| 1 | // Copyright (c) 2001-2010 Hartmut Kaiser |
| 2 | // |
| 3 | // Distributed under the Boost Software License, Version 1.0. (See accompanying |
| 4 | // file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) |
| 5 | |
| 6 | // The purpose of this example is to show, how it is possible to use a lexer |
| 7 | // token definition for two purposes: |
| 8 | // |
| 9 | // . To generate C++ code implementing a static lexical analyzer allowing |
| 10 | // to recognize all defined tokens |
| 11 | // . To integrate the generated C++ lexer into the /Spirit/ framework. |
| 12 | // |
| 13 | |
| 14 | // #define BOOST_SPIRIT_LEXERTL_DEBUG |
| 15 | #define BOOST_VARIANT_MINIMIZE_SIZE |
| 16 | |
| 17 | #include <boost/spirit/include/qi.hpp> |
| 18 | //[wc_static_include |
| 19 | #include <boost/spirit/include/lex_static_lexertl.hpp> |
| 20 | //] |
| 21 | #include <boost/phoenix/operator.hpp> |
| 22 | #include <boost/phoenix/statement.hpp> |
| 23 | #include <boost/phoenix/stl/container.hpp> |
| 24 | |
| 25 | #include <iostream> |
| 26 | #include <string> |
| 27 | |
| 28 | #include "../example.hpp" |
| 29 | #include "word_count_tokens.hpp" // token definition |
| 30 | #include "word_count_static.hpp" // generated tokenizer |
| 31 | |
| 32 | using namespace boost::spirit; |
| 33 | using namespace boost::spirit::ascii; |
| 34 | |
| 35 | /////////////////////////////////////////////////////////////////////////////// |
| 36 | // Grammar definition |
| 37 | /////////////////////////////////////////////////////////////////////////////// |
| 38 | //[wc_static_grammar |
| 39 | // This is an ordinary grammar definition following the rules defined by |
| 40 | // Spirit.Qi. There is nothing specific about it, except it gets the token |
| 41 | // definition class instance passed to the constructor to allow accessing the |
| 42 | // embedded token_def<> instances. |
| 43 | template <typename Iterator> |
| 44 | struct word_count_grammar : qi::grammar<Iterator> |
| 45 | { |
| 46 | template <typename TokenDef> |
| 47 | word_count_grammar(TokenDef const& tok) |
| 48 | : word_count_grammar::base_type(start) |
| 49 | , c(0), w(0), l(0) |
| 50 | { |
| 51 | using boost::phoenix::ref; |
| 52 | using boost::phoenix::size; |
| 53 | |
| 54 | // associate the defined tokens with the lexer, at the same time |
| 55 | // defining the actions to be executed |
| 56 | start = *( tok.word [ ++ref(w), ref(c) += size(_1) ] |
| 57 | | lit('\n') [ ++ref(l), ++ref(c) ] |
| 58 | | qi::token(IDANY) [ ++ref(c) ] |
| 59 | ) |
| 60 | ; |
| 61 | } |
| 62 | |
| 63 | std::size_t c, w, l; // counter for characters, words, and lines |
| 64 | qi::rule<Iterator> start; |
| 65 | }; |
| 66 | //] |
| 67 | |
| 68 | /////////////////////////////////////////////////////////////////////////////// |
| 69 | //[wc_static_main |
| 70 | int main(int argc, char* argv[]) |
| 71 | { |
| 72 | // Define the token type to be used: 'std::string' is available as the type |
| 73 | // of the token value. |
| 74 | typedef lex::lexertl::token< |
| 75 | char const*, boost::mpl::vector<std::string> |
| 76 | > token_type; |
| 77 | |
| 78 | // Define the lexer type to be used as the base class for our token |
| 79 | // definition. |
| 80 | // |
| 81 | // This is the only place where the code is different from an equivalent |
| 82 | // dynamic lexical analyzer. We use the `lexertl::static_lexer<>` instead of |
| 83 | // the `lexertl::lexer<>` as the base class for our token definition type. |
| 84 | // |
| 85 | // As we specified the suffix "wc" while generating the static tables we |
| 86 | // need to pass the type lexertl::static_::lexer_wc as the second template |
| 87 | // parameter below (see word_count_generate.cpp). |
| 88 | typedef lex::lexertl::static_lexer< |
| 89 | token_type, lex::lexertl::static_::lexer_wc |
| 90 | > lexer_type; |
| 91 | |
| 92 | // Define the iterator type exposed by the lexer. |
| 93 | typedef word_count_tokens<lexer_type>::iterator_type iterator_type; |
| 94 | |
| 95 | // Now we use the types defined above to create the lexer and grammar |
| 96 | // object instances needed to invoke the parsing process. |
| 97 | word_count_tokens<lexer_type> word_count; // Our lexer |
| 98 | word_count_grammar<iterator_type> g (word_count); // Our parser |
| 99 | |
| 100 | // Read in the file into memory. |
| 101 | std::string str (read_from_file(infile: 1 == argc ? "word_count.input" : argv[1])); |
| 102 | char const* first = str.c_str(); |
| 103 | char const* last = &first[str.size()]; |
| 104 | |
| 105 | // Parsing is done based on the token stream, not the character stream. |
| 106 | bool r = lex::tokenize_and_parse(first, last, lex: word_count, xpr: g); |
| 107 | |
| 108 | if (r) { // success |
| 109 | std::cout << "lines: " << g.l << ", words: " << g.w |
| 110 | << ", characters: " << g.c << "\n" ; |
| 111 | } |
| 112 | else { |
| 113 | std::string rest(first, last); |
| 114 | std::cerr << "Parsing failed\n" << "stopped at: \"" |
| 115 | << rest << "\"\n" ; |
| 116 | } |
| 117 | return 0; |
| 118 | } |
| 119 | //] |
| 120 | |