Re: Erwin reverse engineering question

"Jerry Gitomer" <jgitomer_at_erols.com> wrote in message news:9snkqq$ksc$1_at_bob.news.rcn.net...
> Tim Mueller wrote:
>
> > As part of a system migration project, I need to reverse
> > engineer a fairly
> > simple schema in an existing Oracle database. There are about
> > half a dozen primary tables, say 10 lookup value tables (the
> > sort of thing that would be used for value assitance in a drop
> > down) and a couple of M:M tables.
> >
> > The problem I'm having is that the original designer/programmer
> > wasn't
> > consistent in his column names for the foreign keys. For
> > example, there is a lookup table called LOCATION that has three
> > columns, CODE, NAME,
> > SHORT_NAME. In the tables that have a relationship with the
> > LOCATION table, the foreign key column is named LOC_CODE,
> > because there are several other xxx_CODE columns there as well.
> >
> > This makes ERwin gag when trying to create the foreign key
> > relationships, and it's compounded by the fact that often the
> > parent table itself has a primary key column called CODE.
> >
> > Anybody have any suggestions on how to define the FK
> > relationships to ERwin,
> > short of renamng columns? Changing the names will craze the
> > programmers who need to refer to the model later.
> >
> > Thanks.
> >
> >
> >
> >
> You are best off biting the bullet and renaming the columns.
> To minimize hardship on your developers prepare a cross
> reference for them with old names and new names.
>
>
> --
> Jerry Gitomer
> Once I learned how to spell DBA, I became one

I believe you can define synonyms in Oracle, for the old names, so the Developers can still use their names without having to change anything. Received on Mon Nov 12 2001 - 20:14:42 CET