oracle SQL集锦

下面的语句都是本人整理出的，并且都在oracle XE上测试通过。下边的函数都没有做说明，相信大家都知道怎么用了，呵呵。

1   普通的查询语句，例如：
      SELECT  last_name, department_id  FROM  employees;
2    查询视图
      SELECT  employee_id, last_name, job_title, department_name, country_name, 
       region_name  FROM  emp_details_view;
3    更换列名查询
      SELECT  employee_id "Employee ID  number ", last_name "Employee last name", 
             first_name "Employee first name"  FROM  employees;
4    联接查询
     自然联接：要求两个表中的公共列必须有相同的名称和结构类型，否则报错。
      SELECT  employee_id, last_name, first_name, department_id, 
             department_name, manager_id   FROM  employees  NATURAL  JOIN  departments;
     查询三张或以上表时用USING，前提是用来连接两张表的列必须名称相同。
      SELECT  e.employee_id, e.last_name, e.first_name, e.manager_id, department_id, 
             d.department_name, d.manager_id  FROM  employees e 
              JOIN  departments d USING (department_id);
     联接时加上WHERE 子句。
   SELECT  e.employee_id, e.last_name, e.first_name, e.department_id,
             d.department_name, d.manager_id, d.location_id, l.country_id  FROM  employees e
              JOIN  departments d  ON  e.department_id  =  d.department_id
              JOIN  locations l  ON  d.location_id  =  l.location_id
              WHERE  l.location_id  =   1700 ;
     自我联接：
      SELECT  e.employee_id emp_id, e.last_name emp_lastname, m.employee_id mgr_id,
             m.last_name mgr_lastname
              FROM  employees e 
              JOIN  employees m  ON  e.manager_id  =  m.employee_id;
       LEFT   OUTER   JOIN ：左表中所有的记录即使在右表中没有联接的都会被查询出来。
      SELECT  e.employee_id, e.last_name, e.department_id, d.department_name
              FROM  employees e  LEFT   OUTER   JOIN  departments d
              ON  (e.department_id  =  d.department_id);
      RIGHT   OUTER   JOIN ：右表中所有的记录即使在左表中没有联接的都会被查询出来。
      SELECT  e.employee_id, e.last_name, e.department_id, d.department_name
             FROM  employees e  RIGHT   OUTER   JOIN  departments d 
             ON  (e.department_id  =  d.department_id);
      FULL   OUTER   JOIN ：左右表中的记录全部被查询出来。
      SELECT  e.employee_id, e.last_name, e.department_id, d.department_name
             FROM  employees e  FULL   OUTER   JOIN  departments d
             ON  (e.department_id  =  d.department_id);

5     绑定变量查询：Oracle将已解析、已编译的SQL连同其他内容存储在共享池中，这是SGA中一个非常重要的存储结构（内存结构主要分为SGA和PGA）。而绑定变量查询在Oracle里执行时只编译一次，随后就会把这个查询计划存储在一个共享池中以便重用，即所谓的软解析。具体的使用方法如下：
         SELECT   *   FROM  employees  WHERE  employee_id  =  :employee_id
6       查询虚拟列：所谓的虚拟列在Oracle里类似表格的列，但并非存储在表中。查询虚拟列时将返回一个值，因此它又类似与函数。Oracle里的虚拟列有：ROWNUM, SYSDATE,  and   USER 。eg：
         SELECT  SYSDATE "NOW"  FROM  DUAL;
         SELECT   USER   FROM  DUAL;
         SELECT  employee_id, hire_date, SYSDATE  FROM  employees  WHERE  ROWNUM  <   10 ;
7     带函数查询：
      Using Numeric Functions：
         SELECT  employee_id,  ROUND (salary / 30 ,  2 ) "Salary per  day "  FROM  employees;
         SELECT  employee_id, TRUNC(salary / 30 ,  0 ) "Salary per  day "  FROM  employees;
         SELECT  employee_id, MOD(employee_id,  2 )  FROM  employees;
      Using  Character  Functions：
         SELECT  employee_id,  UPPER (last_name),  LOWER (first_name)  FROM  employees;
         SELECT  employee_id, INITCAP(first_name), INITCAP(last_name)  FROM  employees;
         SELECT  employee_id,  RTRIM (first_name)  ||   '   '   ||   LTRIM (last_name)  FROM  employees;
         SELECT  employee_id, TRIM(last_name)  ||   ' ,  '   ||  TRIM(first_name)  FROM  employees;
         SELECT  employee_id, RPAD(last_name,  30 ,  '   ' ), first_name  FROM  employees;
         SELECT  employee_id, SUBSTR(last_name,  1 ,  10 )  FROM  employees;
         SELECT  LENGTH(last_name)  FROM  employees;
         SELECT  employee_id,  REPLACE (job_id,  ' SH ' ,  ' SHIPPING ' )  FROM  employees
                     WHERE  SUBSTR(job_id,  1 ,  2 )  =   ' SH ' ;
        Using Date Functions：
         SELECT  employee_id, TRUNC(MONTHS_BETWEEN(SYSDATE, HIRE_DATE)) "Months Employed" 
                      FROM  employees;
         SELECT  employee_id, EXTRACT( YEAR   FROM  hire_date) " Year  Hired"  FROM  employees;
         SELECT  EXTRACT( YEAR   FROM  SYSDATE)  ||  EXTRACT( MONTH   FROM  SYSDATE)  ||  
                   EXTRACT( DAY   FROM  SYSDATE) " Current  Date"  FROM  DUAL;
         SELECT  employee_id, hire_date, ADD_MONTHS(hire_date,  3 )  FROM  employees;
         SELECT  employee_id, hire_date, LAST_DAY(hire_date) "Last  day   of   month " 
                    FROM  employees;
         SELECT  SYSTIMESTAMP  FROM  DUAL;
      Using Conversion Functions：
         SELECT  TO_CHAR(SYSDATE,  ' DD-MON-YYYY AD ' ) "Today"  FROM  DUAL;
         SELECT  TO_CHAR(SYSDATE,  ' FMMonth DD YYYY ' ) "Today"  FROM  DUAL;
         SELECT  TO_CHAR(SYSDATE,  ' MM-DD-YYYY HH24:MI:SS ' ) "Now"  FROM  DUAL;
         SELECT  hire_date, TO_CHAR(hire_date, ' DS ' ) "Short Date"  FROM  employees;
         SELECT  hire_date, TO_CHAR(hire_date, ' DL ' ) " Long  Date"  FROM  employees;
         SELECT  TO_CHAR(EXTRACT( YEAR   FROM  SYSDATE))  ||  
                      TO_CHAR(EXTRACT( MONTH   FROM  SYSDATE), ' FM09 ' )  ||  
                      TO_CHAR(EXTRACT( DAY   FROM  SYSDATE), ' FM09 ' ) " Current  Date"  FROM  DUAL;
         SELECT  TO_CHAR( CURRENT_DATE ,  ' DD-MON-YYYY HH24:MI:SS ' ) " Current  Date"  FROM  DUAL;
         SELECT  TO_CHAR(salary, ' $99,999.99 ' ) salary  FROM  employees;
         SELECT  TO_NUMBER( ' 1234.99 ' )  +   500   FROM  DUAL;
         SELECT  TO_NUMBER( ' 11,200.34 ' ,  ' 99G999D99 ' )  +   1000   FROM  DUAL;
         SELECT  TO_DATE( ' 27-OCT-98 ' ,  ' DD-MON-RR ' )  FROM  DUAL;
         SELECT  TO_DATE( ' 28-Nov-05 14:10:10 ' ,  ' DD-Mon-YY HH24:MI:SS ' )  FROM  DUAL;
         SELECT  TO_DATE( ' January 15, 2006, 12:00 A.M. ' ,  ' Month dd, YYYY, HH:MI A.M. ' )
                      FROM  DUAL;
         SELECT  TO_TIMESTAMP( ' 10-Sep-05 14:10:10.123000 ' ,  ' DD-Mon-RR HH24:MI:SS.FF ' )
                      FROM  DUAL;
        Using Aggregate Functions：
       SELECT   COUNT ( * ) "Employee  Count "  FROM  employees  WHERE  manager_id  =   122 ;
         SELECT   COUNT ( * ) "Employee  Count ", manager_id   FROM  employees 
                     GROUP   BY  manager_id   ORDER   BY  manager_id
         SELECT   COUNT (commission_pct)  FROM  employees;
         SELECT   COUNT ( DISTINCT  department_id)  FROM  employees;
         SELECT   MIN (salary),  MAX (salary),  AVG (salary), job_id  FROM  employees 
                     GROUP   BY  job_id   ORDER   BY  job_id;
         SELECT  RANK( 2600 ) WITHIN  GROUP
                  ( ORDER   BY  salary  DESC ) "Rank  of  $ 2 , 600  among clerks"
                    FROM  employees  WHERE  job_id  LIKE   ' %CLERK ' ;
         SELECT  job_id, employee_id, last_name, salary, DENSE_RANK()  OVER
                  (PARTITION  BY  job_id  ORDER   BY  salary  DESC ) "Salary Rank (Dense)"
                    FROM  employees  WHERE  job_id  =   ' SH_CLERK ' ;
         SELECT  employee_id, salary, hire_date, STDDEV(salary) 
                    OVER  ( ORDER   BY  hire_date) "Std Deviation  of  Salary"
                    FROM  employees  WHERE  job_id  =   ' ST_CLERK ' ;
        Using  NULL  Value Functions：
         SELECT  commission_pct, NVL(commission_pct,  0 )  FROM  employees;
         SELECT  phone_number, NVL(phone_number,  ' MISSING ' )  FROM  employees;
         SELECT  employee_id , last_name,commission_pct, salary, 
                    NVL2(commission_pct, salary  +  (salary  *  commission_pct), salary) income
                     FROM  employees;
         Using Conditional Functions：
          SELECT  employee_id, hire_date , salary,
                     CASE   WHEN  hire_date  <  TO_DATE( ' 01-JAN-90 ' )  THEN  salary * 1.20
              WHEN  hire_date  <  TO_DATE( ' 01-JAN-92 ' )  THEN  salary * 1.15
              WHEN  hire_date  <  TO_DATE( ' 01-JAN-94 ' )  THEN  salary * 1.10  
              ELSE  salary * 1.05   END   "Revised Salary"
                     FROM  employees;
          SELECT  employee_id, job_id , salary,
                    DECODE(job_id,  ' PU_CLERK ' , salary * 1.05 ,
                  ' SH_CLERK ' , salary * 1.10 ,
                  ' ST_CLERK ' , salary * 1.15 ,
                             salary) "Revised Salary"
                     FROM  employees;