[antlr-interest] error recovery

[dav1dr at eml.cc]

Hi,

it's my understanding that antlr will skip tokens in the input / insert
tokens if it comes across
input that it can't parse in an effort to end up with a string that is
parsable. How can I turn this
off? My parser is recognising strings that are not in my language, and I
don't want it to.

(specifically, I'm writing a C-like expression parser, which currently
accepts a string like
"(1 2)" by skipping the 2; this should be an error.

thanks

Dave
-- 
  
  dave_rodgman at fastmail.co.uk