SERIAL data types

Informix®

1. Create the table with a column using SERIAL, or BIGSERIAL.
2. To generate a new serial, no value or a zero value is specified in the INSERT statement:

   INSERT INTO tab1 ( c ) VALUES ( 'aa' )
   INSERT INTO tab1 ( k, c ) VALUES ( 0, 'aa' )

3. After INSERT, the new value of a SERIAL column is provided in SQLCA.SQLERRD[2], while the new value of a BIGSERIAL value must be fetched with a SELECT dbinfo('bigserial') query.

Informix allows you to insert rows with a value different from zero for a serial column. Using an explicit value will automatically increment the internal serial counter, to avoid conflicts with future INSERT statements that are using a zero value:

CREATE TABLE tab ( k SERIAL); -- internal counter = 0
INSERT INTO tab VALUES ( 0 ); -- internal counter = 1
INSERT INTO tab VALUES ( 10 ); -- internal counter = 10
INSERT INTO tab VALUES ( 0 ); -- internal counter = 11
DELETE FROM tab; -- internal counter = 11
INSERT INTO tab VALUES ( 0 ); -- internal counter = 12

PostgreSQL

PostgreSQL SERIAL data type:

• PostgreSQL's SERIAL data type has the same name as in Informix, but it behaves differently.
• You cannot define a start value ( SERIAL(100) ).
• You cannot specify zero as serial value to get a new serial, the PostgreSQL serial is based on default values, thus you must omit the serial column in the INSERT statement.
• When you INSERT a row with a specific value for the serial column, the underlying sequence will not be incremented. As result, the next INSERT that does not specify the serial column may get a new sequence that was already inserted explicitly.
• With some old versions of PostgreSQL, when you drop the table you must drop the sequence too.

PostgreSQL sequences:

• The purpose of sequences is to provide unique integer numbers.
• To create a sequence, you must use the CREATE SEQUENCE statement.
• To get a new sequence value, you must use the nextval() function:

  INSERT INTO tab1 VALUES ( nextval('tab1_seq'), ... )

• To get the last generated number, PostgreSQL provides the currval() function:

  SELECT currval('tab1_seq')

Solution

The Informix SERIAL data type can be emulated with three different methods.

dbi.database.dbname.ifxemul.datatype.serial.emulation = {"native"|"regtable"|"trigseq"}

• native: uses the native PostgreSQL serial data type.
• regtable: uses insert triggers with the SERIALREG table.
• trigseq: uses insert triggers with sequences.

The default emulation technique is "native".

dbi.database.dbname.ifxemul.datatype.serial = {true|false}
dbi.database.dbname.ifxemul.datatype.serial8 = {true|false}
dbi.database.dbname.ifxemul.datatype.bigserial = {true|false}

Using the native serial emulation

The "native" mode is the default serial emulation mode, using the native PostgreSQL SERIAL data type. In this mode, the original type name will be left untouched by the SQL Translator and you will get the behavior of the PostgreSQL SERIAL column type, based on sequences.

See also the PostgreSQL documentation for more details about the native SERIAL type.

Using the regtable serial emulation

With the "regtable" mode, the SERIAL data type is emulated with a PostgreSQL INTEGER data type and INSERT triggers using the table SERIALREG which is dedicated to sequence production. After an insert, sqlca.sqlerrd[2] register holds the last generated serial value. BIGSERIAL and SERIAL8 types can be converted to BIGINT in PostgreSQL, but the sqlca.sqlerrd[2] register cannot be used since it is defined as an INTEGER type.

The triggers can be created manually during the application database installation procedure, or automatically from a BDL program: When a BDL program executes a CREATE [TEMP] TABLE with a SERIAL column, the database interface automatically converts the SERIAL data type to INTEGER and dynamically creates the triggers.

You must create the SERIALREG table as follows:

CREATE TABLE SERIALREG (
   TABLENAME VARCHAR(50) NOT NULL,
   LASTSERIAL DECIMAL(20,0) NOT NULL,
   PRIMARY KEY ( TABLENAME )
)

In database creation scripts, all SERIAL[(n)] data types must be converted to INTEGER data types and you must create one trigger for each table. To know how to write those triggers, you can create a small Genero program that creates a table with a SERIAL column. Set the FGLSQLDEBUG environment variable and run the program. The debug output will show you the native trigger creation command.

INSERT INTO tab ( col1, col2 ) VALUES ( NULL, 'data' )

INSERT INTO tab (col2) VALUES ('data')

Check if your application uses tables with a SERIAL column that can contain a NULL value. Consider removing the serial column from the INSERT statements.

Using the trigseq serial emulation

With "trigseq", the SERIAL data type is emulated with a PostgreSQL INTEGER data type and INSERT triggers using a sequence tablename_seq. After an insert, sqlca.sqlerrd[2] register holds the last generated serial value.

The triggers can be created manually during the application database installation procedure, or automatically from a BDL program: When a BDL program executes a CREATE [TEMP] TABLE with a SERIAL column, the database interface automatically converts the SERIAL data type to INTEGER and dynamically creates the triggers.

In database creation scripts, all SERIAL[(n)] data types must be converted to INTEGER data types and you must create one trigger for each table. To know how to write those triggers, you can create a small Genero program that creates a table with a SERIAL column. Set the FGLSQLDEBUG environment variable and run the program. The debug output will show you the native trigger creation command.

INSERT INTO tab ( col1, col2 ) VALUES ( NULL, 'data' )

INSERT INTO tab (col2) VALUES ('data')

Check if your application uses tables with a SERIAL column that can contain a NULL value. Consider removing the serial column from the INSERT statements.

Notes common to all serial emulation modes

Since sqlca.sqlerrd[2] is defined as an INTEGER, it cannot hold values from BIGSERIAL (BIGINT) auto incremented columns. If you are using BIGSERIAL columns, you must query the sequence pseudo-column CURRVAL() or fetch the LASTSERIAL column from the SERIALREG table if used.

For SQL portability, it is recommended to review INSERT statements to remove the SERIAL column from the list.

For example, the following statement:

INSERT INTO tab (col1,col2) VALUES ( 0 , p_value )

can be converted to:

INSERT INTO tab (col2) VALUES (p_value)

Static SQL INSERT using records defined from the schema file must also be reviewed:

DEFINE rec LIKE tab.*
INSERT INTO tab VALUES ( rec.*) -- will use the serial column

can be converted to:

INSERT INTO tab VALUES rec.* -- without parentheses, serial column is removed