DB2 UDB To PostgreSQL Conversion Guide Version 1.0
DB2 UDB TO POSTGRESQL CONVERSION GUIDE
DB2 UDB To PostgreSQL Migration
DRAFT VERSION : 1.0
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
TABLE OF CONTENTS
1. INTRODUCTION...............................................................................................................4
1.1 Purpose.......................................................................................................................................4
1.2 Scope..........................................................................................................................................4
2 CONVERSION REFERENCE................................................................................................5
2.1 Tools...........................................................................................................................................5
2.2 SQL Components - DB2 Objects...............................................................................................5
2.2.1 Data Types.......................................................................................................................................5
2.2.2 Special Data Types..........................................................................................................................5
2.2.3 Table Constraints.............................................................................................................................7
2.2.4 Sequence Number (Auto generated ID column)............................................................................10
2.2.5 Special Objects...............................................................................................................................12
2.2.6 Views..............................................................................................................................................12
2.2.7 Trigger............................................................................................................................................13
2.2.8 Functions........................................................................................................................................14
2.2.9 Stored Procedures..........................................................................................................................15
2.3 SQL Predicates.........................................................................................................................18
2.3.1 BETWEEN Predicate.....................................................................................................................18
2.3.2 EXISTS / NOT EXISTS Predicate..................................................................................................19
2.3.3 IN / NOT IN Predicate....................................................................................................................20
2.3.4 LIKE Predicate................................................................................................................................20
2.3.5 IS NULL / IS NOT NULL Predicate................................................................................................21
2.4Temporary Tables......................................................................................................................21
2.4.1 Using WITH phrase at the top of the query to define a common table expression.......................21
2.4.2 Full-Select in the FROM part of the query.....................................................................................22
2.4.3 Full-Select in the SELECT part of the query..................................................................................23
2.5 CASE Expression.....................................................................................................................24
2.6 Column Functions.....................................................................................................................24
2.7 OLAP Functions........................................................................................................................25
2.7.1 ROWNUMBER & ROLLUP............................................................................................................25
2.8 Scalar Functions.......................................................................................................................26
2.8.1 Scalar Functions - IBM DB2 vs PostgreSQL.................................................................................26
2.9 ORDER BY, GROUP BY & HAVING.......................................................................................31
2.9.1 ORDER BY.....................................................................................................................................31
2.9.2 GROUP BY.....................................................................................................................................32
2.9.3 HAVING..........................................................................................................................................32
2.10 DYNAMIC Cursors..................................................................................................................33
2.11 Joins .......................................................................................................................................34
2.11.1 Self-Join.......................................................................................................................................34
2.11.2 Left-outer Join...............................................................................................................................34
2.11.3 Right-outer Join............................................................................................................................34
2.12 Sub-Query...............................................................................................................................34
2.13 Manipulating Resultset returned by Called Function (Associate..)........................................35
2.14 UNION & UNION ALL.............................................................................................................39
2.14.1 UNION..........................................................................................................................................39
2.14.2 UNION ALL...................................................................................................................................40
2.15 Dynamic SQL..........................................................................................................................41
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2.16 Condition Handling.................................................................................................................41
2.17 Print Output Messages...........................................................................................................42
2.18 Implicit casting in SQL............................................................................................................42
2.18.1Casting double to integer syntax...................................................................................................42
2.18.2Casting double to integer (Round)................................................................................................42
2.18.3Casting double to integer (lower possible integer)........................................................................42
2.19 Select from SYSIBM.SYSDUMMY1.......................................................................................42
2.20 Variables declaration and assignment...................................................................................42
2.21 Conditional statements and flow control (supported by PostgreSQL)...................................42
3 SUMMARY....................................................................................................................44
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
1. Introduction
Since migrating from DB2 UDB to PostgreSQL requires a certain level of knowledge in both
environments, the purpose of this document is to identify the issues in the process involved
migrating from DB2 UDB to PostgreSQL database.
This document also relates the required information on PostgreSQL equivalents of DB2 UDB and
its syntax of usage.
1.1 Purpose
The intent of this document is to serve as a valid reference - in the near future - for the process of
migrating the structure as well as data from IBM DB2 database to PostgreSQL database .
1.2 Scope
The scope of this document is limited to the extent of identifying the PostgreSQL equivalents of
various SQL components, column / OLAP / Scalar functions, Order by / Group by / Having, Joins,
Sub-queries, Union / Intersect / Except clauses that are currently defined for DB2 database.
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2 Conversion Reference
This section briefly discusses the different steps involved in conversion process from DB2 UDB to
PostgreSQL.
2.1 Tools
The following tools, could be used while migrating data from DB2 to PostgreSQL.
• Aqua Data Studio 4.5.2 and above – Mainly used for exporting DB2 data to csv format and
importing csv format into postgreSQL.
2.2 SQL Components - DB2 Objects
2.2.1 Data Types
Data Types
IBM DB2 PostgreSQL
CHAR(n) CHAR(n)
DATE DATE
Some Valid Inputs:
now, today, tomorrow, yesterday
‘now’::datetime
DECIMAL(m,n) DECIMAL(m,n)
INTEGER INTEGER
SMALLINT SMALLINT
TIMESTAMP TIMESTAMP
Some Valid Inputs:
now, today, tomorrow, yesterday
TIME TIME
Some Valid Inputs:
now
VARCHAR(n) VARCHAR(n)
2.2.2 Special Data Types
Special Data Types
IBM DB2 PostgreSQL
CLOB TEXT (maximum of 1GB)
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
BLOB BYTEA (max 1GB) (Binary data - byte array)
CURRENT TIMESTAMP CURRENT_TIMESTAMP
Example :
CREATE TABLE products (
...
created_date TIMESTAMP DEFAULT
CURRENT_TIMESTAMP,
...
)
;
CURRENT TIME CURRENT_TME
Example :
CREATE TABLE products (
...
reordered_time TIMESTAMP DEFAULT
CURRENT_TIME,
...
);
CURRENT DATE CURRENT_DATE
Example :
CREATE TABLE products (
...
reordered_date TIMESTAMP DEFAULT
CURRENT_DATE,
...
);
GENERATED BY DEFAULT AS IDENTITY Example :
CREATE TABLE products (
product_no INTEGER nextval
(’products_product_no_seq’),
...
);
Using SERIAL
CREATE TABLE products (
product_no SERIAL,
...
)
;
refcursor
This is special data type of CURSOR type.
DECLARE <cursor_name> refcursor;
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2.2.3 Table Constraints
2.2.3.1 Check Constraints
A check constraint is the most generic constraint type. It allows you to specify that the value
in a certain column must satisfy a Boolean (truth-value) expression.
Equivalents / Declaration
IBM DB2 PostgreSQL
CREATE TABLE <table> (
<column1>,
....,
<columnX> CONSTRAINT
<constraints name> CHECK
(<Condition>)
);
CREATE TABLE <table> (
<column1>,
....,
<columnX> CONSTRAINT
<constraints name> CHECK
(<Condition>)
);
Example Usage
CREATE TABLE products (
product_no INTEGER,
name VARCHAR(30),
price INTEGER,
category INTEGER
CONSTRAINT my_catg CHECK
(category IN (1,2,3,4))
);
CREATE TABLE products (
product_no INTEGER,
name TEXT,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0),
category INTEGER
);
2.2.3.2 Not-Null Constraints
A not-null constraint simply specifies that a column must not assume the null value.
Equivalents / Declaration
IBM DB2 PostgreSQL
CREATE TABLE <table> (
<column1> NOT NULL,
....,
<columnX>
);
CREATE TABLE <table> (
<column1> NOT NULL,
....,
<columnX>
);
Example Usage
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
CREATE TABLE products (
product_no INTEGER NOT
NULL,
name VARCHAR(30) NOT
NULL,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0)
);
CREATE TABLE products (
product_no INTEGER NOT
NULL,
name TEXT NOT NULL,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0)
);
2.2.3.3 Unique Constraints
Unique constraints ensure that the data contained in a column or a group of columns is
unique with respect to all the rows in the table.
Equivalents / Declaration
IBM DB2 PostgreSQL
CREATE TABLE <table> (
<column1> NOT NULL,
....,
<columnX>
CONSTRAINT <constraint
name> UNIQUE (<column>)
) DATE CAPTURE NONE IN <Data
tablespace name> INDEX IN <index
tablespace name>
;
CREATE TABLE <table> (
<column1> NOT NULL,
....,
<columnX>
CONSTRAINT <constraint name>
UNIQUE (<column>) USING INDEX
TABLESPACE <Index tablespace name>
) TABLESPACE <Data tablespace
name>
;
Example Usage
CREATE TABLE products (
product_no INTEGER NOT
NULL,
name VARCHAR(30) NOT
NULL,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0),
CONSTRAINT unq_prod_no
UNIQUE (product_no)
) DATA CAPTURE NONE IN
mydataspace INDEX IN myindexspace
;
CREATE TABLE products (
product_no INTEGER NOT
NULL,
name TEXT NOT NULL,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0),
CONSTRAINT unq_prod_no
UNIQUE (product_no) USING INDEX
TABLESPACE myindexspace
) TABLESPACE mydataspace
;
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2.2.3.4 Primary Key Constraints
Technically, a primary key constraint is simply a combination of a unique constraint and a
not-null constraint.
Equivalents / Declaration
IBM DB2 PostgreSQL
CREATE TABLE <table> (
<column1> NOT NULL,
....,
<columnX>
CONSTRAINT <constraint
name> PRIMARY KEY (<column>)
) DATE CAPTURE NONE IN <Data
tablespace name> INDEX IN <index
tablespace name>
;
CREATE TABLE <table> (
<column1> NOT NULL,
....,
<columnX>
CONSTRAINT <constraint name>
PRIMARY KEY (<column>) USING INDEX
TABLESPACE <Index tablespace name>
) TABLESPACE <Data tablespace
name>
;
Example Usage
CREATE TABLE products (
product_no INTEGER NOT
NULL,
name VARCHAR(30) NOT
NULL,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0),
CONSTRAINT pk_prod_no
PRIMARY KEY (product_no)
) DATA CAPTURE NONE IN
mydataspace INDEX IN myindexspace
;
CREATE TABLE products (
product_no INTEGER NOT
NULL,
name TEXT NOT NULL,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0),
CONSTRAINT pk_prod_no
PRIMARY KEY (product_no) USING
INDEX TABLESPACE myindexspace
) TABLESPACE mydataspace
;
2.2.3.5 Foreign Key Constraints
A foreign key constraint specifies that the values in a column (or a group of columns) must
match the values appearing in some row of another table. We say this maintains the
referential integrity between two related tables.
Equivalents / Declaration
IBM DB2 PostgreSQL
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
CREATE TABLE <table> (
<column1> NOT NULL,
....,
<columnX>
CONSTRAINT <constraint
name> FOREIGN KEY (<column>)
REFERENCES <ref table name>
(<column>)
) DATE CAPTURE NONE IN <Data
tablespace name> INDEX IN <index
tablespace name>
;
CREATE TABLE <table> (
<column1> NOT NULL,
....,
<columnX>
CONSTRAINT <constraint name>
FOREIGN KEY (<column>) REFERENCES
<ref table name>(<column>)
) TABLESPACE <Data tablespace
name>
;
Example Usage
CREATE TABLE products (
product_no INTEGER NOT
NULL,
name VARCHAR(30) NOT
NULL,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0),
CONSTRAINT pk_prod_no
PRIMARY KEY (product_no)
) DATA CAPTURE NONE IN
mydataspace INDEX IN myindexspace
;
CREATE TABLE orders (
order_no INTEGER NOT
NULL,
product_no INTEGER,
quantity DECIMAL(12,4),
CONSTRAINT fk_prod_no
FOREIGN KEY (product_no)
REFERENCES products(product_no)
) DATA CAPTURE NONE IN
mydataspace INDEX IN myindexspace
;
CREATE TABLE products (
product_no INTEGER NOT
NULL,
name TEXT NOT NULL,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0),
CONSTRAINT pk_prod_no
PRIMARY KEY (product_no) USING
INDEX TABLESPACE myindexspace
) TABLESPACE mydataspace
;
CREATE TABLE orders (
order_no INTEGER NOT NULL,
product_no INTEGER,
quantity DECIMAL(12,4),
CONSTRAINT fk_prod_no
FOREIGN KEY (product_no)
REFERENCES products(product_no)
) TABLESPACE mydataspace
;
2.2.4 Sequence Number (Auto generated ID column)
The data types serial and bigserial are not true types, but merely a notational convenience for
setting up unique identifier columns (similar to the AUTO_INCREMENT property supported by
some other databases).
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
The <sequence name> should be unique for database level and it minvalue n, is the number
at which the sequence starts.
Note: The sequence is always incremented by 1.
The tables created are later associated with the already created sequence, using nextval
('<sequence_name>') function.
Equivalents / Declaration
IBM DB2 PostgreSQL
CREATE TABLE <table> (
<column1> NOT NULL
GENERATED BY DEFAULT AS IDENTITY
(START WITH n, INCREMENT BY x NO
CACHE),
....,
<columnX>
)
;
CREATE SEQUENCE <sequence_name>
MINVALUE n;
CREATE TABLE <table> (
<column1> DEFAULT nextval
('<sequence_name>'),
....,
<columnX>
)
;
Example Usage
CREATE TABLE products (
product_no INTEGER NOT
NULL GENERATED BY DEFAULT AS
IDENTITY (START WITH 11,
INCREMENT BY 1, NO CACHE),
name VARCHAR(30) NOT
NULL,
price INTEGER
)
;
CREATE SEQUENCE products_seq_prdno
MINVALUE 1;
CREATE TABLE products (
product_no INTEGER nextval
(' products_seq_prdno')
name TEXT NOT NULL,
price INTEGER CONSTRAINT
positive_price CHECK (price > 0),
CONSTRAINT pk_prod_no
PRIMARY KEY (product_no) USING
INDEX TABLESPACE myindexspace
)
TABLESPACE mydataspace
;
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2.2.5 Special Objects
2.2.5.1 CLOB
Equivalents / Declaration
IBM DB2 PostgreSQL
CLOB(n) - n <= 2 GB TEXT (max 1GB)
Example Usage
CREATE TABLE orders
....
notes CLOB(1M),
....
);
CREATE TABLE orders (
...
notes TEXT(1M),
...
);
2.2.5.2 BLOB
Equivalents / Declaration
IBM DB2 PostgreSQL
BLOB(n) - n <= 2 GB BYTEA (maximum 1GB) binary data – byte
array
Example Usage
2.2.6 Views
Equivalents / Declaration
IBM DB2 PostgreSQL
CREATE VIEW <view_name> AS
sql statement
;
CREATE OR REPLACE VIEW <view_name>
AS
sql statement
;
Example Usage
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
CREATE VIEW products_v AS
SELECT x,y,...
FROM products
....
;
CREATE OR REPLACE VIEW products_v
AS
SELECT x,y,...
FROM products
....
;
2.2.7 Trigger
Equivalents / Declaration
IBM DB2 PostgreSQL
CREATE TRIGGER <trigger name>
AFTER INSERT
ON <table name>
REFERENCING
NEW AS N
FOR EACH ROW
MODE DB2SQL
BEGIN ATOMIC
.....
END
;
CREATE TRIGGER <trigger name>
AFTER INSERT
ON <table name>
FOR EACH ROW
EXECUTE PROCEDURE function_name();
Example Usage
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
CREATE TABLE emp_audit(
operation CHAR(1) NOT
NULL,
...
...
);
CREATE TRIGGER process_emp_audit
AFTER INSERT
ON emp_audit
REFERENCING
NEW AS N
FOR EACH ROW
MODE DB2SQL
BEGIN ATOMIC
INSERT INTO emp_audit
SELECT ’I’, now(), user, N.*;
END
;
CREATE TABLE emp_audit(
operation CHAR(1) NOT NULL,
...
...
);
CREATE OR REPLACE FUNCTION
process_emp_audit()
RETURNS TRIGGER
LANGUAGE plpgsql
AS
$emp_audit$
BEGIN
INSERT INTO emp_audit SELECT
’I’, now(), user, NEW.*;
RETURN NEW;
END;
$emp_audit$;
CREATE TRIGGER emp_audit
AFTER INSERT ON emp_audit
FOR EACH ROW EXECUTE
PROCEDURE process_emp_audit();
2.2.8 Functions
Equivalents / Declaration
IBM DB2 PostgreSQL
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
CREATE FUNCTION <function_name>
(
parameter,
....
)
SPECIFIC <function_name>
RETURNS <return_data_type>
NO EXTERNAL ACTION
DETERMINISTIC
RETURN
....
;
CREATE OR REPLACE FUNCTION
<function_name> (
parameter,
....
)
RETURNS <return_data_type>
LANGUAGE PLPGSQL
AS
$$
BEGIN
....
END;
$$
;
Example Usage
CREATE FUNCTION GREATEROF (
V1 INTEGER,
V2 INTEGER
)
SPECIFIC GREATEROF
RETURNS integer
LANGUAGE sql
NO EXTERNAL ACTION
DETERMINISTIC
RETURN
CASE
WHEN V1 > V2 THEN
V1
ELSE V2
END;
;
CREATE OR REPLACE FUNCTION GREATEROF
(
V1 INTEGER,
V2 INTEGER
)
RETURNS integer
LANGUAGE plpgsql
AS
$$
BEGIN
RETURN
CASE
WHEN V1 > V2 THEN V1
ELSE V2
END;
END;
$$
;
2.2.9 Stored Procedures
When creating functions which handles or returns cursors, these points are to be remembered.
• All variable declaration should be done at the top, in other words should be the first
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
few statements.
• Any default values assigned to the variables can be done at the declaration
statement.
• Any assigning of values to the variables should be done within the BEGIN and END
statement.
• Any cursor declaration can be done out side the BEGIN and END statement.
• Any dynamic cursors using dynamic sqls, should be done within BEGIN and END
statement.
• In all the cases OPEN <cursor_name> and returning the cursor RETURN
<cursor_name>, is a must statement for functions returning REFCURSOR.
• The function body block, to be defined within $$ and $$.
Equivalents / Declaration
IBM DB2 PostgreSQL
CREATE PROCEDURE <procedure_name>
(
IN para1 VARCHAR(5),
IN para2 INTEGER
)
SPECIFIC <procedure_name>
DYNAMIC RESULT SETS <number>
LANGUAGE SQL
BEGIN
DECLARE <cursor_name>
CURSOR WITH RETURN TO CLIENT FOR
<sql_statement>;
OPEN <cursor_name>;
END
;
CREATE OR REPLACE FUNCTION
<function_name> (
IN para1 VARCHAR(5),
IN para2 INTEGER
)
RETURNS REFCURSOR
LANGUAGE PLPGSQL
AS
$$
DECLARE <cursor_name> CURSOR
FOR <sql_statement>;
BEGIN
....
OPEN <cursor_name>;
RETURN <cursor_name>;
END;
$$
;
Example Usage
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
1
CREATE PROCEDURE list_orders (
IN prd_no INTEGER
)
SPECIFIC list_orders
DYNAMIC RESULT SETS 1
LANGUAGE SQL
BEGIN
DECLARE lstOrds CURSOR
WITH RETURN TO CLIENT FOR
SELECT * FROM orders WHERE
product_no = prd_no;
OPEN <cursor_name>;
END
;
CREATE OR REPLACE FUNCTION
list_orders (
IN prd_no INTEGER
)
RETURNS REFCURSOR
LANGUAGE plpgsql
AS
$$
DECLARE lstOrds CURSOR FOR
SELECT *
FROM orders
WHERE product_no =
prd_no;
BEGIN
OPEN lstOrds;
RETURN lstOrds;
END;
$$
;
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2
Dynamic Cursor:
CREATE PROCEDURE list_orders (
IN prd_no INTEGER
)
SPECIFIC list_orders
DYNAMIC RESULT SETS 1
LANGUAGE SQL
BEGIN
DECLARE selCur CURSOR WITH
RETURN TO CLIENT FOR
strPrepSelSql;
DECLARE sqlString VARCHAR
(200);
SET sqlString = ' SELECT
* FROM orders WHERE product_no =
' || prd_no;
PREPARE strPrepSelSql FROM
sqlString;
OPEN selCur;
END
;
Dynamic Cursor:
CREATE OR REPLACE FUNCTION
list_orders (
IN prd_no INTEGER
)
RETURNS refcursor
LANGUAGE plpgsql
AS
$$
DECLARE sqlString VARCHAR(200);
selCur refcursor;
BEGIN
sqlString = 'SELECT * FROM
orders WHERE product_no = ' ||
prd_no;
OPEN selCur FOR EXECUTE
sqlString;
RETURN selCur;
END;
$$
;
2.3 SQL Predicates
2.3.1 BETWEEN Predicate
Equivalents / Declaration
IBM DB2 PostgreSQL
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
SELECT x, y
FROM tab1
WHERE
.....
column BETWEEN value1 AND
value2
.....
;
SELECT x, y
FROM tab1
WHERE
column1
.....
column2 BETWEEN value1 AND
value2
.....
.....
;
Example Usage
SELECT *
FROM orders,
WHERE
quantity <= 100
AND order_date BETWEEN
'2005-04-06' AND '2006-04-05';
Note: Both the dates are inclusive, as in DB2.
2.3.2 EXISTS / NOT EXISTS Predicate
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT
column(s),
FROM <table_name>
WHERE
columnx = <value>
AND NOT EXISTS
(SELECT columnx
FROM <table_name>
....)
;
SELECT
column(s),
FROM <table_name>
WHERE
columnx = <value>
AND NOT EXISTS
(SELECT columnx
FROM <table_name>
....)
;
Example Usage
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
SELECT product_no
FROM products
WHERE name LIKE 'A%'
AND category IN (1,2,3,4)
AND NOT EXISTS (
SELECT category_no
FROM categorys
WHERE status = 'D');
2.3.3 IN / NOT IN Predicate
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT *
FROM <table_name>
WHERE
.....
<column> NOT IN ('C','S')
.....
;
SELECT *
FROM <table_name>
WHERE
.....
<column> NOT IN ('C','S')
.....
;
Example Usage
SELECT
product_no,
name,
FROM
products
WHERE
category NOT IN (3,4);
2.3.4 LIKE Predicate
Equivalents / Declaration
IBM DB2 PostgreSQL
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
SELECT x, y
FROM <table_name>
WHERE
.....
tab1.my_name LIKE LCASE
(strName)
;
Same as DB2.
Example Usage
SELECT *
FROM products
WHERE product_no > 125
AND UPPER(name) LIKE 'M%'
;
2.3.5 IS NULL / IS NOT NULL Predicate
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT x, y
FROM tab1
WHERE
.....
column IS NOT NULL
;
Same as DB2.(IS NULL & IS NOT NULL)
Example Usage
SELECT *
FROM products
WHERE product_no > 125
AND category IS NOT NULL;
2.4Temporary Tables
2.4.1 Using WITH phrase at the top of the query to define a common table
expression
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
Equivalents / Declaration
IBM DB2 PostgreSQL
WITH TEMP(
name,
....
) AS (
SELECT
VALUE(id,0)
FROM
....)
;
Ref T121/T122. Yet to be implemented.
2.4.2 Full-Select in the FROM part of the query
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT x, y
FROM tab1
LEFT OUTER JOIN
(SELECT
....
FROM
....)
WHERE
...
;
SELECT x, y
FROM tab1 A
LEFT OUTER JOIN
(SELECT *
FROM ....
....) B
ON A.eid= B.eid
WHERE B.eid < 3
;
Example Usage
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
SELECT
SUM(tot_paid-tot_refund) AS
tot_paid_amount,
...
i.invoice_no
FROM
invoice i
LEFT OUTER JOIN
orders_pending
o
ON i.invoice_no
= o.invoice_no
AND invoice_year =
'20052006'
2.4.3 Full-Select in the SELECT part of the query
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT
<column_name>,
(SELECT <column_name>
FROM <table>
WHERE column = Value)
FROM
<table>
WHERE
<condition>
;
SELECT
<column_name>,
(SELECT <column_name>
FROM <table>
WHERE column = Value)
FROM
<table>
WHERE
<condition>
;
Example Usage
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
SELECT
cust_id,
TO_CHAR((SELECT MAX
(cf.fund_recvd_date)
FROM cust_funding cf
WHERE cf.er_id =
iCuID
...
),'YYYY-MM-DD') AS
fund_date
...
FROM
cust_funding
WHERE
cust_id = iCuID
AND invoice_year =
'20052006'
GROUP BY
cust_id, invoice_year
;
2.5 CASE Expression
Equivalents / Declaration
IBM DB2 PostgreSQL
CASE ctrlVar
WHEN 1 THEN
<statements>;
ELSE <statements>;
END CASE
;
Note : Case expression is not supported in
PostgreSQL. It can used in SELECT
statements. As a workaround, use IF-ELSE
construct.
2.6 Column Functions
Equivalents / Declaration
Column /
Aggregate
Functions
IBM DB2 PostgreSQL
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
AVG
SELECT emp_id, AVG(emp_pay)
FROM emp_payments GROUP BY
emp_id;
Same as DB2
COUNT
SELECT company_id, COUNT
(emp_id) AS employee_count
FROM employee GROUP BY
company_id;
Same as DB2
MAX
SELECT emp_id, MAX
(process_date) AS
last_processed_date FROM
emp_payments GROUP BY emp_id
Same as DB2
MIN
SELECT emp_id, MIN
(process_date) AS
first_processed_date FROM
emp_payments GROUP BY emp_id
Same as DB2
SUM
SELECT emp_id, SUM(emp_pay)
AS total_pay FROM
emp_payments GROUP BY
emp_id;
Same as DB2
2.7 OLAP Functions
2.7.1 ROWNUMBER & ROLLUP
Equivalents / Declaration
IBM DB2 PostgreSQL
ROWNUMBER() Not supported in PostgreSQL
Note : Not used in application. Hence can be
ignored.
Equivalents / Declaration
IBM DB2 PostgreSQL
ROLLUP() There is no direct equivalent for ROLLUP in
PostgreSQL database.
This is could be achieved by using UNION
clause. In some cases, we may end up using
UNION clause along with a required VIEW.
Example Usage
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
SELECT
1 AS cur_row,
cust_id,
cust_name,
fund_date,
cust_funding AS Amount,
invoice_date
FROM customer c, invoice i
WHERE c.cust_id = iCuID
AND c.invoice_no =
i.invoice_no
AND c.invoice_year =
'20052006'
GROUP BY ROLLUP((
cust_id,
cust_name,
cust_funding AS Amount,
invoice_date
)), fund_date
ORDER BY
cur_row,
fund_date
;
SELECT * FROM (
SELECT * FROM (
SELECT
1 AS cur_row,
cust_id,
cust_name,
fund_date,
cust_funding AS Amount,
invoice_date
FROM customer c, invoice i
WHERE c.cust_id = iCuID
AND c.invoice_no =
i.invoice_no
AND c.invoice_year =
'20052006'
) AS LST_RECS
UNION
SELECT
COUNT(*) AS cur_row,
NULL,NULL,NULL,
SUM(cust_funding) AS Amount,
NULL,
FROM customer c,invoice i
WHERE c.cust_id = iCuID
AND c.invoice_no =
i.invoice_no
AND c.invoice_year =
'20052006'
) AS TMP_TAB
ORDER BY cur_row,fund_date
;
2.8 Scalar Functions
Scalar functions act on a single row at a time. This section lists all the IBM DB2 scalar functions
that are used in Able Payroll project & their equivalents in PostgreSQL database.
2.8.1 Scalar Functions - IBM DB2 vs PostgreSQL
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
Scalar
Function
Return
Type
IBM DB2 PostgreSQL Description
CEIL or
CEILING
Same
as input
CEIL
CEILING
Example :
SELECT CEIL(123.89)
FROM SYSIBM.SYSDUMMY1;
SELECT CEILING(123.89)
FROM SYSIBM.SYSDUMMY1;
CEIL
CEILING
Example :
SELECT CEIL(123.89);
SELECT CEILING(123.89);
CEIL or CEILING
returns the next
smallest integer value
that is greater than or
equal to the input
(e.g. CEIL(123.89)
returns 124, also
CEIL(123.19)
returns 124)
CHAR String /
Text
CHAR
Example :
SELECT CHAR(1) FROM
SYSIBM.SYSDUMMY1;
SELECT CHAR(DATE(‘2005-
01-12’), EUR) FROM
SYSIBM.SYSDUMMY1;
TO_CHAR( <timestamp /
interval / int / double precision /
numeric type>, text)
Example :
SELECT TO_CHAR(-212.8,
'999D99S');
Returns character
String of the given input
COALESCE Null or
same
as input
COALESCE(value [, ...])
Example :
SELECT COUNT(*),
MIN(MAIL_ATTACH_ID)
AS min_id,
MAX(MAIL_ATTACH_ID)
AS max_id,
COALESCE(MIN
(MAIL_ATTACH_ID), MAX
(MAIL_ATTACH_ID))
FROM EMAIL_ATTACH_LOG;
COALESCE(value [, ...])
Example : (Same as DB2)
SELECT COUNT(*),
MIN(MAIL_ATTACH_ID)
AS min_id,
MAX(MAIL_ATTACH_ID)
AS max_id,
COALESCE(MIN
(MAIL_ATTACH_ID), MAX
(MAIL_ATTACH_ID))
FROM EMAIL_ATTACH_LOG;
First non-null value in a
list of (compatible) input
expressions (read from
left to right) is returned.
VALUE is a synonym
for COALESCE.
CONCAT or
||
String Example :
SELECT 'A' || 'B' ,
CONCAT('A', 'B'), 'A'
|| 'B' || 'C', CONCAT
(CONCAT('A', 'B'),
'C');
Note : CONCAT is not available
in PostgreSQL, only || works.
A function CONCAT as given
below can be created as a
workaround.
Function :
CREATE OR REPLACE
FUNCTION "concat"
(text,text) RETURNS text
LANGUAGE sql AS $$
SELECT $1 || $2;
$$;
Example : SELECT 'A' ||
'B' , CONCAT('A', 'B'),
'A' || 'B' || 'C',
CONCAT(CONCAT('A', 'B'),
'C');
Joins two strings
together. In IBM DB2,
CONCAT function has
both "infix" and "prefix"
notations. In the former
case, the verb is placed
between the two strings
to be acted upon. In
PostgreSQL, CONCAT
function needs to be
created in order to use
it.
DATE Date Example : SELECT DATE
('2006-09-21') FROM
SYSIBM.SYSDUMMY1;
Example : SELECT TO_DATE
('21-02-2006','DD-MM-
YYYY');
Converts the input to
date value
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
DAY Integer Usage : DAY
(<DATE_FIELD>)
Example : SELECT DAY
(DATE('2006-09-21'))
FROM SYSIBM.SYSDUMMY1;
Usage : DATE_PART(‘day’,
<DATE_FIELD>)
Example : SELECT
DATE_PART('day', '2006-
09-21'::date);
Returns the day (as in
day of the month) part
of a date (or equivalent)
value. The output
format is integer.
DAYS Integer Usage : DAYS
(<DATE_FIELD>)
Example : SELECT (DAYS
(DATE('2006-09-25')) -
DAYS(DATE('2006-09-
21'))) FROM
SYSIBM.SYSDUMMY1;
Note : DAYS is not available in
PostgreSQL.
Example : SELECT TO_DATE
('25-09-2006', 'DD-MM-
YYYY') -TO_DATE('21-09-
2006', 'DD-MM-YYYY');
A function DAYS can be created
as a workaround.
Function :-
CREATE OR REPLACE
FUNCTION DAYS (
V1 DATE
)
RETURNS integer
LANGUAGE plpgsql
AS
$$
BEGIN
RETURN
TO_DATE(V1,'YYYY-MM-DD') -
TO_DATE('4712-01-01','YYYY-
MM-DD');
END;
$$
;
Converts a date (or
equivalent) value into a
number that represents
the number of days
since the date "0001-
01-01" inclusive. The
output format is integer.
DECIMAL /
DEC
Decimal Usage : DECIMAL(<FIELD>)
or DEC(<FIELD>)
Example : SET l_sub4 =
DECIMAL(l_absSub4);
No direct equivalent. Use
TO_NUMBER instead.
Example : SELECT
TO_NUMBER(l_absSub4,
<format_string>);
Converts either
character or numeric
input to decimal.
FLOOR Same
as input
Usage : FLOOR(<FIELD>)
Example : SELECT FLOOR
(5.945) FROM
SYSIBM.SYSDUMMY1;
Usage : FLOOR(<FIELD>)
Example : SELECT FLOOR
(5.945);
Returns the next
largest integer value
that is smaller than or
equal to the input (e.g.
5.945 returns 5.000).
IDENTITY_
VAL_LOCA
L
Integer Example : SET iErID =
IDENTITY_VAL_LOCAL();
Example : CURRVAL
('<<SEQUENCE_NAME>>')
SELECT CURRVAL
('DummySeq');
Returns the most
recently assigned value
(by the current user) to
an identity column.
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
INTEGER Integer Converts either a number or a
valid character value into an
integer. The character input
can have leading and /or
trailing blanks, and a sign
indicator, but it cannot contain
a decimal point. Numeric
decimal input works just fine.
Example : SELECT INTEGER
(234.8817) FROM
SYSIBM.SYSDUMMY1;
Example : TO_NUMBER
(<field>, <format>)
SELECT TO_NUMBER(FLOOR
(234.8817),'999999999');
=> 234
Converts input into an
integer
LCASE or
LOWER
String Usage : LOWER(<FIELD>)
(or)
LCASE(<FIELD>)
Example :
SELECT LCASE(‘LOWER
CASE’), LOWER(‘LOWER
CASE’) FROM
SYSIBM.SYSDUMMY1;
Usage : LOWER(<FIELD>)
Example :
SELECT LOWER('LOWER
CASE');
Converts the mixed or
upper case input string
to lower case
LENGTH Integer Usage : LENGTH(<FIELD>)
Example :
SELECT LENGTH('LOWER
CASE') FROM
SYSIBM.SYSDUMMY1;
Usage : LENGTH(<FIELD>)
Example :
SELECT LENGTH('LOWER
CASE');
Returns an integer
value with the internal
length of the
expression
LTRIM String Usage : LTRIM(<FIELD>)
Example :
SELECT LTRIM(' ABC'),
LENGTH(LTRIM(' ABC')),
LTRIM(' ABC '), LENGTH
(LTRIM(' ABC ')),
LTRIM('ABC '), LENGTH
(LTRIM('ABC ')) FROM
SYSIBM.SYSDUMMY1;
Usage : LTRIM(<FIELD>)
Example :
SELECT LTRIM(' ABC'),
LENGTH(LTRIM(' ABC')),
LTRIM(' ABC '), LENGTH
(LTRIM(' ABC ')),
LTRIM('ABC '), LENGTH
(LTRIM('ABC '));
Removes leading
blanks, but not trailing
blanks, from the
argument.
MOD depend
s on
input
Usage : MOD(<FIELD_1>,
<FIELD_2>)
Example :
SELECT MOD(-31,11) FROM
SYSIBM.SYSDUMMY1;
Usage : MOD(<FIELD_1>,
<FIELD_2)
Example :
SELECT MOD(-31,11);
Returns the remainder
(modulus) for the 1
st
argument divided by
the 2
nd
argument.
MONTH Integer Usage : MONTH
(<DATE_FIELD>)
Example : SELECT MONTH
(DATE('2006-09-21'))
FROM SYSIBM.SYSDUMMY1;
Usage : DATE_PART
('MONTH', <DATE_FIELD>)
Example :
SELECT DATE_PART
('month', '2006-09-
21'::date);
Returns the month part
of the date value. The
output format is integer.
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
POSSTR Integer Usage : POSSTR
(<FIELD_1>, <FIELD_2>)
Example :
SELECT POSSTR('Benefits
and Expenses', 'and')
FROM SYSIBM.SYSDUMMY1;
Usage : POSITION
(<FIELD_1> IN <FIELD_2>)
Example :
SELECT POSITION('and' IN
'Benefits and
Expenses');
Returns the position of
2
nd
string (DB2) / 1
st
string (PostgreSQL) in
1
st
string (DB2) / 2
nd
string (PostgreSQL)
RAND Floating
point
values
Usage : RAND()
Example :
SELECT RAND() FROM
SYSIBM.SYSDUMMY1;
Usage : RANDOM()
Example :
SELECT RANDOM();
Returns a pseudo-
random floating-point
value in the range of
zero to one inclusive.
ROUND Integer Usage : ROUND(<FIELD>,
<precision>)
Example :
SELECT ROUND(216.89, 1)
FROM SYSIBM.SYSDUMMY1;
Usage : ROUND(<FIELD>,
<precision>)
Example :
SELECT ROUND(216.89, 1);
Rounds the rightmost
digits of number (1st
argument). If the
second argument is
positive, it rounds to the
right of the decimal
place. If the second
argument is negative, it
rounds to the left. A
second argument of
zero results rounds to
integer.
RTRIM String Usage : RTRIM
(<TEXT_FIELD>)
Example :
SELECT RTRIM(' ABC'),
LENGTH(RTRIM(' ABC')),
RTRIM(' ABC '), LENGTH
(RTRIM(' ABC ')),
RTRIM('ABC '), LENGTH
(RTRIM('ABC ')) FROM
SYSIBM.SYSDUMMY1;
Usage : RTRIM
(<TEXT_FIELD>)
Example :
SELECT RTRIM(' ABC'),
LENGTH(RTRIM(' ABC')),
RTRIM(' ABC '), LENGTH
(RTRIM(' ABC ')),
RTRIM('ABC '), LENGTH
(RTRIM('ABC '));
Removes trailing
blanks, but not leading
blanks, from the
argument.
SMALLINT Integer Converts either a number or a
valid character value into a
smallint value.
Example :
SELECT SMALLINT(219.89)
FROM SYSIBM.SYSDUMMY1;
Example : TO_NUMBER
(<field>, <format>)
SELECT TO_NUMBER(FLOOR
(234.8817),'999999999');
=> 234
SUBSTR String Usage : SUBSTR
(<TEXT_FIELD>,
<int_position>)
Example :
SELECT SUBSTR('This is
a substring test', 9)
FROM SYSIBM.SYSDUMMY1;
Usage : SUBSTR
(<TEXT_FIELD>,
<int_position>)
Example :
SELECT SUBSTR('This is a
substring test', 9);
Returns part of a string.
If the length is not
provided, the output is
from the start value to
the end of the string.
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
TIMESTAM
P
Timesta
mp
Usage : TIMESTAMP
(<FIELD>)
Example :
SELECT TIMESTAMP('2006-
01-31-
22.44.55.000000'),
TIMESTAMP('2006-01-31-
22.44.55.000'),
TIMESTAMP('2006-01-31-
22.44.55'),
TIMESTAMP
('20060131224455'),
TIMESTAMP('2006-01-
31','22.44.55')
FROM SYSIBM.SYSDUMMY1;
Example : (to get the default
timestamp)
SELECT CURRENT
TIMESTAMP FROM
SYSIBM.SYSDUMMY1;
Usage : TO_TIMESTAMP
(<FIELD>, <format>)
When using as default to
a column, in table
definition.
Default Timestamp :
CURRENT_TIMESTAMP
Example :
SELECT TO_TIMESTAMP
('2006-01-31-
22.44.55.000000', 'YYYY-
MM-DD-HH.MI.SS.MS'),
TO_TIMESTAMP('2006-01-
31-22.44.55.000', 'YYYY-
MM-DD-HH.MI.SS.MS'),
TO_TIMESTAMP('2006-01-
31-22.44.55', 'YYYY-MM-
DD-HH.MI.SS'),
TO_TIMESTAMP
('20060131224455',
'YYYYMMDDHHMISSMS');
Example : (to get the default
timestamp)
SELECT
CURRENT_TIMESTAMP;
Converts the input into
a time value.
UPPER String Usage : UPPER
(<TEXT_FIELD>)
Example :
SELECT UCASE('upper
case'), UPPER('upper
case') FROM
SYSIBM.SYSDUMMY1;
Usage : UPPER
(<TEXT_FIELD>)
Example :
SELECT UPPER('upper
case');
Converts the mixed or
lower case input string
to upper case
VALUE Null or
same
as input
Usage : Same as COALESCE Usage : Same as COALESCE
Refer to COALESCE example
usage
In PostgreSQL, there is
no direct equivalent for
VALUE function. Use
COALESCE instead
YEAR Integer Usage : YEAR
(<DATE_FIELD>)
Example : SELECT YEAR
(DATE('2006-09-21'))
FROM SYSIBM.SYSDUMMY1;
Usage : DATE_PART
('YEAR',<DATE_FIELD>);
SELECT DATE_PART('year',
'2006-09-21'::date);
Returns the year part of
a date value. The
output format is integer.
2.9 ORDER BY, GROUP BY & HAVING
2.9.1 ORDER BY
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT
....
<column>
....
FROM
<table(s)>
WHERE
<condition(s)>
.....
ORDER BY
<column(s)>
;
Same as DB2
2.9.2 GROUP BY
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT
Aggregate_fun(column1),
Aggregate_fun(column2),
<column>
FROM
<table(s)>
WHERE <condition(s)>
GROUP BY <column>
;
Same as DB2
2.9.3 HAVING
Equivalents / Declaration
IBM DB2 PostgreSQL
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
SELECT
Aggregate_fun(column1),
Aggregate_fun(column2),
<column>
FROM <table(s)>
WHERE <condition(s)>
GROUP BY <column>
HAVING <condition>
;
Same as DB2
2.10 DYNAMIC Cursors
In case of defining a dynamic cursor, we need to use refcursor special data type object.
The sample declaration is as follows:
In this sample, we assume the below code is part of a function and the function returns refcursor
special data type and have the following input parameters:
sYear VARCHAR(10),
iCuID INTEGER
....
$$
DECLARE
sqlString VARCHAR(500);
selCur refcursor;
BEGIN
sqlString = 'SELECT product_no,name ' ||
'FROM products ' ||
'WHERE product_no IN (SELECT product_no ' ||
'FROM invoice WHERE cust_id = ' || iCuID || ') ' ||
'AND invoice_year = ''' || sYear || ''') ' ||
'ORDER BY product_no';
OPEN selCur FOR EXECUTE sqlString;
RETURN selCur;
END
;
$$
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2.11 Joins
2.11.1 Self-Join
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT a.emp_id, a.company_id,
b.user_id FROM employee a INNER
JOIN employee b ON a.emp_id=
b.emp_id;
(or)
SELECT a.emp_id, a.company_id,
b.user_id FROM employee a,
employee b WHERE a.emp_id=
b.emp_id;
Same as DB2
2.11.2 Left-outer Join
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT a.company_id,
a.company_name, b.emp_id,
b.company_id FROM company a LEFT
OUTER JOIN employee b ON
a.company_id= b.company_id;
Same as DB2
2.11.3 Right-outer Join
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT a.company_id,
a.company_name, b.emp_id,
b.company_id FROM company a RIGHT
OUTER JOIN employee b ON
a.company_id= b.company_id;
Same as DB2
2.12 Sub-Query
Equivalents / Declaration
IBM DB2 PostgreSQL
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
SELECT title, fname, sname,
forename FROM employee WHERE
emp_id IN (SELECT emp_id FROM
department WHERE company_id =
iCID);
Same as DB2
2.13 Manipulating Resultset returned by Called Function (Associate..)
Equivalents / Declaration
IBM DB2 PostgreSQL
DECLARE result
RESULT_SET_LOCATOR VARYING;
CALL procedure(<params>);
ASSOCIATE RESULT SET LOCATORS
(result) WITH PROCEDURE
procedure;
ALLOCATE cursor CURSOR FOR RESULT
SET result;
FETCH FROM cursor INTO <var
list>;
DECLARE cursor REFCURSOR;
cursor := SELECT
function_returning_cursor();
FETCH ALL IN cursor;
or
FETCH cursor INTO <var list>;
Example Usage
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
1
DECLARE result1
RESULT_SET_LOCATOR VARYING;
CALL SFT_STY_1
(strProcessTaxYear);
ASSOCIATE RESULT SET LOCATORS
(result1) WITH PROCEDURE
SFT_STY_1;
ALLOCATE rsCur CURSOR FOR RESULT
SET result1;
FETCH FROM rsCur INTO var1, var2;
CLOSE rsCur;
CREATE OR REPLACE FUNCTION
func_select()
RETURNS refcursor;
LANGUAGE plpgsql;
AS
$$
DECLARE ref refcursor;
BEGIN
OPEN ref FOR SELECT 'JOHN'
AS name;
RETURN ref;
END;
$$
;
CREATE OR REPLACE FUNCTION
func_fectch()
RETURNS refcursor;
LANGUAGE plpgsql;
AS
$$
BEGIN
DECLARE rsCur REFCURSOR;
rsCur := SELECT func_select
();
FETCH cursor INTO myname;
...
CLOSE rsCur;.
END;
$$
;
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2
Using bound cursor name, that is
cursor name specified.
CREATE TABLE test (col text);
INSERT INTO test VALUES (’123’);
CREATE FUNCTION reffunc(refcursor)
RETURNS refcursor
LANGUAGE plpgsql
AS
$$
BEGIN
OPEN $1 FOR SELECT
col FROM test;
RETURN $1;
END;
$$;
BEGIN;
SELECT reffunc
(’funccursor’);
FETCH ALL IN funccursor;
COMMIT;
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
3
Using unbound cursor, that is
cursor does not have a name,
reference is automatically
generated..
CREATE FUNCTION reffunc2()
RETURNS refcursor
LANGUAGE plpgsql
AS
$$
DECLARE ref refcursor;
BEGIN
OPEN ref FOR SELECT
col FROM test;
RETURN ref;
END;
$$
;
BEGIN;
SELECT reffunc2();
on screen message:
reffunc2
--------------------
<unnamed cursor 1>
(1 row)
FETCH ALL IN "<unnamed
cursor 1>";
COMMIT
;
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
4
Function returning multiple
cursors.
CREATE FUNCTION myfunc(refcursor,
refcursor)
RETURNS SETOF refcursor
LANGUAGE plpgsql
AS
$$
BEGIN
OPEN $1 FOR SELECT * FROM
table_1;
RETURN NEXT $1;
OPEN $2 FOR SELECT * FROM
table_2;
RETURN NEXT $2;
END;
$$
;
-- need to be in a transaction to
use cursors.
BEGIN;
SELECT * FROM myfunc(’a’,
’b’);
FETCH ALL FROM a;
FETCH ALL FROM b;
COMMIT;
2.14 UNION & UNION ALL
2.14.1 UNION
Equivalents / Declaration
IBM DB2 PostgreSQL
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
SELECT emp_id, pay_amt FROM
emp_payments
UNION
SELECT emp_id, pay_amt FROM
emp_absent_payments
Same as DB2
2.14.2 UNION ALL
Equivalents / Declaration
IBM DB2 PostgreSQL
SELECT emp_id, pay_amt FROM
emp_payments
UNION ALL
SELECT emp_id, pay_amt FROM
emp_absent_payments
Same as DB2
(duplicate rows also will be fetched)
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2.15 Dynamic SQL
.....
RETURNS refcursor
LANGUAGE plpgsql
AS
$$
DECLARE
sqlString1 VARCHAR(500);
sqlString2 VARCHAR(500);
selCur refcursor;
BEGIN
sqlString1 = 'SELECT code, list_code, short_description,
description ' ||
'FROM department ' ||
'WHERE code = ''' || strCode || '''';
sqlString2 = 'SELECT code, list_code, short_description,
description ' ||
'FROM payment_master ' ||
'WHERE code IN (''' || strCode || ''')';
IF iwhichCursor = 1 THEN
OPEN selCur FOR EXECUTE sqlString1;
RETURN selCur;
ELSEIF iwhichCursor = 2 THEN
OPEN selCur FOR EXECUTE sqlString2;
RETURN selCur;
END IF;
END;
$$
;
2.16 Condition Handling
EXCEPTION
WHEN division_by_zero or UNIQUE_VIOLATION THEN
RAISE NOTICE 'caught division_by_zero';
RETURN x;
END;
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
Where division_by_zero is a condition which when occurs it comes to the exception block to
execute it.
2.17 Print Output Messages
RAISE NOTICE 'Print any message ';
2.18 Implicit casting in SQL
2.18.1Casting double to integer syntax
SELECT Double_variable::INTEGER;
SELECT 235.22::INTEGER;
2.18.2Casting double to integer (Round)
SELECT 235.674::INTEGER;
This rounds the value to 236.
2.18.3Casting double to integer (lower possible integer)
To cast it to the lower possible integer, use Floor function.
SELECT FLOOR(235.22)::INTEGER;
2.19 Select from SYSIBM.SYSDUMMY1
There is no “SYSIBM.SYSDUMMY1” table equivalent in PostgreSQL. Unlike other RDBMS,
PostgreSQL allows a “select” without the ”from” clause.
SELECT FLOOR(42.2);
2.20 Variables declaration and assignment
Syntax
DECLARE <<Variable_name>> DATATYPE DEFUALT <<DEFUALT_VAL>>;
DECLARE iMaxLen INTEGER DEFAULT 0;
2.21 Conditional statements and flow control (supported by PostgreSQL)
2.21.1 IF – THEN – END IF
IF <boolean-expression> THEN
<statements>
END IF;
2.21.2 IF – THEN – ELSE – END IF
IF <boolean-expression> THEN
<statements>
ELSE
<statements>
END IF;
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
2.21.3 IF – THEN – ELSE IF – END IF
IF statements can be nested, as in the following example:
IF temp.val = ’m’ THEN
gender := ’man’;
ELSE
IF temp.val = ’f’ THEN
gender := ’woman’;
END IF;
END IF;
2.21.4 IF – THEN – ELSIF – THEN – ELSE
IF <boolean-expression> THEN
<statements>
[ ELSIF <boolean-expression> THEN
<statements>
[ ELSIF <boolean-expression> THEN
<statements>
...]]
[ ELSE
<statements> ]
END IF;
2.21.5 IF – THEN – ELSEIF – THEN – ELSE
ELSEIF is an alias for ELSIF & the usage is same as mentioned under IF – THEN –
ELSIF – THEN – ELSE clause
2.21.6 LOOP – statement – END LOOP
[ <<label>> ]
LOOP
statements
END LOOP [ label ];
2.21.7 WHILE condition – LOOP – END LOOP
[ <<label>> ]
WHILE expression LOOP
statements
END LOOP [ label ];
DB2 UDB To PostgreSQL Conversion Guide Version 1.0
3 Summary
Based on the initial experiment, the above similarities & differences are observed between IBM
DB2 & PostgreSQL. The scope of this initial exercise is restricted only to the extent of directly
checking all the IBM DB2 features & their PostgreSQL equivalents at the database level. We may
not though rule out the possibility of any unforeseen issues that can occur at the time of testing the
same from the application layer.
