Modeller and dbgeneration

From: Craig Miskell (cmiskel..lbatross.co.nz)
Date: Tue Nov 26 2002 - 03:01:07 EST

  • Next message: Holger Hoffstätte: "Re: Modeller and dbgeneration"

    Hi,
            I was just today playing with modeller and generating the db. I
    was wondering why the table generation wasn't working quite as I expected
    it to, when dropping and recreating the tables.

    Upon investigation, it turns out that when both drop and create is
    specified, that all tables currently in the map that do exist will be
    dropped, and any that do not currently exist but are in the map will be
    created, but not the ones that were just dropped.

    My gut expectation (fuelled by EOModeller in EOF), is that specifying both
    drop and create for tables will cause all necessary tables to be dropped,
    and *all* tables to be created. Typically this is quite useful during
    development, when the table structure changes periodically.

    Is this an unreasonable expectation, because the code certainly seems to
    me to be quite deliberately written as such (the foreign key generation
    code specifically only handles foreign keys for tables explicitly newly
    generated). Or maybe I'm reading too much into the code.

    The solution is a one liner, but I don't want to break a deliberately
    chosen decision.

    Thoughts?

    Craig



    This archive was generated by hypermail 2.0.0 : Tue Nov 26 2002 - 03:01:11 EST