[Pyrex] Pyrex at Pycon

[Greg Ewing]

> I gave two talks on Pyrex at Pycon. They were both well received. There 
> are many enthusiastic Pyrex users out there and also many people who 
> wanted to learn about it because they plan to use it in the future.

Hi, nice to hear that people are finding Pyrex interesting.

> * C++ is really important to many people. Without C++ Pyrex is a 
>   non-starter

I'm aware of that, and I plan to address it at some point.  Some other
things need to be tidied up first, though.

> * Performance is as much an attraction for people as bridging (maybe 
>   more!) but Pyrex is much better at bridging than at performance.

The way you get performance with Pyrex is *through* bridging.  Pyrex
isn't meant to speed up pure Python code. There are some things I plan
to do to improve it in that area -- I think I can get it so it isn't
ever slower, and might be slightly faster in some cases.

But compiling pure Python with Pyrex is never going to give
order-of-magnitude speedups. Doing that would require something like
type inferencing, and I'm not going there.

Perhaps I ought to comment about this somewhere early on in the docs,
so people don't get unrealistic expectations.

> * The lack of a roadmap makes it harder for people to commit to 
>   learning or developing Pyrex.

People need to keep in mind that Pyrex is still an experiment.  I
can't tell people where it's going, because I don't know myself yet! 
I'm still finding out what works and what doesn't.

Still, I could perhaps include some comments on the web page about
what I'm planning to do and roughly in what order.

> They ask me: "how soon do you think Pyrex will have autogeneration from 
> headers?"

It depends on what they mean by "autogeneration". It probably won't
ever be possible to fully automate translation of C declarations to
Pyrex declarations, but some degree of automation is probably
possible. I know of a couple of people who were experimenting in this
area, but haven't heard anything from them recently about it.
I'll probably tackle the problem myself at some point, but at the
moment it's something for the far future.

Greg Ewing, Computer Science Dept, +--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
greg at cosc.canterbury.ac.nz	   +--------------------------------------+