Before in IRDBImport, when you imported data, it would take whatever data was imported, and once the import statement was finished, it would columnize the result.
If you imported a second or more tables with the same name, it would then uncolumnize the previous result, and append the new data.
This is fine for a few tables, but if you are importing tens or hundreds of separate tables, the columnizing and uncolumnizing starts to slow things down a bit.
So we’ve added the keyword UNCOLUMNIZED that can go after the = in an IMPORT statement, and it will keep the data in an uncolumnized state. This speeds up importing many small tables into the one table.
However while tables are in an uncolumnized state, they will take up more memory, and you cannot execute SQL statements against them.
IMPORT t1 = UNCOLUMNIZED a1.orders
IMPORT t1 = UNCOLUMNIZED a1.orders
COLUMNIZE t1
The above is an example irdbImport script that imports the table orders twice from datasource A1 as table t1. The last statement COLUMNIZE will then columnize the data.