>kind of 'abstract' database class. Unfortunately Database class is tightly
>bound to the semantics of Berkeley DB and will never be good enough
>for a SQL database (it would be good for gdbm/ndbm etc...).
Hmm. I didn't think it was *that* tied to a particular db. Granted,
it doesn't take advantage of everything that database packages
offer--even the Berkeley code offers join functionality. But it seems
like the low-level material, Get, Put, Delete, Exist, Open, Read,
Close, etc. was about all that was there.
>touching 6 files. I don't mind if we get an advantage. Since the advantage
>we have is almost nothing, I'd say we should better go to the simpler
>solution.
Sure. I broke DB2_hash out, mostly because I didn't want to mess up
the regular class with anything that I did. I can certainly see
merging the two classes (and perhaps picking a different name would
be good since "DB2" is a product from IBM).
As for the Database class itself, why don't you sketch out an idea of
what you're thinking. I'm of the opinion that we should have one
'abstract' class and make that as clean as possible--if that means
that there's no code in Database and the present code is moved to the
Berkeley class, great.
-Geoff
------------------------------------
To unsubscribe from the htdig3-dev mailing list, send a message to
[EMAIL PROTECTED] containing the single word "unsubscribe" in
the SUBJECT of the message.
