| Index: trunk/extensions/AbuseFilter/parser_native/parser.h |
| — | — | @@ -22,6 +22,7 @@ |
| 23 | 23 | #include <boost/spirit/core.hpp> |
| 24 | 24 | #include <boost/spirit/utility/confix.hpp> |
| 25 | 25 | #include <boost/spirit/utility/chset.hpp> |
| | 26 | +#include <boost/spirit/utility/loops.hpp> |
| 26 | 27 | #include <boost/spirit/tree/ast.hpp> |
| 27 | 28 | #include <boost/spirit/tree/tree_to_xml.hpp> |
| 28 | 29 | #include <boost/spirit/symbols.hpp> |
| — | — | @@ -211,13 +212,28 @@ |
| 212 | 213 | | string |
| 213 | 214 | ; |
| 214 | 215 | |
| | 216 | + hexchar = chset<>("a-fA-F0-9") |
| | 217 | + ; |
| | 218 | + |
| | 219 | + octchar = chset<>("0-7") |
| | 220 | + ; |
| | 221 | + |
| | 222 | + c_string_char = |
| | 223 | + "\\x" >> hexchar >> hexchar |
| | 224 | + | "\\u" >> repeat_p(4)[hexchar] |
| | 225 | + | "\\U" >> repeat_p(8)[hexchar] |
| | 226 | + | "\\o" >> octchar >> octchar >> octchar |
| | 227 | + | "\\" >> anychar_p - (ch_p('x') | 'u' | 'o') |
| | 228 | + | anychar_p - (ch_p('"') | '\\') |
| | 229 | + ; |
| | 230 | + |
| 215 | 231 | /* |
| 216 | 232 | * config_p can't be used here, because it will rewrite |
| 217 | 233 | * *(c_escape_ch_p[x]) into (*c_escape_ch_p)[x] |
| 218 | 234 | */ |
| 219 | 235 | string = inner_node_d[ |
| 220 | 236 | '"' |
| 221 | | - >> leaf_node_d[ *(lex_escape_ch_p - '"') ] |
| | 237 | + >> leaf_node_d[ *(c_string_char) ] |
| 222 | 238 | >> '"' |
| 223 | 239 | ] |
| 224 | 240 | ; |
| — | — | @@ -341,6 +357,7 @@ |
| 342 | 358 | return tern_expr; |
| 343 | 359 | } |
| 344 | 360 | |
| | 361 | + rule<ScannerT> c_string_char, hexchar, octchar; |
| 345 | 362 | rule<ScannerT, parser_context<>, parser_tag<id_value> > value; |
| 346 | 363 | rule<ScannerT, parser_context<>, parser_tag<id_variable> > variable; |
| 347 | 364 | rule<ScannerT, parser_context<>, parser_tag<id_basic> > basic; |
| Index: trunk/extensions/AbuseFilter/parser_native/ast.h |
| — | — | @@ -12,6 +12,32 @@ |
| 13 | 13 | #ifndef AST_H |
| 14 | 14 | #define AST_H |
| 15 | 15 | |
| | 16 | +namespace { |
| | 17 | + |
| | 18 | +template<typename charT> |
| | 19 | +int |
| | 20 | +hex2int(charT const *str, int ndigits) |
| | 21 | +{ |
| | 22 | + int ret = 0; |
| | 23 | + |
| | 24 | + while (ndigits--) { |
| | 25 | + ret *= 0x10; |
| | 26 | + if (*str >= 'a' && *str <= 'f') |
| | 27 | + ret += 10 + int(*str - 'a'); |
| | 28 | + else if (*str >= 'A' && *str <= 'F') |
| | 29 | + ret += 10 + int(*str - 'A'); |
| | 30 | + else if (*str >= '0' && *str <= '9') |
| | 31 | + ret += int(*str - '0'); |
| | 32 | + |
| | 33 | + str++; |
| | 34 | + } |
| | 35 | + |
| | 36 | + std::cerr << "hex2int: " << ret << '\n'; |
| | 37 | + return ret; |
| | 38 | +} |
| | 39 | + |
| | 40 | +} |
| | 41 | + |
| 16 | 42 | namespace afp { |
| 17 | 43 | |
| 18 | 44 | template<typename T> struct parser_grammar; |
| — | — | @@ -234,6 +260,27 @@ |
| 235 | 261 | case 'v': |
| 236 | 262 | ret.push_back('\v'); |
| 237 | 263 | break; |
| | 264 | + case 'x': |
| | 265 | + if (i + 3 >= end) |
| | 266 | + break; |
| | 267 | + ret.push_back(hex2int(s.data() + i + 2, 2)); |
| | 268 | + i += 2; |
| | 269 | + break; |
| | 270 | + |
| | 271 | + case 'u': |
| | 272 | + if (i + 5 >= end) |
| | 273 | + break; |
| | 274 | + ret.push_back(hex2int(s.data() + i + 2, 4)); |
| | 275 | + i += 4; |
| | 276 | + break; |
| | 277 | + |
| | 278 | + case 'U': |
| | 279 | + if (i + 9 >= end) |
| | 280 | + break; |
| | 281 | + ret.push_back(hex2int(s.data() + i + 2, 8)); |
| | 282 | + i += 8; |
| | 283 | + break; |
| | 284 | + |
| 238 | 285 | default: |
| 239 | 286 | ret.push_back(s[i + 1]); |
| 240 | 287 | break; |