MongoDB 查询总结

1、含日期查询 从起始时间到结束时间

BasicDBObject queryObj = new BasicDBObject();

queryObj.put("date",new BasicDBObject().append("$gte",starttime).append("$lt",endtime));

2、distinct 查询

BasicDBObject queryObj = new BasicDBObject();

coll.distinct("field_name",queryObj);// field_name 为数据库中需要distinct的字段名称，queryObj为查询条件 ，先执行查询条件后在distinct；返回为一个list集合 里面存放所有不重复的 field_name.

这种方式还是有中缺陷，如果数据量过大 list可能超出容量，不好控制，可以使用group by的方法替换

3、exists 查询

BasicDBObject basicObj = new BasicDBObject();
BasicDBObject andSearch = new BasicDBObject();

andSearch.put("$gte", start.getTime());andSearch.put("$lt", end.getTime());basicObj.put("date", andSearch);

basicObj.put("field_name", new BasicDBObject("$exists", true)); //若数据库中 field_name 这个字段有值 (包含null)，则符合条件，反之这条记录没有这个字段则不符合条件

4、group 查询 （分组查询）

Map<String, Integer> map = new HashMap<String,Integer>();
DBCollection coll = MongoDB.getDBCollection("user_processed_corpus");
DBObject fields = new BasicDBObject();
fields.put("cust_name", "$cust_name");
DBObject groupField = new BasicDBObject();
groupField.put("_id", fields);
groupField.put("count", new BasicDBObject("$sum",1));
DBObject group = new BasicDBObject("$group",groupField);
BasicDBObject matchObj = new BasicDBObject();
matchObj.put("date",new BasicDBObject("$lt", end.getTime()).append("$gt",start.getTime()));
matchObj.put("userid", userid);
matchObj.put("from", from);
DBObject match = new BasicDBObject("$match",matchObj);
AggregationOutput output = coll.aggregate(match, group);
Iterable<DBObject> results = output.results();
Iterator<DBObject> it = results.iterator();
while(it.hasNext()){
DBObject obj = it.next();
DBObject name = (DBObject) obj.get("_id");
Object cust_name =name.get("cust_name");
if(CommonUtil.isNotNull(cust_name)){
map.put(cust_name.toString(), Integer.valueOf(obj.get("count").toString()));
}
}

    4.2 group（方式2）

　　public static void group(){
        BasicDBObject group = new BasicDBObject();
        group.put("_id", "$userid");
        group.put("count", new BasicDBObject("$sum",1));
        AggregationOutput output = coll.aggregate(new BasicDBObject("$group",group));
        DBObject obj = output.getCommand();
        System.out.println(obj);
        CommandResult result = output.getCommandResult();
        System.out.println(result.get("result"));
    }

 5、addToSet(只能确保添加到记录，不能影响原有记录数据) 相同的数据则不添加

public static void addToSetAndEachElementInArray(String _id){  
            List list = new ArrayList();  
            list.add("str0");  
            list.add("str1");  
            list.add(new BasicDBObject("arr90",90));
            BasicDBObject idObj = new BasicDBObject("_id", new ObjectId(_id));
            BasicDBObject updateObj = new BasicDBObject();
            updateObj.put("$addToSet", new BasicDBObject("arrays",new BasicDBObject("$each",list)));
            WriteResult result = coll.update(idObj,updateObj);
            System.out.println("update count:"+result.getN());  
 } 

 6、or 查询 将多个查询条件加入一个list中 用$or 连接list中的查询条件即可

public static void test1(){
        BasicDBObject query1 = new BasicDBObject();
        BasicDBObject query2 = new BasicDBObject();
        query1.put("count", new BasicDBObject("$exists",false));
        query2.put("count", new BasicDBObject("$exists",true));
        BasicDBList list = new BasicDBList();
        list.add(query2);
        list.add(query1);
        BasicDBObject query = new BasicDBObject();
        query.put("$or", list);
        int count = coll.find(query).count();
        System.out.println(count);
        
    }

 7、in 查询

 public void findByINQuery(int value1,int value2){  
        coll = getCollection("ParentColl");  
        BasicDBObject query=new BasicDBObject();  
        BasicDBList cond=new BasicDBList();  
        cond.add(value1);  
        cond.add(value2);  
        query.put("intData",new BasicDBObject("$in", cond));  
        DBCursor cur = coll.find(query);  
        while(cur.hasNext()){  
            System.out.println(cur.next());  
        }  
    }  

8、addToSet的另外一种用法（本质相同）  对分组之后的数据在进行操作

userid  question

1　　　　a

1　　　　b

2　　　　c　　

2　　　　a　　　

2　　　　b

1　　　　a　　

3　　　　c　　

3　　　　b　　

3　　　　a

public static void group1(){
            BasicDBObject obj = new BasicDBObject();
            BasicDBObject query = new BasicDBObject();
            BasicDBObject match = new BasicDBObject();
            BasicDBObject group = new BasicDBObject();
            BasicDBObject g1 = new BasicDBObject();
            //第一个分组 （userid 和 question 共同作为分组字段）
            g1.put("userid", "$userid");
            g1.put("question", "$question");
            group.put("_id", g1);
            //可以额外的添加一些字段（以便下次的统计操作）
            group.put("count", new BasicDBObject("$sum",1));
            group.put("userid",new BasicDBObject("$addToSet", "$userid")); 
            group.put("question",new BasicDBObject("$addToSet", "$question")); 
            obj.put("$group", group);
            match.put("$match", query);
            
            //对第一次分组产生的结果进行再次分组（第一次分组的结果见注1）
            DBObject _group = new BasicDBObject();
            _group.put("question", "$question");
            DBObject groupFields1 = new BasicDBObject("_id", _group);
            groupFields1.put("user_sum", new BasicDBObject("$sum", 1));
            BasicDBObject group1 = new BasicDBObject();
            group1.append("$group", groupFields1);
            
            AggregationOutput out =  coll.aggregate(match,obj,group1);
            CommandResult result =  out.getCommandResult();
            BasicDBList record = (BasicDBList) result.get("result");
            System.out.println(record.size());
    }

注1：第一次分组的结果

[{
"_id" : {
"userid" : "0003",
"question" : "how"
},
"count" : 30,
"userid" : ["0003"],
"question" : ["how"]
}, {
"_id" : {
"userid" : "0002",
"question" : "I am comming"
},
"count" : 1,
"userid" : ["0002"],
"question" : ["I am comming"]
}, {
"_id" : {
"userid" : "0002",
"question" : "2323"
},
"count" : 1,
"userid" : ["0002"],
"question" : ["2323"]
}, {
"_id" : {
"userid" : "0002",
"question" : "how"
},
"count" : 38,
"userid" : ["0002"],
"question" : ["how"]
}, {
"_id" : {
"userid" : "0001",
"question" : "I comming"
},
"count" : 1,
"userid" : ["0001"],
"question" : ["I comming"]
}, {
"_id" : {
"userid" : "0002",
"question" : "123"
},
"count" : 1,
"userid" : ["0002"],
"question" : ["123"]
}, {
"_id" : {
"userid" : "0001",
"question" : "I am "
},
"count" : 1,
"userid" : ["0001"],
"question" : ["I am "]
}, {
"_id" : {
"userid" : "0001",
"question" : "I am comming"
},
"count" : 4,
"userid" : ["0001"],
"question" : ["I am comming"]
}
]

结果解释：_id 中的内容为分组的字段是唯一的，故作为_id 后面的count userid 和question都是在分组过程中手动添加上去的，这个其实就当做中间表 以便下面的数据统计和操作

9、模糊查询（pattern）

String like = "...";

Pattern pattern = Pattern.compile("^.*"+like+".*$", Pattern.CASE_INSENSITIVE);

BasicDBObject obj = new BasicDBObject();

obj.put("name",pattern);

coll.find(obj);

10、$eleMatch 查询（数组匹配查询）

DBObject edit = new BasicDBObject();
edit.put("slotname", "content");
query.put("slotbind", new BasicDBObject("$elemMatch", edit));

不难发现主要用到BasicDBObject、BasicDBList、DBCursor这三个类。BasicDBObject好比一个map，好比使用json查询中的{} 。BasicDBList是个list，用于or，nin等条件查询。DBCursor用于遍历结果集。其实只要将对应的json查询使用这3个类转化一下，就能写出对应的java代码了。