Is Unladen Swallow dead?
drsalists at gmail.com
Wed Nov 17 00:40:50 CET 2010
I'm not a contributor to the U-S project, but I have been monitoring
alternative python implementations' progress some, and seem to be adding
something to pypy all of a sudden.
I think unladen swallow has produced performance improvements, and they are
likely to be merged into cpython 3.3.
However, the improvements are not quite as substantial as were hoped for,
and pypy seems to be getting some steam behind it. With pypy looking likely
to even be able to provide source-level compatibility with C extension
modules, the need for unladen swallow is perhaps somewhat lessened. At the
outset, the U-S people looked at pypy and got the impression it was years
from being a fast alternative to cpython for production code, but it sounds
like pypy's coming along nicely now.
Then again, the pypy people don't seem to feel that C extensions are going
to perform that well in pypy (it sounds like an issue of initialization
overhead and infeasibility of JIT compiling C extensions).
I've been testing some of my own code in cpython 2.6, cpython 3.1 and pypy
1.3; with this specific program pypy seems to be about 4x faster than
cpython 2.6, and almost that much faster than cpython 3.1 (I have the same -
albeit unfinished - code running on all 3). This code is I/O intensive and
On Tue, Nov 16, 2010 at 2:30 PM, laspi <lorena.aspiroz at gmail.com> wrote:
> There has been little or no activity at all in this project in the
> last months, and the last comments on their mailing list seem to
> conrfim that it's future is uncertain.
> It's also very strange the lack of updates, news or discussions,
> specially considering that the merging plan has been approved. Or it
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Python-list