Re: Java 5

2007-08-09 Thread Michael Gentry
Another advantage of deciding on 1.5 is we can have native enumeration support. I would definitely like to see this in the modeler and could try to add it (my Swing skills aren't so hot, but I can be stubborn). I'd like to be able to define the enumerated values in the modeler and be able to choos

Re: Java 5

2007-08-07 Thread Andrus Adamchik
--- From: Kevin Menard [mailto:[EMAIL PROTECTED] Sent: Saturday, August 04, 2007 10:39 PM To: dev@cayenne.apache.org Subject: RE: Java 5 I think I'll poll the user list and see how many actually use 1.4 still anyway. It'd be educational. As you've all likely seen, I started up the po

RE: Java 5

2007-08-07 Thread Kevin Menard
> -Original Message- > From: Kevin Menard [mailto:[EMAIL PROTECTED] > Sent: Saturday, August 04, 2007 10:39 PM > To: dev@cayenne.apache.org > Subject: RE: Java 5 > > I think I'll poll the user list and see how many actually use 1.4 still > anyway. It

Re: Java 5

2007-08-06 Thread Mike Kienenberger
http://g.oswego.edu/dl/classes/EDU/oswego/cs/dl/util/concurrent/intro.html On 8/6/07, Craig L Russell <[EMAIL PROTECTED]> wrote: > I don't have details in front of me, but I understand that many of > the popular Concurrent classes have been back-ported to Java 1.4. If > you like, I can get more d

Re: Java 5

2007-08-06 Thread Andrus Adamchik
I think there are ways to fully port 1.5 stuff to 1.4 (yes - except for bytecode format). There is an open source project somewhere that does back-compiling. But I'd rather we avoid more temporary hacks. There is nothing that is pressing us to switch, so we can wait and pick the right time.

Re: Java 5

2007-08-06 Thread Craig L Russell
I don't have details in front of me, but I understand that many of the popular Concurrent classes have been back-ported to Java 1.4. If you like, I can get more details if you can't find it on the Internet. The back-port classes don't use the new Java byte-codes that were added in Java 5 so

Re: Java 5

2007-08-06 Thread Andrus Adamchik
On Aug 5, 2007, at 5:39 AM, Kevin Menard wrote: the new concurrent API could yield significant speed improvements. Being all backwards compatible, I haven't played much with that, but I suspect that things like ConcurrentHashMap should help us to improve throughput of DataRowStore and Eve

Re: Java 5

2007-08-05 Thread Michael Gentry
sure and the new concurrent API could > yield significant speed improvements. > > -- > Kevin > > > -Original Message- > > From: Andrus Adamchik [mailto:[EMAIL PROTECTED] > > Sent: Saturday, August 04, 2007 2:22 PM > > To: dev@cayenne.apache.org > > Subject

RE: Java 5

2007-08-04 Thread Kevin Menard
22 PM > To: dev@cayenne.apache.org > Subject: Re: Java 5 > > Yeah, I'd think we should preserve the 1.4 compatibility for 3.0 > Cayenne Persistence API (vs. JPA that is 1.5 by definition). > Personally I hope I'll be off 1.4 in a matter of months on all my > projects,

Re: Java 5

2007-08-04 Thread Andrus Adamchik
Yeah, I'd think we should preserve the 1.4 compatibility for 3.0 Cayenne Persistence API (vs. JPA that is 1.5 by definition). Personally I hope I'll be off 1.4 in a matter of months on all my projects, but I am sure lots of users will appreciate us preserving it for another year or so. I

Re: Java 5

2007-08-03 Thread Aristedes Maniatis
On 04/08/2007, at 4:07 AM, Kevin Menard wrote: How much longer is 1.4 going to be the target platform? Interesting because I am in the middle of a process of taking our biggest piece of software (around 100,000 lines of code) and generifying [1] it for 1.5. The decision was made to abando

Re: Java 5

2007-08-03 Thread Mike Kienenberger
I wouldn't see a problem with making 3.0 require Java 1.5. JPA already requires Java 1.5. On 8/3/07, Kevin Menard <[EMAIL PROTECTED]> wrote: > Another one to ponder. It should be my last today ;-) > > According to [1], Java 5 was released nearly three years ago. At the > time of its release, it