Smoerk, I totally agree. I would love to have Zope/JPython running on Java. Use java for heavy-lifting (speed a la C++, deluxe garbage collection, GUI, etc.) and JPython for everything else. Servlets would be great if nothing else than to take advantage of all the Java Servlet code that's coming available to tie some servlet(s) into Zope (or just using it/them from Zope). Automatic dovetail with Java object model would an absolute godsend. Get Zope communicating with Java applets hassle free and fairly efficiently (XML-RPC may be pretty hassle-free but requires XML (gulp) and not efficient). XML-RPC is the best standards based approach I know of now. I think dovetailing Zope with Java would make a whole lot of "alternative technology" folk far more comfortable with Zope (think checkbox: "Does it support Java?"). Basically, I think there's an attitude with many technologies, that if you're not for them, you're against them. If it's not that, just the sheer convenience of built-in support would be worth a lot. Plus, Zope could become embeddable in any Java solution. That would be great for Zope. -=- The speed-up might be there. It certainly seems like there should be some. However, JPython is not optimized yet and is really still a work in progress. There's some idea that Python could morph/evolve in two or so years (i.e., a long time). One possibility for this morph would be to dovetail with Java better (i.e., to run on JVM's better). Right now, the Python language is 'defined' (i.e., implemented in CPython and emulated in JPython) so that primitive types in Java are not taken fully advantage of (I believe). Apparently, with a few tweaks (but unfortunately not backwards compatible tweaks) to the Python language, JPython could run magnitudes faster (on the order of 10-100x faster). Needless to say, that would make for a great scripting language. All of this is pie in the sky and a long way a way. The reason it's possible is that Guido (father of Python) is with the advent of JPython considering actually defining Python formally, i.e., separately from CPython. They never needed that before JPython came along. He's also made some noises (as I understand it) that a rev2 Python may be in the works. I assume these two efforts would be related but I could be wrong. -=- Finally, on a coder side, there's been some discussion on this topic before. DC has reflected on the subject and pointed out some non-trivialities of porting to Java (mainly (?) C code). I imagine it's non-trivial, but getting something running might be reasonable. I don't know if there would be any licensing restrictions; I'm afraid (???) the ZODB has some funny license for use only with Zope. I don't know if a Java Zope would still be Zope or what any licensing snafus would do. Also, DC has never sounded enthusastic about Java though it's been mentioned in terms of "strategic direction". I don't really understand what the scoop on that is. That's all. Good post!-) = Joe "Just trying to understand Zope 101 before jumping off the deep end" Grace = "smoerk@gmx.de" wrote:
Did anyone try to port Zope to Java (using JPython)?
But why do I think it's nice to have a Java-Zope?
- using Java classes with Zope - running servlets and/or running Zope as servlet - performance could benefit from JIT compilers or compile Zope as native binary (gcj)
- portability? not sure if Java is more portable than Python ;)
Smoerk
_______________________________________________ Zope maillist - Zope@zope.org http://www.zope.org/mailman/listinfo/zope
(Related lists - please, no cross posts or HTML encoding!
To receive general Zope announcements, see: http://www.zope.org/mailman/listinfo/zope-announce
For developer-specific issues, zope-dev@zope.org - http://www.zope.org/mailman/listinfo/zope-dev )