ElasticSearch Java High Level Client使用备忘

ElasticSearch Java High Level Client

封装类似SQL select * from xxx group by xxx.xxx order by xxx的实现

可以通过top_hits实现

/**
  * @return SearchSourceBuilder 是封装在SearchRequest里的，它里面就是ElasticSearch的查询条件
 */
public SearchSourceBuilder assemblySearchSourceBuilder(){
    SearchSourceBuilder sourceBuilder = SearchSourceBuilder.searchSource();
    //封装query的条件
    QueryBuilder query = ....
    //封装aggregation
    AggregationBuilder aggregationBuilder = AggregationBuilders.terms(xxx).field(xxx);
    //封装top_hits
    TopHitsAggregationBuilder topHit = AggregationBuilders
            .topHits(xxx)
            //source(返回的字段，不包含的字段)
            .source(xxx,xxx)
            //可以多个排序条件,sort(排序字段名,排序规则ASC、DESC)
            .sort(xxx,SortOrder);
    //把top_hits封到外层的AggregationBuilder里
    aggregationBuilder.subAggregation(topHit);

}

Aggregation 获取Data

public void assemblyAggregationData(SearchResponse response){
    Aggregations aggregations = response.getAggregations();
    //key是在aggregations查询时自己设置的(AggregationBuilders.terms(xxx))
    Terms terms = aggregations.get(key);
    
    List<Bucket> bucketList = terms.getBuckets();
    
    for( Bucket b: bucketList ){
        //获取的是aggregations查询时自己设置(AggregationBuilder.field(xxx))
        //xxx对应的是ElasticSearch中存储的字段名称
        //o是ElasticSearch中存储的字段对应的值
        Object o = b.getKey();
    
    }


}

获取Aggregation中Top_Hits中的数据

public void assemblyTopHitsData(SearchResponse response){
//包含上一部分的代码，这里省略
...
for( Bucket b: bucketList ){
    //xxx是aggregations查询时自己设置top_hits的外层的aggregations的名称
    ParsedTopHits  hits = b.getAggregations().get(xxx);
    for(SearchHit hit : hits){
        //这里获取返回数据的json字符串
        String jsonStr = hit.getSourceAsString();
        //后面就是处理获取的json数据了
        ...
    }
    
    

}