Another update:
If I quit playmode on the right side of the rule, the error disappears (it
gives warnings of an unused rule instead), code:
init: INIT side NUMBER { driver.h_init_finished( $2, $3,
(Playmode)0 ); }
;
side: LEFT { $$ = $1; }
| RIGHT { $$ = $1; }
;
playmode: BEFORE_KICK_OFF { $$ = $1; }
| TIME_OVER { $$ = $1; }
| PLAY_ON { $$ = $1; }
;
It only gives warnings, but no compilation errors and creates the
executable:
rcss_parser.y: warning: 1 nonterminal useless in grammar [-Wother]
rcss_parser.y: warning: 3 rules useless in grammar [-Wother]
rcss_parser.y:284.25-32: warning: nonterminal useless in grammar: playmode
[-Wother]
%type < Playmode > playmode
^^^^^^^^
rcss_parser.y:333.17-44: warning: rule useless in grammar [-Wother]
playmode: BEFORE_KICK_OFF { $$ = $1; }
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
rcss_parser.y:334.25-60: warning: rule useless in grammar [-Wother]
| TIME_OVER { $$ = $1; }
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
rcss_parser.y:335.25-52: warning: rule useless in grammar [-Wother]
| PLAY_ON { $$ = $1; }
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
On Thu, Dec 19, 2013 at 12:15 PM, Gilberto Monroy <gilbertomonr...@gmail.com
> wrote:
> Akim:
> 1 Yeah, you are right, I removed the .h and .hh files from the compilation
> line, but the result is the same.
> 2 Actually, the .hh file is included in rcss_driver.h, wich is included in
> rcss_parser.y. John is right, there are inclusion guardas in the .hh file,
> that is not the problem. I changed the code a bit in a way I do not
> (directly or indirectly) include the .hh file in the .y file, but the
> result is similar:
>
> g++ -o parser rcss_parser.tab.cc lex.yy.c rcss_parser_driver.cpp -lfl
> In file included from rcss_parser.tab.cc:53:0:
>
> rcss_parser.tab.hh:583:3: error:
> ‘yy::RcssParser::basic_symbol<Base>::basic_symbol(typename Base::kind_type,
> Playmode, const location_type&)’ cannot be overloaded
> rcss_parser.tab.hh:579:3: error: with
> ‘yy::RcssParser::basic_symbol<Base>::basic_symbol(typename Base::kind_type,
> Playmode, const location_type&)’
> rcss_parser.tab.hh:1896:3: error: redefinition of
> ‘yy::RcssParser::basic_symbol<Base>::basic_symbol(typename Base::kind_type,
> Playmode, const location_type&)’
> rcss_parser.tab.hh:1882:3: error:
> ‘yy::RcssParser::basic_symbol<Base>::basic_symbol(typename Base::kind_type,
> Playmode, const location_type&)’ previously declared here
> In file included from rcss_lexer.l:28:0:
> rcss_parser.tab.hh:583:3: error:
> ‘yy::RcssParser::basic_symbol<Base>::basic_symbol(typename Base::kind_type,
> Playmode, const location_type&)’ cannot be overloaded
> rcss_parser.tab.hh:579:3: error: with
> ‘yy::RcssParser::basic_symbol<Base>::basic_symbol(typename Base::kind_type,
> Playmode, const location_type&)’
> rcss_parser.tab.hh:1896:3: error: redefinition of
> ‘yy::RcssParser::basic_symbol<Base>::basic_symbol(typename Base::kind_type,
> Playmode, const location_type&)’
> rcss_parser.tab.hh:1882:3: error:
> ‘yy::RcssParser::basic_symbol<Base>::basic_symbol(typename Base::kind_type,
> Playmode, const location_type&)’ previously declared here
> make: *** [parser] Error 1
>
>
>
>
>
> On Thu, Dec 19, 2013 at 10:07 AM, John Horigan <j...@glyphic.com> wrote:
>
>> But the generated header has inclusion guards. He should be able to
>> include it multiple times, even if it is a bad idea. There must be another
>> problem.
>>
>> -- john
>>
>>
>> On Thu, Dec 19, 2013 at 3:09 AM, Akim Demaille <a...@lrde.epita.fr>wrote:
>>
>>>
>>> Le 19 déc. 2013 à 03:15, Gilberto Monroy <gilbertomonr...@gmail.com> a
>>> écrit :
>>>
>>> > Hi everybody.
>>> >
>>> > My name is Gilberto, I use the next tools:
>>>
>>> Hi Gilberto,
>>>
>>> > g++ -o parser location.hh position.hh rcss_parser.tab.cc lex.yy.c
>>> rcss_parser_driver.cpp -lfl
>>>
>>> This compilation line is fishy. You are not expected to compile
>>> the headers (*.hh files). Compile only *.cc and *.ccp files.
>>>
>>> > from rcss_parser.y:15:
>>> > rcss_parser.tab.hh:583:3: error:
>>>
>>> If I read this correctly, in your *.y file you #include the generated
>>> parser header. Don't do that, Bison does for you. This would explain
>>> why the symbols are defined twice.
>>>
>>>
>>>
>>
>
