[antlr-interest] Re: alternative prediction mechanism for 3.0
parrt at cs.usfca.edu
Sun Jul 4 14:40:51 PDT 2004
On Jul 4, 2004, at 12:46 PM, antlrbrian wrote:
> --- In antlr-interest at yahoogroups.com, Terence Parr <parrt at c...> wrote:
>> Oh, right...yeah, I rejected that idea because you couldn't step thru
>> the DFAs that way. :( (goto in bytecode won't go back to source) :(
> That was #1 and #2 on my list:
> 1. The DFA textual representation will be easier to read than
> generated Java code.
> 2. Using the new Java debugging API's (JSR 45), the classfiles
> can be generated in such a way that debuggers can trace
> through the DFA's textual representation directly.
> In other words, when you are single-stepping through code in the
> debugger, and you get into the DFA part, the debugger would actually
> open up the files that contain the DFA "source code" (textual
> representation), and single-step through the actual DFA. Presumably,
> the DFA's textual representation would be much easier to read than
> generated Java source code.
> The ANTLR Tool would generate both the textual representation (for the
> debugger to reference) and the classfiles (that contain the references
> to the textual representation) at compile time.
Verrrrry interesting idea. I'll consider that approach. Currently I
have it working with generated DFA objects so I may leave for a while.
CS Professor & Grad Director, University of San Francisco
Creator, ANTLR Parser Generator, http://www.antlr.org
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:
<*> To unsubscribe from this group, send an email to:
antlr-interest-unsubscribe at yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
More information about the antlr-interest