[FRPythoneers] Jython Sprint - this Saturday, August 4

Jim Baker jbaker at zyasoft.com
Wed Aug 1 01:00:20 MDT 2007

 Date/time: August 4, 2007 (Saturday), 9 AM - 5:30 PM
Food/beverages: provided! Just show up, we will keep you going and as
caffeinated as you need to be.
Location: bivio Software, Inc. <http://www.bivio.biz/>, 28th and Iris. Above
Hair Elite in Suite S. Please note, there are in fact two hair salons in
this shopping complex. Google Maps
Topics: IPython on Jython and Jython 2.5 compiler work. Or sprint on the
Python topic of your choice. For example, there will be a critical mass of
people in attendance for example who know the insides and outs of Django.
Participants: People planning to participate, either in person or via IRC on
#jython-dev; please go to the sprint
wiki<http://wiki.python.org/moin/BoulderSprint>to update your

   1. Brian Rosner
   2. Charlie Groves
   3. Chris McClimans
   4. Damien Lejeune
   5. Eric Dobbs
   6. Fernando Perez
   7. Frank Wierzbicki
   8. Ian Kelly
   9. Jim Baker
   10. Kip Lehman
   11. Matt Boersma
   12. Tobias Ivarsson

IPython on Jython: getting IPython to run on Jython 2.2. Fernando Perez will
be attending as a domain expert we can tap, although he plans to squash bugs
in the CPython version.

Jython's ability to explore the large and complex ecosystem that is Java is
one reason it's such a great tool. In particular, I like using Jython on
Jython, as well as to explore functionality we are in the process of adding.
It's much better to use an object shell than the alternatives (mostly
painful). But IPython is a much better object shell, as we know in using it
to explore Django and other projects. Hence the sprint.

Jython 2.5: The upcoming Jython 2.2 release represents a huge milestone for
the Jython community, but in the next step we want to catch up on 2.5 (or
even 2.6) features. The principal reason is to enlarge the community: as
long as there's such a lagging between CPython and Jython, it means that
supporting even core packages is shouldered by Jython implementers instead
of being shared across the Python community. In addition, certain
2.5functionality addresses portability concerns. Resource Allocation
Initialization (RAII), as enabled by the with-statement, makes it much
easier to do the right thing and have deterministic resource destruction
instead of relying on reference counting doing this implicitly, but then
creating an incompatibility with non-refcounting GC as is the case with
Java/Jython. 2.6's Class decorators and function annotations, in addition to
2.5's function decorators and 2.2's import hooking, make it possible for
Jython to consume and produce annotation metadata and type signatures. And
so on, it's a good thing we're doing.

Two of our Google Summer of Code students have confirmed their
participation, Damien Lejeune and Tobias Ivarsson. Damien has been working
on the new AST parser based on Antlr, and Tobias has been working on a new
bytecode compiler based on ASM. In this sprint, we would like to help them
out and especially help with their integration of the compilation pipeline.
In addition one of us (Jim Baker) is planning on finally getting to work on
support for Java annotation metadata.

For more information: http://wiki.python.org/moin/BoulderSprint

Jim Baker
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.community.tummy.com/pipermail/frpythoneers/attachments/20070801/6f041112/attachment.html>

More information about the FRPythoneers mailing list