前言:
Oracle的序列(SEQUENCE)A SCHEMA OBJECT THAT GENERATES A SERIAL LIST OF UNIQUE NUMBERS FOR TABLE COLUMNS.TO USE A SEQUENCE ,YOUR SCHEMA MUST CONTAIN THE SEQUENCE OR YOU MUST HAVE BEEN GRANTED THE SELECT OBJECT PRIVILEGE FOR ANOTHER USER IS SEQUENCE.ONCE A SEQUENCE IS DEFINED,IT CAN BE ACCESSED AND INCREMENTED BY MULTIPLE USERS(WHO HAVE SELECT OBJECT PRIVILEGE FOR THE SEQUENCE CONTAINING THE SEQUENCE)WITH NO WAITING.THE DATABASE DOES NOT WAIT FOR A TRANSACTION THAT HAS INCREMENTED A SEQUENCE TO COMPLETE BEFORE THAT SEQUENCE CAN BE INCREMENTED AGAIN.
简单说:序列一般用于自动递增生成主键值 ..
但是否有一些情况会导致调用SEQ_....NEXTVAL时大于主键最大值呢?
场景:
主键表 -> T表 '100W'数据同步到T1表(同步:TRUNCATE/DELETE或IMPDP...等方式[手动改数据])
例如: T表对应SEQ_T.NEXTVAL= 100W;
T1表对应SEQ_T.NEXTVAL= 10W;
TRUNCATE TABLE T1;
INSERT TABLE T1 SELECT * FROM T;
数据同步但T1表对应SEQ_T.NEXTVAL= 10W序列不变;
此时T1调用序列INSERT到T1表时就会报错 ( ORA-00001 : 违反唯一约束条件 (LOTTERY.PK_T1))
(若批量同步很多表就可能会引起很多表序列需要更新成主键最大值+ 1才能保证不报错
(可以用把源库的SEQUENCE同步过来①或者如下存储解决② ))
①PLSQL 工具的COMPARE USER OBJECTS可以解决SEQUENCE序列对象同步到其他环境...在此就不细说了
②我们环境序列一般都是由SEQ_表名字组成.所以写批量更新存储的时候比较方便~
如下存储针对常用于以上场景的环境,,生产环境不批量导新数据/同步用户数据/表数据 就很少用到...也可只提供参考...
--批量更新序列存储--
CREATE OR REPLACE PROCEDURE P_SYNCSEQ(USERNAME VARCHAR2 /*传入要检查/更新序列的用户*/ ) IS
/*
**@AUTHOR 毛海晴
ORACLE 批量更新SEQUENCE
注释:
批量更新SEQUENCE,
更新序列下一个值 = 主键最大值+1
---序列创建时,属性NOMAXVALUE=最大值是10的28次方
思路:
1、找到每个表主键列 且在该表主键最大值是什么?
2、找到表对应SEQUENCE值 与 表主键最大值去对比。
如果SEQUENCE 下一个值大于表主键最大值就不做更新;
否则需要进行更新(2中更新方式)
1)删除SEQUENCE ,创建新序列开始值为表主键最大值+1; --本文选择此方案...嘿嘿~
(坏处:赶好在DROP SEQUENCE..而程序也恰巧调用依赖它的函数和存储过程将失效
但 后续CREATE SEQUENCE了,再调用了会重新编译 调用..不会报错....有实验过哦~)
2)ALTER SEQUENCE .. INCREMENT BY .. NOCACHE;
SELECT ...NEXTVAL FROM DUAL;
ALTER SEQUENCE .. INCREMENT BY 1 CACHE 20;
.... sequence.nextval其实也可以用user_sequences.last_number字段实现..由于最早存储就这样的写就没改~...谅解~
SEQUENCE和表名长度最大限制是30
SEQUENCE规范的名字SEQ_+表名字 -- 此处规范只是管理维护方便而已 并不是非要这样要求
如果表名长度大小大于26 加上"SEQ_"就大于了SEQUENCE长度限制的30
若表名长度大于26,那对应序列肯定不是规范命名(SEQ_表名字),再由于这样的序列并不多,所以将这些一一处理
在更新前可先注释掉EXECUTE IMMEDIATE,先作下测试看下效果,免得EXECUTE IMMEDIATE DROP .后创建报错,导致在调用 序列不会创建,也校验不到序列
所需权限:
-- 创建序列权限 --
-- Grant/Revoke system privileges
grant create sequence to LOTTERY;
--查询权限--
-- Grant/Revoke object privileges
grant select on DBA_CONSTRAINTS to LOTTERY;
grant select on DBA_CONS_COLUMNS to LOTTERY;
grant select on DBA_SEQUENCES to LOTTERY;
grant select on DBA_TABLES to LOTTERY;
grant select on DBA_TAB_COLUMNS to LOTTERY;
--或者--
-- Grant/Revoke system privileges
grant select any dictionary to LOTTERY;
*/
--变量
MAX_ID NUMBER(12 );
P_SEQ_NUM NUMBER(12 );
P_TABLE_NAME VARCHAR2(50 );
P_COLUMN VARCHAR2(50 );
P_SEQUENCE VARCHAR2(50 );
P_SQL VARCHAR2(500 );
P_SEQ_SQL VARCHAR2(5000 );
P_SQL_SEQ VARCHAR2(30000 );
P_NEW_COUNT NUMBER(12 );
--查询表长度小于26 的表/序列
--游标
CURSOR C_CONS IS -- 查询表长度小于26 的表/序列
SELECT T1.TABLE_NAME TABLE_NAME,
T1.COLUMN_NAME COLUMN_NAME,
T1.SEQUENCE_NAME1 SEQUENCE_NAME
FROM ((SELECT C.TABLE_NAME,
CASE
WHEN C1.DATA_TYPE = 'NUMBER' THEN
C.COLUMN_NAME
ELSE
'TO_NUMBER(' || C.COLUMN_NAME || ')'
END COLUMN_NAME,
C.SEQUENCE_NAME1
FROM (SELECT C.TABLE_NAME,
C.COLUMN_NAME,
'SEQ_' || C.TABLE_NAME SEQUENCE_NAME1
FROM DBA_CONS_COLUMNS C --用户的约束对应的表列信息
WHERE C.OWNER = UPPER (USERNAME)
AND (C.CONSTRAINT_NAME, C.TABLE_NAME) IN
( SELECT S.CONSTRAINT_NAME, S.TABLE_NAME
FROM DBA_CONSTRAINTS S --用户的对象约束信息
WHERE S.OWNER = (UPPER (USERNAME))
AND S.CONSTRAINT_TYPE = 'P' /*CONSTRAINT_TYPE: P:主键,R:外键,C:非空约束/CHECK;*/
---若主键是由多字段'ID1,ID2',该查询会显示成2行分别为(T.ID1 SEQ_T和T.ID2 SEQ_T)
)
--..一个序列被2个表/2字段共用...可以用如下方式进行
UNION
SELECT 'ETL_CS_CUST_INFO_MID' ,
'BATCH_NO', --若数据为VARCHAR类型需要TO_NUMBER转换来取MAX(字段)
'SEQ_ETL_CS_CUST_INFO_MID'
FROM DUAL) C,
DBA_TAB_COLUMNS C1
WHERE C1.OWNER = UPPER (USERNAME)
AND C1.COLUMN_NAME = C.COLUMN_NAME
AND C1.TABLE_NAME = C.TABLE_NAME)
/**
---提供表长度大于26 的表名字/序列 ..再关联DBA_CONS_COLUMNS找到对应的主键字段..和表长度小于26部分的查询进行UNION ALL
CS_BEAR_ALLOWANCE_AND_INJ_DET ---> SEQ_CS_BEAR_ALLOWANCE_INJ_DET
CS_BEAR_ALLOWANCE_AND_INJ_DETS ---> SEQ_CS_BEAR_ALLOWANCE_INJ_DETS...等
*/
UNION ALL (SELECT M1.TABLE_NAME, COLUMN_NAME, M2.SEQUENCE_NAME
FROM (SELECT LENGTH(C.TABLE_NAME) AA,
C.TABLE_NAME,
C.COLUMN_NAME
FROM DBA_CONS_COLUMNS C
WHERE C.OWNER = UPPER (USERNAME)
AND (C.CONSTRAINT_NAME, C.TABLE_NAME) IN
( SELECT S.CONSTRAINT_NAME, S.TABLE_NAME
FROM DBA_CONSTRAINTS S
WHERE S.OWNER = UPPER (USERNAME)
AND S.CONSTRAINT_TYPE = 'P' )) M1 --如果不限制主键 可能找到NOT NULL的列
JOIN (SELECT TABLE_NAME, SEQUENCE_NAME
FROM (SELECT 'CS_BEAR_ALLOWANCE_AND_INJ_DET' TABLE_NAME,
'SEQ_CS_BEAR_ALLOWANCE_INJ_DET' SEQUENCE_NAME
FROM DUAL
UNION ALL
SELECT 'CS_BEAR_ALLOWANCE_AND_INJ_DETS' ,
'SEQ_CS_BEAR_ALLOWANCE_INJ_DETS'
FROM DUAL)) M2
ON M1.TABLE_NAME = M2.TABLE_NAME
WHERE AA > 26 )) T1,
DBA_SEQUENCES SQ, --(列出的序列是否在库中存在)
DBA_TABLES T --(列出的表是否在库中存在)..由于环境不同用到的序列可能也是不同的.若不加可能会报错
WHERE SQ.SEQUENCE_NAME = T1.SEQUENCE_NAME1
AND T.TABLE_NAME = T1.TABLE_NAME
AND SQ.SEQUENCE_OWNER = UPPER (USERNAME)
AND T.OWNER = UPPER (USERNAME);
----------------------以上查询表/对应序列/主键字段 -------------
----------------------以下开始判断序列是否需要更新 -------------
BEGIN
----------------------SEQUENCE判断更新语句 -----------------------------
--~~注释:DBMS_OUTPUT.PUT_LINE(XX)是将这个结果或者查询显示出来
--EXECUTE IMMEDIATE XX; --执行XX的查询
--开始 SEQUENCE.nextval和主键最大值 做比较..
FOR P_C_CONS IN C_CONS LOOP
--利用C_CONS游标对应列值
P_TABLE_NAME := P_C_CONS.TABLE_NAME;
P_COLUMN := P_C_CONS.COLUMN_NAME;
P_SEQUENCE := P_C_CONS.SEQUENCE_NAME;
---每次循环都赋值0 ..
MAX_ID := 0;
--查询表主键中最大值
P_SQL := 'SELECT MAX(' || P_COLUMN || ') FROM ' || P_TABLE_NAME;
--USING MAX_ID
EXECUTE IMMEDIATE P_SQL
INTO MAX_ID;
-- 查询序列.nextval值
P_SEQ_SQL := 'SELECT ' || P_SEQUENCE || '.NEXTVAL FROM DUAL' ;
--USING P_SEQ_SQL
EXECUTE IMMEDIATE P_SEQ_SQL
INTO P_SEQ_NUM;
---SEQUENCE.nextval和主键最大值 做比较..(如果SEQUENCE.nextval当前主键最大值+1 才是SEQUENCE要更新值,才保证主键值再加入的时候不冲突;
P_SQL_SEQ := 'CREATE SEQUENCE ' || P_SEQUENCE ||
' MINVALUE 1 NOMAXVALUE START WITH ' || P_NEW_COUNT ||
' INCREMENT BY 1 CACHE 20'; --创建正确的SEQUENCE语句
/*打印序列创建语句*/
/*DBMS_OUTPUT.PUT_LINE('CREATE SEQUENCE ' || P_SEQUENCE ||
' MINVALUE 1 NOMAXVALUE START WITH ' ||
P_NEW_COUNT || ' INCREMENT BY 1 CACHE 20');*/
--执行创建序列语句
EXECUTE IMMEDIATE P_SQL_SEQ;
--打印错 错误序列对应的表、序列由之前值更新到现在的值;
DBMS_OUTPUT.PUT_LINE( '错误序列对应的表:' || P_TABLE_NAME || '
' || P_SEQUENCE || ' 由' ||
P_SEQ_NUM || '更新到' || P_NEW_COUNT || ';' );
END IF ;
END LOOP;
END P_SYNCSEQ;
--使用步骤:
-- 编辑存储..-->调用存储(Call change_varchar2(username => 'u1' );或者begin..传值.等)
--输出结果:
--DROP SEQUENCE SEQ_T1 --本文中存储打印部分注释掉了.若想看其效果将注释/**/打开.
--CREATE SEQUENCE SEQ_T1 MINVALUE 1 NOMAXVALUE START WITH 1004 INCREMENT BY 1 CACHE 20 --本文中存储打印部分注释掉了.若想看其效果将注释/**/打开.
--错误序列对应的表:T1
SEQ_T1 由1000更新到1004;
祝好~
: