Re: [Tracker] Database access abstraction



On Mon, 2008-11-03 at 09:05 -0500, Jamie McCracken wrote:
On Sat, 2008-11-01 at 01:11 +0000, Martyn Russell wrote:


These are just my thoughts after running over the API. I know your work
might consist mainly of what we already had, but this could be the
perfect opportunity to rename those functions to fit better to what they
actually do.

Great start JÃrg! Thanks for doing this.


have to agree - it looks very promising with better renaming

Thanks, I'll start applying changes to an SVN branch to prepare a merge,
then.

if we fully implemented this could we remove libtracker-db? (I would
assume it would be better combined with your stuff here)

While the database interface abstraction probably gets less important as
we essentially abstract on a higher level, we still need code that keeps
track of the different database files and how they are combined. I don't
mind much whether we integrate libtracker-db into libtracker-data or
keep them separate. Maybe we should postpone the decision until we see
how exactly the decomposed table support will look like.

JÃrg




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]