[antlr-interest] Need help in understanding DFA prediction

[Sven Prevrhal]

Sam,

 It does - 

 

fragment LETTER              :               'a'..'z' | 'A'..'Z';

 

Perhaps the 'fragment' is a problem?

Sven

 

From: Sam Harwell [mailto:sharwell at pixelminegames.com] 
Sent: Sunday, November 09, 2008 10:52 PM
To: sven.prevrhal at ucsf.edu; antlr-interest at antlr.org
Subject: RE: [antlr-interest] Need help in understanding DFA prediction

 

What's the definition for LETTER? If, for example, you have this:

 

LETTER : 'a'..'z' ;

 

Then Cake matches CATEGORYSTART because the capital letter C at the
beginning rules out all options but that. Does your LETTER rule allow
capital letters?

 

Sam

 

From: antlr-interest-bounces at antlr.org
[mailto:antlr-interest-bounces at antlr.org] On Behalf Of Sven Prevrhal
Sent: Monday, November 10, 2008 12:08 AM
To: antlr-interest at antlr.org
Subject: [antlr-interest] Need help in understanding DFA prediction

 

Hello and thanks Gurus:

 

I have a Lexer rule

 

CATEGORYSTART

                :               SPACE* 'Categories:';

 

and the call to DFA.Predict which originates from Lexer.mTokens() predicts
it for input like "Cake", and then of course the call to mCATEGORYSTART() in
the switch statement inside mTokens()after the DFA.Predict call returns
issues an error.

 

Sooo, how do I get DFA.Predict to predict the WORD rule for "Cake" instead
of CATEGORYSTART?

 

Here's my WORD rule:

WORD :                LETTER+;

 

Thank you and cheers,

Sven

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.antlr.org/pipermail/antlr-interest/attachments/20081109/d35a7f16/attachment.html