>I'll guess your changes aren't against the conndef branch which is where
>David and I have been making lots of changes. With the new conndef
>system, you don't pass conn strings to data sets... instead a
>ConnDefinition object, which is abstract and implemented by either a raw
>connection or the old pooling code.
No, they're against the old branch. Is the new branch considered
stable? Is there any functional difference between this branch and
the main branch? Does the ConnDefinition object expose the name of
the database its connected to? One thing that I got a bit frustrated
with in the current branch is what's I'd consider -excessive-
data-hiding. There are private fields that, realistically, should be
protected, or at least exposed with protected accessors. If the new
branch is stable, let me know and I'll move over to working with it.
I'll look through it and find a way to patch the problem there
instead of the main branch.
>Are your changes made on this branch, and do you think they'll work with
>the ConnDefinition object? With this bug worked out and some more
>testing, we might want to make another major version snapshot for Town
>(oh, and review the conn pooling code as well).
Sounds good. :-)
------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Problems?: [EMAIL PROTECTED]
