>
>
> Yup, but I found that it is necessary to set both to their respective
> values instead of null (e.g. for the tests:
>
> http://svn.apache.org/viewcvs.cgi/db/ddlutils/trunk/src/test/jdbc.properties.sqlserver2000?view=markup
> ),
> otherwise too much info was returned. Though all-uppercase is probably
> not required (as opposed to Oracle).
>
> Tom
Oh no. It doesn't work here. I have to set both catalog and schema to null
on SQL Server 2000 with JDBC driver 2005, otherwise I get nothing generated
in the XML file.
Therefore, the extra index problem still exists here.
Anyone else having the same problem?
Another issue, I can get the data outputed into XML for SQL Server, but get
nothing for Oracle using the same codes. Any idea about this? The codes look
like the following:
FileOutputStream fOS = new FileOutputStream(outputFile);
DataWriter dataWriter = new DataWriter(fOS);
// Tables that I want to get data from
List<Table> sysTables = getDataTables(datasource, db);
Table[] tables = new Table[1];
dataWriter.writeDocumentStart();
for (Table tb : sysTables)
{
String sql = "select * from " + tb.getName();
tables[0] = tb;
Iterator dataIt = platform.query(db, sql, tables);
// Write data to xml file
dataWriter.write(dataIt);
}
dataWriter.writeDocumentEnd();
And the way that I read model is like:
Database db = null;
if ("Oracle".equalsIgnoreCase(dbType)) {
db = platform.readModelFromDatabase(
schemaName.toUpperCase(), schemaName.toUpperCase(),
schemaName.toUpperCase(), null);
}
if ("Microsoft SQL Server".equalsIgnoreCase(dbType)) {
db = platform.readModelFromDatabase(schemaName, null, null, null);
}
--
--------------
Cheers,
Jun
