1 | // Copyright (c) 2001-2011 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 | #include <boost/spirit/include/lex_lexertl.hpp> |
7 | #include <boost/spirit/include/lex_lexertl_position_token.hpp> |
8 | |
9 | #include <boost/core/lightweight_test.hpp> |
10 | #include <boost/phoenix/object.hpp> |
11 | #include <boost/phoenix/operator.hpp> |
12 | #include <boost/phoenix/stl/container.hpp> |
13 | |
14 | namespace lex = boost::spirit::lex; |
15 | namespace phoenix = boost::phoenix; |
16 | namespace mpl = boost::mpl; |
17 | |
18 | /////////////////////////////////////////////////////////////////////////////// |
19 | enum tokenids |
20 | { |
21 | ID_INT = 1000, |
22 | ID_DOUBLE |
23 | }; |
24 | |
25 | template <typename Lexer> |
26 | struct token_definitions : lex::lexer<Lexer> |
27 | { |
28 | token_definitions() |
29 | { |
30 | this->self.add_pattern("HEXDIGIT" , "[0-9a-fA-F]" ); |
31 | this->self.add_pattern("OCTALDIGIT" , "[0-7]" ); |
32 | this->self.add_pattern("DIGIT" , "[0-9]" ); |
33 | |
34 | this->self.add_pattern("OPTSIGN" , "[-+]?" ); |
35 | this->self.add_pattern("EXPSTART" , "[eE][-+]" ); |
36 | this->self.add_pattern("EXPONENT" , "[eE]{OPTSIGN}{DIGIT}+" ); |
37 | |
38 | // define tokens and associate them with the lexer |
39 | int_ = "(0x|0X){HEXDIGIT}+|0{OCTALDIGIT}*|{OPTSIGN}[1-9]{DIGIT}*" ; |
40 | int_.id(id: ID_INT); |
41 | |
42 | double_ = "{OPTSIGN}({DIGIT}*\\.{DIGIT}+|{DIGIT}+\\.){EXPONENT}?|{DIGIT}+{EXPONENT}" ; |
43 | double_.id(id: ID_DOUBLE); |
44 | |
45 | whitespace = "[ \t\n]+" ; |
46 | |
47 | this->self = |
48 | double_ |
49 | | int_ |
50 | | whitespace[ lex::_pass = lex::pass_flags::pass_ignore ] |
51 | ; |
52 | } |
53 | |
54 | lex::token_def<lex::omit> int_; |
55 | lex::token_def<lex::omit> double_; |
56 | lex::token_def<lex::omit> whitespace; |
57 | }; |
58 | |
59 | template <typename Lexer> |
60 | struct token_definitions_with_state : lex::lexer<Lexer> |
61 | { |
62 | token_definitions_with_state() |
63 | { |
64 | this->self.add_pattern("HEXDIGIT" , "[0-9a-fA-F]" ); |
65 | this->self.add_pattern("OCTALDIGIT" , "[0-7]" ); |
66 | this->self.add_pattern("DIGIT" , "[0-9]" ); |
67 | |
68 | this->self.add_pattern("OPTSIGN" , "[-+]?" ); |
69 | this->self.add_pattern("EXPSTART" , "[eE][-+]" ); |
70 | this->self.add_pattern("EXPONENT" , "[eE]{OPTSIGN}{DIGIT}+" ); |
71 | |
72 | this->self.add_state(); |
73 | this->self.add_state("INT" ); |
74 | this->self.add_state("DOUBLE" ); |
75 | |
76 | // define tokens and associate them with the lexer |
77 | int_ = "(0x|0X){HEXDIGIT}+|0{OCTALDIGIT}*|{OPTSIGN}[1-9]{DIGIT}*" ; |
78 | int_.id(id: ID_INT); |
79 | |
80 | double_ = "{OPTSIGN}({DIGIT}*\\.{DIGIT}+|{DIGIT}+\\.){EXPONENT}?|{DIGIT}+{EXPONENT}" ; |
81 | double_.id(id: ID_DOUBLE); |
82 | |
83 | whitespace = "[ \t\n]+" ; |
84 | |
85 | this->self("*" ) = |
86 | double_ [ lex::_state = "DOUBLE" ] |
87 | | int_ [ lex::_state = "INT" ] |
88 | | whitespace[ lex::_pass = lex::pass_flags::pass_ignore ] |
89 | ; |
90 | } |
91 | |
92 | lex::token_def<lex::omit> int_; |
93 | lex::token_def<lex::omit> double_; |
94 | lex::token_def<lex::omit> whitespace; |
95 | }; |
96 | |
97 | /////////////////////////////////////////////////////////////////////////////// |
98 | template <typename Token> |
99 | inline bool |
100 | test_token_ids(int const* ids, std::vector<Token> const& tokens) |
101 | { |
102 | for (std::size_t i = 0, len = tokens.size(); i < len; ++i) |
103 | { |
104 | if (*ids == -1) |
105 | return false; // reached end of expected data |
106 | |
107 | if (tokens[i].id() != static_cast<std::size_t>(*ids)) // token id must match |
108 | return false; |
109 | ++ids; |
110 | } |
111 | |
112 | return (*ids == -1) ? true : false; |
113 | } |
114 | |
115 | template <typename Token> |
116 | inline bool |
117 | test_token_states(std::size_t const* states, std::vector<Token> const& tokens) |
118 | { |
119 | for (std::size_t i = 0, len = tokens.size(); i < len; ++i) |
120 | { |
121 | if (*states == std::size_t(-1)) |
122 | return false; // reached end of expected data |
123 | |
124 | if (tokens[i].state() != *states) // token state must match |
125 | return false; |
126 | ++states; |
127 | } |
128 | |
129 | return (*states == std::size_t(-1)) ? true : false; |
130 | } |
131 | |
132 | /////////////////////////////////////////////////////////////////////////////// |
133 | struct position_type |
134 | { |
135 | std::size_t begin, end; |
136 | }; |
137 | |
138 | template <typename Iterator, typename Token> |
139 | inline bool |
140 | test_token_positions(Iterator begin, position_type const* positions, |
141 | std::vector<Token> const& tokens) |
142 | { |
143 | for (std::size_t i = 0, len = tokens.size(); i < len; ++i) |
144 | { |
145 | if (positions->begin == std::size_t(-1) && |
146 | positions->end == std::size_t(-1)) |
147 | { |
148 | return false; // reached end of expected data |
149 | } |
150 | |
151 | boost::iterator_range<Iterator> matched = tokens[i].matched(); |
152 | std::size_t start = std::distance(begin, matched.begin()); |
153 | std::size_t end = std::distance(begin, matched.end()); |
154 | |
155 | // position must match |
156 | if (start != positions->begin || end != positions->end) |
157 | return false; |
158 | |
159 | ++positions; |
160 | } |
161 | |
162 | return (positions->begin == std::size_t(-1) && |
163 | positions->end == std::size_t(-1)) ? true : false; |
164 | } |
165 | |
166 | /////////////////////////////////////////////////////////////////////////////// |
167 | int main() |
168 | { |
169 | typedef std::string::iterator base_iterator_type; |
170 | std::string input(" 01 1.2 -2 0x3 2.3e6 -3.4" ); |
171 | int ids[] = { ID_INT, ID_DOUBLE, ID_INT, ID_INT, ID_DOUBLE, ID_DOUBLE, -1 }; |
172 | std::size_t states[] = { 0, 1, 2, 1, 1, 2, std::size_t(-1) }; |
173 | position_type positions[] = |
174 | { |
175 | { .begin: 1, .end: 3 }, { .begin: 4, .end: 7 }, { .begin: 8, .end: 10 }, { .begin: 11, .end: 14 }, { .begin: 15, .end: 20 }, { .begin: 21, .end: 25 }, |
176 | { .begin: std::size_t(-1), .end: std::size_t(-1) } |
177 | }; |
178 | |
179 | // token type: token id, iterator_pair as token value, no state |
180 | { |
181 | typedef lex::lexertl::token< |
182 | base_iterator_type, mpl::vector<>, mpl::false_> token_type; |
183 | typedef lex::lexertl::actor_lexer<token_type> lexer_type; |
184 | |
185 | token_definitions<lexer_type> lexer; |
186 | std::vector<token_type> tokens; |
187 | base_iterator_type first = input.begin(); |
188 | |
189 | using phoenix::arg_names::_1; |
190 | BOOST_TEST(lex::tokenize(first, input.end(), lexer |
191 | , phoenix::push_back(phoenix::ref(tokens), _1))); |
192 | |
193 | BOOST_TEST(test_token_ids(ids, tokens)); |
194 | } |
195 | |
196 | { |
197 | typedef lex::lexertl::position_token< |
198 | base_iterator_type, mpl::vector<>, mpl::false_> token_type; |
199 | typedef lex::lexertl::actor_lexer<token_type> lexer_type; |
200 | |
201 | token_definitions<lexer_type> lexer; |
202 | std::vector<token_type> tokens; |
203 | base_iterator_type first = input.begin(); |
204 | |
205 | using phoenix::arg_names::_1; |
206 | BOOST_TEST(lex::tokenize(first, input.end(), lexer |
207 | , phoenix::push_back(phoenix::ref(tokens), _1))); |
208 | |
209 | BOOST_TEST(test_token_ids(ids, tokens)); |
210 | BOOST_TEST(test_token_positions(input.begin(), positions, tokens)); |
211 | } |
212 | |
213 | // token type: holds token id, state, iterator_pair as token value |
214 | { |
215 | typedef lex::lexertl::token< |
216 | base_iterator_type, mpl::vector<>, mpl::true_> token_type; |
217 | typedef lex::lexertl::actor_lexer<token_type> lexer_type; |
218 | |
219 | token_definitions_with_state<lexer_type> lexer; |
220 | std::vector<token_type> tokens; |
221 | base_iterator_type first = input.begin(); |
222 | |
223 | using phoenix::arg_names::_1; |
224 | BOOST_TEST(lex::tokenize(first, input.end(), lexer |
225 | , phoenix::push_back(phoenix::ref(tokens), _1))); |
226 | |
227 | BOOST_TEST(test_token_ids(ids, tokens)); |
228 | BOOST_TEST(test_token_states(states, tokens)); |
229 | } |
230 | |
231 | { |
232 | typedef lex::lexertl::position_token< |
233 | base_iterator_type, mpl::vector<>, mpl::true_> token_type; |
234 | typedef lex::lexertl::actor_lexer<token_type> lexer_type; |
235 | |
236 | token_definitions_with_state<lexer_type> lexer; |
237 | std::vector<token_type> tokens; |
238 | base_iterator_type first = input.begin(); |
239 | |
240 | using phoenix::arg_names::_1; |
241 | BOOST_TEST(lex::tokenize(first, input.end(), lexer |
242 | , phoenix::push_back(phoenix::ref(tokens), _1))); |
243 | |
244 | BOOST_TEST(test_token_ids(ids, tokens)); |
245 | BOOST_TEST(test_token_states(states, tokens)); |
246 | BOOST_TEST(test_token_positions(input.begin(), positions, tokens)); |
247 | } |
248 | |
249 | return boost::report_errors(); |
250 | } |
251 | |