[antlr-interest] Re: unicode strings using supplemental char range

[lgcraymer]

Ter--

I have to say that I am not a fan of the "cave man" approach.  ANTLR
2.x is available for backwards compatibility; there is no point in not
requiring 1.4 or maybe even 1.5 and figure out a clean way to
implement the ANTLR runtime with generics.  I've discovered that
unnecessarily reinventing the wheel is usually a waste of effort, and
focusing on backwards compatibility guarantees that you do a lot of
wheel reinvention.  Make it easy on yourself!

--Loring


--- In antlr-interest at yahoogroups.com, Terence Parr <parrt at c...> wrote:
> 
> On Jun 26, 2004, at 2:38 AM, Ric Klaren wrote:
> > For unicode and C++ I'll definitely be looking at ICU (actually have 
> > been
> > looking at it already) Come to think of it hearing the latest 
> > discussions
> > it might definitely be an idea for the java side too. At least they 
> > already
> > figured out the icky bits for several java versions, so maybe it's an 
> > idea
> > to not reinvent stuff (although I hate the extra dependency)
> 
> Yeah, dependency is bad.  Perhaps templates will help as you say.
> 
> > For C++ we'll probably see some templates so ICU or std:string can be
> > plugged in, and/or maybe some lightweight own thing that can deal
with 
> > some
> > simpler uses. (like just UTF8/16)
> >
> > For ANTLR3 I'll probably also try to drop support for some older 
> > compilers.
> > E.g. ones not supporting template members and namespaces.
> 
> Good idea.  Should make stuff cleaner.
> 
> Ric, how will going to IntString from String in ANTLR affect you do you 
> think in terms of code generation?  I want to have the min Java version 
> to be low like 1.1 optimally, but realistically it will be 1.3 to get 
> collections.  That implies though that all this 1.5 UNICODE "fixing" 
> won't be available to ANTLR itself nor the output parsers.  That 
> implies that none of the supplemental codes will work for isUpperCase 
> or isDigit etc...  I'm really at a loss to figure out how to proceed 
> here.  I really appreciate all the feedback from people; eventually 
> we'll figure this out.  Ugh.
> 
> Ter
> --
> CS Professor & Grad Director, University of San Francisco
> Creator, ANTLR Parser Generator, http://www.antlr.org
> Cofounder, http://www.jguru.com
> Cofounder, http://www.knowspam.net enjoy email again!
> Cofounder, http://www.peerscope.com pure link sharing



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
     http://groups.yahoo.com/group/antlr-interest/

<*> To unsubscribe from this group, send an email to:
     antlr-interest-unsubscribe at yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
     http://docs.yahoo.com/info/terms/