Oracle to PostgreSQL translate - oracle

I am migrating Oracle DLL to PostgreSQL, I am having an issue translate the following piece of code to PostgreSQL
-- Generate ID using sequence and trigger
CREATE SEQUENCE partner_seq START WITH 1 INCREMENT BY 1;
CREATE OR REPLACE TRIGGER partner_seq_tr
BEFORE INSERT ON partner FOR EACH ROW
WHEN (NEW.idPartner IS NULL)
BEGIN
SELECT partner_seq.NEXTVAL INTO :NEW.idPartner FROM DUAL;
END;
/
-- CREATE UNIQUE INDEX partner_idxName ON partner (Name);
-- COMMIT;

You don't actually need a trigger in Postgres for that. Simply declare the column as
idpartner integer generated always as identity
And Postgres will use the (automatically created) sequence automatically if the column is not specified as a target column in the INSERT statement.
Alternatively, if you really want a trigger:
create sequence partner_seq;
create function assign_partner_id()
returns trigger
as
$$
begin
if new.idpartner is null then
new.idpartner := nextval('partner_seq');
end if;
return new;
end;
$$
language plpgsql;
create trigger partner_seq_trg
before on partner insert on each row
execute procedure assign_partner_id();

Related

Bind variable in SQL Developer with Oracle 11g

I have problem when trying to excute a sequence with a before insert trigger.
CREATE TABLE personne (ID number, nom varchar2(250 char));
CREATE SEQUENCE s_inc_pers START WITH 1 INCREMENT BY 1;
CREATE TRIGGER tr_inc_pers ON t1 BEFORE INSERT
FOR EACH ROW
DECLARE
BEGIN
select s_inc_pers into :new.t1.ID from DUAL;
END.
Oracle reports a bad bind variable if you try to reference something with :NEW that is not a column in the target table. In this case it may be prompting you for a bind value as the statement is malformed.
The NEW pseudorecord refers to the row in the triggering table that is being affected by the statement. You don't need to (and mustn't) include the table name when you use the pseudorecord field, so :new.t1.ID should just be :new.ID.
To get the next value from the sequence you need to use nextval, you can't only provide the sequence name.
Your clauses are also in the wrong order, you need the DML event (insert) befroe the target table.
CREATE TRIGGER tr_inc_pers BEFORE INSERT ON t1
FOR EACH ROW
BEGIN
select s_inc_pers.nextval into :new.ID from DUAL;
END;
As you are using 11g you don't even need to select from dual, you can just assign the column value:
CREATE TRIGGER tr_inc_pers BEFORE INSERT ON t1
FOR EACH ROW
BEGIN
:new.ID := s_inc_pers.nextval;
END;

Oracle auto increment trigger issue

I am having an issue when creating the auto increment trigger in Oracle 11g. If someone can point out what I am doing wrong, I would really appreciate it. My script for the sequence is this :
CREATE SEQUENCE SPSS_QUOTE_LINE_ITEMS_SEQ start with 1
increment by 1
minvalue 1;
The script for trigger:
CREATE OR REPLACE TRIGGER SPSSQUOTELINEITEMS_ON_INSERT
BEFORE INSERT ON SPSS_QUOTE_LINE_ITEMS
FOR EACH ROW
BEGIN
SELECT SPSS_QUOTE_LINE_ITEMS_SEQ.NEXTVAL
INTO :new.line_num
FROM dual;
END;
The error I am getting:
[Code: 900, SQL State: 42000] ORA-00900: invalid SQL statement
Thanks a lot.
The correct way of doing this to modify your trigger as below:
CREATE OR REPLACE TRIGGER SPSSQUOTELINEITEMS_ON_INSERT
BEFORE INSERT ON SPSS_QUOTE_LINE_ITEMS
FOR EACH ROW
BEGIN
:new.line_num := SPSS_QUOTE_LINE_ITEMS_SEQ.NEXTVAL;
--No need to write any Select Into statement here.
END;
Or if i follow your way then it goes like
CREATE OR REPLACE TRIGGER SPSSQUOTELINEITEMS_ON_INSERT
BEFORE INSERT ON SPSS_QUOTE_LINE_ITEMS
FOR EACH ROW
declare
var number;
BEGIN
SELECT SPSS_QUOTE_LINE_ITEMS_SEQ.NEXTVAL
INTO var
FROM dual;
:new.line_num :=var;
END;
You normally use the terms in a trigger using :old to reference the old value of the column and :new to reference the new value.

ORA-00907: missing right parenthesis on create table query [duplicate]

It appears that there is no concept of AUTO_INCREMENT in Oracle, up until and including version 11g.
How can I create a column that behaves like auto increment in Oracle 11g?
There is no such thing as "auto_increment" or "identity" columns in Oracle as of Oracle 11g. However, you can model it easily with a sequence and a trigger:
Table definition:
CREATE TABLE departments (
ID NUMBER(10) NOT NULL,
DESCRIPTION VARCHAR2(50) NOT NULL);
ALTER TABLE departments ADD (
CONSTRAINT dept_pk PRIMARY KEY (ID));
CREATE SEQUENCE dept_seq START WITH 1;
Trigger definition:
CREATE OR REPLACE TRIGGER dept_bir
BEFORE INSERT ON departments
FOR EACH ROW
BEGIN
SELECT dept_seq.NEXTVAL
INTO :new.id
FROM dual;
END;
/
UPDATE:
IDENTITY column is now available on Oracle 12c:
create table t1 (
c1 NUMBER GENERATED by default on null as IDENTITY,
c2 VARCHAR2(10)
);
or specify starting and increment values, also preventing any insert into the identity column (GENERATED ALWAYS) (again, Oracle 12c+ only)
create table t1 (
c1 NUMBER GENERATED ALWAYS as IDENTITY(START with 1 INCREMENT by 1),
c2 VARCHAR2(10)
);
Alternatively, Oracle 12 also allows to use a sequence as a default value:
CREATE SEQUENCE dept_seq START WITH 1;
CREATE TABLE departments (
ID NUMBER(10) DEFAULT dept_seq.nextval NOT NULL,
DESCRIPTION VARCHAR2(50) NOT NULL);
ALTER TABLE departments ADD (
CONSTRAINT dept_pk PRIMARY KEY (ID));
SYS_GUID returns a GUID-- a globally unique ID. A SYS_GUID is a RAW(16). It does not generate an incrementing numeric value.
If you want to create an incrementing numeric key, you'll want to create a sequence.
CREATE SEQUENCE name_of_sequence
START WITH 1
INCREMENT BY 1
CACHE 100;
You would then either use that sequence in your INSERT statement
INSERT INTO name_of_table( primary_key_column, <<other columns>> )
VALUES( name_of_sequence.nextval, <<other values>> );
Or you can define a trigger that automatically populates the primary key value using the sequence
CREATE OR REPLACE TRIGGER trigger_name
BEFORE INSERT ON table_name
FOR EACH ROW
BEGIN
SELECT name_of_sequence.nextval
INTO :new.primary_key_column
FROM dual;
END;
If you are using Oracle 11.1 or later, you can simplify the trigger a bit
CREATE OR REPLACE TRIGGER trigger_name
BEFORE INSERT ON table_name
FOR EACH ROW
BEGIN
:new.primary_key_column := name_of_sequence.nextval;
END;
If you really want to use SYS_GUID
CREATE TABLE table_name (
primary_key_column raw(16) default sys_guid() primary key,
<<other columns>>
)
In Oracle 12c onward you could do something like,
CREATE TABLE MAPS
(
MAP_ID INTEGER GENERATED ALWAYS AS IDENTITY (START WITH 1 INCREMENT BY 1) NOT NULL,
MAP_NAME VARCHAR(24) NOT NULL,
UNIQUE (MAP_ID, MAP_NAME)
);
And in Oracle (Pre 12c).
-- create table
CREATE TABLE MAPS
(
MAP_ID INTEGER NOT NULL ,
MAP_NAME VARCHAR(24) NOT NULL,
UNIQUE (MAP_ID, MAP_NAME)
);
-- create sequence
CREATE SEQUENCE MAPS_SEQ;
-- create tigger using the sequence
CREATE OR REPLACE TRIGGER MAPS_TRG
BEFORE INSERT ON MAPS
FOR EACH ROW
WHEN (new.MAP_ID IS NULL)
BEGIN
SELECT MAPS_SEQ.NEXTVAL
INTO :new.MAP_ID
FROM dual;
END;
/
Here are three flavors:
numeric. Simple increasing numeric value, e.g. 1,2,3,....
GUID. globally univeral identifier, as a RAW datatype.
GUID (string). Same as above, but as a string which might be easier to handle in some languages.
x is the identity column. Substitute FOO with your table name in each of the examples.
-- numerical identity, e.g. 1,2,3...
create table FOO (
x number primary key
);
create sequence FOO_seq;
create or replace trigger FOO_trg
before insert on FOO
for each row
begin
select FOO_seq.nextval into :new.x from dual;
end;
/
-- GUID identity, e.g. 7CFF0C304187716EE040488AA1F9749A
-- use the commented out lines if you prefer RAW over VARCHAR2.
create table FOO (
x varchar(32) primary key -- string version
-- x raw(32) primary key -- raw version
);
create or replace trigger FOO_trg
before insert on FOO
for each row
begin
select cast(sys_guid() as varchar2(32)) into :new.x from dual; -- string version
-- select sys_guid() into :new.x from dual; -- raw version
end;
/
update:
Oracle 12c introduces these two variants that don't depend on triggers:
create table mytable(id number default mysequence.nextval);
create table mytable(id number generated as identity);
The first one uses a sequence in the traditional way; the second manages the value internally.
Oracle Database 12c introduced Identity, an auto-incremental (system-generated) column.
In the previous database versions (until 11g), you usually implement an Identity by creating a Sequence and a Trigger.
From 12c onward, you can create your own Table and define the column that has to be generated as an Identity.
Assuming you mean a column like the SQL Server identity column?
In Oracle, you use a SEQUENCE to achieve the same functionality. I'll see if I can find a good link and post it here.
Update: looks like you found it yourself. Here is the link anyway:
http://www.techonthenet.com/oracle/sequences.php
Trigger and Sequence can be used when you want serialized number that anyone can easily read/remember/understand. But if you don't want to manage ID Column (like emp_id) by this way, and value of this column is not much considerable, you can use SYS_GUID() at Table Creation to get Auto Increment like this.
CREATE TABLE <table_name>
(emp_id RAW(16) DEFAULT SYS_GUID() PRIMARY KEY,
name VARCHAR2(30));
Now your emp_id column will accept "globally unique identifier value".
you can insert value in table by ignoring emp_id column like this.
INSERT INTO <table_name> (name) VALUES ('name value');
So, it will insert unique value to your emp_id Column.
Starting with Oracle 12c there is support for Identity columns in one of two ways:
Sequence + Table - In this solution you still create a sequence as you normally would, then you use the following DDL:
CREATE TABLE MyTable
(ID NUMBER DEFAULT MyTable_Seq.NEXTVAL,
...)
Table Only - In this solution no sequence is explicitly specified. You would use the following DDL:
CREATE TABLE MyTable (ID NUMBER GENERATED AS IDENTITY, ...)
If you use the first way it is backward compatible with the existing way of doing things. The second is a little more straightforward and is more inline with the rest of the RDMS systems out there.
it is called Identity Columns and it is available only from oracle Oracle 12c
CREATE TABLE identity_test_tab
(
id NUMBER GENERATED ALWAYS AS IDENTITY,
description VARCHAR2 (30)
);
example of insert into Identity Columns as below
INSERT INTO identity_test_tab (description) VALUES ('Just DESCRIPTION');
1 row created.
you can NOT do insert like below
INSERT INTO identity_test_tab (id, description) VALUES (NULL, 'ID=NULL and DESCRIPTION');
ERROR at line 1: ORA-32795: cannot insert into a generated always
identity column
INSERT INTO identity_test_tab (id, description) VALUES (999, 'ID=999 and DESCRIPTION');
ERROR at line 1: ORA-32795: cannot insert into a generated always
identity column
useful link
Here is complete solution w.r.t exception/error handling for auto increment, this solution is backward compatible and will work on 11g & 12c, specifically if application is in production.
Please replace 'TABLE_NAME' with your appropriate table name
--checking if table already exisits
BEGIN
EXECUTE IMMEDIATE 'DROP TABLE TABLE_NAME';
EXCEPTION WHEN OTHERS THEN NULL;
END;
/
--creating table
CREATE TABLE TABLE_NAME (
ID NUMBER(10) PRIMARY KEY NOT NULL,
.
.
.
);
--checking if sequence already exists
BEGIN
EXECUTE IMMEDIATE 'DROP SEQUENCE TABLE_NAME_SEQ';
EXCEPTION WHEN OTHERS THEN NULL;
END;
--creating sequence
/
CREATE SEQUENCE TABLE_NAME_SEQ START WITH 1 INCREMENT BY 1 MINVALUE 1 NOMAXVALUE NOCYCLE CACHE 2;
--granting rights as per required user group
/
GRANT SELECT, INSERT, UPDATE, DELETE ON TABLE_NAME TO USER_GROUP;
-- creating trigger
/
CREATE OR REPLACE TRIGGER TABLE_NAME_TS BEFORE INSERT OR UPDATE ON TABLE_NAME FOR EACH ROW
BEGIN
-- auto increment column
SELECT TABLE_NAME_SEQ.NextVal INTO :New.ID FROM dual;
-- You can also put some other required default data as per need of your columns, for example
SELECT SYS_CONTEXT('USERENV', 'SESSIONID') INTO :New.SessionID FROM dual;
SELECT SYS_CONTEXT('USERENV','SERVER_HOST') INTO :New.HostName FROM dual;
SELECT SYS_CONTEXT('USERENV','OS_USER') INTO :New.LoginID FROM dual;
.
.
.
END;
/
Query to create auto increment in oracle. In below query incrmnt column value will be auto incremented wheneever a new row is inserted
CREATE TABLE table1(
id RAW(16) NOT NULL ENABLE,
incrmnt NUMBER(10,0) GENERATED ALWAYS AS IDENTITY
MINVALUE 1 MAXVALUE 999999999999999999999999999 INCREMENT BY 1 START WITH 1 NOORDER NOCYCLE NOT NULL ENABLE,
CONSTRAINT PK_table1 PRIMARY KEY (id) ENABLE);
This is how I did this on an existing table and column (named id):
UPDATE table SET id=ROWNUM;
DECLARE
maxval NUMBER;
BEGIN
SELECT MAX(id) INTO maxval FROM table;
EXECUTE IMMEDIATE 'DROP SEQUENCE table_seq';
EXECUTE IMMEDIATE 'CREATE SEQUENCE table_seq START WITH '|| TO_CHAR(TO_NUMBER(maxval)+1) ||' INCREMENT BY 1 NOMAXVALUE';
END;
CREATE TRIGGER table_trigger
BEFORE INSERT ON table
FOR EACH ROW
BEGIN
:new.id := table_seq.NEXTVAL;
END;
FUNCTION GETUNIQUEID_2 RETURN VARCHAR2
AS
v_curr_id NUMBER;
v_inc NUMBER;
v_next_val NUMBER;
pragma autonomous_transaction;
begin
CREATE SEQUENCE sequnce
START WITH YYMMDD0000000001
INCREMENT BY 1
NOCACHE
select sequence.nextval into v_curr_id from dual;
if(substr(v_curr_id,0,6)= to_char(sysdate,'yymmdd')) then
v_next_val := to_number(to_char(SYSDATE+1, 'yymmdd') || '0000000000');
v_inc := v_next_val - v_curr_id;
execute immediate ' alter sequence sequence increment by ' || v_inc ;
select sequence.nextval into v_curr_id from dual;
execute immediate ' alter sequence sequence increment by 1';
else
dbms_output.put_line('exception : file not found');
end if;
RETURN 'ID'||v_curr_id;
END;
FUNCTION UNIQUE2(
seq IN NUMBER
) RETURN VARCHAR2
AS
i NUMBER := seq;
s VARCHAR2(9);
r NUMBER(2,0);
BEGIN
WHILE i > 0 LOOP
r := MOD( i, 36 );
i := ( i - r ) / 36;
IF ( r < 10 ) THEN
s := TO_CHAR(r) || s;
ELSE
s := CHR( 55 + r ) || s;
END IF;
END LOOP;
RETURN 'ID'||LPAD( s, 14, '0' );
END;
Creating a Sequence:
CREATE SEQUENCE SEQ_CM_LC_FINAL_STATUS
MINVALUE 1 MAXVALUE 999999999999999999999999999
INCREMENT BY 1 START WITH 1 CACHE 20 NOORDER NOCYCLE;
Adding a Trigger
CREATE OR REPLACE TRIGGER CM_LC_FINAL_STATUS_TRIGGER
BEFORE INSERT
ON CM_LC_FINAL_STATUS
FOR EACH ROW
BEGIN
:NEW.LC_FINAL_STATUS_NO := SEQ_CM_LC_FINAL_STATUS.NEXTVAL;
END;
The first step is to create a SEQUENCE in your database, which is a data object that multiple users can access to automatically generate incremented values. As discussed in the documentation, a sequence in Oracle prevents duplicate values from being created simultaneously because multiple users are effectively forced to “take turns” before each sequential item is generated. –
Finally, we’ll create our SEQUENCE that will be utilized later to actually generate the unique, auto incremented value. –
While we have our table created and ready to go, our sequence is thus far just sitting there but never being put to use. This is where TRIGGERS come in. Similar to an event in modern programming languages, a TRIGGER in Oracle is a stored procedure that is executed when a particular event occurs. Typically a TRIGGER will be configured to fire when a table is updated or a record is deleted, providing a bit of cleanup when necessary. –
In our case, we want to execute our TRIGGER prior to INSERT into our CM_LC_FINAL_STATUS table, ensuring our SEQUENCE is incremented and that new value is passed onto our primary key column.
create trigger t1_trigger
before insert on AUDITLOGS
for each row
begin
select t1_seq.nextval into :new.id from dual;
end;
only I have to just change the table name (AUDITLOGS) with your table name and new.id with new.column_name
oracle has sequences AND identity columns in 12c
http://www.oracle-base.com/articles/12c/identity-columns-in-oracle-12cr1.php#identity-columns
I found this but not sure what rdb 7 is
http://www.oracle.com/technetwork/products/rdb/0307-identity-columns-128126.pdf
Maybe just try this simple script:
http://www.hlavaj.sk/ai.php
Result is:
CREATE SEQUENCE TABLE_PK_SEQ;
CREATE OR REPLACE TRIGGER TR_SEQ_TABLE BEFORE INSERT ON TABLE FOR EACH ROW
BEGIN
SELECT TABLE_PK_SEQ.NEXTVAL
INTO :new.PK
FROM dual;
END;

Oracle trigger keywoard :new not found

my Oracle SqlDeveloper (or Oracle Database?) doesn't know the :NEW keyword.
For instance, if I enter the following sample from Oracles website,
when I execute the "create or replace trigger" paragraph, a window "Enter bind variable" pops up and asks for the bind variable ":new".
Shouldn't this ":new" variable be predefined?
(Oracle SQL Developer 4.0.1.14, Oracle DB 11gR2, Windows)
drop table tab1;
create table tab1 (c1 clob);
insert into tab1 values ('testtext');
create or replace trigger trg1
before update on tab1
for each row
begin
dbms_output.put_line('Old value of CLOB column: '||:OLD.c1);
dbms_output.put_line('Proposed new value of CLOB column: '||:NEW.c1);
-- Previously, we couldn't change the new value for a LOB.
-- Now, we can replace it, or construct a new value using SUBSTR, INSTR...
-- operations for a CLOB, or DBMS_LOB calls for a BLOB.
:NEW.c1 := :NEW.c1 || to_clob('<hr><p>Standard footer paragraph.');
dbms_output.put_line('Final value of CLOB column: '||:NEW.c1);
end;
/
set serveroutput on;
update tab1 set c1 = '<h1>Different Document Fragment</h1><p>Different text.';
select * from tab1;
It turned out I just had to press "Apply" in the Bind dialog, and the trigger was created. Obviously a bug in SqlDeveloper. At least there is a workaround...
I added the 'REFERENCING' statement, and tested the code, seems to work just fine--- BTW, you can just turn DBMS output on in SQL dev w/o having to run the command..
CREATE OR REPLACE TRIGGER trg1
BEFORE UPDATE
ON tab1
REFERENCING NEW AS new OLD AS old
FOR EACH ROW
BEGIN
DBMS_OUTPUT.put_line ('Old value of CLOB column: ' || :old.c1);
DBMS_OUTPUT.put_line ('Proposed new value of CLOB column: ' || :new.c1);
-- Previously, we couldn't change the new value for a LOB.
-- Now, we can replace it, or construct a new value using SUBSTR, INSTR...
-- operations for a CLOB, or DBMS_LOB calls for a BLOB.
:new.c1 := :new.c1 || TO_CLOB ('<hr><p>Standard footer paragraph.');
DBMS_OUTPUT.put_line ('Final value of CLOB column: ' || :new.c1);
END;

No Data Found Error in Oracle

I'm trying to insert a data into oracle database(version 11g xe).But when i'm trying to execute the procedure using toad i'm getting the error as 'ORA-01403: no data found'.
Here's my code
CREATE OR REPLACE PROCEDURE ACTSINFO.sp_Insert_WorkDetails
(p_workname IN varchar ,
p_workaddress IN varchar)
IS
BEGIN
insert into workdetails (workname,workaddress) values (p_workname,p_workaddress);
END sp_Insert_WorkDetails;
I tried to execute the procedure using the below statememt
EXEC sp_Insert_WorkDetails('test','test');
Also i've defined a trigger and sequence for the autoincrement of workdetailsid in table workdetails
Sequence is as follows
ALTER SEQUENCE ACTSINFO.WORKDETAILS_WORKID_SEQ
INCREMENT BY 1
MINVALUE 0
MAXVALUE 9999999999999999999999999999
NOCACHE
NOCYCLE
NOORDER
Trigger is as follows
DROP TRIGGER ACTSINFO.WORKDETAILS_INSERT;
CREATE OR REPLACE TRIGGER ACTSINFO.WORKDETAILS_INSERT
BEFORE INSERT
ON ACTSINFO.WORKDETAILS
REFERENCING NEW AS New OLD AS Old
FOR EACH ROW
BEGIN
SELECT WORKDETAILS_WORKID_SEQ.NEXTVAL INTO :NEW.WORKID FROM WORKDETAILS;
END;
I'm new to oracle.Pls help me...
Your trigger is the problem:
SELECT WORKDETAILS_WORKID_SEQ.NEXTVAL INTO :NEW.WORKID FROM WORKDETAILS;
If there are no rows in workdetails the select will return nothing. Even worse, if your workdetails table has more than one row this will also fail miserably.
You really want the following:
SELECT WORKDETAILS_WORKID_SEQ.NEXTVAL INTO :NEW.WORKID FROM dual;
or - if you are on 11g - then you can use:
:NEW.WORKID := WORKDETAILS_WORKID_SEQ.NEXTVAL;

Resources