elasticsearch高级搜索功能多维度分享

Dear 丶 2023-10-03 22:54 18阅读 0赞

**目录**

一、业务搜索核心功能

二、高级搜索匹配功能

三、搜索排序功能

--------------------

elasticsearch高级搜索功能多维度分享，这也是实战的比较之路，此次我们全面分享常用的业务情景，全覆盖功能分享，让大家有一览众山小的感觉，废话少数，现在开始！

## 一、业务搜索核心功能 ##

1、返回指定字段

GET /nandao_scenic/_search
    {
      "_source": ["title","city"], 
      "query": {
        "term": {
          "city": {
            "value": "北京市"
          }
        }
      }
    }

java 语言

public List<Scenic> getScenicField(String keyword) {
            SearchRequest searchRequest = new SearchRequest("scenic");//客户端请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();//创建搜索builder
            searchSourceBuilder.query(new TermQueryBuilder("city", keyword));//构建query
            searchSourceBuilder.fetchSource(new String[]{"title", "city"}, null);//设定希望返回的字段数组
         //   searchSourceBuilder.from(20);
           // searchSourceBuilder.size(10);
            searchRequest.source(searchSourceBuilder);
            List<Scenic> resultList = new ArrayList<Scenic>();
            try {
                SearchResponse searchResponse = client.search(searchRequest, RequestOptions.DEFAULT);//执行搜索
                return resultList;
            } catch (Exception e) {
                e.printStackTrace();
            }
            return null;
        }

2、结果计数

GET /nandao_scenic/_count
    {
       "query": {
        "term": {
          "city": {
            "value": "北京市"
          }
        }
      }
    }

java语言

public long getCityCount() {
            CountRequest countRequest = new CountRequest("scenic");//客户端count请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();//创建搜索builder
            searchSourceBuilder.query(new TermQueryBuilder("city", "北京"));//构建query
            countRequest.source(searchSourceBuilder);//设置查询
            try {
                CountResponse countResponse = client.count(countRequest, RequestOptions.DEFAULT);//执行count
                return countResponse.getCount();//返回count结果
            } catch (Exception e) {
                e.printStackTrace();
            }
            return 0;
        }

3、结果分页

GET /nandao_scenic/_search
    {
       "from": 0,
       "size": 20, 
      "query": {
        "term": {
          "city": {
            "value": "北京市"
          }
        }
      }
    }

设置分页最大值：

PUT /nandao_scenic/_settings
    {
      "index":{
        "max_result_window":10001
      }
    }

es不适合深度分页，影响性能。

java语言

public List<Scenic> getScenicField(String keyword) {
            SearchRequest searchRequest = new SearchRequest("scenic");//客户端请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();//创建搜索builder
            searchSourceBuilder.query(new TermQueryBuilder("city", keyword));//构建query
            searchSourceBuilder.fetchSource(new String[]{"title", "city"}, null);//设定希望返回的字段数组
            searchSourceBuilder.from(20);
            searchSourceBuilder.size(10);
            searchRequest.source(searchSourceBuilder);
            List<Scenic> resultList = new ArrayList<Scenic>();
            try {
                SearchResponse searchResponse = client.search(searchRequest, RequestOptions.DEFAULT);//执行搜索
                return resultList;
            } catch (Exception e) {
                e.printStackTrace();
            }
            return null;
        }

4、性能分析：开启

GET /nandao_scenic/_search
    {
      "profile": "true", 
      "query": {
        "term": {
          "city": {
            "value": "北京市"
          }
        }
      }
    }

5、评分分析

GET /nandao_scenic/_explain/003
    {
      "query": {
        "term": {
          "city": {
            "value": "北京市"
          }
        }
      }
    }

## 二、高级搜索匹配功能 ##

1、查询所有文档

GET /nandao_scenic/_search
    {
      "_source": ["title","city"], 
      "query": {
        "match_all": {
          "boost": 2
        }
      }
    }

java客户端

public void matchAllSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//新建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            MatchAllQueryBuilder matchAllQueryBuilder = QueryBuilders.matchAllQuery().boost(2.0f);//新建match_all查询，并设置boost值为2.0
            searchSourceBuilder.query(matchAllQueryBuilder);
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }
    
    
    public void printResult(SearchRequest searchRequest) {
            try {
                SearchResponse searchResponse = client.search(searchRequest, RequestOptions.DEFAULT);//执行搜索
                SearchHits searchHits = searchResponse.getHits();//获取搜索结果集
                for (SearchHit searchHit : searchHits) {//遍历搜索结果集
                    String index = searchHit.getIndex();//获取索引名称
                    String id = searchHit.getId();//获取文档_id
                    Float score = searchHit.getScore();//获取得分
                    String source = searchHit.getSourceAsString();//获取文档内容
                    System.out.println("index=" + index + ",id=" + id + ",score=" + score + ",source=" + source);//打印数据
                }
            } catch (Exception e) {
                e.printStackTrace();
            }
        }

2、term 级别查询

2.1、term 查询

GET /nandao_scenic/_search
    {
     "query": {
       "term": {
         "city": {
           "value": "北京市"
         }
       }
     }
    }

2.2、terms查询

GET /nandao_scenic/_search
    {
     "query": {
       "terms": {
         "city": [
           "北京市",
           "河南"
           ]
       }
     }
    }

java客户端

public void termsSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//创建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            searchSourceBuilder.query(QueryBuilders.termsQuery("city", "北京", "天津"));//构建terms查询
            searchRequest.source(searchSourceBuilder);//设置查询请求
            printResult(searchRequest);//打印搜索结果
        }

2.3、range查询

GET /nandao_scenic/_search
    {
     "query": {
       "range": {
         "price": {
           "gte":50,
           "lte":100
         }
       }
     }
    }

java语言客户端

public void rangeSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//创建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            //构建range查询
            QueryBuilder queryBuilder = QueryBuilders.rangeQuery("create_time").gte("20220115120000").lte("20220116120000");
            searchSourceBuilder.query(queryBuilder);
            searchRequest.source(searchSourceBuilder);//设置查询请求
            printResult(searchRequest);//打印搜索结果
        }

2. 4、exists 查询

GET /nandao_scenic/_search
    {
     "query": {
       "exists": {
        "field": "tag"
       }
     }
    }

java语言

public void existsSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            searchSourceBuilder.query(QueryBuilders.existsQuery("tag"));
            searchRequest.source(searchSourceBuilder);
            printResult(searchRequest);
        }

3、布尔类型

3.1、must查询

GET /nandao_scenic/_search
    {
     "query": {
        "bool": {
          "must": [
            {
              "term": {
                "city": {
                  "value": "北京市"
                }
              }
            },
            {
              "range": {
                "price": {
                  "gte": 10,
                  "lte": 220
                }
              }
            }
          ]
        }
     }
    }

java语言

public void mustSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//新建请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
            TermQueryBuilder termQueryIsReady = QueryBuilders.termQuery("city", "北京");//构建城市term查询
            RangeQueryBuilder rangeQueryBuilder = QueryBuilders.rangeQuery("price").gte(350).lte(400);//构建价格range查询
            boolQueryBuilder.must(termQueryIsReady).must(rangeQueryBuilder);//进行关系“与”查询
            searchSourceBuilder.query(boolQueryBuilder);
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }

3.2、should查询

GET /nandao_scenic/_search
    {
     "query": {
        "bool": {
          "should": [
            {
              "term": {
                "city": {
                  "value": "北京市"
                }
              }
            },
            {
               "term": {
                "city": {
                  "value": "北京市"
                }
              }
            }
          ]
        }
     }
    }

java语言

public void shouldSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//新建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
            TermQueryBuilder termQueryIsReady = QueryBuilders.termQuery("city", "北京");//构建城市为“北京”的term查询
            TermQueryBuilder termQueryWritter = QueryBuilders.termQuery("city", "天津");//构建城市为“天津”的term查询
            boolQueryBuilder.should(termQueryIsReady).should(termQueryWritter);//进行关系“或”查询
            searchSourceBuilder.query(boolQueryBuilder);
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }

3.3、must not 查询

GET /nandao_scenic/_search
    {
     "query": {
        "bool": {
          "must_not": [
            {
              "term": {
                "city": {
                  "value": "北京市"
                }
              }
            },
            {
               "term": {
                "city": {
                  "value": "河南"
                }
              }
            }
          ]
        }
     }
    }

java客户端

public void mustNotSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//新建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
            TermQueryBuilder termQueryIsReady = QueryBuilders.termQuery("city", "北京");//构建城市为“北京”的term查询
            TermQueryBuilder termQueryWritter = QueryBuilders.termQuery("city", "天津");//构建城市为“天津”的term查询
            boolQueryBuilder.mustNot(termQueryIsReady).mustNot(termQueryWritter);//进行关系“必须不”查询
            searchSourceBuilder.query(boolQueryBuilder);
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }

3.4、filter 查询

GET /nandao_scenic/_search
    {
     "query": {
        "bool": {
          "filter": [
            {
              "term": {
                "city": {
                  "value": "北京市"
                }
              }
            },
            {
               "term": {
                "full": true
              }
            }
          ]
        }
     }
    }

java客户端

public void filterSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
            boolQueryBuilder.filter(QueryBuilders.termQuery("city", "北京"));
            boolQueryBuilder.filter(QueryBuilders.termQuery("full", false));
            searchSourceBuilder.query(boolQueryBuilder);
            searchRequest.source(searchSourceBuilder);
            printResult(searchRequest);
        }

4、Constant Scope 查询

GET /nandao_scenic/_search
    {
      "_source": ["city"], 
     "query": {
       "constant_score": {
         "filter": {
           "match":{
             "city":"北京市"
           }
         }
       }
     }
    }

java语言

public void constantScoreSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//新建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            ConstantScoreQueryBuilder constantScoreQueryBuilder = new ConstantScoreQueryBuilder(QueryBuilders.termQuery("amenities", "停车场"));//构建城市为“北京”的term查询
            searchSourceBuilder.query(constantScoreQueryBuilder);
            constantScoreQueryBuilder.boost(2.0f);
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }

5、Function score 查询

GET /nandao_scenic/_search
    {
      "_source": ["city"], 
     "query": {
       "function_score": {
         "query": {
           "term": {
             "city": {
               "value": "北京市"
             }
           }
         },
         "functions": [
           {
             "random_score": {}
           }
         ],
         "score_mode": "sum"
       }
     }
    }

java语言

public void functionScoreSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//创建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            TermQueryBuilder termQuery = QueryBuilders.termQuery("city", "北京");//构建term查询
            ScoreFunctionBuilder<?> scoreFunction = ScoreFunctionBuilders.randomFunction();//构建随机函数
            //构建function_score查询
            FunctionScoreQueryBuilder funcQuery = QueryBuilders.functionScoreQuery(termQuery, scoreFunction).boostMode(CombineFunction.SUM);
            searchSourceBuilder.query(funcQuery);
            searchRequest.source(searchSourceBuilder);//设置查询请求
            printResult(searchRequest);//打印搜索结果
        }

6、全文搜索

6.1、match 查询：只要包含一个字就能搜出来

GET /nandao_scenic/_search
    {
      "_source": ["title"], 
         "query": {
           "match": {
               "title": "水山寺"
           }
         }
    }

java客户端

public void matchSearch() {
            SearchRequest searchRequest = new SearchRequest();//新建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            searchSourceBuilder.query(QueryBuilders.matchQuery("title", "山水寺").operator(Operator.AND));//新建match查询，并设置operator值为and
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }

6.2、multi\_match 查询

GET /nandao_scenic/_search
    {
      "_source": ["title","city"], 
         "query": {
           "multi_match": {
             "query": "山市",
             "fields": [
               "city",
               "title"
               ]
           }
         }
    }

java客户端

public void multiMatchSearch() {
            SearchRequest searchRequest = new SearchRequest();//新建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            searchSourceBuilder.query(QueryBuilders.multiMatchQuery("山市", "title", "city"));//新建multi_match查询，从"title"和"amenities"字段查询"假日"
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }

6.3、multi\_phrase 查询

GET /nandao_scenic/_search
    {
       "query": {
         "match_phrase": {
           "title": "台山"
         }
       }
    }

java语言

public void matchPhraseSearch() {
            SearchRequest searchRequest = new SearchRequest();//新建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            searchSourceBuilder.query(QueryBuilders.matchPhraseQuery( "title", "台山").slop(2)); 
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }

7、基于地理位置查询

GET /nandao_scenic/_search
    {
      "_source": ["title","city"],
       "query": {
         "geo_distance": {
           "distance": "5km",
           "location":{
             "lat":"34.333",
             "lon":"23.56"
           }
         }
       }
    }

java 客户端

public void geoDistanceSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//新建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            //新建geo_distance查询，设置基准点坐标和周边距离
            searchSourceBuilder.query(QueryBuilders.geoDistanceQuery("location").distance(5, DistanceUnit.KILOMETERS).point(40.026919, 116.47473));
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }

多边形：

GET /nandao_scenic/_search
    {
      
       "query": {
         "geo_polygon": {
           "location":{
             "points":[
               {
                   "lat":"39.959829",
             "lon":"116.417088"
               },
               {
                   "lat":"39.960272",
             "lon":"116.432035"
               },
               {
                   "lat":"39.965802",
             "lon":"116.421399"
               }
               ]
           }
         }
       }
    }

java语言

public void geoPolygonSearch() {
            SearchRequest searchRequest = new SearchRequest("scenic");//新建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            //新建geo_distance查询，设置基准点坐标和周边距离
            List<GeoPoint> geoPointList = new ArrayList<GeoPoint>();//新建多边形顶点列表
            //添加多边形顶点
            geoPointList.add(new GeoPoint(39.959829, 116.417088));
            geoPointList.add(new GeoPoint(39.960272, 116.432035));
            geoPointList.add(new GeoPoint(39.965802, 116.421399));
            searchSourceBuilder.query(QueryBuilders.geoPolygonQuery("location", geoPointList));//新建geo_polygon查询
            searchRequest.source(searchSourceBuilder);//设置查询
            printResult(searchRequest);//打印结果
        }

8、实战场景之一：京东搜索商品，前缀查询

PUT /nandao_scenic_sug
    {
      "mappings": {
        "properties": {
          "query_word":{
            "type": "completion"
          }
        }
      }
    }

查询

GET /nandao_scenic_sug/_search
    {
      "suggest": {
        "scenic_zh_sug": {
          "prefix":"北京",
          "completion": {
            "field": "query_word"
          }
        }
      }
    }

java语言

public void suggestSearch() throws IOException {
            SearchRequest searchRequest = new SearchRequest("nandao_scenic_sug");//创建搜索请求,指定索引名称为scenic_sug
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            //创建completion类型搜索建议
            CompletionSuggestionBuilder comSuggest = SuggestBuilders.completionSuggestion("query_word").prefix("北京");
            SuggestBuilder suggestBuilder = new SuggestBuilder();
            suggestBuilder.addSuggestion("scenic_zh_sug", comSuggest);//添加搜索建议，"scenic_zh_sug"为自定义名称
            searchSourceBuilder.suggest(suggestBuilder);//设置suggest请求
            searchRequest.source(searchSourceBuilder);//设置查询请求
            SearchResponse response = client.search(searchRequest, RequestOptions.DEFAULT);//进行搜索,获取搜索结果
            CompletionSuggestion suggestion = response.getSuggest().getSuggestion("scenic_zh_sug");//获取suggest结果
            System.out.println("sug result:");
            //遍历suggest结果，并进行打印
            for (CompletionSuggestion.Entry.Option option : suggestion.getOptions()) {
                System.out.println("sug:" + option.getText().string());
            }
        }

## 三、搜索排序功能 ##

1、按照普通字段排序

GET /nandao_scenic/_search
    {
      "_source": ["title","city","price"],
      "query": {
        "match": {
          "title": "山"
        }
      },
      "sort": [
        {
          "price": {
            "order": "desc"
          },
          "city": {
            "order": "desc"
          }
        }
      ]
    }

java客户端

public void commonSort() {
            SearchRequest searchRequest = new SearchRequest("nandao_scenic");//创建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            searchSourceBuilder.query(QueryBuilders.matchQuery("title", "山"));//构建match查询
            searchRequest.source(searchSourceBuilder);//设置查询请求
            searchSourceBuilder.sort("price", SortOrder.DESC);//设置按照价格降序
            searchSourceBuilder.sort("praise", SortOrder.DESC);//设置按照口碑值降序
            printResult(searchRequest);//打印搜索结果
        }

2、按照地理位置排序

GET /nandao_scenic/_search
    {
      "_source": ["title","city","location"],
      "query": {
        "geo_distance": {
           "distance": "5km",
           "location":{
             "lat":"34.333",
             "lon":"23.56"
           }
         }
      },
      "sort": [
        {
          "geo_distance": {
           "location":{
             "lat":"34.333",
             "lon":"23.56"
           },
           "order": "desc",
           "unit":"km",
           "distance_type":"plane"
         }
        }
      ]
    }

java客户端

public void geoDistanceSearchSort() throws IOException {
            SearchRequest searchRequest = new SearchRequest("nandao_scenic");//创建搜索请求
            SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
            //创建geo_distance查询，搜索距离中心点5公里范围内的酒店
            searchSourceBuilder.query(QueryBuilders.geoDistanceQuery("location").distance(5, DistanceUnit.KILOMETERS).point(39.915143, 116.4039));
            //创建geo_distance_sort排序，设定按照与中心点的距离升序排序
            GeoDistanceSortBuilder geoDistanceSortBuilder = SortBuilders.geoDistanceSort("location", 39.915143, 116.4039)
                    .point(39.915143, 116.4039).unit(DistanceUnit.KILOMETERS).order(SortOrder.ASC);
            searchSourceBuilder.sort(geoDistanceSortBuilder);//设置排序规则
            searchRequest.source(searchSourceBuilder);//设置查询
            //开始搜索
            SearchResponse searchResponse = client.search(searchRequest, RequestOptions.DEFAULT);
            SearchHits searchHits = searchResponse.getHits();//获取搜索结果
            System.out.println("search result distance sort:");
            //开始遍历搜索结果
            for (SearchHit searchHit : searchHits) {
                //得到酒店距离中心点的距离
                double geoDistance = (double) searchHit.getSortValues()[0];
                //以Map形式获取文档_source内容
                Map<String, Object> sourceMap = searchHit.getSourceAsMap();
                Object title = sourceMap.get("title");
                Object city = sourceMap.get("city");
                //打印结果
                System.out.println("title=" + title + ",city=" + city + ",geoDistance:" + geoDistance);
            }
        }

到此，es高级搜索功能分享完毕，大家一定要多多练习，定会很快掌握，下篇我们分享文本搜索相关功能，敬请期待！