Hi, On Sat, Sep 11, 2010 at 12:51:07AM +0200, Samuel Thibault wrote: > olafbuddenha...@gmx.net, le Fri 10 Sep 2010 02:31:25 +0200, a écrit : > > On Thu, Sep 09, 2010 at 02:03:28AM +0200, Samuel Thibault wrote:
> > > Or put another way, I'm really not at ease adding an RPC just for > > > this kind of information, because later on we'll probably want > > > some other information, and then another, etc. > > > > I'm not sure why you consider this a problem?... > > Bloat. Indeed, that's a valid concern, and one I care about deeply myself. However, it seems to me that in this case adding specific RPCs is actually the approach adding the *least* bloat... > > *If* ten year down the road we indeed end up with half a dozen > > miscallaneous info queries, we can *still* replace them by a new RPC > > covering all of it... > > But still have to keep the old calls for old clients. For a transitional period, yes. Compatibility interfaces are not pretty... But they are pretty much inevitable when creating any complex software systems. It's just something we have to accept. > > Also note that there is precedenct for extending the proc info query > > by adding new RPCs. > > Ok, but I guess it wasn't just for _one_ piece information. It is. -antrik-
