[antlr-interest] Why no links to ANTLR 3.0 on www.antlr.org??

[Nigel Sheridan-Smith]

> -----Original Message-----
> From: antlr-interest-bounces at antlr.org [mailto:antlr-interest-
> bounces at antlr.org] On Behalf Of Terence Parr
> Sent: Saturday, 11 June 2005 4:15 AM
> To: ANTLR Interest
> Subject: Re: [antlr-interest] Why no links to ANTLR 3.0 on www.antlr.org??
> 
> 
> On Jun 9, 2005, at 4:25 PM, Nigel Sheridan-Smith wrote:
> > Is this the course notes, Ter?
> >
> > http://www.cs.usfca.edu/~parrt/course/652/
> >
> > I don't know that it *is* actually directly linked from the site.
> > Only the
> > following (as a "tutorial" for beginners):
> >
> > http://www.cs.usfca.edu/~parrt/course/652/lectures/antlr.html
> 
> Well, it's the first thing in the list of tutorials in the getting
> started doc.  Should I move to the front page?  perhaps it's more
> useful than the glossary?
> 

No not exactly... 

My suggestion is to have both links - keep the existing one (as the
"tutorial") and add another that refers to the whole set of course notes.
Add a title to this one to say "General theory of compilers and interpreters
(Ter's course notes)" or something like that and make it distinctly
separate. It can be on the same "getting started" page. 

It just wasn't obvious that the "Intro to ANTLR" was part of a broader set
of course notes, with more examples and information about the principles of
compilers/interpreters.

Personally, I agree (to some degree) with the other remarks made. The
documents and web links that are available tend to have a big chasm leap
from the very simple (walkthrough tutorials) to the very difficult (the
reference manual). Not that it prevents people learning if they sit down and
try it out, but I would like to see something a bit more in the middle.

If I ever finish this PhD (this year hopefully!), then I will sit down and
try to build a few more examples of AST usage, for others to use and learn
from. I still personally opt to build my own heterogenous ASTs using the
parser (not tree parser), but that might change with ANTLR 3.0's dynamic
attributes. 

In particular I would like to see more examples of how to use ANTLR tree
walking to carry out particular things (interpreters and execution,
byte-code/native compilers, JIT compilers, pre- and post- code
optimisations, ?) so that we can build a bit of a knowledge-base about
common ways of dealing with these issues. Igor's little C-to-asm compiler
looks a good place to start, or perhaps someone should finish off Scott
Stanchfield's tutorial (for his "complete compiler"). I'm thinking about
starting a Wiki for this sort of stuff, but it's on the backburner for the
time being. 

Nigel

--
Nigel Sheridan-Smith
PhD research student

Faculty of Engineering
University of Technology, Sydney
Phone: 02 9514 7946
Fax: 02 9514 2435