ES 两次group by,先按照每天的时间力度，在统计每天的指定字段的值出现的次数

本文链接：https://blog.youkuaiyun.com/zhanghe687/article/details/133762662

文章介绍了如何在Elasticsearch中使用DateHistogram聚合器按天对数据进行分组，并获取agg_tw_id字段的子聚合结果。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

    public void addPubtimeAgg(){

        DateHistogramBuilder dateHistogramBuilder = AggregationBuilders.dateHistogram("pubtime_histogram")
                .field("pubtime").interval(DateHistogram.Interval.days(1))
                .minDocCount(0L);
        dateHistogramBuilder.format("yyyy-MM-dd");

        TermsBuilder field = AggregationBuilders.terms("agg_tw_id").field("tw_id").size(30);
        dateHistogramBuilder.subAggregation(field);

        searchbuilder.addAggregation(dateHistogramBuilder);
        if (this.query != null) {
            this.searchbuilder.setQuery(this.query);
        }
        SearchResponse sr = searchbuilder.execute().actionGet();

        Map<String, Aggregation> aggMap = sr.getAggregations().asMap();
        DateHistogram classTerms = (DateHistogram)aggMap.get("pubtime_histogram");
        Iterator<org.elasticsearch.search.aggregations.bucket.histogram.DateHistogram.Bucket> classBucketIt = (Iterator<Bucket>) classTerms.getBuckets().iterator();
        while(classBucketIt.hasNext()){
            org.elasticsearch.search.aggregations.bucket.histogram.DateHistogram.Bucket classBucket = classBucketIt.next();
            System.out.println("时间"+classBucket.getKey());
            StringTerms agg_type = (StringTerms) classBucket.getAggregations().asMap().get("agg_tw_id");
            List<Terms.Bucket> buckets = (List<Terms.Bucket>) agg_type.getBuckets();
            for (Terms.Bucket bucket : buckets) {
                String[] str = new String[2];
                str[0] = bucket.getKey().trim();//日期
                str[1] = String.valueOf(bucket.getDocCount());//数量
                System.out.println( str[0]  +"======"   + str[1] );
            }
        }
    }