In extracttable.java,

getCheckSum

getCreateSetType

: we parse the creation string to extract the value, problem is we currently parse the string into objects using " ", but you can createa table in many diffrent ways returning

strings like the following

CREATE TABLE CHECK_SUM_DEFAULT\n\tCHECKSUM = DEFAULT (\n\tQITS    TIMESTAMP(6) FORMAT 'YYYY-MM-DDBHH:MI:SS.S(6)' DEFAULT CURRENT_TIMESTAMP(6),\n\tA       CHAR(5) CHAR SET LATIN NOT NULL \n\t)

in this example you can see that the object doesn't break down into the correct peaces do to use of \n\t, in order to create a qualified parser for this we would need to know all of the possible characters that can go into the text and then create a much more complex parser to extract the correct data...

could we fill a bug for teradata to store this information somewhere in a catalog table?