We just discussed timelines on the mailing list: http://sourceforge.net/p/jython/mailman/message/32116603/Next I will try to answer your specific
On Thu, Mar 20, 2014 at 8:07 AM, Matthias F. Brandstetter <email@example.com> wrote:
1.) Are there any plans or is there a roadmap for when you estimate to release a stable version of Jython 2.7? I understand if you can't give any exact dates, but is it more like "about 1 month", "somewhere in 2014", or anything else?However, as I can see the last beta release of Jython 2.7 has been from over a year now, and I can't find any recent news on this. So I have two questions:Hello all,I am both a Python and Java developer, so am *really* thankful for your idea of putting both guys into one boat. That's why I am using Jython for several smaller projects now.
In 2014 - I cannot see why this will not be the case. The next beta really should be in the next few weeks at most. After that we should have at least one more beta, plus release candidates. As I mentioned in the quote thread above, fully supporting socket/select/ssl has really delayed things for us. But it's almost there. As of yesterday, I was getting 83%+ of the tests in test_socket running - and these really do test corner cases. (Yes, it is possible to use UDP in a connected fashion through the socket API...) So I'm still on target for 100% by end of this week.2.) Until then, how "save" is it to use the last beta version of Jython 2.7, can I use it for production systems? And is there also a documentation section for 2.7 beta? I have only found docs for 2.5 so far.
You may find Jython 2.7b1 suitable for your usage now. Perhaps you don't need pip. But we consider support of pip and the overall Python ecosystem what makes Jython quite useful: it's compatible with regular Python (factoring out any nonported libraries based on C extensions, at least before we have JyNI) and its tools, but you get to run it on the Java environment.One thing you can do to figure out your decision is to look at the commits:http://hg.python.org/jython; there has been a fair amount of work since "Prepare for 2.7b1" 13 months ago.
Given that nothing is committed against trunk that has not improved Jython and has corresponding unit tests to back up that claim, you may find using trunk makes sense. This is especially the case if you are interested exploring the suitability of Jython for a larger project.That's the case for me. I want a stable release for any production usage (unit tests are great but not perfect), but I know 2.7.0 is going to be extraordinarily close to what it is now. That means it's reasonable to develop against Jython trunk now.