Typing system vs. Java

Chris Barker chrishbarker at home.net
Tue Jul 31 20:12:32 CEST 2001

Courageous wrote:
> By and large this is nothing a good optimizing JIT can't handle
> without static typing information. QKS is getting a 10-100X improvement
> in Python execution speeds without modifying the Python language.
> http:\\www.smallscript.com. Execution times within 1-2X of C.

I have a hard time believing that this can be done without some sort of
type declarations, or homogenous sequences, or something. How do you get
around the fact that every time you see a:

z = x+y

x and y could be anything? 

I've seen the smallscript reference before, but the only mention of
Python I see on that page is a link back to python.org. Does anyone know
where we can find info about this miracualous JIT compiler?

> >For performances, sake another nice addition would be the introduction
> >of the "homogenous" sequence.

> Sure. Pretty easy to implement yourself, of course, saving the first-class
> language support. Although one can imagine something like:

Yes, and that's what NumPy is. However, without the first class language
support, it is pretty limited.

Steve Holden wrote:

> It's a good idea, but given that the compiler we use daily doesn't yet even
> do numerical constant folding, and adds two integers together every time it
> uses (2+3) there is clearly a long road to hoe before we see progress.

Good point, but if it was in the language, someone might make use of it.
If nothing else, you could use it in extension modules. Maybe it would
get used in Py2C. (is that project dead? it seems to be sleeping, but I
thought it was pretty promising)


Christopher Barker,
ChrisHBarker at home.net                 ---           ---           ---
http://members.home.net/barkerlohmann ---@@       -----@@       -----@@
                                   ------@@@     ------@@@     ------@@@
Oil Spill Modeling                ------   @    ------   @   ------   @
Water Resources Engineering       -------      ---------     --------    
Coastal and Fluvial Hydrodynamics --------------------------------------

More information about the Python-list mailing list