With a UTF-8 database, CODEUNITS32 is not correct with create table script. The values are 4x higher than they should be. (4243739)

With a UTF-8 database, CODEUNITS32 is not correct with create table script. The values are 4x higher than they should be.

With a UTF-8 database, CODEUNITS32 is not correct with create table script. The values are 4x higher than they should be.


If you create a table for example

CREATE TABLE "TEST "."Table" (
                 "C1" VARCHAR(16 CODEUNITS32) NOT NULL ,
                 "C2" VARCHAR(32 CODEUNITS32) NOT NULL ,
                 "C3" CHAR(3 CODEUNITS32) NOT NULL ,
                 "C4" TIMESTAMP NOT NULL ,
                 "C5" TIMESTAMP ,
                 "C6" CHAR(1 CODEUNITS32) ,
                 "C7" VARCHAR(200 CODEUNITS32) )
                DISTRIBUTE BY HASH("C1")
                  IN "TEST"
                ORGANIZE BY ROW;

Then, right click on a table -> Generate SQL -> Creation script -> in editor

It then shows:

CREATE TABLE TEST.TABLE (
 C1   VARCHAR(64 CODEUNITS32)   NOT NULL,
 C2   VARCHAR(128 CODEUNITS32)   NOT NULL,
 C3   CHARACTER(12 CODEUNITS32)   NOT NULL,
 C4   TIMESTAMP   NOT NULL,
 C5   TIMESTAMP,
 C6   CHARACTER(4 CODEUNITS32),
 C7   VARCHAR(800 CODEUNITS32)
 )
 IN TEST
 DISTRIBUTE BY HASH (C1)
 ORGANIZE BY ROW;

The value is 4 times bigger then original.