存储过程mergeinto

① mysql 存储过程 若主键冲突则更新，不冲突则插入数据

你可以分两部份导入
先导入在表里有IDDATE的更新部份

mysql> UPDATE 表1 SET 列名=(SELECT 列名 FROM 表2
WHERE 表1.IDDATE=表2.IDDATE)
WHERE IDDATE in (SELECT IDDATE FROM 表2);

再导入插入部份
INSERT INTO 表1 (SELECT IDDATE,列名 FROM 表2
WHERE DDATE not in (SELECT IDDATE FROM 表2));

② Oracle使用merge into 编写存储过程 遇编译错误：PL/SQL: ORA-00926: 缺失 VALUES 关键字

Oracle使用猜码掘merge into 编写存储过程 遇编译错误，是设置错误造成的，解决方法如下：

1、首先穗核新建存储过程，Create or ReplaceProcere CHK_SYS_EMP 检查系统工号。

③ 在存储过程里面使用mergeinto出错了，看不出来哪里错了

merge into RoomType a using #RoomType b
on b.TypeName=a.TypeName
when matched then
update a --这个a不能要
set a.TypePrice=b.TypePrice
where b.TypeName=a.TypeName
when not matched then
insert into RoomType (这老燃个表好含氏名也不能友散加) values(@TypeName,@TypePrice,@IsTv,@IsKongTiao,
@Remark)

④ 怎么在存储过程中动态加入索引

create or replace procere sp_crt_stg_index( p_src_tbl_name varchar2,p_tgt_tbl_name varchar2,p_tx_date varchar2)
is
v_crt_ind_sql varchar2(4000) := '';
v_drop_ind_sql varchar2(4000) := '';
v_ind_name varchar2(100) := '';
v_tgt_ind_name varchar2(100) := '';
v_col_list varchar2(1000) := '';
v_col_name varchar2(100) := '';
i_ind_col_cnt number(2,0) := 0;
i_tbl_cnt number(2,0) := 0;
i_ind_cnt number(2,0) := 0;
v_step_no varchar2(3) :='';
begin
--step 1 检查索引定义是否存在
v_step_no :='1';
select count(1) into i_ind_col_cnt from all_ind_columns
where table_name = p_src_tbl_name;
if(i_ind_col_cnt = 0 ) then
sp_log('sp_crt_stg_index',v_step_no,'3',p_tx_date,p_src_tbl_name||'不存在索引定义');
return;
end if;
--step 2 检查目标表是否存在
v_step_no :='2';
select count(1) into i_tbl_cnt from tabs
where table_name = p_tgt_tbl_name;
if(i_tbl_cnt = 0 ) then
sp_log('sp_crt_stg_index',v_step_no,'3',p_tx_date,p_src_tbl_name||'不存在');
return;
end if;
--step 3 创建索引
v_step_no :='3';
for cur_ind in (select index_name from all_indexes where table_name = p_src_tbl_name) loop
v_ind_name := cur_ind.index_name;
v_col_name := '';
v_col_list := '(';
for cur_ind_col in (select column_name from all_ind_columns where index_name = v_ind_name order by column_position) loop
v_col_name := cur_ind_col.column_name;
v_col_list := v_col_list||v_col_name||',';
end loop;
v_col_list := v_col_list||')';
v_col_list := replace(v_col_list,',)',')');
v_tgt_ind_name :=replace(v_ind_name,p_src_tbl_name,p_tgt_tbl_name);
v_tgt_ind_name := v_tgt_ind_name||substr(p_tx_date,7,21);
select count(1) into i_ind_cnt from all_indexes where index_name = v_tgt_ind_name;
if(i_ind_cnt > 0) then
v_drop_ind_sql := 'drop index '||v_tgt_ind_name;
execute immediate v_drop_ind_sql;
end if;
sp_log('sp_crt_stg_index',v_step_no,'1',p_tx_date,v_tgt_ind_name||'创建开始');
v_crt_ind_sql := 'create index '||v_tgt_ind_name||' on '||p_tgt_tbl_name||v_col_list||' nologging';
if(v_crt_ind_sql is not null) then
execute immediate v_crt_ind_sql;
end if;
sp_log('sp_crt_stg_index',v_step_no,'1',p_tx_date,v_tgt_ind_name||'创建结束');
end loop;
exception
when others then
sp_log('sp_crt_stg_index',v_step_no,'3',p_tx_date,v_tgt_ind_name||'创建异常:'||SQLERRM);
end;

#########################

---- 日期类型转换
to_char(sysdate,'YYYY-MM-DD HH24:MI:SS');
----Oracle检查分区
select count(1)
from ALL_TAB_PARTITIONS
where table_name = p_tgt_tbl_name
and PARTITION_NAME = v_prt_name;
----码表
Select * From ict_s_dic Where opttype = 'ID_TYPE';
decode 内部字段拼接
decode(Date_Nextvisit,'','下次跟进时间：['||Date_Nextvisit||']') RMK1,

----拼接字段,删除表
SELECT 'DROP TABLE '||TABLE_NAME||';' FROM TABS WHERE TABLE_NAME LIKE 'M_ICT%';
Select * From tabs ;
----交换分区，分区交换后数据是否交换？
'ALTER TABLE '|| p_tgt_tbl_name ||' EXCHANGE PARTITION '||v_prt_name||' WITH TABLE '||p_src_tbl_name;
----查询某表是否存在
select count(1) from tabs
where table_name = p_src_tbl_name;
----重建索引
ALTER index ind_id_idx rebuild
----创建索引nologging
create index I_ICT_CUST_INFO__0 on ICT_CUST_INFO_20141222 (CUST_NO)
nologging;
----查询分区内数据
select count(1) from stg_ict_trade_info partition(ICT_PRT_2014005);
Select * From User_Ind_Partitions;
Select * From User_Part_Indexes;

-----查询索引
select index_name from ALL_INDEXES WHERE TABLE_NAME=p_src_tbl_name;
-----查询分区
select * from ALL_TAB_PARTITIONS
where table_name = p_tgt_tbl_name and PARTITION_NAME = v_prt_name;
-----查询表名
select * from tabs where table_name = p_tgt_tbl_name;
-----清空表分区数据
'ALTER TABLE '||p_tgt_tbl_name||' TRUNCATE PARTITION ' || v_prt_name;
-----增加表分区
'ALTER TABLE '||p_tgt_tbl_name||' ADD PARTITION ' || v_prt_name||' VALUES LESS THAN (''' ||v_monthend||''') TABLESPACE ICLIENT_O_DATA01 ';
-----重建索引
'ALTER INDEX '||cur_ind.index_name|| ' REBUILD PARALLEL 128 COMPUTE STATISTICS NOLOGGING';
-----交换分区
'ALTER TABLE '|| p_tgt_tbl_name ||' EXCHANGE PARTITION '||v_prt_name||' WITH TABLE '||p_src_tbl_name||' INCLUDING INDEXES';

-----oralce创建同义词------------

create or replace public SYNONYM ICT_ORG_BPH for iclientodata.ICT_ORG_BPH;

-----oracle 赋权限
grant select, insert, update, delete on ICT_ORG_BPH to ICLIENTOOPR; --赋权限

---------------DBA查看表空间------------
select a.tablespace_name,
a.bytes / 1024 / 1024 / 1024 "Sum G",
(a.bytes - b.bytes) / 1024 / 1024 / 1024 "used G",
b.bytes / 1024 / 1024 / 1024 "free G",
round(((a.bytes - b.bytes) / a.bytes) * 100, 2) "percent_used"
from (select tablespace_name, sum(bytes) bytes
from dba_data_files
group by tablespace_name) a,
(select tablespace_name, sum(bytes) bytes, max(bytes) largest
from dba_free_space
group by tablespace_name) b
where a.tablespace_name = b.tablespace_name
order by ((a.bytes - b.bytes) / a.bytes) desc

-------------------------------------------------------------------------------------
--oracle 生成删表语句，
SELECT 'DROP TABLE '||TABLE_NAME||';' FROM TABS WHERE TABLE_NAME LIKE 'M_ICT%'
--查询oracle 中ICT开头的
SELECT * FROM tabs WHERE TABLE_NAME LIKE 'ICT%'

----oracle 表分析语句
'ANALYZE TABLE ' || v_ana_tbl_name || ' estimate system statistics';

例如：ANALYZE TABLE ICT_SUM_AST_DBT_CUST ESTIMATE SYSTEM STATISTICS;
-----------查看锁定对象及会话
SELECT OBJECT_NAME,MACHINE,S.SID,S.SERIAL#
FROM GV$LOCKED_OBJECT I,DBA_OBJECTS O,GV$SESSION S
WHERE I.OBJECT_ID=O.OBJECT_ID AND I.SESSION_ID=S.SID;
----------oralce解锁
ALTER SYSTEM KILL SESSION '280,219';

授权脚本生成方法：
select 'grant select on table dmccrm.'||tbl_name||' to public,ex_sdbods; ' from t_ict_tbl_type where tbl_type in('ALL','GP') and ETL_DIR IN( 'GP->ORACLE','无需同步');

-----------oracle批量生成同义词
SELECT 'create or replace public synonym '||table_name ||' for iclientodata.'||table_name FROM tabs WHERE table_name like 'ICT_%'

-----------oracle批量生成修改表字段的长度
select 'ALTER TABLE '||TABLE_NAME||' MODIFY '||COLUMN_NAME||' NUMBER(30,8);' from cols t
where t.DATA_TYPE = 'NUMBER'
AND T.DATA_SCALE >0 AND SUBSTR(TABLE_NAME,-8,8)<>'20140531'

---------oracle 批量创建同义词
select 'create or replace public synonym '||table_name||' for iclientodata.'||table_name||';' from user_tables WHERE table_name LIKE 'ICT_%' AND table_name NOT LIKE '%20140531';

-----------oracle 交换分区语句
ALTER TABLE ICT_CUST_LEVEL_HIS ADD PARTITION ICT_PRT_20140531 VALUES LESS THAN ('2014-06-01')
--------MERGE使用方法
MERGE INTO ICT_CUST_INFO_ALL a
USING
(
SELECT t.CUST_MNG_UM_NO ,t.CUST_NO
FROM ict_cust_mnger_rel t
WHERE t.CUST_NO IN
('600021562650', '600037441214', '600036874754', '600038507516',
'600020226746', '600038089420', '600041030403', '600038952992',
'600039468303')
)b
ON( a.cust_no= b.cust_no)
WHEN MATCHED THEN
UPDATE SET a.MAX_ASSET_INTRO_NO=b.CUST_MNG_UM_NO;

-------------oracle 树形查询，查询机构编号为‘9902’的和其下级子机构
select org_id,org_name,org_level from (
SELECT rownum rn, ioi.org_id , ioi.org_name,ioi.org_level
FROM ict_org_info ioi
START WITH ioi.org_id ='9902'
CONNECT BY PRIOR ioi.org_id = ioi.parent_org_id_b
order by ioi.org_level desc
) where rn=1

-------------oracle 树形查询，查询机构编号为‘9902’的和上级机构

select org_id,org_name,org_level from (
SELECT rownum rn, ioi.org_id , ioi.org_name,ioi.org_level
FROM ict_org_info ioi
START WITH ioi.org_id ='9902'
CONNECT BY ioi.org_id = PRIOR ioi.parent_org_id_b
order by ioi.org_level desc
) where rn=1

-------------oracle 树形查询，查询机构编号为‘9902’的上级机构

select org_id,org_name,org_level from (
SELECT rownum rn, ioi.org_id , ioi.org_name,ioi.org_level
FROM ict_org_info ioi
START WITH ioi.org_id ='9902'
CONNECT BY ioi.org_id = PRIOR ioi.parent_org_id_b
order by ioi.org_level desc
) where rn=1

----------查看表名与表空间
Select * From user_tables Where table_name = 'ICT_RMT_APPO';
Select * From User_Tablespaces;

----------查询SQL 预估时间
SELECT SE.SID,
OPNAME,
TRUNC(SOFAR / TOTALWORK * 100, 2) || '%' AS PCT_WORK,
ELAPSED_SECONDS ELAPSED,
ROUND(ELAPSED_SECONDS * (TOTALWORK - SOFAR) / SOFAR) REMAIN_TIME,
SQL_TEXT
FROM V$SESSION_LONGOPS SL, V$SQLAREA SA, V$SESSION SE
WHERE SL.SQL_HASH_VALUE = SA.HASH_VALUE
AND SL.SID = SE.SID
AND SOFAR != TOTALWORK
ORDER BY START_TIME;
------TYPE opty_cur IS REF CURSOR;
整体的意思是“创建一个类型变量cur，它引用游标”，除了cur外，其余全是关键字。
TYPE cur：定义类型变量 ,is ref cursor：相当于数据类型，不过是引用游标的数据类型。
这种变量通常用于存储过程和函数返回结果集时使用，
因为PL/SQL不允许存储过程或函数直接返回结果集，
但可以返回类型变量，于是引用游标的类型变量作为输出参数或返回值就应运而生了。
----查杀进程
SELECT dob.OBJECT_NAME Table_Name,
lo.LOCKED_MODE,
lo.SESSION_ID,
vss.SERIAL#,
vps.spid,
vss.action Action,
vss.osuser OSUSER,
vss.process AP_PID,
VPS.SPID DB_PID,
'alter system kill session ' || '''' || lo.SESSION_ID || ',' ||
vss.SERIAL# || ''';' kill_command
from v$locked_object lo, dba_objects dob, v$session vss, V$PROCESS VPS
where lo.OBJECT_ID = dob.OBJECT_ID
and lo.SESSION_ID = vss.SID
AND VSS.paddr = VPS.addr
order by 2, 3, DOB.object_name;

⑤ Oracle 存储过程面试题

提供参考
DECLARE
BEGIN
mergeINTOt17_organt1
USINGt40_organt2
ONt1.organno=t2.organno
WHENmatchedTHEN
UPDATE
SETt1.organname=t2.organname,
t1.organlevel=t2.organlevel,
t1.uporgankey=t2.uporgankey,
t1.flag=t2.flag;
COMMIT;
END;
/

⑥ sqlserver 存储过程大批量数据插入

可使用Merge Into 语句完成操作。
具体用法参考文章：http://www.cnblogs.com/biwork/p/3370335.html

⑦ 存储过程优化oracle，merge into 表全表和查询表一部分哪个效率高

merge into主要是要将a与using的表进行一一对比，随着a表的增大，using后的表查询的改如渣速度在变慢，主要优化select *from a inner join b on a.id=b.id这条查询，因为a表在不断增大，b表的大小如何橡毕，二者的连接关系核悄是什么，通过执行计划可以看出。

⑧ oracle 用存储过程将某一些表里面的数据查出来之后，批量插入另一张大表中，并判断，若有些已经存在的数据

MERGE INTO table1 p
USING table2 np
ON (p.proct_id = np.proct_id)
WHEN MATCHED THEN
UPDATE
SET p.proct_name = np.proct_name,

WHEN NOT MATCHED THEN
INSERT
VALUES (np.proct_id, np.proct_name, np.category)
备注：
table1：是要插入数据的目标表
table2 ：是数据的来源表
针对你提出的需求是不用写存储过程的，上面的sql就完全可以实现；当然，你将上面的sql代码
放到过程体里面就OK了，你试下吧