JAVA Rest High Level Client如何取聚合后的數據

對於剛剛學習es的童鞋來說，很容易不清楚怎么獲取客戶端對es文檔的聚合結果，下面就演示一下模仿DSL寫聚合，然后獲取到聚合的結果。

一， 對於下面這個簡單的聚合，目的是對於文檔全文匹配，聚合顏色字段。把滿足匹配的文檔放入自定義名稱的colors桶中，參照《elasticsearch權威指南》的范圍限定聚合

{
    "size" : 0,
    "query" : {
        "match_all" : {}
    },
    "aggs" : {
        "colors" : {
            "terms" : {
              "field" : "color"
            }
        }
    }
}

二， 對照着這個DSL寫我們的java rest client api的調用方法，供業務層調用

    /**
     * ES中查詢所有color
     *
     * @param indices
     * @return
     */
    public SearchResponse getAllColor(String... indices) {
        SearchRequest searchRequest = new SearchRequest(indices);
        SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder().size(0);
        searchSourceBuilder.query(QueryBuilders.matchAllQuery());
        TermsAggregationBuilder builder = AggregationBuilders.terms("colors").field("color");
        searchSourceBuilder.aggregation(builder);
        searchRequest.source(searchSourceBuilder);

        try {
            return client.getHighLevelClient().search(searchRequest);
        } catch (IOException e) {
            throw new BizException("聚合失敗：{}", e.getCause());
        }
    }

三， 在我們的業務層調用上文中的getAllColor方法獲取response，對response解析，獲取聚合的結果

     * 顏色信息
     *
     * @return
     */
    public List<String> getAllColorInfo() {
        SearchResponse response = highRestHelper.getAllColor(getIndices());
        Aggregations aggregations = response.getAggregations();
        ParsedStringTerms colorTerms = aggregations.get("app_group");
        List<String> colors = new ArrayList<>();
        List<? extends Terms.Bucket> buckets = colorTerms.getBuckets();
        for (Terms.Bucket bucket : buckets) {
            String appName = bucket.getKey().toString();
            colors.add(appName);
        }   
        return colors;
    } 

至此即可獲取所有的聚合結果，只要能夠參照DSL，轉換成相應的api調用，那么解析過程完全可以復用。