http://permalink.gmane.org/gmane.comp.parsers.spirit.general hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13896 -{--bjam-I-- -*- mode: compilation; default-directory: "~/prog_dev/boost-svn/ro/boost_1_37_0/Spirit2x/libs/spirit/example/qi/" -*- Compilation started at Wed Dec 3 22:02:21 bjam -d+2 mini_xml1 gcc.compile.c++ bin/gcc-4.2/debug/mini_xml1.o "/usr/bin/g++-4.2" -ftemplate-depth-300 -O0 -fno-inline -Wall -g -fPIC -I"../../../.." -I"../../../../.." -c -o "bin/gcc-4.2/debug/mini_xml1.o" "mini_xml1.cpp" -}--bjam-I-- So the -I option arguments faithfully reflect the Jamroot: -{--Jamroot-- use-project boost : ../../../../.. ; project : requirements <include>../../../.. <include>../../../../.. ; -}--Jamroot-- So, I'm stumped! I'll investigate further. Thanks for the help. -regards. ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the wo Larry Evans 2008-12-04T04:14:11 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13895 ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/_______________________________________________ Spirit-general mailing list Spirit-general< at >lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/spirit-general Carl Barron 2008-12-04T04:06:12 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13894 Yep, not a problem here: bjam gcc msvc mini_xml1 gcc.compile.c++ bin\gcc-4.3\debug\mini_xml1.o gcc.link bin\gcc-4.3\debug\mini_xml1.exe compile-c-c++ bin\msvc-9.0\debug\threading-multi\mini_xml1.obj mini_xml1.cpp using native typeof msvc.link bin\msvc-9.0\debug\threading-multi\mini_xml1.exe msvc.manifest bin\msvc-9.0\debug\threading-multi\mini_xml1.exe ...updated 4 targets... Best, Joel de Guzman 2008-12-04T03:36:19 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13893 ie, //========================================================================= int red=0, green=0, blue=0, alpha=0; rule_t junk = lexeme_d[*(anychar_p - eol_p) >> eol_p]; rule_t option = (str_p("colour") >> '=' >> int_p[assign(red)] >> int_p[assign(green)] >> int_p[assign(blue)] >> int_p[assign(alpha)]) | (str_p("alpha") >> '=' >> int_p[assign(settings.alpha_)]) | (str_p("name") >> '=' >> lexeme_d['"' >> (*(anychar_p - '"'))[assign(settings.name_)] >> '"']) | junk ; //------------------------------------------------------------------------- but what i want to know now is, how can i stop a rule that recognizes colour = 1 2 3 4 from recognizing colour = 1 2 3 4 5 ???? if i add terminal symbol, eg //=========================================== rule_t term = ch_p(';') >> eol_p; rule_t option = (str_p("colour") >> ... int_p[assign(alpha)] >> term) | ... //------------------------------------------- and make the file syntax colour = 1 2 3 4 ; then the grammar still a jono 2008-12-04T03:12:37 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13892 works fine here with g++ 4.3.2. the error seems wrong you shouldn't be getting the fusion::vector any more (instead the latest code uses fusion::vectorN). But I am not sure. There *might* be some problems with your include paths confusing spirit2 and spirit2x? Regards, Joel de Guzman 2008-12-04T03:32:39 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13891 [snip] Thanks Joel. I tried just mini_xml1, but: -{--bjam error-- -*- mode: compilation; default-directory: "~/prog_dev/boost-svn/ro/boost_1_37_0/Spirit2x/libs/spirit/example/qi/" -*- Compilation started at Wed Dec 3 21:11:05 bjam mini_xml1 ...patience... ...patience... ...found 1636 targets... ...updating 2 targets... gcc.compile.c++ bin/gcc-4.2/debug/mini_xml1.o ../../../../boost/spirit/home/qi/nonterminal/rule.hpp: In member function 'boost::spirit::qi::rule<Iterator, T1, T2, T3>& boost::spirit::qi::rule<Iterator, T1, T2, T3>::operator=(const Expr&) [with Expr = boost::proto::exprns_::expr<boost::proto::tag::shift_right, boost::proto::argsns_::list2<const boost::proto::exprns_::expr<boost::proto::tag::shift_right, boost::proto::argsns_::list2<const boost::proto::exprns_::expr<boost::proto::tag::shift_right, boost::proto::argsns_::list2<boost::proto::exprns_::expr<boost::proto::tag::terminal, boost::proto::argsns_::term<const char&>, 0l>, const boost::proto::exprns_::expr<boost::proto::tag: Larry Evans 2008-12-04T03:19:35 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13890 Alright, I cleaned up some including the mini-xml samples. I suggest you bjam only the parts that you want to compile. Not all examples are clean yet. Regards, Joel de Guzman 2008-12-04T02:44:31 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13889 I don't think so. +(char_) will gladly eat anything in its path. You might want to constrain that further. Remember Qi is PEG. Regards, Joel de Guzman 2008-12-04T02:41:43 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13888 Larry and Joel - thank you both for the quick reply. Now I see the error, thanks. With the rule start %= lexeme[+(char_)] >> ruleFrequency; Am I correct in the expectation that with the above rule I should be able to parse the input string up to the first empty character and then the rest of the string will be handled by "ruleFrequency" rule? -----Original Message----- From: Joel de Guzman <joel< at >boost-consulting.com> To: spirit-general< at >lists.sourceforge.net Date: Thu, 04 Dec 2008 10:02:21 +0800 Subject: Re: [Spirit-general] Rule compilation error ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ QPlace 2008-12-04T02:13:47 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13887 ------------------------^^ Isn't that a bogus >> ? There is no LHS. How about: start %= lexeme[+(char_)] >> ruleFrequency; Regards, Joel de Guzman 2008-12-04T02:02:21 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13886 >> is a binary operator. Without the eps, there's no left arg to 1st >>. ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ Larry Evans 2008-12-04T02:00:27 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13885 Grammar below does not compile, complaining about "syntax error : '>>'" under VC2008. It only compiles when eps is added as in: start %= eps >> lexeme[+(char_)] >> ruleFrequency; Can someone help me understand why? template <typename Iterator> struct command_parser : grammar<Iterator, commandselect(), space_type> { command_parser() : command_parser::base_type(start) { ruleFrequency %= eps >> int_ >> frequencies; start %= >> lexeme[+(char_)] >> ruleFrequency; } rule<Iterator, frequency(), space_type> ruleFrequency; rule<Iterator, commandselect(), space_type> start; }; ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ QuitePlace 2008-12-04T01:44:01 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13884 If you want the dash, you can make it the last. Regards, Joel de Guzman 2008-12-04T00:17:18 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13883 Yep. Only the ones that Karma uses. Regards, Joel de Guzman 2008-12-04T00:09:52 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13882 [snip] What if I want to accept either 'a' or '-' or 'c'. Judging from the */qi/char.cpp test driver, char_("a-c") accepts 'a' or 'b' or 'c'. OTOH, the small chr_set.cpp test driver included in my previous post indicates: char_("<(") accepts either '<' or '('. Is - treated specially so that to allow accepting '-' it must be somehow escaped, as in, for example, char_("\-")? ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ Larry Evans 2008-12-03T23:36:52 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13881 Great! I highly appreciate this as I always felt like sprinkling the "using" directive over my code until it worked... Matthias Matthias Vallentin 2008-12-03T17:39:21 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13880 These components also exist in karma, right? So will you hoist them (or a subset) into karma in a similar fashion? Matthias Matthias Vallentin 2008-12-03T17:36:31 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13879 Joel de Guzman-2 wrote: I'll take a look at those. Joel de Guzman-2 wrote: I agree! For what it's worth, I too would rather not rely on ICU, but I also agree it would be expedient to start with it (particularly as boost::regex supports it too). Regards, Ben Ben.Hanson 2008-12-03T17:15:41 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13878 The new name is "labels". You can bring all the labels in using the using directive without worrying about bringing in too much. As consistent with 2 below, this namespace is also available in the qi namespace. Thus: using namespace boost::spirit::labels; is mirrored in using namespace boost::spirit::qi::labels; Again, holler if you see something you dislike. Best, Joel de Guzman 2008-12-03T16:47:57 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13877 Oh definitely! I was thinking the same, when I got to it. Anyway, we should at least have level-1 unicode support for the lexer: http://www.unicode.org/reports/tr18/ This might help: http://www.regular-expressions.info/unicode.html Let us coordinate our efforts. We need a common core library support. My thinking is to make it not reliant on one single library (like ICU) so we can plug in something better in the future. At any rate, it would be practical to start with ICU. Regards, Joel de Guzman 2008-12-03T16:30:56 http://permalink.gmane.org/gmane.comp.parsers.spirit.general/13876 You might be wondering why I need some time to port the examples. Well, it's not just porting. I took it to task to make the examples real exemplars. For example, I am taking out all using namespaces, avoid putting things in global scope, etc. Regards, Joel de Guzman 2008-12-03T16:23:56 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.parsers.spirit.general