[antlr-interest] Grammar problem with ~

fridi fridi70 at gmx.de
Tue Jun 26 14:22:15 PDT 2012


Hi all,

I have a grammar that I have to extend in order to meet changes in the 
files we have to parse.
There is a code (which can be an int or a letter followed by int ) that 
is followed by a description.

i.e.

     3007508   This is the description of the code
     3024400   A/B bla bla bla & bla
     A1234       This is another description
     B4711        asdf ghjz-kk with Test w/o

I am interested in the code only and I tried something like

in parser
(
     INT REST
)+

in lexer

REST
   :
   ~('\n' | '\r')
   ;

which works  for some instances with the code as int, but not for a code 
that starts with a letter because my rule for REST now consumes other 
tokens.
I have used the "all except NL" approach for some other comments to read 
until NL, but there was always a constant string at the beginning of the 
line so that the rule didn't consume any other part of my text. I think 
that REST is to common and it might kill my other rules.

Is there any other way to parse the descriptions?

Any help is very appreciated,

   fridi




More information about the antlr-interest mailing list