilders.cardinality("distinct_count_uid").field("uid");
5. （3）聚合过滤
6. FilterAggregationBuilder fab= AggregationBuilders.filter("uid_filter").filter(QueryBuilders.queryStringQuery("uid:001"));
7. （4）按某个字段分组
8. TermsBuilder tb= AggregationBuilders.terms("group_name").field("name");
9. （5）求和
10. SumBuilder sumBuilder= AggregationBuilders.sum("sum_price").field("price");
11. （6）求平均
12. AvgBuilder ab= AggregationBuilders.avg("avg_price").field("price");
13. （7）求最大值
14. MaxBuilder mb= AggregationBuilders.max("max_price").field("price");
15. （8）求最小值
16. MinBuilder min= AggregationBuilders.min("min_price").field("price");
17. （9）按日期间隔分组
18. DateHistogramBuilder dhb= AggregationBuilders.dateHistogram("dh").field("date");
19. （10）获取聚合里面的结果
20. TopHitsBuilder thb= AggregationBuilders.topHits("top_result");
21. （11）嵌套的聚合
22. NestedBuilder nb= AggregationBuilders.nested("negsted_path").path("quests");
23. （12）反转嵌套
24. AggregationBuilders.reverseNested("res_negsted").path("kps ");