From: Robert Feldt <feldt@...>
Date: 2001-02-25T17:13:28+09:00
Subject: [ruby-talk:11517] Re: RFC: RubyVM (long)

On Mon, 19 Feb 2001, Mathieu Bouchard wrote:

> > I'll tell you what I've done so far on this parser gen:
> > 
> > * Specified a grammar for grammars (its similar to SableCC's format, but
> > even simpler with Ruby RegExp'sfor the tokens)
> > * Hand-coded this grammar
> > * Implemented the LALR(1)-gen algs in Dragon book
> > * A really simple lexer gen that simply applies the token regexps in
> > order until a match is found (This may not give good enough performance
> > but we can do a more traditional DFA-based gen later...)
> >  * Generated a grammar parser from the handcoded grammar
> 
> Could you send me that?
> 
It's been a busy week but I'll try to pack it up and send it in the coming
days.

> > Next is extensive testing (Ok, in the name of pragmatism I wrote
> > some tests prior to implementation but sadly they dont cover
> > everything... ;-)
> 
> Don't forget peer review. Automated tests are not sufficient. They never
> cover everything. They are there to ensure that you've got a minimum of
> things working.
> 
Yeah, you're right but I don't even have tests to show the minimum is
working... ;-)

> > Next step is getting a Ruby parser by converting parse.y to the grammar
> > format of the parser gen. 
> 
> Oh. I thought you had started doing this as well. This would be a good
> thing to do soon. Whatever. What I recommend is that you publish what you
> have ASAP. =)
> 
Yeah, I'll pack it up and then start working on the Ruby parser.

> > > Actually, a strict subset of Ruby syntax, in such a way that the Ruby
> > > parser class can inherit from the Ruby-- parser class.
> > The Grammar class I've used supports this kind of "merging" of grammars
> > (and thus of parsers).
> 
> Yippee!!! nau zat iz kewl!
> 
> (sorry for this burst of enthusiasm)
> 
Wel,, don't expect too much; I simply add the productions of one grammar
to the ones of another. No checking that the merge is possible or that
they "connect" (through a nonterminal or whatever). Do you know a
"right"/better/more useful way to merge grammars?

> > Basically, Ruby-- would lack at least: > > > * for, case..when
> > > * flip-flop operator
> > > * reversed form of if, unless, while, until
> > > * multiple assignment
> > > * yield (use & declarator and .call() instead)
> 
> > Why these? I guess yield and the flip-flop may not be straightforward to
> > translate to C but IMHO the other ones should be possible?
> 
> It's possible that people reimplement all the features of Array#[],
> Array#[]= for their own class; it's possible, but it's not funny.
> 
> Look at the amount of code that was necessary to write BitArray (sample
> usage of MetaRuby). This supports most of Array's features.
> 
I agree, its both beautiful, cool and important. Keep up the good work!

Minor comment on ArrayMixin:
(I haven't checked it thoroughly so maybe you already do this) Wouldn't it
be better if put_many and get_many wasn't part of the Interface but was
defined in ArrayMixin in terms of put/get. Then if people want better
performance they can override these default version by defining their own
and faster ones?

 > I see no reason why the
language itself should not have a
restricted > version and a full version. The pattern is exactly the same. 
> 
I see your point and now agrees.

> > I think excluding yield would be a major hindrance since it is
> > frequently used in iterators.
> 
> Well it's not a hindrance at all for me. I just use the & declarator to
> pick up the special parameter, and then perform a .call on it where i
> could do yield.
> 
You're right; that should be sufficient.

> > I've translated gc.c (not all of it but doing a OO design
> > based on the existing stuff) so far.
> 
> This will be interesting to see.
> 
I'll post it. Aleksi mentioned maybe this should go into a cvs some
place. Whats the eval criteria for being allowed to sit on the ruby
cvs? How do you get "listed"?

> > memory. I think it would not be natural for a Ruby developer to use
> > something different than an iterator for this => yield is needed in
> > RootSet implementation => yield is needed in sRuby/Ruby--.
> 
> Ok, see above. "Iterators" are a library feature. "yield" is just
> sugar. As long as we have closures (and the & declarator) I don't think
> we've got a problem.
> 
I agree.

Regards,

Robert