This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
cp/parse.y problem
- To: Thomas dot Koenig at ciw dot uni-karlsruhe dot de (Thomas König)
- Subject: cp/parse.y problem
- From: Mark Mitchell <mmitchell at usa dot net>
- Date: Wed, 1 Oct 1997 09:26:44 GMT
- Cc: egcs at cygnus dot com
- References: <199710011346.OAA18407@mvmap66.ciw.uni-karlsruhe.de>
- Reply-To: mmitchell at usa dot net
>>>>> "Thomas" == Thomas Koenig <ig25@mvmap66.ciw.uni-karlsruhe.de> writes:
Thomas> Even with the patch below applied, I still have the
Thomas> following problem:
Thomas> make[2]: Entering directory `/usr/src/egcs-970929/gcc/cp'
Thomas> [...]
Thomas> expect 18 shift/reduce conflicts and 39 reduce/reduce
Thomas> conflicts. cd .; bison -d -o parse.c parse.y parse.y
Thomas> contains 19 shift/reduce conflicts and 39 reduce/reduce
Thomas> conflicts.
Thomas> so I suspect something's rotten in the state of Denmark.
Yes, but it's just me. I neglected to change the expected output
numbers: my recent changes introduced one additional shift/reduce
conflict, but it's a harmless one.
Thomas> Wed Oct 1 15:37:58 CEST 1997 Thomas Koenig
Thomas> (ig25@rz.uni-karlsruhe.de)
Thomas> * cp/parse.y: Remove duplicate definition of PFUNCNAME
Thomas> --- cp/parse.y.orig Wed Oct 1 15:11:37 1997 +++ cp/parse.y
Thomas> Wed Oct 1 15:35:19 1997 @@ -196,7 +196,6 @@ %type <code>
Thomas> unop
Thomas> %type <ttype> identifier IDENTIFIER TYPENAME CONSTANT
Thomas> expr nonnull_exprlist -%type <ttype> PFUNCNAME %type
Thomas> <ttype> paren_expr_or_null nontrivial_exprlist SELFNAME
Thomas> %type <ttype> expr_no_commas cast_expr unary_expr primary
Thomas> string STRING %type <ttype> reserved_declspecs
Thomas> boolean.literal
Thomas> -- Thomas Koenig, Thomas.Koenig@ciw.uni-karlsruhe.de,
Thomas> ig25@dkauni2.bitnet. The joy of engineering is to find a
Thomas> straight line on a double logarithmic diagram.
--
Mark Mitchell mmitchell@usa.net
Stanford University http://www.stanford.edu