Wonderful! I'd be interested in the alternatives you consider and review. Also, I'll state up front that I'm a stickler for keeping this stuff simple. I'm the one of two initial authors of Python's DB-API. IMO, it is clean and simple for both user and implementor, yet provides ways for both to crank out serious performance. The capability negotiation is transparent and implied, so it works really well.
I would suggest that the set of databases that PHP can access as a target point for APRUTIL. I look forward to your draft. Thanks! Cheers, -g On Tue, Mar 13, 2001 at 09:18:49AM +0000, [EMAIL PROTECTED] wrote: > Since it looks like this is new ground and no one has any objection, > I'd like to go ahead an draft something up. I'll email the list again > when I have something worth looking at. > > john. > > On Mon, 12 Mar 2001 13:57:25 -0800 Greg Stein <[EMAIL PROTECTED]> allegedly > said: > > On Mon, Mar 12, 2001 at 05:52:08PM +0000, [EMAIL PROTECTED] wrote: > > > Hi, > > > > > > I was wondering if there were any plans or if there should be > any > > > plans for adding a relational database abstraction layer to apr > (similar > > > to Perl's DBI)? I noticed there is dbm in apr-util but it doesn't > > > look like it's meant for relational databases. > > > > No plans that I've heard of. > > > > I would love to see it, but won't be starting any time soon. One > day, when I > > go and try to build SQL support into the back-side of Subversion, > then I > > might look into something like this. I'm sure there are also some > existing > > libraries/designs that we can take advantage of (esp. if the licensing > is > > decent). > > > > Cheers, > > -g > > > > -- > > Greg Stein, http://www.lyra.org/ > > -- Greg Stein, http://www.lyra.org/
