Question

那里：
我是两张桌子a）＆＃34; incidents_all＆＃34;和＆＃34;事件＆＃34;和一个序列＆＃34; incident_ids＆＃34; .ii）table，事件有一个额外的列id作为主键约束，而不是table，incidents_all。 iii）＆＃34; incidents_all＆＃34;包含重复记录。

目标：将唯一事件记录插入表格，＆＃34;事件＆＃34;来自表格，＆＃34; incidents_all＆＃34;以及序列号。

我正在尝试插入唯一记录，如下所示：
a）使用以下方法在需要唯一性的列上创建唯一索引：

CREATE UNIQUE INDEX "evnt_time_hst_flt_code_idx"  ON "incidents"
(
"host_name",
"down_since",
"fault_code"
);

b）填充＆＃34;事件＆＃34;通过从＆＃34; ids＆＃34;中选择序列号来获得具有唯一记录的表以及来自＆＃34; incidents_all＆＃34;。

的记录

INSERT INTO "incidents"
(
"id",
"sl_no",
"host_name"
"down_since",
"category",
"fault_code"
)
SELECT
"incident_ids".NEXTVAL AS "id",
"incidents_all"."sl_no" AS "sl_no",
"incidents_all"."host_name" AS "host_name",
"incidents_all"."down_since" AS "down_since",
"incidents_all"."category" AS "category",
"incidents_all"."fault_code" AS "fault_code"
FROM "incidents_all", "incident_ids";

但是，获得Oracle例外，＆＃34;序列不允许在这里！＆＃34;。

请引导我插入唯一记录和序列号！

Answer 1

试试这个：

INSERT INTO "incidents"
(
"id",
"sl_no",
"host_name"
"down_since",
"category",
"fault_code"
)
SELECT
"ids".NEXTVAL AS "id",
"incidents_all"."sl_no" AS "sl_no",
"incidents_all"."host_name" AS "host_name",
"incidents_all"."down_since" AS "down_since",
"incidents_all"."category" AS "category",
"incidents_all"."fault_code" AS "fault_code"
FROM "incidents_all";

您不需要在from子句中包含序列。

Answer 2

试试这个：

INSERT INTO "incidents"
(
"id",
"sl_no",
"host_name"
"down_since",
"category",
"fault_code"
)
SELECT
incident_ids.NEXTVAL AS "id",   -- ->!!!!
"incidents_all"."sl_no" AS "sl_no",
"incidents_all"."host_name" AS "host_name",
"incidents_all"."down_since" AS "down_since",
"incidents_all"."category" AS "category",
"incidents_all"."fault_code" AS "fault_code"
FROM "incidents_all";

如果需要序列的下一个值，则必须使用sequence_name.nextval。

祝你好运。

Answer 3

假设您有一个具有重复值的表，基于列a, b和另一个具有相同结构的表，您希望在其中插入非重复值。

create table duplicateTable(id number, a number, b number, c number);
create table uniqueTable ( id number primary key, a number, b number, c number);

假设您要使用序列作为第二个表的id

create sequence id_seq start with 1;

假设数据如下

insert into duplicateTable(id, a, b, c) values (1, 100, 50, 1000);
insert into duplicateTable(id, a, b, c) values (2, 100, 50, 2000);
insert into duplicateTable(id, a, b, c) values (3, 100, 99, 9000);
insert into duplicateTable(id, a, b, c) values (4, 999, 50, 7777);
insert into duplicateTable(id, a, b, c) values (5, 999, 99, 0);

您必须决定如何处理不属于您的唯一键的列;例如，假设您希望避免基于a, b上的键的重复项，并且您希望为给定的一对c获得(a, b)的最大值，则可以使用：< / p>

insert into uniqueTable ( id, a, b, c)
select id_seq.nextVal, a, b, c
from (
        select a, b, max(c) as c
        from duplicateTable
        group by a, b
     )

SQL> select * from uniqueTable;

        ID          A          B          C
---------- ---------- ---------- ----------
         1        999         99          0
         2        100         50       2000
         3        100         99       9000
         4        999         50       7777

Answer 4

参考，http://www.dba-oracle.com/t_packages_dbms_errlog.htm和Bhavesh Ghodasara的回答，

a）创建错误记录表：

CREATE TABLE err$_log_table
  (
    ora_err_number$ NUMBER,
    ora_err_mesg$ VARCHAR2(2000),
    ora_err_rowid$ rowid,
    ora_err_optyp$ VARCHAR2(2),
    ora_err_tag$   VARCHAR2(2000)
  );

b）填充＆＃34;事件＆＃34;来自＆＃34; incidents_all＆＃34;的记录表并从＆＃34; ids＆＃34;中选择序列号：

INSERT INTO "incidents"
(
"id",
"sl_no",
"host_name"
"down_since",
"category",
"fault_code"
)
SELECT
"incident_ids".NEXTVAL AS "id",
"incidents_all"."sl_no" AS "sl_no",
"incidents_all"."host_name" AS "host_name",
"incidents_all"."down_since" AS "down_since",
"incidents_all"."category" AS "category",
"incidents_all"."fault_code" AS "fault_code"
FROM "incidents_all"
LOG ERRORS INTO err$_log_table ('duplicate') REJECT LIMIT UNLIMITED;

为表＆＃34;事件添加了独特的记录＆＃34;并在遇到重复记录时记录表中的错误，错误$ _log_table。

如何在Oracle中使用INSERT INTO ... SELECT ... FROM ...去重复记录？

4 个答案: