[antlr-interest] Any syntax predicate can be removed.

mzukowski at yci.com mzukowski at yci.com
Thu Aug 29 12:55:05 PDT 2002


No, you need syntactic predicates if you need to make a decision that has to
look past the k tokens buffered.

Monty

> -----Original Message-----
> From: Ruslan Zasukhin [mailto:sunshine at public.kherson.ua]
> Sent: Thursday, August 29, 2002 12:46 PM
> To: antlr-interest at yahoogroups.com
> Cc: lgcraymer
> Subject: [antlr-interest] Any syntax predicate can be removed.
> 
> 
> Hi Lorin,
> 
> Once you have write on list
> 
> > If you are writing a new grammar, then "predicates are your 
> friend" is
> > good practice for getting past the "Omigod--look at all these
> > warnings" stage.  Wherever you have an ambiguity warning that you
> > cannot figure out quickly, insert a (syntactic) predicate to
> > disambiguate alternatives.  Also, consider inserting semantic
> > predicates as "assertions".  Once you get a warning-free antlr pass,
> > verify that the generated language processor properly 
> recognizes your
> > test input (catch exceptions in the debugger at the point where they
> > are thrown).  You may want to have a test version of your language
> > processor which produces some sort of text output that can be fed to
> > "diff" for regression testing.  Once you get the translator working,
> > then go back and refactor on a predicate-by-predicate basis--it's a
> > lot easier to clean up ambiguities when you know where to look, and
> > having a test suite to run makes it easier to verify 
> correct operation
> > after each change.
> 
> So I want to clarify.
> 
> If ANY syntax predicate can be removed from a grammar ?
> 
> Or sometimes we must shut up, and use them ?
> 
> -- 
> Best regards,
> Ruslan Zasukhin      [ I feel the need...the need for speed ]
> -------------------------------------------------------------
> e-mail: ruslan at paradigmasoft.com
> web: http://www.paradigmasoft.com
> 
> To subscribe to the Valentina mail list
> send a letter to valentina-on at lists.macserve.net
> -------------------------------------------------------------
> 
> 
>  
> 
> Your use of Yahoo! Groups is subject to 
http://docs.yahoo.com/info/terms/ 


 

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



More information about the antlr-interest mailing list