Solr API例子详解

package test;import java.io.IOException;  import java.net.MalformedURLException;  import java.text.SimpleDateFormat;import java.util.ArrayList;  import java.util.Comparator;import java.util.Date;import java.util.List;   import java.util.Map;import java.util.TreeMap;import org.apache.commons.collections.CollectionUtils;import org.apache.lucene.analysis.Analyzer;import org.apache.lucene.document.Document;import org.apache.lucene.document.Field;import org.apache.lucene.document.FieldType;import org.apache.lucene.util.CollectionUtil;import org.apache.solr.client.solrj.SolrQuery;  import org.apache.solr.client.solrj.SolrRequest;import org.apache.solr.client.solrj.SolrRequest.METHOD;import org.apache.solr.client.solrj.SolrServer;import org.apache.solr.client.solrj.SolrServerException;  import org.apache.solr.client.solrj.impl.HttpSolrServer;import org.apache.solr.client.solrj.response.FacetField;import org.apache.solr.client.solrj.response.FacetField.Count;  import org.apache.solr.client.solrj.response.FieldStatsInfo;import org.apache.solr.client.solrj.response.Group;import org.apache.solr.client.solrj.response.GroupCommand;import org.apache.solr.client.solrj.response.GroupResponse;import org.apache.solr.client.solrj.response.PivotField;import org.apache.solr.client.solrj.response.QueryResponse;  import org.apache.solr.common.SolrDocument;  import org.apache.solr.common.SolrDocumentList;    import org.apache.solr.common.params.FacetParams;import org.apache.solr.common.params.GroupParams;import org.apache.solr.common.util.NamedList;import org.junit.Test;  import com.xd.iis.se.pojos.IContentCommon;@SuppressWarnings("deprecation")  public class SolrTest {          //指定solr服务器的地址  /*    private final static String URL = "http://192.168.20.195:8084/solr/collection1";*/      private final static String URL = "http://192.168.20.195:8084/solr";  //微名人    private final static String BigSolrURL="http://solr-24h.wyq.cn/solr";    /*private final static String URL = "http://192.168.20.196:8081/solr" ;   //事件分析*/    @Test      public void slorQuery() throws SolrServerException, MalformedURLException{          /*         * //项目取得HttpSolrServer方式 HttpSolrServer         *          * url配置在weibo_analysis.properties文件中         * server=SolrWbFactory.getSolrServerComment();         * weibo_comment_solr=http://192.168.20.195:8084/solr/comment         * weibo_content_comment_solr=http://192.168.20.195:8084/solr/all         * weibo_content_slor=http://192.168.20.195:8084/solr/collection1         */        HttpSolrServer server = new HttpSolrServer(BigSolrURL);        /*         * server.setConnectionTimeout(180 * 1000); server.setSoTimeout(240 *         * 1000); server.setMaxTotalConnections(1200);         * server.setDefaultMaxConnectionsPerHost(100); server.setMaxRetries(3);         * server.setAllowCompression(true);         */        // 定义查询字符串        SolrQuery query = new SolrQuery("*:*");                // 设置查询关键字        /* query.setQuery("Content:* AND Spare3:1 "); */        // 指定查询返回字段        /* query.setParam("fl", "Content,IndexTime"); */        // 设置高亮        query.setHighlight(true).setHighlightSimplePre("<span class='red'>")                .setHighlightSimplePost("</span>");        query.setParam("hl.fl", "Content");//设置高亮字段        query.setParam("fl", "ID,Published");                //排除条件      - NOT        //wbQuery.addFilterQuery("OriginType:wb -Spare3:0");        //wbQuery.addFilterQuery("OriginType:wb NOT Spare3:0");        // 时间条件过滤        /* query.addFilterQuery("Content:超哥"); */        /*         * query.addFilterQuery(         * "Published:[1995-12-31T23:59:59.999Z TO 2016-03-06T00:00:00Z]");         */        query.addFilterQuery("Published:[* TO NOW]");                // 实现分页的查询        query.setStart(0);        query.setRows(10);        // 设定排序，如果需要对field进行排序就必须在schema.xml中对该field配置stored="true"属性        //set会清空原来的sort条件，add不会清空原来的，会在原来的基础上添加 sort=Published asc,Author asc(多条件排序)        query.setSort(IContentCommon.IndexField.Published.getName(),                SolrQuery.ORDER.asc);                query.addSort(IContentCommon.IndexField.Published.getName(),                SolrQuery.ORDER.asc);        QueryResponse res = server.query(query);        System.out.println(query);        // 查询出来的结果都保存在SolrDocumentList中        SolrDocumentList sdl = res.getResults();        System.out.println("总数：" + sdl.getNumFound());        System.out.println(sdl.getMaxScore());        for (SolrDocument sd : sdl) {            Object id = sd.get("ID");            // 打印高亮信息            System.out.println(res.getHighlighting().get(id).get("Content"));            //            System.out.println(sd.get("ID") + "#" + sd.get("Content") + "#"                    + sd.get("WeiboId") + "#" + sd.get("Published") + "#"                    + sd.get("OriginType"));            /* System.out.println(sd.getFieldValue("ID")); */            Date date = (Date) sd.get("Published");            SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd");            String dstring = sdf.format(date);            System.out.println(dstring);        }          }               /*    java对象 <---> xml文件之间的转换   JAXB注解JAXB能够使用Jackson对JAXB注解的支持实现(jackson-module-jaxb-annotations)，既方便生成XML，也方便生成JSON    @XmlRootElement對类标识自动转化类中有get 、set方法的属性，没有get、set方法的属性无法转化    @XmlElement对属性标识，属性无需get、set方法      public Analyzer analyzer = new IKAnalyzer(true);//true开启智能分词         拿IKAnalyzer分词器为例，IKAnalyzer的切分方式是细粒度切分，当不需要智能处理时，     其就把切出的所有词输出，但若启动了智能处理，那么接下来就是要进行消歧工作。   智能分词结果:    张三|说的|确实|在理 最细粒度分词结果:    张三|三|说的|的确|的|确实|实在|在理(1) TermAttribute： 表示token的字符串信息。比如"I'm"(2) TypeAttribute： 表示token的类别信息(在上面讲到)。比如 I'm 就属于<APOSTROPHE>，有撇号的类型(3) OffsetAttribute：表示token的首字母和尾字母在原文本中的位置。比如 I'm 的位置信息就是(0,3)(4) PositionIncrementAttribute：这个有点特殊，它表示tokenStream中的当前token与前一个token在实际的原文本中相隔的词语数量。              Math.round 四舍五入Math.round(1.4)=1  Math.round(1.5)=2    Math.ceil向上取整Math.ceil(1.4)=2.0    Math.floor向下取整      邮箱：[e-mail]账号密码账号：yuanhai密码：cmpz0cUx0phbase设置hbase_server1=192.168.1.120hbase_server2=192.168.1.121hbase_server3=192.168.1.122192.168.1.120 jintai-test-0192.168.1.121 jintai-test-1192.168.1.122 jintai-test-2solr地址：http://192.168.20.195:8084/solr/#/http://blog.csdn.net/zwx19921215/article/details/41820483http://www.cnblogs.com/chenz/articles/3229997.htmlhttp://blog.csdn.net/huoyunshen88/article/details/38082455--------slor日期格式存储----------TrieDateField  DateField 表示一个精确到毫秒的时间，值的格式是：YYYY-MM-DD T hh:mm:ss Z--------------来源类型-------------BasicContent-------字符串转换成日期工具类-----------------DateUtilsIis.parse(strDate,strPattern)------------错误码类型类-------------------ErrorCodeConstant//ArrayList线程安全List<Map<String,Object>> data=Collections.synchronizedList(new ArrayList<Map<String,Object>>());-------------监听器------------------InitDataListener --> springmvc配置的监听器iis_application_platform_common.category_type_keywords表中获取title keyword filter_keyword01449763832357389257242----------LinkedHashMap-----------LinkedHashMap实体虽然是以Hash的顺序存放在Map的数组table里面，但是实体之间却用链表的形式保持了存入的先后关系。 LinkedHashMap保存了记录的插入顺序，所以当你需要输出的顺序和输入的相同,那么用LinkedHashMap可以实现，它还可以按读取顺序来排列。 Collections.sort方法对arraylist排序           -------字段--------    Spare3:微博的的层级    Spare5：父微博的用户名    CustomFlag1=1疑似负面    CustomFlag1=2 负面    CustomFlag1=4 非敏感        TitleHs 标题哈希   对于微博而言，标题就是内容         UserScreenName 用户名     ClusterContent            select参数对应着SearchCondition里面的 newlstSelect属性      1代表模糊信息列表=全部信息    2代表精确信息列表=智能推荐    默认为2       comblineflg   相同文章合并   true:(默认)合并 false:不合并      splitTypeFlg   false合并(xw,app合并到xw)      true不合并(xw app分开显示)            filterKeyword 排除关键字，即不包括    *    *    */            /**      * 测试分组     */     @Test      public void facet() throws MalformedURLException, SolrServerException{          HttpSolrServer server = new HttpSolrServer(URL);        SolrQuery query = new SolrQuery("*:*");        query.setIncludeScore(false);        query.setFacet(true);        /* query.addFacetField("Province"); */        query.addFacetField("Spare3","UserLocation");//两个域有各自独立的结果            /*  query.addFacetField(new String[] {"salary","publishDate","educateBackground","jobExperience","companytype","jobsType" });//设置需要facet的字段*/        query.addFacetQuery("Spare3:[4 TO 7]");        /*         * FacetComponet有两种排序选择，分别是count和index，         * count是按每个词出现的次数，index是按词的字典顺序。如果查询参数不指定facet.sort，solr默认是按count排序。         */        /* query.setFacetSort("index"); */        query.setFacetSort("count");        /*query.setFacetLimit(101);  */ // 设置返回结果条数 ,-1表示返回所有,默认值为100        /* query.setParam(FacetParams.FACET_OFFSET, "100");*/   //开始条数,偏移量,它与facet.limit配合使用可以达到分页的效果        query.setFacetSort(FacetParams.FACET_SORT_COUNT);        query.setFacetMinCount(1);//设置 限制 count的最小返回值，默认为0         query.setFacetMissing(false);//不统计null的值        /* query.setFacetPrefix("湖");//设置前缀 */                                  System.out.println(query.toString());        QueryResponse res = server.query(query);                System.out.println("-------单个facet结果--------");        List<Count> spare3List = res.getFacetField("Spare3").getValues();        for (Count count : spare3List) {            System.out.println(count.getName() + "#" + count.getCount());        }                System.out.println("-------所有facet结果--------");        // 得到所有Facet结果         List<FacetField> facets = res.getFacetFields();//返回的facet列表          for (FacetField facet :facets) {             System.out.println(facet.getName());             System.out.println("----------------");             List<Count> counts = facet.getValues();                     int i=1;//计数器             for (Count count : counts){                System.out.println(i+count.getName()+":"+ count.getCount());                i++;             }             System.out.println();          }                          System.out.println("-------FacetQuery结果--------");        // 得到FacetQuery结果        Map<String, Integer> facetQueryResult = res.getFacetQuery();        for (Map.Entry<String, Integer> fqr : facetQueryResult.entrySet()) {            System.out.println(fqr.getKey() + ":" + fqr.getValue());        }                                            res.getFacetDate(null);        res.getFacetDates();        res.getFacetRanges();                /*                           query.set("group", "true");                         query.set("group.field", groupField);                        query.set("group.format", "grouped"); // default:simple,                        query.set("group.main", "false"); // when                        // /*group.format=simple                        query.set("group.cache.percent", "50"); // default is 0;                        query.set("rows", topCount);                        query.setParam("shards.tolerant", true);         * */        /**          *          *          *          * 控制台输出结果：          *               *             q=Usergender%3A*&fl=*&facet=true&facet.field=Spare3&facet.query=Spare3%3A%5B4+TO+7%5D&facet.sort=count&facet.limit=10            0#35389            4#22778            3#22333            2#16567            1#15571            6#13161            5#12339            7#11550            8#2659            9#1399            Spare3:[4 TO 7]:59828                        ---------对应slor图形界面http访问url-----            http://192.168.20.195:8084/solr/collection1/select?q=Usergender%3A*&            wt=json&indent=true&facet=true&facet.query=Spare3%3A%5B4+TO+7%5D&facet.field=Spare3                                                                 -------------- slor图形界面结果------------                                              "facet_counts": {                "facet_queries": {                  "Spare3:[4 TO 7]": 59828                },                "facet_fields": {                  "Spare3": [                    "0",                    35389,                    "4",                    22778,                    "3",                    22333,                    "2",                    16567,                    "1",                    15571,                    "6",                    13161,                    "5",                    12339,                    "7",                    11550,                    "8",                    2659,                    "9",                    1399,                    "10",                    500,                    "11",                    187,                    "12",                    115,                    "13",                    51,                    "14",                    34,                    "15",                    30,                    "16",                    3                  ]                },                "facet_dates": {},                "facet_ranges": {}              }                                          facet.rangehttp://…/select?&facet=true&facet.range=price&facet.range.start=5000&facet.range.end=8000&facet.range.gap=1000solrconfig.xml配置<requestHandler name="/browse" class="solr.SearchHandler" name="/select"><lst name="defaults"><str name="echoParams">explicit</str><int name="rows">10</int><str name="df">text</str>str name="facet">on</str><str name="facet.range">Price</str><int name="f.Price.facet.range.start">0</int><int name="f.Price.facet.range.end">5000</int><int name="f.Price.facet.range.gap">1000</int></lst></requestHandler>        facet.range节点中表示按范围分段的字段为Price   f.Price.facet.range.start表示起始值为0  f.Price.facet.range.end表示最大值为 5000  f.Price.facet.range.gap表示每次间隔1000进行分段 ，         */                              }                            @Test  public  void group()  {        try {            HttpSolrServer server = new HttpSolrServer(URL);            SolrQuery query = new SolrQuery("*:*");                        /*-----增加group-----------*/            query.setParam(GroupParams.GROUP,true);              query.setParam(GroupParams.GROUP_FIELD,"Spare3","UserLocation");             query.setParam(GroupParams.GROUP_LIMIT,"3");//每个分组返回的文档数量            query.setParam(GroupParams.GROUP_FORMAT, "grouped");// 默认为 grouped所有文档分组显示 ,simple的话所有文档显示在一个集合中            query.setParam(GroupParams.GROUP_MAIN,"false");//format为grouped的时候，必须为false   query.setParam(GroupParams.GROUP_TOTAL_COUNT, true);//返回分组个数ngroups：               query.setRows(100);//限制返回group个数                          System.out.println(query.toString());            QueryResponse res = server.query(query);System.out.println("--------group分组结果----------------");         /* ------得到分组结果-----*/      GroupResponse groupResponse =res.getGroupResponse();               if(groupResponse !=null) {    //得到每个分组字段集合（多个分组字段）     List<GroupCommand> groupList =groupResponse.getValues();     //遍历每个字段分组结果     for(GroupCommand groupCommand : groupList){                 System.out.println("--------group字段----------------");                  System.out.println(  groupCommand.getName());              System.out.println("--------单个group字段结果----------------");                  List<Group> groups =groupCommand.getValues();                 for(Group group : groups) {                  System.out.println(group.getGroupValue()+"数量为："+group.getResult().getNumFound());                  System.out.println(group.getResult());                 System.out.println("第一个文档:"+group.getResult().get(0));                  System.out.println("第二个文档:"+group.getResult().get(1));                  System.out.println("第三个文档:"+group.getResult().get(2));                  System.out.println("第一个文档内容:"+group.getResult().get(0).get("Content"));                  System.out.println("第二个文档内容:"+group.getResult().get(1).get("Content"));                  System.out.println("第三个文档内容:"+group.getResult().get(2).get("Content"));              }        }               }        } catch (Exception e) {                        e.printStackTrace();        }}        // 高亮部分(自动摘要)，lucene中是通过highlighter来操作高亮和摘要的    @Test    public void Highlight() {        HttpSolrServer server = new HttpSolrServer(URL);        SolrQuery query = new SolrQuery("Content:\"财经周刊\"  OR  UserLocation:北京 ");        // 设置高亮，以下两种方式都行（相当于开启高亮功能）        // query.setHighlight(true);        query.setParam("hl", "true"); // highlighting        /*         * 给多个字段开启高亮功能         * query.addHighlightField("Content");         * query.addHighlightField("UserLocation");         */                //hl.fl: 用空格或逗号隔开的字段列表。要启用某个字段的highlight功能，就得保证该字段在schema中是stored。        //如果该参数未被给出，那么就会高亮默认字段 standard handler会用df参数，dismax字段用qf参数。        //你可以使用星号去方便的高亮所有字段。如果你使用了通配符，那么要考虑启用hl.requiredFieldMatch选项。        query.setParam("hl.fl", "Content", "UserLocation");        // 高亮显示字段前后添加html代码        query.setHighlightSimplePre("<font color=\"red\">");        query.setHighlightSimplePost("</font>");//高亮时显示的格式，默认是<em></em>        query.setHighlightSnippets(3);        //这是highlighted片段的最大数。默认值为1，也几乎不会修改。获取高亮分片数，一般搜索词可能分布在文章中的不同位置，其所在一定长度的语句即为一个片段，默认为1，但根据业务需要有时候需要多取出几个分片。        //如果某个特定的字段的该值被置为0，这就表明该字段被禁用高亮了        query.setHighlightFragsize(0);// 每个snippet返回的最大字符数。默认是100.如果为0，那么该字段不会被fragmented且整个字段的值会被返回。大字段时不会这么做        query.set("hl.usePhraseHighlighte", true);//如果一个查询中含有短语（引号框起来的）那么会保证一定要完全匹配短语的才会被高亮。                //有时我们查询根据条件“java OR (empId:1000 AND empId:1001)”搜索时，结果如果高亮显示，可能出现1000,1001数字也会高亮，        //但是我们只希望java关键字高亮，这个时候可以用下面的方法,只对lucene和solr关键字进行高亮显示（solr不作为搜索条件也可以）       query.setParam("hl.q", "lucene solr");        query.setRows(10);        SolrDocumentList list = new SolrDocumentList();        SolrDocument document = null;        System.out.println(query);        try {            QueryResponse response = server.query(query);            SolrDocumentList documents = response.getResults();                        System.out.println("----原文档-----");            System.out.println(documents);            for (SolrDocument doc : documents) {                System.out.println(doc.getFieldValue("Content"));                System.out.println(doc.getFieldValue("UserLocation"));            }            // 第一个Map的键是文档的ID，第二个Map的键是高亮显示的字段名,List<String>中封装的就是高亮与摘要处理过后的内容了            Map<String, Map<String, List<String>>> map = response                    .getHighlighting();           //得到的结果documents替换为高亮内容或者加入一个新的list中            for (int i = 0; i < documents.size(); i++) {                document = documents.get(i);                document.setField("Content",                        map.get(document.getFieldValue("ID")).get("Content"));                document.setField(                        "UserLocation",                        map.get(document.getFieldValue("ID")).get(                                "UserLocation"));                list.add(document);            }            System.out.println(list);            System.out.println(documents);                        for (SolrDocument doc : documents) {                System.out.println(doc.getFieldValue("Content"));                System.out.println(doc.getFieldValue("UserLocation"));            }        } catch (Exception e) {            e.printStackTrace();        }    }                   //事件分析    @Test    public void test(){                HttpSolrServer server = new HttpSolrServer("http://192.168.20.196:8081/solr");            SolrQuery query = new SolrQuery("Content:我们");        QueryResponse response;        try {            response = server.query(query);                        System.out.println(response.getResults());        } catch (SolrServerException e) {                        e.printStackTrace();        }                     }     @Test    public void lucene(){                Field field=new Field("id","11",Field.Store.YES,Field.Index.NOT_ANALYZED_NO_NORMS);        field.setBoost(2);        Document document = new Document();        //DefaultSimilarity继承TFIDFSimilarity        //VSM(向量空间模型)   余弦相似度  TF/IDF(term frequency/inverse document frequency)是一种统计方法        //默认的评分类                //EnumMap是专门为枚举类型量身定做的Map实现。虽然使用其它的Map实现（如HashMap）也能完成枚举类型实例到值得映射，但是使用EnumMap会更加高效：        //它只能接收同一枚举类型的实例作为键值，并且由于枚举类型实例的数量相对固定并且有限，所以EnumMap使用数组来存放与枚举类型对应的值。这使得EnumMap的效率非常高。    }            @Test     public void treeMap(){                //传入的比较器只能根据key来排序，TreeMap如不指定排序器，默认将按照key值进行升序排序        //指定排序器按照key值降序排列 ，        //Comparator中泛型必须传入key类型的的超类TreeMap(Comparator<? super K> comparator)         TreeMap<String, Integer> treeMap=new TreeMap<String, Integer>(new Comparator<Object>() {            @Override            public int compare(Object o1, Object o2) {                return o2.hashCode()-(o1.hashCode());                                //如果key是String类型   return o2.compareTo(o1);            }        }) ;        treeMap.put("2", 1);          treeMap.put("b", 1);         treeMap.put("1", 1);          treeMap.put("a", 1);          System.out.println("treeMap="+treeMap);          }            //单个字段分组统计    //参考文章：http://shiyanjun.cn/archives/78.html    @Test    public void stats(){                SolrQuery query = new SolrQuery("Content:盗墓笔记");        HttpSolrServer server = new HttpSolrServer(BigSolrURL);        query.setGetFieldStatistics("ForwardNumber");        query.setGetFieldStatistics("RepeatNum");        try {            System.out.println(query );            QueryResponse response = server.query(query);            Map<String, FieldStatsInfo> map=    response.getFieldStatsInfo();                                    for( String  key :map.keySet()){                                FieldStatsInfo     fieldStatsInfo=    map.get(key);                System.out.println(fieldStatsInfo);                System.out.println(fieldStatsInfo.getName());                System.out.println(fieldStatsInfo.getCount());                System.out.println(fieldStatsInfo.getMax());                System.out.println(fieldStatsInfo.getMin());                System.out.println(fieldStatsInfo.getSum());                System.out.println(fieldStatsInfo.getMean());                System.out.println(fieldStatsInfo.getMissing());                System.out.println(fieldStatsInfo.getStddev());//标准差            }                    } catch (SolrServerException e) {                        e.printStackTrace();        }    }                //一次对多个字段独立分组统计    //这相当于执行两个带有GROUP BY(facet)子句的SQL，这两个GROUP BY分别只对一个字段进行汇总统计    @Test    public void statsFacet(){                SolrQuery query = new SolrQuery("Content:盗墓笔记");        HttpSolrServer server = new HttpSolrServer(BigSolrURL);            //第一个参数是要统计的字段(stats)，第二个参数是分组(Facet)字段        //setGetFieldStatistics不能少        query.setGetFieldStatistics("ForwardNumber");        query.setGetFieldStatistics("RepeatNum");        query.addStatsFieldFacets("ForwardNumber", "Spare3","Province");          query.addStatsFieldFacets("RepeatNum", "Spare3","Province");    System.out.println(query);        try {            QueryResponse response = server.query(query);                            Map<String, FieldStatsInfo> map=    response.getFieldStatsInfo();                    for( String  key :map.keySet()){            System.out.println("stats:"+key);                FieldStatsInfo     fieldStatsInfo=    map.get(key);                                System.out.println(fieldStatsInfo);                System.out.println(fieldStatsInfo.getName());                System.out.println(fieldStatsInfo.getCount());                System.out.println(fieldStatsInfo.getMax());                System.out.println(fieldStatsInfo.getMin());                System.out.println(fieldStatsInfo.getSum());                System.out.println(fieldStatsInfo.getMean());                System.out.println(fieldStatsInfo.getMissing());                System.out.println(fieldStatsInfo.getStddev());//标准差                                 Map<String, List<FieldStatsInfo>>        maps=    fieldStatsInfo.getFacets();//得到每个(facet)分组                 for ( String  facet: maps.keySet()){                     System.out.println("facet:"+facet);                     List<FieldStatsInfo> lists=     maps.get(facet);                     for(FieldStatsInfo facetFieldStatsInfo: lists ){                                                  System.out.println(facetFieldStatsInfo);                         System.out.println(facetFieldStatsInfo.getName());                            System.out.println(facetFieldStatsInfo.getCount());                            System.out.println(facetFieldStatsInfo.getMax());                            System.out.println(facetFieldStatsInfo.getMin());                            System.out.println(facetFieldStatsInfo.getSum());                            System.out.println(facetFieldStatsInfo.getMean());                            System.out.println(facetFieldStatsInfo.getMissing());                            System.out.println(facetFieldStatsInfo.getStddev());//标准差                     }                 }                }                                                                                    } catch (SolrServerException e) {                        e.printStackTrace();        }            }        //参考文章：http://blog.csdn.net/a925907195/article/details/42241265    //多个字段分组只支持(count函数)    @Test  public  void facetpivot()  {                try {            HttpSolrServer solrServer = new HttpSolrServer(URL);                        SolrQuery query = new SolrQuery("Content:盗墓笔记");                        query.setFacet(true);              //  query.add("facet.pivot", "Spare3,UserLocation");//根据这两维度来分组查询                query.addFacetPivotField("Spare3,Province");                System.out.println(query);                QueryResponse response = solrServer.query(query,SolrRequest.METHOD.POST);                     NamedList<List<PivotField>> namedList = response.getFacetPivot();                System.out.println(namedList);//底下为啥要这样判断，把这个值打印出来，你就明白了                if(namedList != null){                    List<PivotField> pivotList = null;                    for(int i=0;i<namedList.size();i++){                        pivotList = namedList.getVal(i);                        if(pivotList != null){                            for(PivotField pivot:pivotList){                                int pos = 0;                                int neg = 0;                                List<PivotField> fieldList = pivot.getPivot();                                if(fieldList != null){                                    for(PivotField field:fieldList){                                        int proValue = (Integer) field.getValue();                                        int count = field.getCount();                                        if(proValue == 1){                                            pos = count;                                        }else{                                            neg = count;                                        }                                    }                                }                            }                        }                    }                }        } catch (Exception e) {            e.printStackTrace();        }    }}

本文出自 “点滴积累” 博客，请务必保留此出处http://tianxingzhe.blog.51cto.com/3390077/1748191