基于Elasticsearch的自定义评分算法扩展

实现思路：重写评分方法，调整计算文档得分的过程，然后根据function_score或script_sort进行排序检索。 实现步骤：1、新建java项目TestProject，引入Elasticsearch的jar包2、新建package：es.testscript3、新建类TestScriptFactory，继承NativeScriptFactory，示例：package es.testscript;import java.util.Map; import org.elasticsearch.common.Nullable;import org.elasticsearch.script.ExecutableScript;import org.elasticsearch.script.NativeScriptFactory; /** * Created by lijunhao on 2016/3/29. */public class TestScriptFactory implements NativeScriptFactory {    @Override    public ExecutableScript newScript(@Nullable Map<String, Object> params) {        return new TestScript(params);    }} 4、新建类TestScript，假设计算double类型的得分，继承AbstractDoubleSearchScript，并重写runAsDouble方法，示例：package es.testscript; import java.util.Iterator;import java.util.Map;import java.util.Set; import org.elasticsearch.common.Nullable;import org.elasticsearch.index.fielddata.ScriptDocValues;import org.elasticsearch.script.AbstractDoubleSearchScript;import org.elasticsearch.script.AbstractLongSearchScript; import java.util.Map; public class TestScript extends AbstractDoubleSearchScript {    //客户端传递的参与动态计算得分的参数    private String[] paramArray;     /**     * 构造函数 获取传入的参数     *     * @param params     */    public TestScript(@Nullable Map<String, Object> params) {        if (params == null || params.size() == 0) {            return;        }        Set<String> keys = params.keySet();        Iterator<String> iterator = keys.iterator();        while (iterator.hasNext()) {            String key = iterator.next();            String val = params.get(key).toString();            System.out.println(“key:” + key + ” val:” + val + “\r\n”);        }        if (params.get(“fields”) == null) {            return;        }        paramArray = params.get(“fields”).toString().split(“,”);        System.out.println(“fields:” + params.get(“fields”).toString() + “\r\n”);    }     /**     * 排序方法，计算得分     *     * @return     */    @Override    public double runAsDouble() {        double defaultReturnVal = Double.parseDouble(String.valueOf(((ScriptDocValues.Longs)doc().get(“id”)).getValue()));        if (paramArray == null || paramArray.length == 0) {            return defaultReturnVal;        }        //根据传入的paramArray计算得分        defaultReturnVal=defaultReturnVal+1000;        System.out.println(“score:” + defaultReturnVal + “\r\n”);        return defaultReturnVal;    }} 5、打包输出jar文件TestProject.jar6、将TestProject.jar拷贝至ES目录的lib下7、修改ES配置文件elasticsearch.yml，添加：script.native:    mynativescript.type: es.testscript.TestScriptFactory注：mynativescript为自定义的脚本别名。8、重启ES服务9、执行检索：function_score方式{  “query”: {    “function_score”: {      “query”: {        “match_all”: {}      },      “functions”: [        {          “script_score”: {            “script”: “mynativescript”,            “lang”: “native”,            “params”: {              “p1”: 1,              “fields”: “p1,p2”            }          }        }      ]    }  }}10、执行检索：script_sort方式{  “query”: {    “match_all”: {}  },  “sort”: {    “_script”: {      “script”: “mynativescript”,      “lang”: “native”,      “order”: “asc”,      “type”: “string”,      “params”: {        “p1”: 1,        “p2”: 2,        “p3”: 3      }    }  }}11、执行检索：Nest方式之Linqvar s = new SearchDescriptor<ModelTest>().From(0).Size(20).MatchAll().SortScript(sort => sort                     .Descending()                     .Script(“mynativescript”)                     .Descending()                     .Params(p => p                         .Add(“p1”, 1.1).Add(“p2”, 2.2)                     )                     .Language(“native”)                     .Type(“string”)                 );//获取请求的json字符串string reqStr = Encoding.UTF8.GetString(client.Serializer.Serialize(s));ISearchResponse<ModelTest> resp = client.Search<ModelTest>(s);ModelTest[] result = resp.Documents.ToArray(); 12. Nest方式之Query对象

 QueryContainer mainQuery = null; 
 FunctionScoreQuery funcQuery = new FunctionScoreQuery();
 funcQuery.ScoreMode = FunctionScoreMode.Sum;
 funcQuery.BoostMode = FunctionBoostMode.Replace;
 funcQuery.MaxBoost = 1000.0f;
 IFunctionScoreFunction func = new FunctionScoreFunctionsDescriptor<DTOCarInfoIndexField>().ScriptScore(s => s.Lang("native").Script("mynativescript"));
 IList<IFunctionScoreFunction> list = new List<IFunctionScoreFunction>();
 list.Add(func);
 funcQuery.Functions = list;
 mainQuery &= funcQuery;