On Feb 17, 2009, at 1:21 PM, Uwe Grauer wrote:
> Often when i criticize something, you seem to take it personally.
> Isn't discussion something we should do?
It's not personal, and it certainly isn't a discussion when you make
sweeping pronouncements about the consequences of design decisions
first, and then ask for discussion.
> When it comes to real problems, i have to tell you that i can't use
> dConnectInfo any more, because it doesn't let me specify which
> connection file i want to use.
> I know how to work around this, but it is a problem because
> now dConnectInfo is useless for me.
This is an excellent example: you want to use the same name for
connections, and distinguish by file. After the design is explained to
you, you insist on following your concepts instead of working with the
framework design. I've stated many times that connection names must be
unique, but you stubbornly insist that it should be connection file +
connection name that should be unique.
> Explicit is better than implicit - you know that!
And yet you insist that explicitly naming your connections is not a
good thing.
Let me explain this once more, explicitly: the concept of file has
absolutely nothing to do with connection definitions. You can define
connections in code, or you can use a tool to define them in XML. If
you choose the latter, you can define multiple connections in one
file, or one per file, or any combination, because *files don't
matter*. If you want the Dabo application object to manage your
connection definitions, each connection must be named uniquely.
-- Ed Leafe
_______________________________________________
Post Messages to: [email protected]
Subscription Maintenance: http://leafe.com/mailman/listinfo/dabo-dev
Searchable Archives: http://leafe.com/archives/search/dabo-dev
This message:
http://leafe.com/archives/byMID/[email protected]
