Hello all,
I'm developing a tool to move data from a SQL server database to a Derby
database using Java.
Both databases have been constructed using through nearly identical (as much as
possible) SQL scripts run through the appropriate means for each engine. One of
the initial steps of my process is to confirm that the schema of the two
databases is identically before transferring data.
This has uncovered some issues.
1) I want to compare the column names in each table to ensure they match. It
seems that derby does not necessarily store column names in the order they were
declared at the time of the tables creation. Ex.
SQL:
CREATE TABLE someTable (colA int, colB int, colC int);
JAVA:
For (int i = 1; I < results.getMetaData().getColumnCount(); i++) {
System.out.print(results.getMetaData().getColumnName(i) + " ");
}
The output of the java code is not always "colA colB colC", for some tables it
is appearing as "colB colA colC" for example.
When the table is viewed via the ij tool, the columns are in the expected order
(the order they were declared).
2) in a similar vein, .getMetaData().getColumnCount() sometimes returns more
columns then were specified in creation script.
Any help would be much appreciated, and allow me to avoid developing a
complicated work-around.
Thanks,
Alain Kuchta
