That might be a problem for us... We're about to use AHC (which is based on mina 1.1.x) in a production environment. Switching to mina 2.0 now would set us back in terms of invested time (testing, regression, etc.)... If at all possible, it would be great if we could support the current AHC as is, while continuing the work on rewriting the client. Thoughts? Regards, Sangjin
On Wed, Mar 5, 2008 at 4:05 PM, Alex Karasulu <[EMAIL PROTECTED]> wrote: > You might though want to use MINA 2.0 the move is not that big and it > might > be the best option. > > Alex > > On Wed, Mar 5, 2008 at 6:39 PM, Sangjin Lee <[EMAIL PROTECTED]> wrote: > > > OK, thanks... I like the suggestion. +1 from me. :) > > Sangjin > > > > On Wed, Mar 5, 2008 at 2:27 PM, Mike Heath <[EMAIL PROTECTED]> wrote: > > > > > Sangjin Lee wrote: > > > > That sounds like a good idea. Just so I understand, your proposal > is > > to > > > > move the existing AHC in the Geronimo sandbox based on mina 1.1.xover > > > to > > > > asyncweb under a branch and keep up the maintenance and support on > it, > > > > right? > > > > Thanks, > > > > Sangjin > > > > > > Yes, that's exactly what I'm suggesting. > > > > > > -Mike > > > > > > > > > > > > > > > On Tue, Mar 4, 2008 at 2:16 PM, Mike Heath <[EMAIL PROTECTED]> > wrote: > > > > > > > >> Sangjin Lee wrote: > > > >>> I would also like to see asyncweb make progress as quickly as > > > possible, > > > >> and > > > >>> I'd like to contribute to that effect as well. As Mike pointed > out > > in > > > a > > > >>> different thread, however, there are some challenges to this. > It's > > > >> looking > > > >>> more likely that this is not going to be a simple "merge" of code > > but > > > >>> substantial rework. I think part of it stems from the fact that > the > > > old > > > >> AHC > > > >>> relies on its own codec (based on mina 1.1.x) and the asyncweb > > already > > > >> has a > > > >>> good codec that's completely different from AHC's. > > > >>> We do have an immediate need to use AHC *now*, and critical bug > > fixes > > > >> need > > > >>> to happen, as we're using it right now. But we're making a > > conscious > > > >> effort > > > >>> to limit the changes to mostly bug fixes, and we're trying to > > > propagate > > > >> the > > > >>> changes to asyncweb whenever it is applicable. Those are the > things > > > >> we're > > > >>> doing (or trying to do) to make sure things don't diverge or get > out > > > of > > > >>> hand. > > > >> Why don't we put AHC in a branch in the AsyncWeb Subversion > > repository? > > > >> This way AHC can continue using its own codec and we can support > and > > > >> maintain it without going through a lot of work. Once it gets > > > >> stabilized we could even cut a release. > > > >> > > > >> In the mean time, we can continue working toward a revised "2.0" > > client > > > >> that uses the AsyncWeb codec. > > > >> > > > >> WDYT? > > > >> > > > >> -Mike > > > >> > > > >>> On Tue, Mar 4, 2008 at 6:49 AM, Alex Karasulu < > [EMAIL PROTECTED]> > > > >> wrote: > > > >>>> I am in agreement as well. I would like to see this merge happen > > > >> quickly > > > >>>> so > > > >>>> the users see progress and there's no longer any need to keep the > G > > > >> branch > > > >>>> alive. Someone said to me you need to get cookin in the kitchen > > when > > > >> the > > > >>>> guests arrive :). Then we can just start releasing some > milestones > > > >> that > > > >>>> people can use and we can track/patch etc. > > > >>>> > > > >>>> It's nice now that MINA 2.0-m1 is out. This means we can release > > an > > > >>>> Asyncweb milestone as a whole. > > > >>>> > > > >>>> Also another thing I want people to think about is that this > > project > > > is > > > >>>> one > > > >>>> unit rather than just a client. There's a server in there too > and > > > we > > > >> can > > > >>>> release it together. The community around this is coming > together > > > fast > > > >>>> and > > > >>>> that's just great which means there's a good potential for > > graduating > > > >> this > > > >>>> project eventually. > > > >>>> > > > >>>> These are my hopes for Asyncweb. > > > >>>> > > > >>>> Alex > > > >>>> > > > >>>> On Tue, Mar 4, 2008 at 9:33 AM, Jeff Genender < > [EMAIL PROTECTED] > > > > > > >>>> wrote: > > > >>>> > > > >>>>> I agree with Alan...I understood that the G version was going > away > > > now > > > >>>>> that we built community over here on this. Comments? > > > >>>>> > > > >>>>> Jeff > > > >>>>> > > > >>>>> Alan Cabrera wrote: > > > >>>>>> On Mar 1, 2008, at 8:12 PM, Alan D. Cabrera wrote: > > > >>>>>> > > > >>>>>>> AsyncHttpClient was changed w/ the last checkin on 2/26 and > now > > > the > > > >>>>>>> build is broken. > > > >>>>>> I looked at the actual changes. I'm just trying to grok the > > > changes > > > >>>>>> because I realize that I am new here. It seems that the "old" > > > >>>>>> AsyncHttpClient is still evolving? How does this fit in with > the > > > >>>> plans > > > >>>>>> for the "old" AsyncHttpClient, the "new" Geronimo > > AsyncHttpClient, > > > >> and > > > >>>>>> the new API that's currently in discussion? > > > >>>>>> > > > >>>>>> I had thought, maybe naively, that we were going to roll the > > "old" > > > >> two > > > >>>>>> into the new one. > > > >>>>>> > > > >>>>>> > > > >>>>>> Regards, > > > >>>>>> Alan > > > >> > > > > > > > > > > > > >
