管理Oracle约束与分区表

管理Oracle约束与分区表

1．约束

作用：约束用于确保数据库数据的完整性,在oracle数据库中,可以使用约束,触发器和应用代码(过程,函数)3种方法实现数据完整性,这3种方法中,因为约束易于维护,并且具有最好的性能,所以实现数据完整性首选约束.分类：约束的种类有：not null,unique,primary key,foreign key,checkNot null确保字段值不能为空Unique:确保字段值唯一性Primary key,最常用的约束（主键约束），主键约束的列值不仅不能重复，也不能为NULL，注意一张表最多只能有一个主键约束，当定义主键约束后oracle自动建立一个以主键为关键字段的索引。Foreign key:定义了主从表之间的关系，foreign要定义在从表上，但主表必须具有主键约束或唯一约束，当定义froeign key后外部键列的数据必须在主表的主键列中存在，或者为NULLCheck::用于强制表行数据必须满足的条件，如工资表，工人工资必须在2000-5000之间约束状态enable validate:是默认,新旧数据同时满足约束规则enable novalidate:旧数据可以不满足,检新插入的数据要满足约束disable validate:不允许在表上执行任何DML操作,主要用在分区表,对于主键和唯一约事,会删除相应的唯一索引,但约束状态任可用disable novalidate数据可不满足约束规则,对于主键和唯一约事,会删除相应的唯一索引,约束常用语句

create table t(i number,v mubmer not null)create table t(i number,v mubmer unique)create table t(i number constraint pk_i primary key,v number)create table t2(c number,d number,constraint fk_d foreign key(c),references t1(v));alter table t add constraint pk_i primary key (i)alter table t modify i not null;alter table t add constraint t_i unique(i)[(create index ind_name on t(i))];alter table t add constraint t_i check(i in (1,2,3,4,5));alter table t disable novalidate constraint ialter table t enable novalidate constraint check_ialter table t drop constraint i;alter table t drop primary key i;#常用的数据字典dba_constraintsdba_cons_columnsuser_cons_columnsuser_constraints

简单应用检验当为一个表建立主键索时后，这个字段是否满足约束非空，唯一性，而且自动建立一个索引，并查看当把约束状态关闭再次插入相同的记录,是否还能把把约束设为enable ividate状态。

SQL> create table t(i number constraint pk_i primary key,v number);SQL> insert into t values(1,2);SQL> insert into t values(3,4);SQL> commit;SQL> select * from t; I V---------- --------------------------- 1 2 3 4

现在表中有两条记录，然后给它插主键为空或相同的值

SQL> insert into t values('',10);ERROR at line 1:ORA-01400: cannot insert NULL into ("Y"."T"."I")SQL> insert into t values(1,10);ERROR at line 1:ORA-00001: unique constraint (Y.PK_I) violated

可以看到全部报错，此时主键不能为空或重复查看是否建立索引

SQL> select index_name from user_indexes;INDEX_NAME------------------------------PK_I

把约束关闭再次做同样的操用

SQL> alter table t disable novalidate constraint pk_i;Table altered.SQL> insert into t values('',10);1 row created.SQL> insert into t values(1,10);1 row created.SQL> commit;Commit complete.SQL> select * from t; I V---------- ---------- 1 2 3 4 10 1 10SQL> select index_name from user_indexes;no rows selected

可见当把约束关闭后就可以何意给表插数据了，而具索引也自动删除了。现在激活约束

SQL> alter table t enable validate constraint pk_i;alter table t enable validate constraint pk_iERROR at line 1:ORA-02437: cannot validate (SYS.PK_I) - primary key violated

因为表中主键有相同的值所以不能恢复到enable validate状态了再次测试回复到enable novalidate

SQL> alter table t enable novalidate constraint pk_i;alter table t enable validate constraint pk_iERROR at line 1:ORA-02437: cannot validate (SYS.PK_I) - primary key violated

也失败了,因为表中主键有了空值和相同的值,所以恢复不到enable validate状态,但enable novalidate不检查旧数据所以应该还能恢复到enable novalidate.要想恢复到enable novalidate必须建立主键索引(关闭约束时自动删除的那个索引)如下:

SQL> create index pk_i on t(i);Index created.

然后恢复到enable disvalidate,以后再插数据不能为空,主键也不能重复了.

SQL> alter table t enable novalidate constraint pk_i;Table altered.SQL> insert into t values(1,14);insert into t values(1,14)ERROR at line 1:ORA-00001: unique constraint (SYS.PK_I) violated

2.修正约束数据当给一个表作主键约束时,因为已存数据不满足约束规则,会提示错误信息,些时必须对数据进行修正要修正数据先找出不满足约束的数据如下表,有不满足约束的数据

SQL> select * from t; I V---------- ------------------------ 1 2 3 4 15 12 15 10

如果一个表数据量多可通过如下方法查找

SQL> alter table t drop constraint pk_i; Table altered.SQL>conn y / 123SQL> @$ORACLE_HOME/rdbms/admin/utlexcpt.sqlTable created.SQL> alter table t add constraint pk_i primary key (i) exceptions into exceptions;select * from t where rowid in (select row_id from exceptions) I V---------- ------------------------ 15 12 15 10

找到了重复的记录修正

SQL>update t set i=10 where v=12;SQL> select * from t; I V---------- ---------- 1 2 3 4 10 12 15 10

再建主键约束

alter table t add constraint pk_i primary key (i)Table altered.

成功了!!!

二:分区表管理

作用:将在张大表的数据分布到多个表分区段,不同分区彼此独立,从而提高了表的可用性和性能种类:范围分区,散列分区(使用HASH算法,最常使用),列表分区,范围/散列组合分区,范围/列表组合分区范围分区表创建菬围分区表

create table t(v number,b number)partition by range(v) (partition p1 values less than ('11') tablespace test1,partition p2 values less than ('21') tablespace test2);

增加与删除分区

#增加分区alter table t add partition p3 values less than ('31') tablespace test3;alter table t drop partition p3

一个时间分区的例子

alter session set nls_data_lanage=AMERICAN;alter session set nls_data_format='DD-MON-YYYY'create table t(v_date date,b number)partition by range(v_date)(partition p1 values less than ('01-APR-2009') tablespace test1,partition p2 values less than ('01-JUN-2009') tablespace test2);

2.散列分区表(最常用)创建

create table t1(v number,b number)partition by hash(v)(partition p1 tablespace test1,partition p2 tablespace test2);

增加分区

alter table t add partition p3 tablespace test3;

删除分区

alter table t drop coalesce partition;

3.列表分区建列表分区

create table t(v varchar2(10),b number)partition by list(v)(partition p1 values('a','b') tablespace test1,partition p2 values('c','d') tablespace test2);

#插入数据

SQL> insert into t values('a',10);SQL> insert into t values('d',20);#注意,插入数据时第一个字段只能为a,b,c,d SQL> insert into t values('f',30);ERROR at line 1: ORA-14400: inserted partition key does not map to any partition

#查询

select * from t;select * from t partition(p1);select * from t partition(p2);select * from t where v=XXX

增加分区

alter table t add partition p3 values('31','32') tablespace test3;

删除分区

alter table t drop partition p3

4.范围/散列组合分区建立散列组合分区

create table t(v number,b number)partition by range(v)subpartition by hash(b) subpartitions 2store in (test1,test2)(partition p1 values less than ('11'),partition p2 values less than ('21'));

查询

select * from t;select * from t partition(p1);select * from t where ....

增加主分区和子分区

alter table t add partition p3 values less than ('31') tablespace test3;alter table t modify partition p3 add subpartition;

删除分区

alter table t coalesce partition;alter table t modify partition p1 coalesce subpartition;

5.范围/列表组合分区创建

create table t(v number,b number)partition by range(v)subpartition by list(b)(partition p1 values less than ('11') tablespace test1(subpartition p1_1 values('1','3'),subpartition p1_2 values('5','6')),partition p2 values less than ('21') tablespace test2(subpartition p2_1 values('13','14'),subpartition p2_2 values('15','16')));

查询

select * from tselect * from t partition(p1)select * from t subpartition(p1_1)select * from t where .....select segment_name,partition_name,tablespace_namefrom user_segments where segment_name='T';

增加分区和子分区

alter table t add partition p3 values less than ('31') tablespace test3(subpartition p3_1 values('25','26'),subpartition p3_2 values('22','23'));alter table t modify partition r3add subpartition r3_3 tablespace test3 values('28','29');

删除分区

alter table t modify partition p1 coalesce subpartition;

其它设置

交换分区数据alter table t exchange partition p1 with table tt;载断分区alter table t truncate partition p1;修改分区名alter table t rename partition p2_1 to p2;合并分区alter table t merge partitions p1,p2 into partition p01重组分区alter table t move partition p1 tablespace test04为列表分区和子分区加值alter table t modify partition p1 add values('111');alter table t modify subpartition p3_1 add values('111');从列表分区和子分区中删除值alter table t modify partition p1 drop values('111')alter table t modify subpartition p3_1 drop values('111')

分区表常用的数据字典

分区表信息: dba_part_tables显示分区: dba_tab_partitions显示子分区: dba_tab_subpartitions显示分区列: dba_part_key_columns显示子分区列:dba_subpart_dey_columns显示分区索引:dba_part_indexes显示索引分区:dba_ind_partitions

本文出自 “追求” 博客，请务必保留此出处http://chenxy.blog.51cto.com/729966/169250