我们有两张table。第一个表是员工，第二个表是书籍。

CREATE TABLE  EMPLOYEE
   (    "EMPLOYEE_ID" NUMBER(19,0) not null,
    "NAME" VARCHAR2(256),
     CONSTRAINT "ID_PK" PRIMARY KEY ("EMPLOYEE_ID") ENABLE
   );
   
   CREATE TABLE  BOOKS
   (    "ID" VARCHAR2(32) NOT NULL ENABLE,
    "BOOKNAME" VARCHAR2(256),
  "EMP_ID" NUMBER(19,0) not null,
     CONSTRAINT "BOOK_ID_PK" PRIMARY KEY ("ID") ENABLE,
     CONSTRAINT "EMP_FK1" FOREIGN KEY (EMP_ID) REFERENCES EMPLOYEE
   );
insert into employee  values (1, 'James');
insert into employee  values (2, 'John');

insert into BOOKS  values (1, 'JAVA', 1);
insert into BOOKS  values (2, 'CProgramming', 1);
insert into BOOKS  values (3, 'JSP', 1);
insert into BOOKS  values (4, 'Accountancy', 2);
insert into BOOKS  values (5, 'History', 2);
insert into BOOKS  values (6, 'Geography', 2);

在上面的代码中，可能有很多书籍，我们不想加载书籍行。因此，我们在where子句中使用rownum。有没有一种方法，我们可以使用rownum，并获得正确的计数，从雇员表只，虽然我们做了一个连接？

select distinct b.EMP_ID from EMPLOYEE e , Books b
where e.EMPLOYEE_ID= b.EMP_ID
and rownum <= 2;

select count(distinct b.EMP_ID) from EMPLOYEE e , Books b
where e.EMPLOYEE_ID= b.EMP_ID
and rownum <= 2;

**Expected output:**

     EMP_ID
----------
         1
         2