|YACC with infinite lookahead firstname.lastname@example.org (Daniele Benegiamo) (1999-05-16)|
|Re: YACC with infinite lookahead email@example.com (Bob Buckley) (1999-05-20)|
|Re: YACC with infinite lookahead firstname.lastname@example.org (Chris F Clark) (1999-05-20)|
|Re:YACC with infinite lookahead email@example.com (Salvador V. Cavadini) (1999-05-20)|
|Re: YACC with infinite lookahead firstname.lastname@example.org (Akim Demaille) (1999-05-21)|
|Re: YACC with infinite lookahead email@example.com (Vladimir Makarov) (1999-05-21)|
|Re: YACC with infinite lookahead firstname.lastname@example.org.OZ.AU (1999-05-27)|
|From:||email@example.com.OZ.AU (Andrew Bromage)|
|Date:||27 May 1999 23:21:23 -0400|
|Organization:||Computer Science, The University of Melbourne|
|Keywords:||parse, yacc, C++|
Akim Demaille <firstname.lastname@example.org> writes:
>What exactly are the problems given to LALR(1) by C++?
The biggie is the famous rule of C++ grammar which goes something like
If it looks like a declaration, then it is.
Otherwise, if it looks like a statement, then it is.
Otherwise, it's a syntax error.
Naturally, such a rule would only be needed if the rules for
declarations and statements are so ambiguously overlapped that the
usual arsenal of lexical feedback hacks won't work.
Return to the
Search the comp.compilers archives again.