Thanks, Gregg! That is interesting. Unfortunately I have not much time for experimenting and had to focus on get things done.
Zsolt On Tue, Jan 13, 2015 at 3:39 PM, Gregg Wonderly <[email protected]> wrote: > This is also a space where queuing can make things easier. It some cases > like these, there are natural flows of state that are better represented > that way. A long time ago, I created a new Jini service that I called > griddle. It is still visible on java.net. That service is a follow on > to java spaces, but has the ability to plug in “matching” algorithms as > well as separate “keys” from data. Keys are intended to always be > non-downloaded classes. So, they would typically be native classes, or > they would be classes that you would deploy with your griddle instance, in > it’s class path. The same with the matching algorithms, in general. It > wasn’t ever dramatically tested and proven, but it did function as I > recall. That was about the time that economy tanked and I had to work on > some other things which did then and continue, today, to keep me away from > Jini development. > > Gregg Wonderly > > > On Jan 7, 2015, at 6:51 AM, Zsolt Kúti <[email protected]> wrote: > > > > Tom, > > > > Thank you for what you have outlined. This will help. > > > > Zsolt > > > > > > On Wed, Jan 7, 2015 at 1:28 PM, Tom Hobbs <[email protected]> > wrote: > > > >> An approach that comes immediately to mind is implementing a lock > >> mechanism. Write a "lock object" to the space (either a global one or a > > > > > > ... > >
