科技行者

行者学院 转型私董会 科技行者专题报道 网红大战科技行者

知识库

知识库 安全导航

至顶网软件频道Oracle中使用的若干技术经典总结

Oracle中使用的若干技术经典总结

  • 扫一扫
    分享文章到微信

  • 扫一扫
    关注官方公众号
    至顶头条

  怎么样让我的用户名和密码不泄漏?    =====================   在unix下,我用sqlplus sys/sys登陆。

作者:中国IT实验室 来源:中国IT实验室 2007年10月7日

关键字: 备份 数据库 ORACLE

  • 评论
  • 分享微博
  • 分享邮件

  怎么样让我的用户名和密码不泄漏? 
  =====================
  在unix下,我用sqlplus sys/sys登陆,别的用户很容易就能看到我的密码:怎么办? 
   
  $ ps -ef|grep sqlplus 
  oracle 3787 3781 1 22:05:34 pts/3 0:00 sqlplus sys/sys 
  oracle 3789 3772 0 22:05:44 pts/2 0:00 grep sqlplus 
  采用sqlplus /nolog 
   
  sql>connect sys/sys,这样别的用户就看不到你的密码啦。 
   
  怎样生成建表的完整的DDL语句? 
  ====================
  用exp ,再Imp,show=y可以看到。 
   
  或者使用某些Oracle 的小工具,比如quest的toad和sql*navigator. 
   
  truncate table和delete table有些什么区别? 
  ==========================
  truncate: DDL ,no rollback possibility and no rollback segment usage, quick ,release space used by the table except the original one. 
   
  delete: dml, can rollback, use rollback space, not release space, slow, delete large table may cause ora-1555 error. 
   
  如何删除重复的记录: 
  =============
  第一个办法: 1。生成建表的完整DDL语句,并且生成tab_bak的表名。 
   
  2。insert into tab_bak select distinct * from tab_name; 
   
  3。drop table tab_name, rename tab_bak to tab_name; 
   
  第二个办法: 
   
  DELETE FROM table_name A WHERE ROWID > 
  ( SELECT min(rowid) FROM table_name B 
   
  WHERE A.key_values = B.key_values); 
   
  第三个办法: 
   
  Delete from my_table where rowid not in 
  ( SQL> select max(rowid) from my_table 
   
  group by my_column_name ); 
   
  第四个办法: 
   
  delete from my_table t1 
  where exists (select 'x' from my_table t2 
   
  where t2.key_value1 = t1.key_value1 
   
  and t2.key_value2 = t1.key_value2 
   
  and t2.rowid > t1.rowid); 
   
  如何快速为已有的表加上一个主键? 
  =====================
  加上一个非空的列,比如seqno,然后: 
   
  update table_name set seqno=rownum; 
   
  或者: 
   
  CREATE SEQUENCE testseq START WITH 1 INCREMENT BY 1; 
  update table_name set seqno=testseq.nextval; 
   
  SQL排序问题:我怎么才能选择出按照某个列排序后前N行来? 
  =====================================
  在SQL*Server 里面,可以用这样的语句:select top 10 col1,col2 from table_name; 
   
  从Oracle8i开始,支持这样的语法(在子查询里面使用order by语句) 
   
  select * from (select col1,col2 from table_name order by col1,col2) 
   
  where rownum<11; 
   
  这样就能够起到同样的效果。 
   
  在Oracle8或者以下,可以这样: 
   
  SELECT col1,col2 FROM 
  (SELECT /*+ INDEX_DESC (table_name index_name) */ col1,col2 FROM table_name) 
  WHERE rownum < 6; 
   
  使用提示可以让Oracle在子查询返回结果之前先对他进行排序,一般可以使用hintINDEX_DESC(TABLE_NAME,INDEX_NAME)来起到这个作用。 
   
  我们可以分别查看两个SQL的执行计划: 
   
  scott@testdb> select * from sort_sample; 
   
  ID NAME 
  ---------- ---------------------------------------- 
  1 aa 
  5 33 
  90 23s 
  23 fdisk 
  746 2343 
  24 format 
  3 low format 
  7 rows selected. 
   
  scott@testdb> create index sort_id_idx on sort_sample(id); 
   
  Index created. 
  scott@testdb> set autotrace on explain 
  scott@testdb> --way 1: 
  scott@testdb> select * from (select * from sort_sample order by id desc) where rownum<3; 
   
  ID NAME 
  ---------- ---------------------------------------- 
  746 2343 
  90 23s 
   
  Execution Plan 
  ------------------------------------- 
  0 SELECT STATEMENT Optimizer=CHOOSE 
  1 0 COUNT (STOPKEY) 
  2 1 VIEW 
  3 2 SORT (ORDER BY STOPKEY) 
  4 3 TABLE ACCESS (FULL) OF 'SORT_SAMPLE' 
   
  scott@testdb> --way 2 :wrong result 
  scott@testdb> select * from sort_sample where rownum<3; 
   
  ID NAME 
  ---------- ---------------------------------------- 
  1 aa 
  5 33 
   
  Execution Plan 
  ---------------------------------------------------------- 
  0 SELECT STATEMENT Optimizer=CHOOSE 
  1 0 COUNT (STOPKEY) 
  2 1 TABLE ACCESS (FULL) OF 'SORT_SAMPLE' 
   
  scott@testdb> ANALYZE TABLE SORT_SAMPLE COMPUTE STATISTICS; 
   
  Table analyzed. 
   
  scott@testdb> ANALYZE INDEX SORT_ID_IDX COMPUTE STATISTICS; 
   
  Index analyzed. 
   
  scott@testdb> --way 3: can work in oracle8 and oracle7 
  scott@testdb> select * from (select /*+index_desc(sort_sample sort_id_idx)*/ * from sort_sample) 
  2 where rownum<3; 
   
  ID NAME 
  ---------- ---------------------------------------- 
  1 aa 
  5 33 
   
  //原因:col sort_id_idx列为nullable,所以CBO不能确定,加上not null约束即可达到目的。 
   
  用group by可以生成从小开始的排序: 
   
  scott@testdb> SELECT ID,NAME FROM 
  2 (SELECT ID,NAME,COUNT(*) FROM SORT_SAMPLE GROUP BY ID, NAME) 
  3 WHERE ROWNUM<3; 
   
  ID NAME 
  ---------- ---------------------------------------- 
  1 aa 
  3 low format 
   
  Execution Plan 
  ---------------------------------------------------------- 
  0 SELECT STATEMENT Optimizer=CHOOSE (Cost=3 Card=7 Bytes=175) 
  1 0 COUNT (STOPKEY) 
  2 1 VIEW (Cost=3 Card=7 Bytes=175) 
  3 2 SORT (GROUP BY STOPKEY) (Cost=3 Card=7 Bytes=56) 
  4 3 TABLE ACCESS (FULL) OF 'SORT_SAMPLE' (Cost=1 Card=7 
   
  6。怎么每隔N条记录获得一条记录?比如第3,6,9等? 
  =================================
  CHAO@PING>select * from testseq; 
   
  ID NAME 
  ---------- ------------------------------------- 
  1 this is 1th record 
  2 this is 2th record 
  3 this is 3th record 
  4 this is 4th record 
  5 this is 5th record 
  6 this is 6th record 
  7 this is 7th record 
  8 this is 8th record 
  9 this is 9th record 
  10 this is 10th record 
   
  10 rows selected. 
   
  CHAO@PING>select id, name from 
  2 (select id, name, rownum rz from testseq) temp 
  3 where mod(rz,3)=0; 
   
  ID NAME 
  ---------- ------------------------------------------- 
  3 this is 3th record 
  6 this is 6th record 
  9 this is 9th record 
   
  CHAO@PING> 
   
  如何删除一个列? 
  ===========
  从Oracle8i开始,Oracle支持一个列的删除,语法如下: 
   
  alter table tab_name drop column col1; 
   
  7。如何重命名一个列? 
  ==============
  CHAO@PING> create table testrename(id number, nama varchar2(30)); 
   
  Table created. 
   
  CHAO@PING> begin 
  2 for x in 1..10 loop 
  3 insert into testrename values(x,'this is '||to_char(x)||'th record'); 
  4 end loop; 
  5 end; 
  6 / 
   
  PL/SQL procedure successfully completed. 
   
  CHAO@PING> commit; 
   
  Commit complete. 
   
  CHAO@PING> alter table testrename add name varchar2(30); 
   
  Table altered. 
   
  CHAO@PING> update testrename set name=nama; 
   
  10 rows updated. 
   
  CHAO@PING> alter table testrename drop column nama; 
   
  Table altered. 
   
  CHAO@PING> select * from testrename; 
   
  ID NAME 
  ---------- --------------------------------------------- 

查看本文来源

    • 评论
    • 分享微博
    • 分享邮件