[Jruby-devel] Bravo on JRuby

2006-04-30 Thread Michael Garrels
I just wanted to tell you that I think your JRuby program will have a huge affect.  This Friday I downloaded Google's new Sketchup like many other people.  I've spent the past day or two fiddling with it, including learning the scripting language that it uses: Ruby.    I've written the occasiona

Re: [Jruby-devel] Another socket patch

2006-04-30 Thread Thomas E Enebo
On Sun, 30 Apr 2006, Evan Buswell defenestrated me: > > The select() in question is on TCPServer objects, so doing a workaround > with InputStream.available isn't going to work. As far as I can tell, we > need to break out nio. Yep. As Charlie noted, we need nio. It is just a matter of time.

Re: [Jruby-devel] Another socket patch

2006-04-30 Thread Charles O Nutter
If you were to tackle a NIO IOHandler, you'd be a lot of people's hero. The move to NIO is not a maybe, it's a must...there's simple too much in Ruby we can't implement without it. We've already started to use NIO in a few places for things like file locking, so we're on the way there already. In

[Jruby-devel] Linux rubicon results

2006-04-30 Thread Charles O Nutter
I'm on linux now and just ran rubicon with some modifications of mine; does the following look like a correct result? All 71 files FAIL 105015317 64 114 64 fail/114 error -- Charles Oliver Nutter @ headius.blogspot.com JRuby Developer @ jruby.sourceforge.net Application Arc

Re: [Jruby-devel] Another socket patch

2006-04-30 Thread Evan Buswell
Poking around a bit more, I defined AI_PASSIVE to some dummy value and now WEBrick::GenericServer runs up to the point of select(...). Of course, this is no big deal since we already knew it was going to fail on select and that's the sticky point here anyway. The select() in question is on TCPSer