Q: Python 2.0 preliminary features?

[Michael Hudson]

On Mon, 18 Oct 1999, Jeremy Hylton wrote:

> >>>>> "MH" == Michael Hudson <mwh21 at cam.ac.uk> writes:
> 
>   MH> PS: Given the bizarre reluctance of my university to give me any
>   MH> work, can I ask if the source to viperi (at least) will be
>   MH> perusable anytime soon?
> 
> Suggestion for an interesting project if you don't have anything else
> to do:  A Python compiler written in Python.  For starters, it could
> just generate Python bytecode.  But it would seem much easier to do
> analysis of Python code if we didn't have to muck with any C code to
> do it.

Uhh.. two problems. 

1) I'm assuming I will have something to do, pretty soon.
2) I really don't get on with parsing & high level semantic analysis. I've
never poured much effort into it, but I certainly don't have any natural
flair for it.

Something I had been meaning to play with was extend the Python core to
incorporate a register machine as well as the stack machine we have at the
moment. This also strays far from my experience, but sounds more appealing
to me.

I was thinking of working from the existing bytecode and recompiling it to
my new vm. Not sure if this is wise.

But your idea is definitely a nice one. I think I'm more suited to
backends, me.

speculative-ly y'rs Michael