solr创建查询索引

1、 创建solrserver对象：

try {

              solr = newCommonsHttpSolrServer("http://localhost:8983/solr");

              solr.setConnectionTimeout(100);

              solr.setDefaultMaxConnectionsPerHost(100);

              solr.setMaxTotalConnections(100);

       } catch (Exception e) {

              System.out.println("请检查tomcat服务器或端口是否开启!");

              e.printStackTrace();

       }

 

2、 添加索引

Collection<SolrInputDocument>docs = newArrayList<SolrInputDocument>();

       for (int i = 0; i < list.size(); i++) {

              Item item = list.get(i);

              //设置每个字段不得为空，可以在提交索引前进行检查

              if (CheckItem(item)) {

                     SolrInputDocument doc = new SolrInputDocument();

                     //在这里请注意date的格式，要进行适当的转化，上文已提到

                     doc.addField("id", item.getId());

                     …………

                     docs.add(doc);

              }

       }

 

       try {

              solr.add(docs);

              //对索引进行优化

              solr.optimize();

              solr.commit();

       } catch (Exception e) {

              e.printStackTrace();

       }

3、 使用bean对象添加索引

      创建一个对应于solr索引的类别：

      public class Item {

            @Field

            private String id;

            …………

            public void setId(String id) {

                  this.id = id;

            }

            public String getId() {

                  return id;

            }

            …………

            public Item(){

            }

      }

 

      使用数据创建bean对象列表，

       try{

             solr.addBeans(beansList);

             } catch (Exception e) {

                    e.printStackTrace();

             } finally {

                    try {

                           solr.optimize();

                           solr.commit();

                    } catch (Exception e) {

                           e.printStackTrace();

                    }

             }

      索引提交完毕。

 

4、 删除索引

      据查询结果删除：      

       try{

                    //删除所有的索引

                    solr.deleteByQuery("*:*");

                    solr.commit();

             } catch (Exception e) {

                    e.printStackTrace();

             }

 

      根据索引号删除索引：

      try {

                    solr.deleteById(ids);

                    solr.commit();

             } catch (Exception e) {

                    e.printStackTrace();

             }

 

5、 查询

      SolrJ提供的查询功能比较强大，可以进行结果中查询、范围查询、排序等。

      下面是笔者封装的一个查询函数：

      

      public static QueryResponse Search(String[] field, String[] key,int start,

                    int count, String[] sortfield, Boolean[] flag, Boolean hightlight){

             //检测输入是否合法

             if (null == field || null == key || field.length != key.length){

                    return null;

             }

             if (null == sortfield || null == flag || sortfield.length !=flag.length) {

                    return null;

             }

             

             SolrQuery query = null;

             try {

                    //初始化查询对象

                    query = new SolrQuery(field[0] + ":" + key[0]);

                    for (int i = 0; i < field.length; i++) {

                           query.addFilterQuery(field[i] + ":" + key[i]);

                    }

                    //设置起始位置与返回结果数

                    query.setStart(start);

                    query.setRows(count);

                    //设置排序

                    for(int i=0; i<sortfield.length; i++){

                           if (flag[i]) {

                                  query.addSortField(sortfield[i], SolrQuery.ORDER.asc);

                           } else {

                                  query.addSortField(sortfield[i], SolrQuery.ORDER.desc);

                           }

                    }

                    //设置高亮

                    if (null != hightlight) {

                           query.setHighlight(true); // 开启高亮组件

                           query.addHighlightField("title");// 高亮字段

                           query.setHighlightSimplePre("<fontcolor="red">");// 标记

                           query.setHighlightSimplePost("</font>");

                           query.setHighlightSnippets(1);//结果分片数，默认为1

                           query.setHighlightFragsize(1000);//每个分片的最大长度，默认为100

                    }

             } catch (Exception e) {

                    e.printStackTrace();

             }

 

             QueryResponse rsp = null;

             try {

                    rsp = solr.query(query);

             } catch (Exception e) {

                    e.printStackTrace();

                    return null;

             }

             //返回查询结果

             return rsp;

      }

      补充一下范围查询的格式：[star t TOend]，start与end是相应数据格式的值的字符串形式，“TO”    一定要保持大写！

 

6、 读取查询结果

      DocList读取返回结果：

      SolrDocumentList solrList = rsp.getResults();

 

      Beans读取返回结果：

      List<Item> tmpLists =rsp.getBeans(Item.class);

 

      读取高亮显示结果：

                    rsp = Search(field, key, 0, 10, sortfield, flag, true);

                    if(null == rsp){

                           return;

                    }

                    Map<String, Map<String,List<String>>>hightlight = rsp.getHighlighting();

                    //Item即为上面定义的bean类

                    List<Item> tmpLists =rsp.getBeans(Item.class);

                    for (int i = 0; i < tmpLists.size(); i++){

                           //hightlight的键为Item的id，值唯一，我们设置的高亮字段为title

                           String hlString =hightlight.get(tmpLists.get(i).getId()).get("title").toString();

                           if (null != hlString) {

                                  System.out.println(hlString);

                           }

                    }