Elasticsearch(ES6)------（5）kibana的es查询、mysql查询转换和对应javaAPI使用（一）

准备数据测试数据

我这里提前在mysql中造好了一些数据，通过javaAPI导入到es中

mysql建表语句

CREATE TABLE `product_item` (`product_id` int NOT NULL AUTO_INCREMENT,`product_name` varchar(256) NOT NULL COMMENT '产品名称',`price` double(18,2) NOT NULL COMMENT '价格',`count` int NOT NULL COMMENT '数量',`remark` varchar(500) DEFAULT NULL COMMENT '备注',`supplier_id` bigint DEFAULT NULL COMMENT '供应商ID',PRIMARY KEY (`product_id`) USING BTREE,KEY `idx_product_name` (`product_name`) USING BTREE
) ENGINE=InnoDB AUTO_INCREMENT=10001 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci ROW_FORMAT=DYNAMIC COMMENT='商品表';

es索引信息，这种创建方式不考虑分词

mysql数据导入ES，java使用BulkProcessor做ES批量导入，这里就不多介绍了

@PostMapping(value = "/db-to-es")public BaseResponse dbToEs() throws InterruptedException {List<ProductItem> productItemList = productItemService.getProductItemList();EsManager esManager = new EsManager();BulkProcessor bulkProcessor = null;try {bulkProcessor = esManager.bulkProcessor();} catch (UnknownHostException e) {e.printStackTrace();}for (ProductItem var1 : productItemList) {String s = new Gson().toJson(var1);bulkProcessor.add(new IndexRequest("product_item", "_doc", var1.getProductId().toString()).source(s, XContentType.JSON));}bulkProcessor.awaitClose(30, TimeUnit.SECONDS);return BaseResponse.ok("保存数据成功:", productItemList.size());}

mysql数据

ES数据查看

进入正题

1. mysql的 = 查询对应es的term

select *from product_item where product_id='8001'

ES 查询

GET product_item/_search
{"query": {"term": {"productId": {"value": "8001"}}}
}

java代码

@PostMapping(value = "/search/term")public BaseResponse term(){//Term查询TermQueryBuilder termQueryBuilder = QueryBuilders.termQuery("productId","8001");//在这里输入索引名称和type类型SearchResponse response = transportClient.prepareSearch("product_item").setTypes("_doc")//设置查询类型java.setSearchType(SearchType.DFS_QUERY_THEN_FETCH)//设置查询关键词.setQuery(termQueryBuilder)//返回搜索响应信息.get();//打印下查询条件System.out.println("termQueryBuilder="+termQueryBuilder);SearchHits hits = response.getHits();List<ProductItem> list = new ArrayList<>();ProductItem productItem;for (SearchHit searchHit : hits) {Map<String, Object> sourceAsMap = searchHit.getSourceAsMap();productItem = new ProductItem();System.out.printf("searchHit="+searchHit);productItem.setProductId(Long.valueOf(sourceAsMap.get("productId").toString()));productItem.setPrice(Double.valueOf(sourceAsMap.get("price").toString()));productItem.setProductName(sourceAsMap.get("productName").toString());productItem.setSupplierId(Long.valueOf(sourceAsMap.get("supplierId").toString()));productItem.setCount(Integer.valueOf(sourceAsMap.get("count").toString()));list.add(productItem);}return BaseResponse.ok("success:", list);}

2. like 查询，对应 es的wildcard 并对结果聚合 group by

ES查询 from=0 ,size=0 就是不显示hits结果内容，通过结果图可以看出来，这里着重说一下，from=0 并不是从第几页，而是从第几行，后续写分页查询会在用到这里。aggs是对结果进行聚合，聚合器的内容也很多，后续有时间会单独写一篇

GET product_item/_search
{"query": {"wildcard" : { "productName" : "*显示器*" }},"from": 0,"size": 0, "aggs": {"distinct": {"terms": {"field": "productName"}}}
}

buckets中的doc_conut 就是总数,和count(*)一样