[antlr-interest] Transformation phylosophy : AST trans./ pattern matching

Andy Tripp antlr at jazillian.com
Fri Oct 27 07:02:09 PDT 2006


Terence Parr wrote:

>
> On Oct 18, 2006, at 2:53 PM, Mathieu Clabaut wrote:
>
>> Hello all,
>>
>> I've read some of the interesting mails about the different way to do
>> transformation.
>>
>> Just for reference, a collegue of mine pointed me to the cocktail95
>> tool suite :
>>   http://www.iste.uni-stuttgart.de/ps/cocktail/
>> which seems to allow AST transformation but as also a Transformation
>> Tool based on Pattern Matching (puma).... No, I didn't give a try to
>> it, but I may have a deeper look in a near futur...
>
>
> Interesting; just looked at Puma.  Syntax is not obvious.  Seems to  
> allow tree grammar specification due to recursive references but  
> doesn't call it that.  Allows patterns and predicates to match.  Has  
> some interesting functionality similar to other tools.  My thoughts  
> would lean more towards what Andy did but with some twists.
>
I had looked into Cocktail and a bunch of similar tools when I was 
starting out. IIRC, Terence is being kind when he says
the syntax is "not obvious". "Not obvious" is almost an anagram for 
"obnoxios" or "oblivious" :)
I don't remember the details, but I would guess that my attitude was "I 
just can't work with this syntax."

Andy



More information about the antlr-interest mailing list