> How are you going to assign the database identifier? We do a
> similar
> thing in one of our systems. The primary key is made up of a 3 char
> database prefix, 2 chars for the table and 10 chars for a unique
> number (from a generator). This gives us a char(15) key.

Each database has an id which is needed for the incremental data
transfer (transport changes into the central database).
When setting this id wrong, the data transfer won't work,
so they are forced to set it correct.

Additionally, we have a table of all ids in our application hardcoded
to check them. This requires a new app when adding a database, but
saves a lot of support time.

> Also a guid can I believe be stored as a char(8) which is smaller.

I do not know any algorithm which does this.

> You seem to be making a very complex system for key generation when
> guids (or uuids) are available in most operating systems and many
> databases (MSSQL server, Oracle etc).
>
> Having tried both approaches, I would use a guid. These can also be
> generated in the application rather than the database if desired,
> requiring no access to the database (unlike your method which
> requires the high id to be generated via the database).

I am not sure if GUID is unique for us, since some customer use
Microsoft Terminal Server which would result in the same MAC-adress
for severall clients. With our approach I can handle this situation.

The high id is stored in a table and is read and incremented with the
old value in the where-clause. When no row has been updated, another
user did an update at the same time and I have to start it over.

I tried to shrink the needed bits and would need at the moment
64 bits. Fortunately, interbase supports this via INT64 and NUMERIC
(18,0). Numeric is portable, so I could use this key also on other
databases like Oracle.

Unfortunately, the BDE doesn't support large ints (or I am to stupid
to get it working)!!!! Does anyone know how to get this working?

Thanks in advance,
CL