Our system requires that people can use a variety of locations for their data including using more that one set of data. We wish to define a dbt file for the most frequently used set but it appears that DEFINE_DATABASE will only operate in the development version of OI (we use 3.14). Is this really the case and if so, is there a straightforward workaround?
Thanks
You could try programmatically creating a .DBT file… If you look at one in AREV's editor you could fairly easily (IMHO) reverse-engineer its structure and write out your own.
Do a search for "Swap_Database" on this discussion group too; you'll see a few things you might be interested in.
Firstly, thanks Alex for the suggestion - I will have a look into that.
What I don't understand though is why it should need a development version to do such a basic task that might be expected of any end users, after all it doesn't allow compilation of programs and even that you can do in a small way with dictionary items (or at least you could in AREV - I haven't tried in OI).
Just keep multiple DBTs and overwrite them….
World Leaders in all things RevSoft (Except VIP)
That is all very well if you know in advance what the paths will be. We have at least one user wanting to put the data on one part of a server and the app on another. We also allow for multiple data sets to be used and users might make more of their own.
David, there's another possibility you may have
missed… Why bother using hardcoded database
schema (.DBT files) at all?
If you want your data to be available from diverse
network locations, why not just attach whatever
tables you need using Attach_Table every time the
application starts? There may be some performance
implications of this I'm not aware of, but it
definitely works. Keep a list of which
users/accounts/etc. need which tables/locations,
and attach the appropriate ones in the CREATE
event of your application's main window.
Good luck…
Alex
Yes, that is what we used to do, but thought that using define database was meant to be quite a bit faster for starting up (particularly across relatively slow networks). The idea was to allow users to choose which set was to be fast and save that and then use ordinary attaches for any other set that they wished to attach e.g. to search a remote dataset.
So Osread and Swap…
World Leaders in all things RevSoft (Except VIP)