Mercurial > projects > ddmd
annotate dmd/Lexer.d @ 162:438eaa11eed4
updated build script to use dmd2.039
some missing methods implemented
| author | korDen |
|---|---|
| date | Tue, 21 Sep 2010 14:59:56 +0400 |
| parents | b7b61140701d |
| children | fe932c1a9563 |
| rev | line source |
|---|---|
| 0 | 1 module dmd.Lexer; |
| 2 | |
| 114 | 3 import dmd.common; |
| 0 | 4 import dmd.StringTable; |
| 5 import dmd.OutBuffer; | |
| 6 import dmd.Token; | |
| 7 import dmd.Loc; | |
| 8 import dmd.Module; | |
| 9 import dmd.Identifier; | |
| 10 import dmd.TOK; | |
| 11 import dmd.Keyword; | |
| 12 import dmd.StringValue; | |
| 13 import dmd.Global; | |
| 14 import dmd.Util; | |
| 15 import dmd.Id; | |
| 16 import dmd.Dchar; | |
| 17 import dmd.Utf; | |
| 18 | |
| 19 import std.stdio : writeln; | |
| 20 | |
| 4 | 21 import core.memory; |
| 2 | 22 |
| 0 | 23 import core.stdc.ctype; |
| 24 import core.stdc.stdlib; | |
| 25 import core.stdc.string; | |
| 26 import core.stdc.stdio; | |
| 27 import core.stdc.time; | |
| 28 import core.stdc.errno; | |
| 29 | |
| 30 enum LS = 0x2028; // UTF line separator | |
| 31 enum PS = 0x2029; // UTF paragraph separator | |
| 32 | |
| 33 extern (C) extern | |
| 34 { | |
| 35 __gshared char* __locale_decpoint; | |
| 36 } | |
| 37 | |
| 162 | 38 bool isUniAlpha(uint u) |
| 0 | 39 { |
| 162 | 40 static ushort table[][2] = |
| 41 [ | |
| 42 [ 0x00AA, 0x00AA ], | |
| 43 [ 0x00B5, 0x00B5 ], | |
| 44 [ 0x00B7, 0x00B7 ], | |
| 45 [ 0x00BA, 0x00BA ], | |
| 46 [ 0x00C0, 0x00D6 ], | |
| 47 [ 0x00D8, 0x00F6 ], | |
| 48 [ 0x00F8, 0x01F5 ], | |
| 49 [ 0x01FA, 0x0217 ], | |
| 50 [ 0x0250, 0x02A8 ], | |
| 51 [ 0x02B0, 0x02B8 ], | |
| 52 [ 0x02BB, 0x02BB ], | |
| 53 [ 0x02BD, 0x02C1 ], | |
| 54 [ 0x02D0, 0x02D1 ], | |
| 55 [ 0x02E0, 0x02E4 ], | |
| 56 [ 0x037A, 0x037A ], | |
| 57 [ 0x0386, 0x0386 ], | |
| 58 [ 0x0388, 0x038A ], | |
| 59 [ 0x038C, 0x038C ], | |
| 60 [ 0x038E, 0x03A1 ], | |
| 61 [ 0x03A3, 0x03CE ], | |
| 62 [ 0x03D0, 0x03D6 ], | |
| 63 [ 0x03DA, 0x03DA ], | |
| 64 [ 0x03DC, 0x03DC ], | |
| 65 [ 0x03DE, 0x03DE ], | |
| 66 [ 0x03E0, 0x03E0 ], | |
| 67 [ 0x03E2, 0x03F3 ], | |
| 68 [ 0x0401, 0x040C ], | |
| 69 [ 0x040E, 0x044F ], | |
| 70 [ 0x0451, 0x045C ], | |
| 71 [ 0x045E, 0x0481 ], | |
| 72 [ 0x0490, 0x04C4 ], | |
| 73 [ 0x04C7, 0x04C8 ], | |
| 74 [ 0x04CB, 0x04CC ], | |
| 75 [ 0x04D0, 0x04EB ], | |
| 76 [ 0x04EE, 0x04F5 ], | |
| 77 [ 0x04F8, 0x04F9 ], | |
| 78 [ 0x0531, 0x0556 ], | |
| 79 [ 0x0559, 0x0559 ], | |
| 80 [ 0x0561, 0x0587 ], | |
| 81 [ 0x05B0, 0x05B9 ], | |
| 82 [ 0x05BB, 0x05BD ], | |
| 83 [ 0x05BF, 0x05BF ], | |
| 84 [ 0x05C1, 0x05C2 ], | |
| 85 [ 0x05D0, 0x05EA ], | |
| 86 [ 0x05F0, 0x05F2 ], | |
| 87 [ 0x0621, 0x063A ], | |
| 88 [ 0x0640, 0x0652 ], | |
| 89 [ 0x0660, 0x0669 ], | |
| 90 [ 0x0670, 0x06B7 ], | |
| 91 [ 0x06BA, 0x06BE ], | |
| 92 [ 0x06C0, 0x06CE ], | |
| 93 [ 0x06D0, 0x06DC ], | |
| 94 [ 0x06E5, 0x06E8 ], | |
| 95 [ 0x06EA, 0x06ED ], | |
| 96 [ 0x06F0, 0x06F9 ], | |
| 97 [ 0x0901, 0x0903 ], | |
| 98 [ 0x0905, 0x0939 ], | |
| 99 [ 0x093D, 0x093D ], | |
| 100 [ 0x093E, 0x094D ], | |
| 101 [ 0x0950, 0x0952 ], | |
| 102 [ 0x0958, 0x0963 ], | |
| 103 [ 0x0966, 0x096F ], | |
| 104 [ 0x0981, 0x0983 ], | |
| 105 [ 0x0985, 0x098C ], | |
| 106 [ 0x098F, 0x0990 ], | |
| 107 [ 0x0993, 0x09A8 ], | |
| 108 [ 0x09AA, 0x09B0 ], | |
| 109 [ 0x09B2, 0x09B2 ], | |
| 110 [ 0x09B6, 0x09B9 ], | |
| 111 [ 0x09BE, 0x09C4 ], | |
| 112 [ 0x09C7, 0x09C8 ], | |
| 113 [ 0x09CB, 0x09CD ], | |
| 114 [ 0x09DC, 0x09DD ], | |
| 115 [ 0x09DF, 0x09E3 ], | |
| 116 [ 0x09E6, 0x09EF ], | |
| 117 [ 0x09F0, 0x09F1 ], | |
| 118 [ 0x0A02, 0x0A02 ], | |
| 119 [ 0x0A05, 0x0A0A ], | |
| 120 [ 0x0A0F, 0x0A10 ], | |
| 121 [ 0x0A13, 0x0A28 ], | |
| 122 [ 0x0A2A, 0x0A30 ], | |
| 123 [ 0x0A32, 0x0A33 ], | |
| 124 [ 0x0A35, 0x0A36 ], | |
| 125 [ 0x0A38, 0x0A39 ], | |
| 126 [ 0x0A3E, 0x0A42 ], | |
| 127 [ 0x0A47, 0x0A48 ], | |
| 128 [ 0x0A4B, 0x0A4D ], | |
| 129 [ 0x0A59, 0x0A5C ], | |
| 130 [ 0x0A5E, 0x0A5E ], | |
| 131 [ 0x0A66, 0x0A6F ], | |
| 132 [ 0x0A74, 0x0A74 ], | |
| 133 [ 0x0A81, 0x0A83 ], | |
| 134 [ 0x0A85, 0x0A8B ], | |
| 135 [ 0x0A8D, 0x0A8D ], | |
| 136 [ 0x0A8F, 0x0A91 ], | |
| 137 [ 0x0A93, 0x0AA8 ], | |
| 138 [ 0x0AAA, 0x0AB0 ], | |
| 139 [ 0x0AB2, 0x0AB3 ], | |
| 140 [ 0x0AB5, 0x0AB9 ], | |
| 141 [ 0x0ABD, 0x0AC5 ], | |
| 142 [ 0x0AC7, 0x0AC9 ], | |
| 143 [ 0x0ACB, 0x0ACD ], | |
| 144 [ 0x0AD0, 0x0AD0 ], | |
| 145 [ 0x0AE0, 0x0AE0 ], | |
| 146 [ 0x0AE6, 0x0AEF ], | |
| 147 [ 0x0B01, 0x0B03 ], | |
| 148 [ 0x0B05, 0x0B0C ], | |
| 149 [ 0x0B0F, 0x0B10 ], | |
| 150 [ 0x0B13, 0x0B28 ], | |
| 151 [ 0x0B2A, 0x0B30 ], | |
| 152 [ 0x0B32, 0x0B33 ], | |
| 153 [ 0x0B36, 0x0B39 ], | |
| 154 [ 0x0B3D, 0x0B3D ], | |
| 155 [ 0x0B3E, 0x0B43 ], | |
| 156 [ 0x0B47, 0x0B48 ], | |
| 157 [ 0x0B4B, 0x0B4D ], | |
| 158 [ 0x0B5C, 0x0B5D ], | |
| 159 [ 0x0B5F, 0x0B61 ], | |
| 160 [ 0x0B66, 0x0B6F ], | |
| 161 [ 0x0B82, 0x0B83 ], | |
| 162 [ 0x0B85, 0x0B8A ], | |
| 163 [ 0x0B8E, 0x0B90 ], | |
| 164 [ 0x0B92, 0x0B95 ], | |
| 165 [ 0x0B99, 0x0B9A ], | |
| 166 [ 0x0B9C, 0x0B9C ], | |
| 167 [ 0x0B9E, 0x0B9F ], | |
| 168 [ 0x0BA3, 0x0BA4 ], | |
| 169 [ 0x0BA8, 0x0BAA ], | |
| 170 [ 0x0BAE, 0x0BB5 ], | |
| 171 [ 0x0BB7, 0x0BB9 ], | |
| 172 [ 0x0BBE, 0x0BC2 ], | |
| 173 [ 0x0BC6, 0x0BC8 ], | |
| 174 [ 0x0BCA, 0x0BCD ], | |
| 175 [ 0x0BE7, 0x0BEF ], | |
| 176 [ 0x0C01, 0x0C03 ], | |
| 177 [ 0x0C05, 0x0C0C ], | |
| 178 [ 0x0C0E, 0x0C10 ], | |
| 179 [ 0x0C12, 0x0C28 ], | |
| 180 [ 0x0C2A, 0x0C33 ], | |
| 181 [ 0x0C35, 0x0C39 ], | |
| 182 [ 0x0C3E, 0x0C44 ], | |
| 183 [ 0x0C46, 0x0C48 ], | |
| 184 [ 0x0C4A, 0x0C4D ], | |
| 185 [ 0x0C60, 0x0C61 ], | |
| 186 [ 0x0C66, 0x0C6F ], | |
| 187 [ 0x0C82, 0x0C83 ], | |
| 188 [ 0x0C85, 0x0C8C ], | |
| 189 [ 0x0C8E, 0x0C90 ], | |
| 190 [ 0x0C92, 0x0CA8 ], | |
| 191 [ 0x0CAA, 0x0CB3 ], | |
| 192 [ 0x0CB5, 0x0CB9 ], | |
| 193 [ 0x0CBE, 0x0CC4 ], | |
| 194 [ 0x0CC6, 0x0CC8 ], | |
| 195 [ 0x0CCA, 0x0CCD ], | |
| 196 [ 0x0CDE, 0x0CDE ], | |
| 197 [ 0x0CE0, 0x0CE1 ], | |
| 198 [ 0x0CE6, 0x0CEF ], | |
| 199 [ 0x0D02, 0x0D03 ], | |
| 200 [ 0x0D05, 0x0D0C ], | |
| 201 [ 0x0D0E, 0x0D10 ], | |
| 202 [ 0x0D12, 0x0D28 ], | |
| 203 [ 0x0D2A, 0x0D39 ], | |
| 204 [ 0x0D3E, 0x0D43 ], | |
| 205 [ 0x0D46, 0x0D48 ], | |
| 206 [ 0x0D4A, 0x0D4D ], | |
| 207 [ 0x0D60, 0x0D61 ], | |
| 208 [ 0x0D66, 0x0D6F ], | |
| 209 [ 0x0E01, 0x0E3A ], | |
| 210 [ 0x0E40, 0x0E5B ], | |
| 211 // { 0x0E50, 0x0E59 }, | |
| 212 [ 0x0E81, 0x0E82 ], | |
| 213 [ 0x0E84, 0x0E84 ], | |
| 214 [ 0x0E87, 0x0E88 ], | |
| 215 [ 0x0E8A, 0x0E8A ], | |
| 216 [ 0x0E8D, 0x0E8D ], | |
| 217 [ 0x0E94, 0x0E97 ], | |
| 218 [ 0x0E99, 0x0E9F ], | |
| 219 [ 0x0EA1, 0x0EA3 ], | |
| 220 [ 0x0EA5, 0x0EA5 ], | |
| 221 [ 0x0EA7, 0x0EA7 ], | |
| 222 [ 0x0EAA, 0x0EAB ], | |
| 223 [ 0x0EAD, 0x0EAE ], | |
| 224 [ 0x0EB0, 0x0EB9 ], | |
| 225 [ 0x0EBB, 0x0EBD ], | |
| 226 [ 0x0EC0, 0x0EC4 ], | |
| 227 [ 0x0EC6, 0x0EC6 ], | |
| 228 [ 0x0EC8, 0x0ECD ], | |
| 229 [ 0x0ED0, 0x0ED9 ], | |
| 230 [ 0x0EDC, 0x0EDD ], | |
| 231 [ 0x0F00, 0x0F00 ], | |
| 232 [ 0x0F18, 0x0F19 ], | |
| 233 [ 0x0F20, 0x0F33 ], | |
| 234 [ 0x0F35, 0x0F35 ], | |
| 235 [ 0x0F37, 0x0F37 ], | |
| 236 [ 0x0F39, 0x0F39 ], | |
| 237 [ 0x0F3E, 0x0F47 ], | |
| 238 [ 0x0F49, 0x0F69 ], | |
| 239 [ 0x0F71, 0x0F84 ], | |
| 240 [ 0x0F86, 0x0F8B ], | |
| 241 [ 0x0F90, 0x0F95 ], | |
| 242 [ 0x0F97, 0x0F97 ], | |
| 243 [ 0x0F99, 0x0FAD ], | |
| 244 [ 0x0FB1, 0x0FB7 ], | |
| 245 [ 0x0FB9, 0x0FB9 ], | |
| 246 [ 0x10A0, 0x10C5 ], | |
| 247 [ 0x10D0, 0x10F6 ], | |
| 248 [ 0x1E00, 0x1E9B ], | |
| 249 [ 0x1EA0, 0x1EF9 ], | |
| 250 [ 0x1F00, 0x1F15 ], | |
| 251 [ 0x1F18, 0x1F1D ], | |
| 252 [ 0x1F20, 0x1F45 ], | |
| 253 [ 0x1F48, 0x1F4D ], | |
| 254 [ 0x1F50, 0x1F57 ], | |
| 255 [ 0x1F59, 0x1F59 ], | |
| 256 [ 0x1F5B, 0x1F5B ], | |
| 257 [ 0x1F5D, 0x1F5D ], | |
| 258 [ 0x1F5F, 0x1F7D ], | |
| 259 [ 0x1F80, 0x1FB4 ], | |
| 260 [ 0x1FB6, 0x1FBC ], | |
| 261 [ 0x1FBE, 0x1FBE ], | |
| 262 [ 0x1FC2, 0x1FC4 ], | |
| 263 [ 0x1FC6, 0x1FCC ], | |
| 264 [ 0x1FD0, 0x1FD3 ], | |
| 265 [ 0x1FD6, 0x1FDB ], | |
| 266 [ 0x1FE0, 0x1FEC ], | |
| 267 [ 0x1FF2, 0x1FF4 ], | |
| 268 [ 0x1FF6, 0x1FFC ], | |
| 269 [ 0x203F, 0x2040 ], | |
| 270 [ 0x207F, 0x207F ], | |
| 271 [ 0x2102, 0x2102 ], | |
| 272 [ 0x2107, 0x2107 ], | |
| 273 [ 0x210A, 0x2113 ], | |
| 274 [ 0x2115, 0x2115 ], | |
| 275 [ 0x2118, 0x211D ], | |
| 276 [ 0x2124, 0x2124 ], | |
| 277 [ 0x2126, 0x2126 ], | |
| 278 [ 0x2128, 0x2128 ], | |
| 279 [ 0x212A, 0x2131 ], | |
| 280 [ 0x2133, 0x2138 ], | |
| 281 [ 0x2160, 0x2182 ], | |
| 282 [ 0x3005, 0x3007 ], | |
| 283 [ 0x3021, 0x3029 ], | |
| 284 [ 0x3041, 0x3093 ], | |
| 285 [ 0x309B, 0x309C ], | |
| 286 [ 0x30A1, 0x30F6 ], | |
| 287 [ 0x30FB, 0x30FC ], | |
| 288 [ 0x3105, 0x312C ], | |
| 289 [ 0x4E00, 0x9FA5 ], | |
| 290 [ 0xAC00, 0xD7A3 ], | |
| 291 ]; | |
| 292 | |
| 293 debug { | |
| 294 for (int i = 0; i < table.length; i++) | |
| 295 { | |
| 296 //printf("%x\n", table[i][0]); | |
| 297 assert(table[i][0] <= table[i][1]); | |
| 298 if (i < table.length - 1) | |
| 299 assert(table[i][1] < table[i + 1][0]); | |
| 300 } | |
| 301 } | |
| 302 | |
| 303 if (u > 0xD7A3) | |
| 304 goto Lisnot; | |
| 305 | |
| 306 // Binary search | |
| 307 int mid; | |
| 308 int low; | |
| 309 int high; | |
| 310 | |
| 311 low = 0; | |
| 312 high = table.length - 1; | |
| 313 while (low <= high) | |
| 314 { | |
| 315 mid = (low + high) >> 1; | |
| 316 if (u < table[mid][0]) | |
| 317 high = mid - 1; | |
| 318 else if (u > table[mid][1]) | |
| 319 low = mid + 1; | |
| 320 else | |
| 321 goto Lis; | |
| 322 } | |
| 323 | |
| 324 Lisnot: | |
| 325 debug { | |
| 326 for (int i = 0; i < table.length; i++) | |
| 327 { | |
| 328 assert(u < table[i][0] || u > table[i][1]); | |
| 329 } | |
| 330 } | |
| 331 return false; | |
| 332 | |
| 333 Lis: | |
| 334 debug { | |
| 335 for (int i = 0; i < table.length; i++) | |
| 336 { | |
| 337 if (u >= table[i][0] && u <= table[i][1]) | |
| 338 return 1; | |
| 339 } | |
| 340 assert(0); // should have been in table | |
| 341 } | |
| 342 return true; | |
| 0 | 343 } |
| 344 | |
| 345 class Lexer | |
| 346 { | |
| 347 static StringTable stringtable; | |
| 348 static OutBuffer stringbuffer; | |
| 349 static Token* freelist; | |
| 350 | |
| 351 Loc loc; // for error messages | |
| 352 | |
| 353 ubyte* base; // pointer to start of buffer | |
| 354 ubyte* end; // past end of buffer | |
| 355 ubyte* p; // current character | |
| 356 Token token; | |
| 357 Module mod; | |
| 358 int doDocComment; // collect doc comment information | |
| 359 int anyToken; // !=0 means seen at least one token | |
| 360 int commentToken; // !=0 means comments are TOKcomment's | |
| 361 | |
| 362 static this() | |
| 363 { | |
| 364 stringtable = new StringTable(); | |
| 365 stringbuffer = new OutBuffer(); | |
| 366 } | |
| 367 | |
| 368 static ~this() | |
| 369 { | |
| 34 | 370 //delete stringtable; |
| 0 | 371 } |
| 372 | |
| 373 this(Module mod, ubyte* base, uint begoffset, uint endoffset, int doDocComment, int commentToken) | |
| 374 { | |
| 375 loc = Loc(mod, 1); | |
| 376 | |
| 377 memset(&token,0,token.sizeof); | |
| 378 this.base = base; | |
| 379 this.end = base + endoffset; | |
| 380 p = base + begoffset; | |
| 381 this.mod = mod; | |
| 382 this.doDocComment = doDocComment; | |
| 383 this.anyToken = 0; | |
| 384 this.commentToken = commentToken; | |
| 385 //initKeywords(); | |
| 386 | |
| 387 /* If first line starts with '#!', ignore the line | |
| 388 */ | |
| 389 | |
| 390 if (p[0] == '#' && p[1] =='!') | |
| 391 { | |
| 392 p += 2; | |
| 393 while (1) | |
| 394 { | |
| 395 ubyte c = *p; | |
| 396 switch (c) | |
| 397 { | |
| 398 case '\n': | |
| 399 p++; | |
| 400 break; | |
| 401 | |
| 402 case '\r': | |
| 403 p++; | |
| 404 if (*p == '\n') | |
| 405 p++; | |
| 406 break; | |
| 407 | |
| 408 case 0: | |
| 409 case 0x1A: | |
| 410 break; | |
| 411 | |
| 412 default: | |
| 413 if (c & 0x80) | |
| 414 { | |
| 415 uint u = decodeUTF(); | |
| 416 if (u == PS || u == LS) | |
| 417 break; | |
| 418 } | |
| 419 p++; | |
| 420 continue; | |
| 421 } | |
| 422 break; | |
| 423 } | |
| 424 loc.linnum = 2; | |
| 425 } | |
| 426 } | |
| 427 | |
| 428 version (DMDV2) { | |
| 429 static Keyword[] keywords = | |
| 430 [ | |
| 431 // { "", TOK }, | |
| 432 | |
| 433 { "this", TOK.TOKthis }, | |
| 434 { "super", TOK.TOKsuper }, | |
| 435 { "assert", TOK.TOKassert }, | |
| 436 { "null", TOK.TOKnull }, | |
| 437 { "true", TOK.TOKtrue }, | |
| 438 { "false", TOK.TOKfalse }, | |
| 439 { "cast", TOK.TOKcast }, | |
| 440 { "new", TOK.TOKnew }, | |
| 441 { "delete", TOK.TOKdelete }, | |
| 442 { "throw", TOK.TOKthrow }, | |
| 443 { "module", TOK.TOKmodule }, | |
| 444 { "pragma", TOK.TOKpragma }, | |
| 445 { "typeof", TOK.TOKtypeof }, | |
| 446 { "typeid", TOK.TOKtypeid }, | |
| 447 | |
| 448 { "template", TOK.TOKtemplate }, | |
| 449 | |
| 450 { "void", TOK.TOKvoid }, | |
| 451 { "byte", TOK.TOKint8 }, | |
| 452 { "ubyte", TOK.TOKuns8 }, | |
| 453 { "short", TOK.TOKint16 }, | |
| 454 { "ushort", TOK.TOKuns16 }, | |
| 455 { "int", TOK.TOKint32 }, | |
| 456 { "uint", TOK.TOKuns32 }, | |
| 457 { "long", TOK.TOKint64 }, | |
| 458 { "ulong", TOK.TOKuns64 }, | |
| 459 { "cent", TOK.TOKcent, }, | |
| 460 { "ucent", TOK.TOKucent, }, | |
| 461 { "float", TOK.TOKfloat32 }, | |
| 462 { "double", TOK.TOKfloat64 }, | |
| 463 { "real", TOK.TOKfloat80 }, | |
| 464 | |
| 465 { "bool", TOK.TOKbool }, | |
| 466 { "char", TOK.TOKchar }, | |
| 467 { "wchar", TOK.TOKwchar }, | |
| 468 { "dchar", TOK.TOKdchar }, | |
| 469 | |
| 470 { "ifloat", TOK.TOKimaginary32 }, | |
| 471 { "idouble", TOK.TOKimaginary64 }, | |
| 472 { "ireal", TOK.TOKimaginary80 }, | |
| 473 | |
| 474 { "cfloat", TOK.TOKcomplex32 }, | |
| 475 { "cdouble", TOK.TOKcomplex64 }, | |
| 476 { "creal", TOK.TOKcomplex80 }, | |
| 477 | |
| 478 { "delegate", TOK.TOKdelegate }, | |
| 479 { "function", TOK.TOKfunction }, | |
| 480 | |
| 481 { "is", TOK.TOKis }, | |
| 482 { "if", TOK.TOKif }, | |
| 483 { "else", TOK.TOKelse }, | |
| 484 { "while", TOK.TOKwhile }, | |
| 485 { "for", TOK.TOKfor }, | |
| 486 { "do", TOK.TOKdo }, | |
| 487 { "switch", TOK.TOKswitch }, | |
| 488 { "case", TOK.TOKcase }, | |
| 489 { "default", TOK.TOKdefault }, | |
| 490 { "break", TOK.TOKbreak }, | |
| 491 { "continue", TOK.TOKcontinue }, | |
| 492 { "synchronized", TOK.TOKsynchronized }, | |
| 493 { "return", TOK.TOKreturn }, | |
| 494 { "goto", TOK.TOKgoto }, | |
| 495 { "try", TOK.TOKtry }, | |
| 496 { "catch", TOK.TOKcatch }, | |
| 497 { "finally", TOK.TOKfinally }, | |
| 498 { "with", TOK.TOKwith }, | |
| 499 { "asm", TOK.TOKasm }, | |
| 500 { "foreach", TOK.TOKforeach }, | |
| 501 { "foreach_reverse", TOK.TOKforeach_reverse }, | |
| 502 { "scope", TOK.TOKscope }, | |
| 503 | |
| 504 { "struct", TOK.TOKstruct }, | |
| 505 { "class", TOK.TOKclass }, | |
| 506 { "interface", TOK.TOKinterface }, | |
| 507 { "union", TOK.TOKunion }, | |
| 508 { "enum", TOK.TOKenum }, | |
| 509 { "import", TOK.TOKimport }, | |
| 510 { "mixin", TOK.TOKmixin }, | |
| 511 { "static", TOK.TOKstatic }, | |
| 512 { "final", TOK.TOKfinal }, | |
| 513 { "const", TOK.TOKconst }, | |
| 514 { "typedef", TOK.TOKtypedef }, | |
| 515 { "alias", TOK.TOKalias }, | |
| 516 { "override", TOK.TOKoverride }, | |
| 517 { "abstract", TOK.TOKabstract }, | |
| 518 { "volatile", TOK.TOKvolatile }, | |
| 519 { "debug", TOK.TOKdebug }, | |
| 520 { "deprecated", TOK.TOKdeprecated }, | |
| 521 { "in", TOK.TOKin }, | |
| 522 { "out", TOK.TOKout }, | |
| 523 { "inout", TOK.TOKinout }, | |
| 524 { "lazy", TOK.TOKlazy }, | |
| 525 { "auto", TOK.TOKauto }, | |
| 526 | |
| 527 { "align", TOK.TOKalign }, | |
| 528 { "extern", TOK.TOKextern }, | |
| 529 { "private", TOK.TOKprivate }, | |
| 530 { "package", TOK.TOKpackage }, | |
| 531 { "protected", TOK.TOKprotected }, | |
| 532 { "public", TOK.TOKpublic }, | |
| 533 { "export", TOK.TOKexport }, | |
| 534 | |
| 535 { "body", TOK.TOKbody }, | |
| 536 { "invariant", TOK.TOKinvariant }, | |
| 537 { "unittest", TOK.TOKunittest }, | |
| 538 { "version", TOK.TOKversion }, | |
| 539 //{ "manifest", TOK.TOKmanifest }, | |
| 540 | |
| 541 // Added after 1.0 | |
| 542 { "ref", TOK.TOKref }, | |
| 543 { "macro", TOK.TOKmacro }, | |
| 544 { "pure", TOK.TOKpure }, | |
| 545 { "nothrow", TOK.TOKnothrow }, | |
| 546 { "__thread", TOK.TOKtls }, | |
| 547 { "__gshared", TOK.TOKgshared }, | |
| 548 { "__traits", TOK.TOKtraits }, | |
| 549 { "__overloadset", TOK.TOKoverloadset }, | |
| 550 { "__FILE__", TOK.TOKfile }, | |
| 551 { "__LINE__", TOK.TOKline }, | |
| 552 { "shared", TOK.TOKshared }, | |
| 553 { "immutable", TOK.TOKimmutable }, | |
| 554 ]; | |
| 555 } else { | |
| 556 static Keyword[] keywords = | |
| 557 [ | |
| 558 // { "", TOK }, | |
| 559 | |
| 560 { "this", TOK.TOKthis }, | |
| 561 { "super", TOK.TOKsuper }, | |
| 562 { "assert", TOK.TOKassert }, | |
| 563 { "null", TOK.TOKnull }, | |
| 564 { "true", TOK.TOKtrue }, | |
| 565 { "false", TOK.TOKfalse }, | |
| 566 { "cast", TOK.TOKcast }, | |
| 567 { "new", TOK.TOKnew }, | |
| 568 { "delete", TOK.TOKdelete }, | |
| 569 { "throw", TOK.TOKthrow }, | |
| 570 { "module", TOK.TOKmodule }, | |
| 571 { "pragma", TOK.TOKpragma }, | |
| 572 { "typeof", TOK.TOKtypeof }, | |
| 573 { "typeid", TOK.TOKtypeid }, | |
| 574 | |
| 575 { "template", TOK.TOKtemplate }, | |
| 576 | |
| 577 { "void", TOK.TOKvoid }, | |
| 578 { "byte", TOK.TOKint8 }, | |
| 579 { "ubyte", TOK.TOKuns8 }, | |
| 580 { "short", TOK.TOKint16 }, | |
| 581 { "ushort", TOK.TOKuns16 }, | |
| 582 { "int", TOK.TOKint32 }, | |
| 583 { "uint", TOK.TOKuns32 }, | |
| 584 { "long", TOK.TOKint64 }, | |
| 585 { "ulong", TOK.TOKuns64 }, | |
| 586 { "cent", TOK.TOKcent, }, | |
| 587 { "ucent", TOK.TOKucent, }, | |
| 588 { "float", TOK.TOKfloat32 }, | |
| 589 { "double", TOK.TOKfloat64 }, | |
| 590 { "real", TOK.TOKfloat80 }, | |
| 591 | |
| 592 { "bool", TOK.TOKbool }, | |
| 593 { "char", TOK.TOKchar }, | |
| 594 { "wchar", TOK.TOKwchar }, | |
| 595 { "dchar", TOK.TOKdchar }, | |
| 596 | |
| 597 { "ifloat", TOK.TOKimaginary32 }, | |
| 598 { "idouble", TOK.TOKimaginary64 }, | |
| 599 { "ireal", TOK.TOKimaginary80 }, | |
| 600 | |
| 601 { "cfloat", TOK.TOKcomplex32 }, | |
| 602 { "cdouble", TOK.TOKcomplex64 }, | |
| 603 { "creal", TOK.TOKcomplex80 }, | |
| 604 | |
| 605 { "delegate", TOK.TOKdelegate }, | |
| 606 { "function", TOK.TOKfunction }, | |
| 607 | |
| 608 { "is", TOK.TOKis }, | |
| 609 { "if", TOK.TOKif }, | |
| 610 { "else", TOK.TOKelse }, | |
| 611 { "while", TOK.TOKwhile }, | |
| 612 { "for", TOK.TOKfor }, | |
| 613 { "do", TOK.TOKdo }, | |
| 614 { "switch", TOK.TOKswitch }, | |
| 615 { "case", TOK.TOKcase }, | |
| 616 { "default", TOK.TOKdefault }, | |
| 617 { "break", TOK.TOKbreak }, | |
| 618 { "continue", TOK.TOKcontinue }, | |
| 619 { "synchronized", TOK.TOKsynchronized }, | |
| 620 { "return", TOK.TOKreturn }, | |
| 621 { "goto", TOK.TOKgoto }, | |
| 622 { "try", TOK.TOKtry }, | |
| 623 { "catch", TOK.TOKcatch }, | |
| 624 { "finally", TOK.TOKfinally }, | |
| 625 { "with", TOK.TOKwith }, | |
| 626 { "asm", TOK.TOKasm }, | |
| 627 { "foreach", TOK.TOKforeach }, | |
| 628 { "foreach_reverse", TOK.TOKforeach_reverse }, | |
| 629 { "scope", TOK.TOKscope }, | |
| 630 | |
| 631 { "struct", TOK.TOKstruct }, | |
| 632 { "class", TOK.TOKclass }, | |
| 633 { "interface", TOK.TOKinterface }, | |
| 634 { "union", TOK.TOKunion }, | |
| 635 { "enum", TOK.TOKenum }, | |
| 636 { "import", TOK.TOKimport }, | |
| 637 { "mixin", TOK.TOKmixin }, | |
| 638 { "static", TOK.TOKstatic }, | |
| 639 { "final", TOK.TOKfinal }, | |
| 640 { "const", TOK.TOKconst }, | |
| 641 { "typedef", TOK.TOKtypedef }, | |
| 642 { "alias", TOK.TOKalias }, | |
| 643 { "override", TOK.TOKoverride }, | |
| 644 { "abstract", TOK.TOKabstract }, | |
| 645 { "volatile", TOK.TOKvolatile }, | |
| 646 { "debug", TOK.TOKdebug }, | |
| 647 { "deprecated", TOK.TOKdeprecated }, | |
| 648 { "in", TOK.TOKin }, | |
| 649 { "out", TOK.TOKout }, | |
| 650 { "inout", TOK.TOKinout }, | |
| 651 { "lazy", TOK.TOKlazy }, | |
| 652 { "auto", TOK.TOKauto }, | |
| 653 | |
| 654 { "align", TOK.TOKalign }, | |
| 655 { "extern", TOK.TOKextern }, | |
| 656 { "private", TOK.TOKprivate }, | |
| 657 { "package", TOK.TOKpackage }, | |
| 658 { "protected", TOK.TOKprotected }, | |
| 659 { "public", TOK.TOKpublic }, | |
| 660 { "export", TOK.TOKexport }, | |
| 661 | |
| 662 { "body", TOK.TOKbody }, | |
| 663 { "invariant", TOK.TOKinvariant }, | |
| 664 { "unittest", TOK.TOKunittest }, | |
| 665 { "version", TOK.TOKversion }, | |
| 666 //{ "manifest", TOK.TOKmanifest }, | |
| 667 | |
| 668 // Added after 1.0 | |
| 669 { "ref", TOK.TOKref }, | |
| 670 { "macro", TOK.TOKmacro }, | |
| 671 ]; | |
| 672 } | |
| 673 | |
| 674 static ubyte cmtable[256]; | |
| 675 enum CMoctal = 0x1; | |
| 676 enum CMhex = 0x2; | |
| 677 enum CMidchar = 0x4; | |
| 678 | |
| 679 ubyte isoctal (ubyte c) { return cmtable[c] & CMoctal; } | |
| 680 ubyte ishex (ubyte c) { return cmtable[c] & CMhex; } | |
| 681 ubyte isidchar(ubyte c) { return cmtable[c] & CMidchar; } | |
| 682 | |
| 683 static void cmtable_init() | |
| 684 { | |
| 685 for (uint c = 0; c < cmtable.length; c++) | |
| 686 { | |
| 687 if ('0' <= c && c <= '7') | |
| 688 cmtable[c] |= CMoctal; | |
| 689 if (isdigit(c) || ('a' <= c && c <= 'f') || ('A' <= c && c <= 'F')) | |
| 690 cmtable[c] |= CMhex; | |
| 691 if (isalnum(c) || c == '_') | |
| 692 cmtable[c] |= CMidchar; | |
| 693 } | |
| 694 } | |
| 695 | |
| 696 static void initKeywords() | |
| 697 { | |
| 698 uint nkeywords = keywords.length; | |
| 699 | |
| 700 if (global.params.Dversion == 1) | |
| 701 nkeywords -= 2; | |
| 702 | |
| 703 cmtable_init(); | |
| 704 | |
| 705 for (uint u = 0; u < nkeywords; u++) | |
| 706 { | |
| 34 | 707 //printf("keyword[%d] = '%.*s'\n",u, keywords[u].name); |
| 0 | 708 string s = keywords[u].name; |
| 709 TOK v = keywords[u].value; | |
| 710 StringValue* sv = stringtable.insert(s); | |
| 711 sv.ptrvalue = cast(void*) new Identifier(sv.lstring.string_, v); | |
| 712 | |
| 713 //printf("tochars[%d] = '%s'\n",v, s); | |
| 714 Token.tochars[v] = s; | |
| 715 } | |
| 716 | |
| 717 Token.tochars[TOK.TOKeof] = "EOF"; | |
| 718 Token.tochars[TOK.TOKlcurly] = "{"; | |
| 719 Token.tochars[TOK.TOKrcurly] = "}"; | |
| 720 Token.tochars[TOK.TOKlparen] = "("; | |
| 721 Token.tochars[TOK.TOKrparen] = ")"; | |
| 722 Token.tochars[TOK.TOKlbracket] = "["; | |
| 723 Token.tochars[TOK.TOKrbracket] = "]"; | |
| 724 Token.tochars[TOK.TOKsemicolon] = ";"; | |
| 725 Token.tochars[TOK.TOKcolon] = ":"; | |
| 726 Token.tochars[TOK.TOKcomma] = ","; | |
| 727 Token.tochars[TOK.TOKdot] = "."; | |
| 728 Token.tochars[TOK.TOKxor] = "^"; | |
| 729 Token.tochars[TOK.TOKxorass] = "^="; | |
| 730 Token.tochars[TOK.TOKassign] = "="; | |
| 731 Token.tochars[TOK.TOKconstruct] = "="; | |
| 732 version (DMDV2) { | |
| 733 Token.tochars[TOK.TOKblit] = "="; | |
| 734 } | |
| 735 Token.tochars[TOK.TOKlt] = "<"; | |
| 736 Token.tochars[TOK.TOKgt] = ">"; | |
| 737 Token.tochars[TOK.TOKle] = "<="; | |
| 738 Token.tochars[TOK.TOKge] = ">="; | |
| 739 Token.tochars[TOK.TOKequal] = "=="; | |
| 740 Token.tochars[TOK.TOKnotequal] = "!="; | |
| 741 Token.tochars[TOK.TOKnotidentity] = "!is"; | |
| 742 Token.tochars[TOK.TOKtobool] = "!!"; | |
| 743 | |
| 744 Token.tochars[TOK.TOKunord] = "!<>="; | |
| 745 Token.tochars[TOK.TOKue] = "!<>"; | |
| 746 Token.tochars[TOK.TOKlg] = "<>"; | |
| 747 Token.tochars[TOK.TOKleg] = "<>="; | |
| 748 Token.tochars[TOK.TOKule] = "!>"; | |
| 749 Token.tochars[TOK.TOKul] = "!>="; | |
| 750 Token.tochars[TOK.TOKuge] = "!<"; | |
| 751 Token.tochars[TOK.TOKug] = "!<="; | |
| 752 | |
| 753 Token.tochars[TOK.TOKnot] = "!"; | |
| 754 Token.tochars[TOK.TOKtobool] = "!!"; | |
| 755 Token.tochars[TOK.TOKshl] = "<<"; | |
| 756 Token.tochars[TOK.TOKshr] = ">>"; | |
| 757 Token.tochars[TOK.TOKushr] = ">>>"; | |
| 758 Token.tochars[TOK.TOKadd] = "+"; | |
| 759 Token.tochars[TOK.TOKmin] = "-"; | |
| 760 Token.tochars[TOK.TOKmul] = "*"; | |
| 761 Token.tochars[TOK.TOKdiv] = "/"; | |
| 762 Token.tochars[TOK.TOKmod] = "%"; | |
| 763 Token.tochars[TOK.TOKslice] = ".."; | |
| 764 Token.tochars[TOK.TOKdotdotdot] = "..."; | |
| 765 Token.tochars[TOK.TOKand] = "&"; | |
| 766 Token.tochars[TOK.TOKandand] = "&&"; | |
| 767 Token.tochars[TOK.TOKor] = "|"; | |
| 768 Token.tochars[TOK.TOKoror] = "||"; | |
| 769 Token.tochars[TOK.TOKarray] = "[]"; | |
| 770 Token.tochars[TOK.TOKindex] = "[i]"; | |
| 771 Token.tochars[TOK.TOKaddress] = "&"; | |
| 772 Token.tochars[TOK.TOKstar] = "*"; | |
| 773 Token.tochars[TOK.TOKtilde] = "~"; | |
| 774 Token.tochars[TOK.TOKdollar] = "$"; | |
| 775 Token.tochars[TOK.TOKcast] = "cast"; | |
| 776 Token.tochars[TOK.TOKplusplus] = "++"; | |
| 777 Token.tochars[TOK.TOKminusminus] = "--"; | |
| 778 Token.tochars[TOK.TOKtype] = "type"; | |
| 779 Token.tochars[TOK.TOKquestion] = "?"; | |
| 780 Token.tochars[TOK.TOKneg] = "-"; | |
| 781 Token.tochars[TOK.TOKuadd] = "+"; | |
| 782 Token.tochars[TOK.TOKvar] = "var"; | |
| 783 Token.tochars[TOK.TOKaddass] = "+="; | |
| 784 Token.tochars[TOK.TOKminass] = "-="; | |
| 785 Token.tochars[TOK.TOKmulass] = "*="; | |
| 786 Token.tochars[TOK.TOKdivass] = "/="; | |
| 787 Token.tochars[TOK.TOKmodass] = "%="; | |
| 788 Token.tochars[TOK.TOKshlass] = "<<="; | |
| 789 Token.tochars[TOK.TOKshrass] = ">>="; | |
| 790 Token.tochars[TOK.TOKushrass] = ">>>="; | |
| 791 Token.tochars[TOK.TOKandass] = "&="; | |
| 792 Token.tochars[TOK.TOKorass] = "|="; | |
| 793 Token.tochars[TOK.TOKcatass] = "~="; | |
| 794 Token.tochars[TOK.TOKcat] = "~"; | |
| 795 Token.tochars[TOK.TOKcall] = "call"; | |
| 796 Token.tochars[TOK.TOKidentity] = "is"; | |
| 797 Token.tochars[TOK.TOKnotidentity] = "!is"; | |
| 798 | |
| 799 Token.tochars[TOK.TOKorass] = "|="; | |
| 800 Token.tochars[TOK.TOKidentifier] = "identifier"; | |
| 801 Token.tochars[TOK.TOKat] = "@"; | |
|
130
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
802 Token.tochars[TOK.TOKpow] = "^^"; |
| 135 | 803 Token.tochars[TOK.TOKpowass] = "^^="; |
|
130
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
804 |
| 0 | 805 // For debugging |
| 73 | 806 Token.tochars[TOKerror] = "error"; |
| 0 | 807 Token.tochars[TOK.TOKdotexp] = "dotexp"; |
| 808 Token.tochars[TOK.TOKdotti] = "dotti"; | |
| 809 Token.tochars[TOK.TOKdotvar] = "dotvar"; | |
| 810 Token.tochars[TOK.TOKdottype] = "dottype"; | |
| 811 Token.tochars[TOK.TOKsymoff] = "symoff"; | |
| 812 Token.tochars[TOK.TOKarraylength] = "arraylength"; | |
| 813 Token.tochars[TOK.TOKarrayliteral] = "arrayliteral"; | |
| 814 Token.tochars[TOK.TOKassocarrayliteral] = "assocarrayliteral"; | |
| 815 Token.tochars[TOK.TOKstructliteral] = "structliteral"; | |
| 816 Token.tochars[TOK.TOKstring] = "string"; | |
| 817 Token.tochars[TOK.TOKdsymbol] = "symbol"; | |
| 818 Token.tochars[TOK.TOKtuple] = "tuple"; | |
| 819 Token.tochars[TOK.TOKdeclaration] = "declaration"; | |
| 820 Token.tochars[TOK.TOKdottd] = "dottd"; | |
| 821 Token.tochars[TOK.TOKon_scope_exit] = "scope(exit)"; | |
| 822 Token.tochars[TOK.TOKon_scope_success] = "scope(success)"; | |
| 823 Token.tochars[TOK.TOKon_scope_failure] = "scope(failure)"; | |
| 824 } | |
| 825 | |
| 826 static Identifier idPool(string s) | |
| 827 { | |
| 828 StringValue* sv = stringtable.update(s); | |
| 829 Identifier id = cast(Identifier) sv.ptrvalue; | |
| 830 if (id is null) | |
| 831 { | |
| 832 id = new Identifier(sv.lstring.string_, TOK.TOKidentifier); | |
| 833 sv.ptrvalue = cast(void*)id; | |
| 834 } | |
| 835 | |
| 836 return id; | |
| 837 } | |
| 838 | |
| 839 static Identifier uniqueId(string s) | |
| 840 { | |
| 841 static int num; | |
| 842 return uniqueId(s, ++num); | |
| 843 } | |
| 844 | |
| 845 /********************************************* | |
| 846 * Create a unique identifier using the prefix s. | |
| 847 */ | |
| 848 static Identifier uniqueId(string s, int num) | |
| 849 { | |
| 850 char buffer[32]; | |
| 851 size_t slen = s.length; | |
| 852 | |
| 853 assert(slen + num.sizeof * 3 + 1 <= buffer.sizeof); | |
| 854 int len = sprintf(buffer.ptr, "%.*s%d", s, num); | |
| 855 | |
| 856 return idPool(buffer[0..len].idup); | |
| 857 } | |
| 858 | |
| 859 TOK nextToken() | |
| 860 { | |
| 861 Token *t; | |
| 862 | |
| 863 if (token.next) | |
| 864 { | |
| 865 t = token.next; | |
| 866 memcpy(&token, t, Token.sizeof); | |
| 867 t.next = freelist; | |
| 868 freelist = t; | |
| 869 } | |
| 870 else | |
| 871 { | |
| 872 scan(&token); | |
| 873 } | |
| 874 | |
| 875 //token.print(); | |
| 876 return token.value; | |
| 877 } | |
| 878 | |
| 879 /*********************** | |
| 880 * Look ahead at next token's value. | |
| 881 */ | |
| 882 TOK peekNext() | |
| 883 { | |
| 884 return peek(&token).value; | |
| 885 } | |
| 886 | |
| 887 TOK peekNext2() | |
| 888 { | |
| 889 assert(false); | |
| 890 } | |
| 891 | |
| 892 void scan(Token* t) | |
| 893 { | |
| 894 uint lastLine = loc.linnum; | |
| 895 uint linnum; | |
| 896 | |
| 897 t.blockComment = null; | |
| 898 t.lineComment = null; | |
| 899 while (1) | |
| 900 { | |
| 901 t.ptr = p; | |
| 902 //printf("p = %p, *p = '%c'\n",p,*p); | |
| 903 switch (*p) | |
| 904 { | |
| 905 case 0: | |
| 906 case 0x1A: | |
| 907 t.value = TOK.TOKeof; // end of file | |
| 908 return; | |
| 909 | |
| 910 case ' ': | |
| 911 case '\t': | |
| 912 case '\v': | |
| 913 case '\f': | |
| 914 p++; | |
| 915 continue; // skip white space | |
| 916 | |
| 917 case '\r': | |
| 918 p++; | |
| 919 if (*p != '\n') // if CR stands by itself | |
| 920 loc.linnum++; | |
| 921 continue; // skip white space | |
| 922 | |
| 923 case '\n': | |
| 924 p++; | |
| 925 loc.linnum++; | |
| 926 continue; // skip white space | |
| 927 | |
| 928 case '0': case '1': case '2': case '3': case '4': | |
| 929 case '5': case '6': case '7': case '8': case '9': | |
| 930 t.value = number(t); | |
| 931 return; | |
| 932 | |
| 933 version (CSTRINGS) { | |
| 934 case '\'': | |
| 935 t.value = charConstant(t, 0); | |
| 936 return; | |
| 937 | |
| 938 case '"': | |
| 939 t.value = stringConstant(t,0); | |
| 940 return; | |
| 941 | |
| 942 case 'l': | |
| 943 case 'L': | |
| 944 if (p[1] == '\'') | |
| 945 { | |
| 946 p++; | |
| 947 t.value = charConstant(t, 1); | |
| 948 return; | |
| 949 } | |
| 950 else if (p[1] == '"') | |
| 951 { | |
| 952 p++; | |
| 953 t.value = stringConstant(t, 1); | |
| 954 return; | |
| 955 } | |
| 956 } else { | |
| 957 case '\'': | |
| 958 t.value = charConstant(t,0); | |
| 959 return; | |
| 960 | |
| 961 case 'r': | |
| 962 if (p[1] != '"') | |
| 963 goto case_ident; | |
| 964 p++; | |
| 965 case '`': | |
| 966 t.value = wysiwygStringConstant(t, *p); | |
| 967 return; | |
| 968 | |
| 969 case 'x': | |
| 970 if (p[1] != '"') | |
| 971 goto case_ident; | |
| 972 p++; | |
| 973 t.value = hexStringConstant(t); | |
| 974 return; | |
| 975 | |
| 976 version (DMDV2) { | |
| 977 case 'q': | |
| 978 if (p[1] == '"') | |
| 979 { | |
| 980 p++; | |
| 981 t.value = delimitedStringConstant(t); | |
| 982 return; | |
| 983 } | |
| 984 else if (p[1] == '{') | |
| 985 { | |
| 986 p++; | |
| 987 t.value = tokenStringConstant(t); | |
| 988 return; | |
| 989 } | |
| 990 else | |
| 991 goto case_ident; | |
| 992 } | |
| 993 | |
| 994 case '"': | |
| 995 t.value = escapeStringConstant(t,0); | |
| 996 return; | |
| 997 version (TEXTUAL_ASSEMBLY_OUT) { | |
| 998 } else { | |
| 999 case '\\': // escaped string literal | |
| 1000 { uint c; | |
| 1001 ubyte* pstart = p; | |
| 1002 | |
| 1003 stringbuffer.reset(); | |
| 1004 do | |
| 1005 { | |
| 1006 p++; | |
| 1007 switch (*p) | |
| 1008 { | |
| 1009 case 'u': | |
| 1010 case 'U': | |
| 1011 case '&': | |
| 1012 c = escapeSequence(); | |
| 1013 stringbuffer.writeUTF8(c); | |
| 1014 break; | |
| 1015 | |
| 1016 default: | |
| 1017 c = escapeSequence(); | |
| 1018 stringbuffer.writeByte(c); | |
| 1019 break; | |
| 1020 } | |
| 1021 } while (*p == '\\'); | |
| 1022 t.len = stringbuffer.offset; | |
| 1023 stringbuffer.writeByte(0); | |
| 2 | 1024 char* cc = cast(char*)GC.malloc(stringbuffer.offset); |
| 0 | 1025 memcpy(cc, stringbuffer.data, stringbuffer.offset); |
| 1026 t.ustring = cc; | |
| 1027 t.postfix = 0; | |
| 1028 t.value = TOK.TOKstring; | |
| 1029 if (!global.params.useDeprecated) | |
| 1030 error("Escape String literal %.*s is deprecated, use double quoted string literal \"%.*s\" instead", p - pstart, pstart, p - pstart, pstart); | |
| 1031 return; | |
| 1032 } | |
| 1033 } | |
| 1034 case 'l': | |
| 1035 case 'L': | |
| 1036 } | |
| 1037 case 'a': case 'b': case 'c': case 'd': case 'e': | |
| 1038 case 'f': case 'g': case 'h': case 'i': case 'j': | |
| 1039 case 'k': case 'm': case 'n': case 'o': | |
| 1040 version (DMDV2) { | |
| 1041 case 'p': /*case 'q': case 'r':*/ case 's': case 't': | |
| 1042 } else { | |
| 1043 case 'p': case 'q': /*case 'r':*/ case 's': case 't': | |
| 1044 } | |
| 1045 case 'u': case 'v': case 'w': /*case 'x':*/ case 'y': | |
| 1046 case 'z': | |
| 1047 case 'A': case 'B': case 'C': case 'D': case 'E': | |
| 1048 case 'F': case 'G': case 'H': case 'I': case 'J': | |
| 1049 case 'K': case 'M': case 'N': case 'O': | |
| 1050 case 'P': case 'Q': case 'R': case 'S': case 'T': | |
| 1051 case 'U': case 'V': case 'W': case 'X': case 'Y': | |
| 1052 case 'Z': | |
| 1053 case '_': | |
| 1054 case_ident: | |
| 135 | 1055 { |
| 1056 ubyte c; | |
| 0 | 1057 |
| 135 | 1058 while (1) |
| 1059 { | |
| 1060 c = *++p; | |
| 1061 if (isidchar(c)) | |
| 1062 continue; | |
| 1063 else if (c & 0x80) | |
| 1064 { | |
| 1065 ubyte *s = p; | |
| 1066 uint u = decodeUTF(); | |
| 1067 if (isUniAlpha(u)) | |
| 1068 continue; | |
| 1069 error("char 0x%04x not allowed in identifier", u); | |
| 1070 p = s; | |
| 1071 } | |
| 1072 break; | |
| 1073 } | |
| 1074 | |
| 1075 StringValue *sv = stringtable.update((cast(immutable(char)*)t.ptr)[0.. p - t.ptr]); | |
| 1076 Identifier id = cast(Identifier) sv.ptrvalue; | |
| 1077 | |
| 0 | 1078 if (id is null) |
| 1079 { id = new Identifier(sv.lstring.string_, TOK.TOKidentifier); | |
| 1080 sv.ptrvalue = cast(void*)id; | |
| 1081 } | |
| 1082 t.ident = id; | |
| 1083 t.value = cast(TOK) id.value; | |
| 1084 anyToken = 1; | |
| 1085 if (*t.ptr == '_') // if special identifier token | |
| 1086 { | |
| 1087 static char date[11+1]; | |
| 1088 static char time[8+1]; | |
| 1089 static char timestamp[24+1]; | |
| 1090 | |
| 1091 if (!date[0]) // lazy evaluation | |
| 1092 { time_t tm; | |
| 1093 char *p; | |
| 1094 | |
| 1095 .time(&tm); | |
| 1096 p = ctime(&tm); | |
| 1097 assert(p); | |
| 1098 sprintf(date.ptr, "%.6s %.4s", p + 4, p + 20); | |
| 1099 sprintf(time.ptr, "%.8s", p + 11); | |
| 1100 sprintf(timestamp.ptr, "%.24s", p); | |
| 1101 } | |
| 1102 | |
| 1103 ///version (DMDV1) { | |
| 1104 /// if (mod && id == Id.FILE) | |
| 1105 /// { | |
| 1106 /// t.ustring = cast(ubyte*)(loc.filename ? loc.filename : mod.ident.toChars()); | |
| 1107 /// goto Lstr; | |
| 1108 /// } | |
| 1109 /// else if (mod && id == Id.LINE) | |
| 1110 /// { | |
| 1111 /// t.value = TOK.TOKint64v; | |
| 1112 /// t.uns64value = loc.linnum; | |
| 1113 /// } | |
| 1114 /// else | |
| 1115 ///} | |
| 1116 if (id == Id.DATE) | |
| 1117 { | |
| 1118 t.ustring = date.ptr; | |
| 1119 goto Lstr; | |
| 1120 } | |
| 1121 else if (id == Id.TIME) | |
| 1122 { | |
| 1123 t.ustring = time.ptr; | |
| 1124 goto Lstr; | |
| 1125 } | |
| 1126 else if (id == Id.VENDOR) | |
| 1127 { | |
| 1128 t.ustring = "Digital Mars D".ptr; | |
| 1129 goto Lstr; | |
| 1130 } | |
| 1131 else if (id == Id.TIMESTAMP) | |
| 1132 { | |
| 1133 t.ustring = timestamp.ptr; | |
| 1134 Lstr: | |
| 1135 t.value = TOK.TOKstring; | |
| 1136 Llen: | |
| 1137 t.postfix = 0; | |
| 1138 t.len = strlen(cast(char*)t.ustring); | |
| 1139 } | |
| 1140 else if (id == Id.VERSIONX) | |
| 1141 { | |
| 1142 uint major = 0; | |
| 1143 uint minor = 0; | |
| 1144 | |
| 1145 foreach (char cc; global.version_[1..$]) | |
| 1146 { | |
| 1147 if (isdigit(cc)) | |
| 1148 minor = minor * 10 + cc - '0'; | |
| 1149 else if (cc == '.') | |
| 1150 { | |
| 1151 major = minor; | |
| 1152 minor = 0; | |
| 1153 } | |
| 1154 else | |
| 1155 break; | |
| 1156 } | |
| 1157 t.value = TOK.TOKint64v; | |
| 1158 t.uns64value = major * 1000 + minor; | |
| 1159 } | |
| 1160 ///version (DMDV2) { | |
| 1161 else if (id == Id.EOFX) | |
| 1162 { | |
| 1163 t.value = TOK.TOKeof; | |
| 1164 // Advance scanner to end of file | |
| 1165 while (!(*p == 0 || *p == 0x1A)) | |
| 1166 p++; | |
| 1167 } | |
| 1168 ///} | |
| 1169 } | |
| 1170 //printf("t.value = %d\n",t.value); | |
| 1171 return; | |
| 1172 } | |
| 1173 | |
| 1174 case '/': | |
| 1175 p++; | |
| 1176 switch (*p) | |
| 1177 { | |
| 1178 case '=': | |
| 1179 p++; | |
| 1180 t.value = TOK.TOKdivass; | |
| 1181 return; | |
| 1182 | |
| 1183 case '*': | |
| 1184 p++; | |
| 1185 linnum = loc.linnum; | |
| 1186 while (1) | |
| 1187 { | |
| 1188 while (1) | |
| 1189 { | |
| 1190 ubyte c = *p; | |
| 1191 switch (c) | |
| 1192 { | |
| 1193 case '/': | |
| 1194 break; | |
| 1195 | |
| 1196 case '\n': | |
| 1197 loc.linnum++; | |
| 1198 p++; | |
| 1199 continue; | |
| 1200 | |
| 1201 case '\r': | |
| 1202 p++; | |
| 1203 if (*p != '\n') | |
| 1204 loc.linnum++; | |
| 1205 continue; | |
| 1206 | |
| 1207 case 0: | |
| 1208 case 0x1A: | |
| 1209 error("unterminated /* */ comment"); | |
| 1210 p = end; | |
| 1211 t.value = TOK.TOKeof; | |
| 1212 return; | |
| 1213 | |
| 1214 default: | |
| 1215 if (c & 0x80) | |
| 1216 { uint u = decodeUTF(); | |
| 1217 if (u == PS || u == LS) | |
| 1218 loc.linnum++; | |
| 1219 } | |
| 1220 p++; | |
| 1221 continue; | |
| 1222 } | |
| 1223 break; | |
| 1224 } | |
| 1225 p++; | |
| 1226 if (p[-2] == '*' && p - 3 != t.ptr) | |
| 1227 break; | |
| 1228 } | |
| 1229 if (commentToken) | |
| 1230 { | |
| 1231 t.value = TOK.TOKcomment; | |
| 1232 return; | |
| 1233 } | |
| 1234 else if (doDocComment && t.ptr[2] == '*' && p - 4 != t.ptr) | |
| 1235 { // if /** but not /**/ | |
| 1236 getDocComment(t, lastLine == linnum); | |
| 1237 } | |
| 1238 continue; | |
| 1239 | |
| 1240 case '/': // do // style comments | |
| 1241 linnum = loc.linnum; | |
| 1242 while (1) | |
| 1243 { ubyte c = *++p; | |
| 1244 switch (c) | |
| 1245 { | |
| 1246 case '\n': | |
| 1247 break; | |
| 1248 | |
| 1249 case '\r': | |
| 1250 if (p[1] == '\n') | |
| 1251 p++; | |
| 1252 break; | |
| 1253 | |
| 1254 case 0: | |
| 1255 case 0x1A: | |
| 1256 if (commentToken) | |
| 1257 { | |
| 1258 p = end; | |
| 1259 t.value = TOK.TOKcomment; | |
| 1260 return; | |
| 1261 } | |
|
157
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
1262 if (doDocComment && t.ptr[2] == '/' || t.ptr[2] == '!') // '///' or '//!' |
| 0 | 1263 getDocComment(t, lastLine == linnum); |
| 1264 p = end; | |
| 1265 t.value = TOK.TOKeof; | |
| 1266 return; | |
| 1267 | |
| 1268 default: | |
| 1269 if (c & 0x80) | |
| 1270 { uint u = decodeUTF(); | |
| 1271 if (u == PS || u == LS) | |
| 1272 break; | |
| 1273 } | |
| 1274 continue; | |
| 1275 } | |
| 1276 break; | |
| 1277 } | |
| 1278 | |
| 1279 if (commentToken) | |
| 1280 { | |
| 1281 p++; | |
| 1282 loc.linnum++; | |
| 1283 t.value = TOK.TOKcomment; | |
| 1284 return; | |
| 1285 } | |
|
157
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
1286 if (doDocComment && t.ptr[2] == '/' || t.ptr[2] == '!') // '///' or '//!' |
| 0 | 1287 getDocComment(t, lastLine == linnum); |
| 1288 | |
| 1289 p++; | |
| 1290 loc.linnum++; | |
| 1291 continue; | |
| 1292 | |
| 1293 case '+': | |
| 1294 { | |
| 1295 int nest; | |
| 1296 | |
| 1297 linnum = loc.linnum; | |
| 1298 p++; | |
| 1299 nest = 1; | |
| 1300 while (1) | |
| 1301 { ubyte c = *p; | |
| 1302 switch (c) | |
| 1303 { | |
| 1304 case '/': | |
| 1305 p++; | |
| 1306 if (*p == '+') | |
| 1307 { | |
| 1308 p++; | |
| 1309 nest++; | |
| 1310 } | |
| 1311 continue; | |
| 1312 | |
| 1313 case '+': | |
| 1314 p++; | |
| 1315 if (*p == '/') | |
| 1316 { | |
| 1317 p++; | |
| 1318 if (--nest == 0) | |
| 1319 break; | |
| 1320 } | |
| 1321 continue; | |
| 1322 | |
| 1323 case '\r': | |
| 1324 p++; | |
| 1325 if (*p != '\n') | |
| 1326 loc.linnum++; | |
| 1327 continue; | |
| 1328 | |
| 1329 case '\n': | |
| 1330 loc.linnum++; | |
| 1331 p++; | |
| 1332 continue; | |
| 1333 | |
| 1334 case 0: | |
| 1335 case 0x1A: | |
| 1336 error("unterminated /+ +/ comment"); | |
| 1337 p = end; | |
| 1338 t.value = TOK.TOKeof; | |
| 1339 return; | |
| 1340 | |
| 1341 default: | |
| 1342 if (c & 0x80) | |
| 1343 { uint u = decodeUTF(); | |
| 1344 if (u == PS || u == LS) | |
| 1345 loc.linnum++; | |
| 1346 } | |
| 1347 p++; | |
| 1348 continue; | |
| 1349 } | |
| 1350 break; | |
| 1351 } | |
| 1352 if (commentToken) | |
| 1353 { | |
| 1354 t.value = TOK.TOKcomment; | |
| 1355 return; | |
| 1356 } | |
| 1357 if (doDocComment && t.ptr[2] == '+' && p - 4 != t.ptr) | |
| 1358 { // if /++ but not /++/ | |
| 1359 getDocComment(t, lastLine == linnum); | |
| 1360 } | |
| 1361 continue; | |
| 1362 } | |
| 1363 | |
| 1364 default: | |
| 1365 break; /// | |
| 1366 } | |
| 1367 t.value = TOK.TOKdiv; | |
| 1368 return; | |
| 1369 | |
| 1370 case '.': | |
| 1371 p++; | |
| 1372 if (isdigit(*p)) | |
| 1373 { /* Note that we don't allow ._1 and ._ as being | |
| 1374 * valid floating point numbers. | |
| 1375 */ | |
| 1376 p--; | |
| 1377 t.value = inreal(t); | |
| 1378 } | |
| 1379 else if (p[0] == '.') | |
| 1380 { | |
| 1381 if (p[1] == '.') | |
| 1382 { p += 2; | |
| 1383 t.value = TOK.TOKdotdotdot; | |
| 1384 } | |
| 1385 else | |
| 1386 { p++; | |
| 1387 t.value = TOK.TOKslice; | |
| 1388 } | |
| 1389 } | |
| 1390 else | |
| 1391 t.value = TOK.TOKdot; | |
| 1392 return; | |
| 1393 | |
| 1394 case '&': | |
| 1395 p++; | |
| 1396 if (*p == '=') | |
| 1397 { p++; | |
| 1398 t.value = TOK.TOKandass; | |
| 1399 } | |
| 1400 else if (*p == '&') | |
| 1401 { p++; | |
| 1402 t.value = TOK.TOKandand; | |
| 1403 } | |
| 1404 else | |
| 1405 t.value = TOK.TOKand; | |
| 1406 return; | |
| 1407 | |
| 1408 case '|': | |
| 1409 p++; | |
| 1410 if (*p == '=') | |
| 1411 { p++; | |
| 1412 t.value = TOK.TOKorass; | |
| 1413 } | |
| 1414 else if (*p == '|') | |
| 1415 { p++; | |
| 1416 t.value = TOK.TOKoror; | |
| 1417 } | |
| 1418 else | |
| 1419 t.value = TOK.TOKor; | |
| 1420 return; | |
| 1421 | |
| 1422 case '-': | |
| 1423 p++; | |
| 1424 if (*p == '=') | |
| 1425 { p++; | |
| 1426 t.value = TOK.TOKminass; | |
| 1427 } | |
| 1428 /// #if 0 | |
| 1429 /// else if (*p == '>') | |
| 1430 /// { p++; | |
| 1431 /// t.value = TOK.TOKarrow; | |
| 1432 /// } | |
| 1433 /// #endif | |
| 1434 else if (*p == '-') | |
| 1435 { p++; | |
| 1436 t.value = TOK.TOKminusminus; | |
| 1437 } | |
| 1438 else | |
| 1439 t.value = TOK.TOKmin; | |
| 1440 return; | |
| 1441 | |
| 1442 case '+': | |
| 1443 p++; | |
| 1444 if (*p == '=') | |
| 1445 { p++; | |
| 1446 t.value = TOK.TOKaddass; | |
| 1447 } | |
| 1448 else if (*p == '+') | |
| 1449 { p++; | |
| 1450 t.value = TOK.TOKplusplus; | |
| 1451 } | |
| 1452 else | |
| 1453 t.value = TOK.TOKadd; | |
| 1454 return; | |
| 1455 | |
| 1456 case '<': | |
| 1457 p++; | |
| 1458 if (*p == '=') | |
| 1459 { p++; | |
| 1460 t.value = TOK.TOKle; // <= | |
| 1461 } | |
| 1462 else if (*p == '<') | |
| 1463 { p++; | |
| 1464 if (*p == '=') | |
| 1465 { p++; | |
| 1466 t.value = TOK.TOKshlass; // <<= | |
| 1467 } | |
| 1468 else | |
| 1469 t.value = TOK.TOKshl; // << | |
| 1470 } | |
| 1471 else if (*p == '>') | |
| 1472 { p++; | |
| 1473 if (*p == '=') | |
| 1474 { p++; | |
| 1475 t.value = TOK.TOKleg; // <>= | |
| 1476 } | |
| 1477 else | |
| 1478 t.value = TOK.TOKlg; // <> | |
| 1479 } | |
| 1480 else | |
| 1481 t.value = TOK.TOKlt; // < | |
| 1482 return; | |
| 1483 | |
| 1484 case '>': | |
| 1485 p++; | |
| 1486 if (*p == '=') | |
| 1487 { p++; | |
| 1488 t.value = TOK.TOKge; // >= | |
| 1489 } | |
| 1490 else if (*p == '>') | |
| 1491 { p++; | |
| 1492 if (*p == '=') | |
| 1493 { p++; | |
| 1494 t.value = TOK.TOKshrass; // >>= | |
| 1495 } | |
| 1496 else if (*p == '>') | |
| 1497 { p++; | |
| 1498 if (*p == '=') | |
| 1499 { p++; | |
| 1500 t.value = TOK.TOKushrass; // >>>= | |
| 1501 } | |
| 1502 else | |
| 1503 t.value = TOK.TOKushr; // >>> | |
| 1504 } | |
| 1505 else | |
| 1506 t.value = TOK.TOKshr; // >> | |
| 1507 } | |
| 1508 else | |
| 1509 t.value = TOK.TOKgt; // > | |
| 1510 return; | |
| 1511 | |
| 1512 case '!': | |
| 1513 p++; | |
| 1514 if (*p == '=') | |
| 1515 { p++; | |
| 1516 if (*p == '=' && global.params.Dversion == 1) | |
| 1517 { p++; | |
| 1518 t.value = TOK.TOKnotidentity; // !== | |
| 1519 } | |
| 1520 else | |
| 1521 t.value = TOK.TOKnotequal; // != | |
| 1522 } | |
| 1523 else if (*p == '<') | |
| 1524 { p++; | |
| 1525 if (*p == '>') | |
| 1526 { p++; | |
| 1527 if (*p == '=') | |
| 1528 { p++; | |
| 1529 t.value = TOK.TOKunord; // !<>= | |
| 1530 } | |
| 1531 else | |
| 1532 t.value = TOK.TOKue; // !<> | |
| 1533 } | |
| 1534 else if (*p == '=') | |
| 1535 { p++; | |
| 1536 t.value = TOK.TOKug; // !<= | |
| 1537 } | |
| 1538 else | |
| 1539 t.value = TOK.TOKuge; // !< | |
| 1540 } | |
| 1541 else if (*p == '>') | |
| 1542 { p++; | |
| 1543 if (*p == '=') | |
| 1544 { p++; | |
| 1545 t.value = TOK.TOKul; // !>= | |
| 1546 } | |
| 1547 else | |
| 1548 t.value = TOK.TOKule; // !> | |
| 1549 } | |
| 1550 else | |
| 1551 t.value = TOK.TOKnot; // ! | |
| 1552 return; | |
| 1553 | |
| 1554 case '=': | |
| 1555 p++; | |
| 1556 if (*p == '=') | |
| 1557 { p++; | |
| 1558 if (*p == '=' && global.params.Dversion == 1) | |
| 1559 { p++; | |
| 1560 t.value = TOK.TOKidentity; // === | |
| 1561 } | |
| 1562 else | |
| 1563 t.value = TOK.TOKequal; // == | |
| 1564 } | |
| 1565 else | |
| 1566 t.value = TOK.TOKassign; // = | |
| 1567 return; | |
| 1568 | |
| 1569 case '~': | |
| 1570 p++; | |
| 1571 if (*p == '=') | |
| 1572 { p++; | |
| 1573 t.value = TOK.TOKcatass; // ~= | |
| 1574 } | |
| 1575 else | |
| 1576 t.value = TOK.TOKtilde; // ~ | |
| 1577 return; | |
|
130
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1578 |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1579 version(DMDV2) { |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1580 case '^': |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1581 p++; |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1582 if (*p == '^') |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1583 { p++; |
| 135 | 1584 if (*p == '=') |
| 1585 { p++; | |
| 1586 t.value = TOKpowass; // ^^= | |
| 1587 } | |
| 1588 else | |
|
130
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1589 t.value = TOKpow; // ^^ |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1590 } |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1591 else if (*p == '=') |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1592 { p++; |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1593 t.value = TOKxorass; // ^= |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1594 } |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1595 else |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1596 t.value = TOKxor; // ^ |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1597 return; |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1598 } |
|
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1599 |
| 0 | 1600 /* |
| 1601 #define SINGLE(c,tok) case c: p++; t.value = tok; return; | |
| 1602 | |
| 1603 SINGLE('(', TOKlparen) | |
| 1604 SINGLE(')', TOKrparen) | |
| 1605 SINGLE('[', TOKlbracket) | |
| 1606 SINGLE(']', TOKrbracket) | |
| 1607 SINGLE('{', TOKlcurly) | |
| 1608 SINGLE('}', TOKrcurly) | |
| 1609 SINGLE('?', TOKquestion) | |
| 1610 SINGLE(',', TOKcomma) | |
| 1611 SINGLE(';', TOKsemicolon) | |
| 1612 SINGLE(':', TOKcolon) | |
| 1613 SINGLE('$', TOKdollar) | |
| 1614 SINGLE('@', TOKat) | |
| 1615 | |
| 1616 #undef SINGLE | |
| 1617 | |
| 1618 #define DOUBLE(c1,tok1,c2,tok2) \ | |
| 1619 case c1: \ | |
| 1620 p++; \ | |
| 1621 if (*p == c2) \ | |
| 1622 { p++; \ | |
| 1623 t.value = tok2; \ | |
| 1624 } \ | |
| 1625 else \ | |
| 1626 t.value = tok1; \ | |
| 1627 return; | |
| 1628 | |
| 1629 DOUBLE('*', TOKmul, '=', TOKmulass) | |
| 1630 DOUBLE('%', TOKmod, '=', TOKmodass) | |
|
130
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1631 #if DMDV1 |
| 0 | 1632 DOUBLE('^', TOKxor, '=', TOKxorass) |
|
130
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1633 #endif |
| 0 | 1634 #undef DOUBLE |
| 1635 */ | |
| 1636 | |
| 1637 case '(': p++; t.value = TOK.TOKlparen; return; | |
| 1638 case ')': p++; t.value = TOK.TOKrparen; return; | |
| 1639 case '[': p++; t.value = TOK.TOKlbracket; return; | |
| 1640 case ']': p++; t.value = TOK.TOKrbracket; return; | |
| 1641 case '{': p++; t.value = TOK.TOKlcurly; return; | |
| 1642 case '}': p++; t.value = TOK.TOKrcurly; return; | |
| 1643 case '?': p++; t.value = TOK.TOKquestion; return; | |
| 1644 case ',': p++; t.value = TOK.TOKcomma; return; | |
| 1645 case ';': p++; t.value = TOK.TOKsemicolon; return; | |
| 1646 case ':': p++; t.value = TOK.TOKcolon; return; | |
| 1647 case '$': p++; t.value = TOK.TOKdollar; return; | |
| 1648 case '@': p++; t.value = TOK.TOKat; return; | |
| 1649 | |
| 1650 case '*': | |
| 1651 p++; | |
| 1652 if (*p == '=') { | |
| 1653 p++; | |
| 1654 t.value = TOK.TOKmulass; | |
| 1655 } else { | |
| 1656 t.value = TOK.TOKmul; | |
| 1657 } | |
| 1658 return; | |
| 1659 | |
| 1660 case '%': | |
| 1661 p++; | |
| 1662 if (*p == '=') { | |
| 1663 p++; | |
| 1664 t.value = TOK.TOKmodass; | |
| 1665 } else { | |
| 1666 t.value = TOK.TOKmod; | |
| 1667 } | |
| 1668 return; | |
|
130
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1669 version(DMDV1) { |
| 0 | 1670 case '^': |
| 1671 p++; | |
| 1672 if (*p == '=') { | |
| 1673 p++; | |
| 1674 t.value = TOK.TOKxorass; | |
| 1675 } else { | |
| 1676 t.value = TOK.TOKxor; | |
| 1677 } | |
| 1678 return; | |
|
130
60bb0fe4563e
dmdfe 2.037 first main iteration
Eldar Insafutdinov <e.insafutdinov@gmail.com>
parents:
114
diff
changeset
|
1679 } |
| 0 | 1680 case '#': |
| 1681 p++; | |
| 1682 pragma_(); | |
| 1683 continue; | |
| 1684 | |
| 1685 default: | |
| 135 | 1686 { uint c = *p; |
| 0 | 1687 |
| 1688 if (c & 0x80) | |
| 135 | 1689 { c = decodeUTF(); |
| 0 | 1690 |
| 1691 // Check for start of unicode identifier | |
| 135 | 1692 if (isUniAlpha(c)) |
| 0 | 1693 goto case_ident; |
| 1694 | |
| 135 | 1695 if (c == PS || c == LS) |
| 0 | 1696 { |
| 1697 loc.linnum++; | |
| 1698 p++; | |
| 1699 continue; | |
| 1700 } | |
| 1701 } | |
| 135 | 1702 if (c < 0x80 && isprint(c)) |
| 0 | 1703 error("unsupported char '%c'", c); |
| 1704 else | |
| 1705 error("unsupported char 0x%02x", c); | |
| 1706 p++; | |
| 1707 continue; | |
| 1708 } | |
| 1709 } | |
| 1710 } | |
| 1711 } | |
| 1712 | |
| 1713 Token* peek(Token* ct) | |
| 1714 { | |
| 1715 Token* t; | |
| 1716 | |
| 1717 if (ct.next) | |
| 1718 t = ct.next; | |
| 1719 else | |
| 1720 { | |
| 1721 t = new Token(); | |
| 1722 scan(t); | |
| 1723 t.next = null; | |
| 1724 ct.next = t; | |
| 1725 } | |
| 1726 return t; | |
| 1727 } | |
| 1728 | |
| 1729 Token* peekPastParen(Token* tk) | |
| 1730 { | |
| 1731 //printf("peekPastParen()\n"); | |
| 1732 int parens = 1; | |
| 1733 int curlynest = 0; | |
| 1734 while (1) | |
| 1735 { | |
| 1736 tk = peek(tk); | |
| 1737 //tk.print(); | |
| 1738 switch (tk.value) | |
| 1739 { | |
| 1740 case TOK.TOKlparen: | |
| 1741 parens++; | |
| 1742 continue; | |
| 1743 | |
| 1744 case TOK.TOKrparen: | |
| 1745 --parens; | |
| 1746 if (parens) | |
| 1747 continue; | |
| 1748 tk = peek(tk); | |
| 1749 break; | |
| 1750 | |
| 1751 case TOK.TOKlcurly: | |
| 1752 curlynest++; | |
| 1753 continue; | |
| 1754 | |
| 1755 case TOK.TOKrcurly: | |
| 1756 if (--curlynest >= 0) | |
| 1757 continue; | |
| 1758 break; | |
| 1759 | |
| 1760 case TOK.TOKsemicolon: | |
| 1761 if (curlynest) | |
| 1762 continue; | |
| 1763 break; | |
| 1764 | |
| 1765 case TOK.TOKeof: | |
| 1766 break; | |
| 1767 | |
| 1768 default: | |
| 1769 continue; | |
| 1770 } | |
| 1771 return tk; | |
| 1772 } | |
| 1773 } | |
| 1774 | |
| 1775 /******************************************* | |
| 1776 * Parse escape sequence. | |
| 1777 */ | |
| 1778 uint escapeSequence() | |
| 1779 { | |
| 1780 uint c = *p; | |
| 1781 | |
| 1782 version (TEXTUAL_ASSEMBLY_OUT) { | |
| 1783 return c; | |
| 1784 } | |
| 1785 int n; | |
| 1786 int ndigits; | |
| 1787 | |
| 1788 switch (c) | |
| 1789 { | |
| 1790 case '\'': | |
| 1791 case '"': | |
| 1792 case '?': | |
| 1793 case '\\': | |
| 1794 Lconsume: | |
| 1795 p++; | |
| 1796 break; | |
| 1797 | |
| 1798 case 'a': c = 7; goto Lconsume; | |
| 1799 case 'b': c = 8; goto Lconsume; | |
| 1800 case 'f': c = 12; goto Lconsume; | |
| 1801 case 'n': c = 10; goto Lconsume; | |
| 1802 case 'r': c = 13; goto Lconsume; | |
| 1803 case 't': c = 9; goto Lconsume; | |
| 1804 case 'v': c = 11; goto Lconsume; | |
| 1805 | |
| 1806 case 'u': | |
| 1807 ndigits = 4; | |
| 1808 goto Lhex; | |
| 1809 case 'U': | |
| 1810 ndigits = 8; | |
| 1811 goto Lhex; | |
| 1812 case 'x': | |
| 1813 ndigits = 2; | |
| 1814 Lhex: | |
| 1815 p++; | |
| 1816 c = *p; | |
| 1817 if (ishex(cast(ubyte)c)) | |
| 1818 { | |
| 1819 uint v; | |
| 1820 | |
| 1821 n = 0; | |
| 1822 v = 0; | |
| 1823 while (1) | |
| 1824 { | |
| 1825 if (isdigit(c)) | |
| 1826 c -= '0'; | |
| 1827 else if (islower(c)) | |
| 1828 c -= 'a' - 10; | |
| 1829 else | |
| 1830 c -= 'A' - 10; | |
| 1831 v = v * 16 + c; | |
| 1832 c = *++p; | |
| 1833 if (++n == ndigits) | |
| 1834 break; | |
| 1835 if (!ishex(cast(ubyte)c)) | |
| 1836 { error("escape hex sequence has %d hex digits instead of %d", n, ndigits); | |
| 1837 break; | |
| 1838 } | |
| 1839 } | |
| 1840 if (ndigits != 2 && !utf_isValidDchar(v)) | |
| 1841 { error("invalid UTF character \\U%08x", v); | |
| 1842 v = '?'; // recover with valid UTF character | |
| 1843 } | |
| 1844 c = v; | |
| 1845 } | |
| 1846 else | |
| 1847 error("undefined escape hex sequence \\%c\n",c); | |
| 1848 break; | |
| 1849 | |
| 1850 case '&': // named character entity | |
| 1851 for (ubyte* idstart = ++p; true; p++) | |
| 1852 { | |
| 1853 switch (*p) | |
| 1854 { | |
| 1855 case ';': | |
| 1856 c = HtmlNamedEntity(idstart, p - idstart); | |
| 1857 if (c == ~0) | |
| 1858 { | |
| 1859 error("unnamed character entity &%s;", idstart[0..(p - idstart)]); | |
| 1860 c = ' '; | |
| 1861 } | |
| 1862 p++; | |
| 1863 break; | |
| 1864 | |
| 1865 default: | |
| 1866 if (isalpha(*p) || | |
| 1867 (p != idstart + 1 && isdigit(*p))) | |
| 1868 continue; | |
| 1869 error("unterminated named entity"); | |
| 1870 break; | |
| 1871 } | |
| 1872 break; | |
| 1873 } | |
| 1874 break; | |
| 1875 | |
| 1876 case 0: | |
| 1877 case 0x1A: // end of file | |
| 1878 c = '\\'; | |
| 1879 break; | |
| 1880 | |
| 1881 default: | |
| 1882 if (isoctal(cast(ubyte)c)) | |
| 1883 { | |
| 1884 uint v; | |
| 1885 | |
| 1886 n = 0; | |
| 1887 v = 0; | |
| 1888 do | |
| 1889 { | |
| 1890 v = v * 8 + (c - '0'); | |
| 1891 c = *++p; | |
| 1892 } while (++n < 3 && isoctal(cast(ubyte)c)); | |
| 1893 c = v; | |
| 1894 if (c > 0xFF) | |
| 1895 error("0%03o is larger than a byte", c); | |
| 1896 } | |
| 1897 else | |
| 1898 error("undefined escape sequence \\%c\n",c); | |
| 1899 break; | |
| 1900 } | |
| 1901 return c; | |
| 1902 } | |
| 1903 | |
| 1904 TOK wysiwygStringConstant(Token* t, int tc) | |
| 1905 { | |
|
8
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1906 uint c; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1907 Loc start = loc; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1908 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1909 p++; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1910 stringbuffer.reset(); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1911 while (true) |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1912 { |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1913 c = *p++; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1914 switch (c) |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1915 { |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1916 case '\n': |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1917 loc.linnum++; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1918 break; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1919 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1920 case '\r': |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1921 if (*p == '\n') |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1922 continue; // ignore |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1923 c = '\n'; // treat EndOfLine as \n character |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1924 loc.linnum++; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1925 break; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1926 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1927 case 0: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1928 case 0x1A: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1929 error("unterminated string constant starting at %s", start.toChars()); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1930 t.ustring = "".ptr; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1931 t.len = 0; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1932 t.postfix = 0; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1933 return TOKstring; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1934 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1935 case '"': |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1936 case '`': |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1937 if (c == tc) |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1938 { |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1939 t.len = stringbuffer.offset; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1940 stringbuffer.writeByte(0); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1941 char* tmp = cast(char*)GC.malloc(stringbuffer.offset); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1942 memcpy(tmp, stringbuffer.data, stringbuffer.offset); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1943 t.ustring = tmp; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1944 stringPostfix(t); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1945 return TOKstring; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1946 } |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1947 break; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1948 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1949 default: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1950 if (c & 0x80) |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1951 { p--; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1952 uint u = decodeUTF(); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1953 p++; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1954 if (u == PS || u == LS) |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1955 loc.linnum++; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1956 stringbuffer.writeUTF8(u); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1957 continue; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1958 } |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1959 break; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1960 } |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1961 stringbuffer.writeByte(c); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1962 } |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
1963 |
| 0 | 1964 assert(false); |
| 1965 } | |
| 1966 | |
| 51 | 1967 /************************************** |
| 1968 * Lex hex strings: | |
| 1969 * x"0A ae 34FE BD" | |
| 1970 */ | |
| 0 | 1971 TOK hexStringConstant(Token* t) |
| 1972 { | |
| 51 | 1973 uint c; |
| 1974 Loc start = loc; | |
| 1975 uint n = 0; | |
| 1976 uint v; | |
| 1977 | |
| 1978 p++; | |
| 1979 stringbuffer.reset(); | |
| 1980 while (1) | |
| 1981 { | |
| 1982 c = *p++; | |
| 1983 switch (c) | |
| 1984 { | |
| 1985 case ' ': | |
| 1986 case '\t': | |
| 1987 case '\v': | |
| 1988 case '\f': | |
| 1989 continue; // skip white space | |
| 1990 | |
| 1991 case '\r': | |
| 1992 if (*p == '\n') | |
| 1993 continue; // ignore | |
| 1994 // Treat isolated '\r' as if it were a '\n' | |
| 1995 case '\n': | |
| 1996 loc.linnum++; | |
| 1997 continue; | |
| 1998 | |
| 1999 case 0: | |
| 2000 case 0x1A: | |
| 2001 error("unterminated string constant starting at %s", start.toChars()); | |
| 2002 t.ustring = "".ptr; | |
| 2003 t.len = 0; | |
| 2004 t.postfix = 0; | |
| 2005 return TOKstring; | |
| 2006 | |
| 2007 case '"': | |
| 2008 if (n & 1) | |
| 2009 { | |
| 2010 error("odd number (%d) of hex characters in hex string", n); | |
| 2011 stringbuffer.writeByte(v); | |
| 2012 } | |
| 2013 t.len = stringbuffer.offset; | |
| 2014 stringbuffer.writeByte(0); | |
| 2015 void* mem = malloc(stringbuffer.offset); | |
| 2016 memcpy(mem, stringbuffer.data, stringbuffer.offset); | |
| 2017 t.ustring = cast(const(char)*)mem; | |
| 2018 stringPostfix(t); | |
| 2019 return TOKstring; | |
| 2020 | |
| 2021 default: | |
| 2022 if (c >= '0' && c <= '9') | |
| 2023 c -= '0'; | |
| 2024 else if (c >= 'a' && c <= 'f') | |
| 2025 c -= 'a' - 10; | |
| 2026 else if (c >= 'A' && c <= 'F') | |
| 2027 c -= 'A' - 10; | |
| 2028 else if (c & 0x80) | |
| 2029 { p--; | |
| 2030 uint u = decodeUTF(); | |
| 2031 p++; | |
| 2032 if (u == PS || u == LS) | |
| 2033 loc.linnum++; | |
| 2034 else | |
| 135 | 2035 error("non-hex character \\u%04x", u); |
| 51 | 2036 } |
| 2037 else | |
| 2038 error("non-hex character '%c'", c); | |
| 2039 if (n & 1) | |
| 2040 { v = (v << 4) | c; | |
| 2041 stringbuffer.writeByte(v); | |
| 2042 } | |
| 2043 else | |
| 2044 v = c; | |
| 2045 n++; | |
| 2046 break; | |
| 2047 } | |
| 2048 } | |
| 0 | 2049 } |
| 2050 | |
| 2051 version (DMDV2) { | |
| 51 | 2052 /************************************** |
| 2053 * Lex delimited strings: | |
| 2054 * q"(foo(xxx))" // "foo(xxx)" | |
| 2055 * q"[foo(]" // "foo(" | |
| 2056 * q"/foo]/" // "foo]" | |
| 2057 * q"HERE | |
| 2058 * foo | |
| 2059 * HERE" // "foo\n" | |
| 2060 * Input: | |
| 2061 * p is on the " | |
| 2062 */ | |
| 0 | 2063 TOK delimitedStringConstant(Token* t) |
| 2064 { | |
| 51 | 2065 uint c; |
| 2066 Loc start = loc; | |
| 2067 uint delimleft = 0; | |
| 2068 uint delimright = 0; | |
| 2069 uint nest = 1; | |
| 2070 uint nestcount; | |
| 2071 Identifier hereid = null; | |
| 2072 uint blankrol = 0; | |
| 2073 uint startline = 0; | |
| 2074 | |
| 2075 p++; | |
| 2076 stringbuffer.reset(); | |
| 2077 while (1) | |
| 2078 { | |
| 2079 c = *p++; | |
| 2080 //printf("c = '%c'\n", c); | |
| 2081 switch (c) | |
| 2082 { | |
| 2083 case '\n': | |
| 2084 Lnextline: | |
| 2085 loc.linnum++; | |
| 2086 startline = 1; | |
| 2087 if (blankrol) | |
| 2088 { blankrol = 0; | |
| 2089 continue; | |
| 2090 } | |
| 2091 if (hereid) | |
| 2092 { | |
| 2093 stringbuffer.writeUTF8(c); | |
| 2094 continue; | |
| 2095 } | |
| 2096 break; | |
| 2097 | |
| 2098 case '\r': | |
| 2099 if (*p == '\n') | |
| 2100 continue; // ignore | |
| 2101 c = '\n'; // treat EndOfLine as \n character | |
| 2102 goto Lnextline; | |
| 2103 | |
| 2104 case 0: | |
| 2105 case 0x1A: | |
| 2106 goto Lerror; | |
| 2107 | |
| 2108 default: | |
| 2109 if (c & 0x80) | |
| 2110 { p--; | |
| 2111 c = decodeUTF(); | |
| 2112 p++; | |
| 2113 if (c == PS || c == LS) | |
| 2114 goto Lnextline; | |
| 2115 } | |
| 2116 break; | |
| 2117 } | |
| 2118 if (delimleft == 0) | |
| 2119 { | |
| 2120 delimleft = c; | |
| 2121 nest = 1; | |
| 2122 nestcount = 1; | |
| 2123 if (c == '(') | |
| 2124 delimright = ')'; | |
| 2125 else if (c == '{') | |
| 2126 delimright = '}'; | |
| 2127 else if (c == '[') | |
| 2128 delimright = ']'; | |
| 2129 else if (c == '<') | |
| 2130 delimright = '>'; | |
| 2131 else if (isalpha(c) || c == '_' || (c >= 0x80 && isUniAlpha(c))) | |
| 2132 { | |
| 2133 // Start of identifier; must be a heredoc | |
| 2134 Token t2; | |
| 2135 p--; | |
| 2136 scan(&t2); // read in heredoc identifier | |
| 2137 if (t2.value != TOKidentifier) | |
| 2138 { | |
| 2139 error("identifier expected for heredoc, not %s", t2.toChars()); | |
| 2140 delimright = c; | |
| 2141 } | |
| 2142 else | |
| 2143 { | |
| 2144 hereid = t2.ident; | |
| 2145 //printf("hereid = '%s'\n", hereid.toChars()); | |
| 2146 blankrol = 1; | |
| 2147 } | |
| 2148 nest = 0; | |
| 2149 } | |
| 2150 else | |
| 2151 { | |
| 2152 delimright = c; | |
| 2153 nest = 0; | |
| 2154 if (isspace(c)) | |
| 2155 error("delimiter cannot be whitespace"); | |
| 2156 } | |
| 2157 } | |
| 2158 else | |
| 2159 { | |
| 2160 if (blankrol) | |
| 2161 { | |
| 2162 error("heredoc rest of line should be blank"); | |
| 2163 blankrol = 0; | |
| 2164 continue; | |
| 2165 } | |
| 2166 if (nest == 1) | |
| 2167 { | |
| 2168 if (c == delimleft) | |
| 2169 nestcount++; | |
| 2170 else if (c == delimright) | |
| 2171 { nestcount--; | |
| 2172 if (nestcount == 0) | |
| 2173 goto Ldone; | |
| 2174 } | |
| 2175 } | |
| 2176 else if (c == delimright) | |
| 2177 goto Ldone; | |
| 2178 if (startline && isalpha(c) && hereid) | |
| 2179 { | |
| 2180 Token t2; | |
| 2181 ubyte* psave = p; | |
| 2182 p--; | |
| 2183 scan(&t2); // read in possible heredoc identifier | |
| 2184 //printf("endid = '%s'\n", t2.ident.toChars()); | |
| 2185 if (t2.value == TOKidentifier && t2.ident.equals(hereid)) | |
| 2186 { | |
| 2187 /* should check that rest of line is blank | |
| 2188 */ | |
| 2189 goto Ldone; | |
| 2190 } | |
| 2191 p = psave; | |
| 2192 } | |
| 2193 stringbuffer.writeUTF8(c); | |
| 2194 startline = 0; | |
| 2195 } | |
| 2196 } | |
| 2197 | |
| 2198 Ldone: | |
| 2199 if (*p == '"') | |
| 2200 p++; | |
| 2201 else | |
| 2202 error("delimited string must end in %c\"", delimright); | |
| 2203 t.len = stringbuffer.offset; | |
| 2204 stringbuffer.writeByte(0); | |
| 2205 void* mem = malloc(stringbuffer.offset); | |
| 2206 memcpy(mem, stringbuffer.data, stringbuffer.offset); | |
| 2207 t.ustring = cast(const(char)*)mem; | |
| 2208 stringPostfix(t); | |
| 2209 return TOKstring; | |
| 2210 | |
| 2211 Lerror: | |
| 2212 error("unterminated string constant starting at %s", start.toChars()); | |
| 2213 t.ustring = "".ptr; | |
| 2214 t.len = 0; | |
| 2215 t.postfix = 0; | |
| 2216 return TOKstring; | |
| 0 | 2217 } |
| 2218 | |
|
8
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2219 /************************************** |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2220 * Lex delimited strings: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2221 * q{ foo(xxx) } // " foo(xxx) " |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2222 * q{foo(} // "foo(" |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2223 * q{{foo}"}"} // "{foo}"}"" |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2224 * Input: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2225 * p is on the q |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2226 */ |
| 0 | 2227 TOK tokenStringConstant(Token* t) |
| 2228 { | |
|
8
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2229 uint nest = 1; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2230 Loc start = loc; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2231 ubyte* pstart = ++p; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2232 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2233 while (true) |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2234 { |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2235 Token tok; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2236 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2237 scan(&tok); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2238 switch (tok.value) |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2239 { |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2240 case TOKlcurly: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2241 nest++; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2242 continue; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2243 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2244 case TOKrcurly: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2245 if (--nest == 0) |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2246 goto Ldone; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2247 continue; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2248 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2249 case TOKeof: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2250 goto Lerror; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2251 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2252 default: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2253 continue; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2254 } |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2255 } |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2256 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2257 Ldone: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2258 t.len = p - 1 - pstart; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2259 char* tmp = cast(char*)GC.malloc(t.len + 1); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2260 memcpy(tmp, pstart, t.len); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2261 tmp[t.len] = 0; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2262 t.ustring = tmp; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2263 stringPostfix(t); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2264 return TOKstring; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2265 |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2266 Lerror: |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2267 error("unterminated token string constant starting at %s", start.toChars()); |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2268 t.ustring = "".ptr; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2269 t.len = 0; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2270 t.postfix = 0; |
|
d42cd5917df4
wysiwyg strings, alias this, templates, TypeSlice implementation
dkoroskin <>
parents:
4
diff
changeset
|
2271 return TOKstring; |
| 0 | 2272 } |
| 2273 } | |
| 2274 TOK escapeStringConstant(Token* t, int wide) | |
| 2275 { | |
| 2276 uint c; | |
| 2277 Loc start = loc; | |
| 2278 | |
| 2279 p++; | |
| 2280 stringbuffer.reset(); | |
| 2281 while (true) | |
| 2282 { | |
| 2283 c = *p++; | |
| 2284 switch (c) | |
| 2285 { | |
| 2286 version (TEXTUAL_ASSEMBLY_OUT) { | |
| 2287 } else { | |
| 2288 case '\\': | |
| 2289 switch (*p) | |
| 2290 { | |
| 2291 case 'u': | |
| 2292 case 'U': | |
| 2293 case '&': | |
| 2294 c = escapeSequence(); | |
| 2295 stringbuffer.writeUTF8(c); | |
| 2296 continue; | |
| 2297 | |
| 2298 default: | |
| 2299 c = escapeSequence(); | |
| 2300 break; | |
| 2301 } | |
| 2302 break; | |
| 2303 } | |
| 2304 case '\n': | |
| 2305 loc.linnum++; | |
| 2306 break; | |
| 2307 | |
| 2308 case '\r': | |
| 2309 if (*p == '\n') | |
| 2310 continue; // ignore | |
| 2311 c = '\n'; // treat EndOfLine as \n character | |
| 2312 loc.linnum++; | |
| 2313 break; | |
| 2314 | |
| 2315 case '"': | |
| 2316 t.len = stringbuffer.offset; | |
| 2317 stringbuffer.writeByte(0); | |
| 2 | 2318 char* tmp = cast(char*)GC.malloc(stringbuffer.offset); |
| 0 | 2319 memcpy(tmp, stringbuffer.data, stringbuffer.offset); |
| 2320 t.ustring = tmp; | |
| 2321 stringPostfix(t); | |
| 2322 return TOK.TOKstring; | |
| 2323 | |
| 2324 case 0: | |
| 2325 case 0x1A: | |
| 2326 p--; | |
| 2327 error("unterminated string constant starting at %s", start.toChars()); | |
| 2328 t.ustring = "".ptr; | |
| 2329 t.len = 0; | |
| 2330 t.postfix = 0; | |
| 2331 return TOK.TOKstring; | |
| 2332 | |
| 2333 default: | |
| 2334 if (c & 0x80) | |
| 2335 { | |
| 2336 p--; | |
| 2337 c = decodeUTF(); | |
| 2338 if (c == LS || c == PS) | |
| 2339 { c = '\n'; | |
| 2340 loc.linnum++; | |
| 2341 } | |
| 2342 p++; | |
| 2343 stringbuffer.writeUTF8(c); | |
| 2344 continue; | |
| 2345 } | |
| 2346 break; | |
| 2347 } | |
| 2348 stringbuffer.writeByte(c); | |
| 2349 } | |
| 2350 | |
| 2351 assert(false); | |
| 2352 } | |
| 2353 | |
| 2354 TOK charConstant(Token* t, int wide) | |
| 2355 { | |
| 2356 uint c; | |
| 2357 TOK tk = TOKcharv; | |
| 2358 | |
| 2359 //printf("Lexer.charConstant\n"); | |
| 2360 p++; | |
| 2361 c = *p++; | |
| 2362 switch (c) | |
| 2363 { | |
| 2364 version (TEXTUAL_ASSEMBLY_OUT) { | |
| 2365 } else { | |
| 2366 case '\\': | |
| 2367 switch (*p) | |
| 2368 { | |
| 2369 case 'u': | |
| 2370 t.uns64value = escapeSequence(); | |
| 2371 tk = TOKwcharv; | |
| 2372 break; | |
| 2373 | |
| 2374 case 'U': | |
| 2375 case '&': | |
| 2376 t.uns64value = escapeSequence(); | |
| 2377 tk = TOKdcharv; | |
| 2378 break; | |
| 2379 | |
| 2380 default: | |
| 2381 t.uns64value = escapeSequence(); | |
| 2382 break; | |
| 2383 } | |
| 2384 break; | |
| 2385 } | |
| 2386 case '\n': | |
| 2387 L1: | |
| 2388 loc.linnum++; | |
| 2389 case '\r': | |
| 2390 case 0: | |
| 2391 case 0x1A: | |
| 2392 case '\'': | |
| 2393 error("unterminated character constant"); | |
| 2394 return tk; | |
| 2395 | |
| 2396 default: | |
| 2397 if (c & 0x80) | |
| 2398 { | |
| 2399 p--; | |
| 2400 c = decodeUTF(); | |
| 2401 p++; | |
| 2402 if (c == LS || c == PS) | |
| 2403 goto L1; | |
| 2404 if (c < 0xD800 || (c >= 0xE000 && c < 0xFFFE)) | |
| 2405 tk = TOKwcharv; | |
| 2406 else | |
| 2407 tk = TOKdcharv; | |
| 2408 } | |
| 2409 t.uns64value = c; | |
| 2410 break; | |
| 2411 } | |
| 2412 | |
| 2413 if (*p != '\'') | |
| 2414 { | |
| 2415 error("unterminated character constant"); | |
| 2416 return tk; | |
| 2417 } | |
| 2418 p++; | |
| 2419 return tk; | |
| 2420 } | |
| 2421 | |
| 2422 /*************************************** | |
| 2423 * Get postfix of string literal. | |
| 2424 */ | |
| 2425 void stringPostfix(Token* t) | |
| 2426 { | |
| 2427 switch (*p) | |
| 2428 { | |
| 2429 case 'c': | |
| 2430 case 'w': | |
| 2431 case 'd': | |
| 2432 t.postfix = *p; | |
| 2433 p++; | |
| 2434 break; | |
| 2435 | |
| 2436 default: | |
| 2437 t.postfix = 0; | |
| 2438 break; | |
| 2439 } | |
| 2440 } | |
| 2441 | |
| 2442 uint wchar_(uint u) | |
| 2443 { | |
| 2444 assert(false); | |
| 2445 } | |
| 2446 | |
| 2447 /************************************** | |
| 2448 * Read in a number. | |
| 2449 * If it's an integer, store it in tok.TKutok.Vlong. | |
| 2450 * integers can be decimal, octal or hex | |
| 2451 * Handle the suffixes U, UL, LU, L, etc. | |
| 2452 * If it's double, store it in tok.TKutok.Vdouble. | |
| 2453 * Returns: | |
| 2454 * TKnum | |
| 2455 * TKdouble,... | |
| 2456 */ | |
| 2457 | |
| 2458 TOK number(Token* t) | |
| 2459 { | |
| 2460 // We use a state machine to collect numbers | |
| 2461 enum STATE { STATE_initial, STATE_0, STATE_decimal, STATE_octal, STATE_octale, | |
| 2462 STATE_hex, STATE_binary, STATE_hex0, STATE_binary0, | |
| 2463 STATE_hexh, STATE_error }; | |
| 2464 STATE state; | |
| 2465 | |
| 2466 enum FLAGS | |
| 2467 { | |
| 2468 FLAGS_undefined = 0, | |
| 2469 FLAGS_decimal = 1, // decimal | |
| 2470 FLAGS_unsigned = 2, // u or U suffix | |
| 2471 FLAGS_long = 4, // l or L suffix | |
| 2472 }; | |
| 2473 | |
| 2474 FLAGS flags = FLAGS.FLAGS_decimal; | |
| 2475 | |
| 2476 int i; | |
| 2477 int base; | |
| 2478 uint c; | |
| 2479 ubyte *start; | |
| 2480 TOK result; | |
| 2481 | |
| 2482 //printf("Lexer.number()\n"); | |
| 2483 state = STATE.STATE_initial; | |
| 2484 base = 0; | |
| 2485 stringbuffer.reset(); | |
| 2486 start = p; | |
| 2487 while (1) | |
| 2488 { | |
| 2489 c = *p; | |
| 2490 switch (state) | |
| 2491 { | |
| 2492 case STATE.STATE_initial: // opening state | |
| 2493 if (c == '0') | |
| 2494 state = STATE.STATE_0; | |
| 2495 else | |
| 2496 state = STATE.STATE_decimal; | |
| 2497 break; | |
| 2498 | |
| 2499 case STATE.STATE_0: | |
| 2500 flags = (flags & ~FLAGS.FLAGS_decimal); | |
| 2501 switch (c) | |
| 2502 { | |
| 2503 version (ZEROH) { | |
| 2504 case 'H': // 0h | |
| 2505 case 'h': | |
| 2506 goto hexh; | |
| 2507 } | |
| 2508 case 'X': | |
| 2509 case 'x': | |
| 2510 state = STATE.STATE_hex0; | |
| 2511 break; | |
| 2512 | |
| 2513 case '.': | |
| 2514 if (p[1] == '.') // .. is a separate token | |
| 2515 goto done; | |
| 2516 case 'i': | |
| 2517 case 'f': | |
| 2518 case 'F': | |
| 2519 goto real_; | |
| 2520 version (ZEROH) { | |
| 2521 case 'E': | |
| 2522 case 'e': | |
| 2523 goto case_hex; | |
| 2524 } | |
| 2525 case 'B': | |
| 2526 case 'b': | |
| 2527 state = STATE.STATE_binary0; | |
| 2528 break; | |
| 2529 | |
| 2530 case '0': case '1': case '2': case '3': | |
| 2531 case '4': case '5': case '6': case '7': | |
| 2532 state = STATE.STATE_octal; | |
| 2533 break; | |
| 2534 | |
| 2535 version (ZEROH) { | |
| 2536 case '8': case '9': case 'A': | |
| 2537 case 'C': case 'D': case 'F': | |
| 2538 case 'a': case 'c': case 'd': case 'f': | |
| 2539 case_hex: | |
| 2540 state = STATE.STATE_hexh; | |
| 2541 break; | |
| 2542 } | |
| 2543 case '_': | |
| 2544 state = STATE.STATE_octal; | |
| 2545 p++; | |
| 2546 continue; | |
| 2547 | |
| 2548 case 'L': | |
| 2549 if (p[1] == 'i') | |
| 2550 goto real_; | |
| 2551 goto done; | |
| 2552 | |
| 2553 default: | |
| 2554 goto done; | |
| 2555 } | |
| 2556 break; | |
| 2557 | |
| 2558 case STATE.STATE_decimal: // reading decimal number | |
| 2559 if (!isdigit(c)) | |
| 2560 { | |
| 2561 version (ZEROH) { | |
| 2562 if (ishex(c) | |
| 2563 || c == 'H' || c == 'h' | |
| 2564 ) | |
| 2565 goto hexh; | |
| 2566 } | |
| 2567 if (c == '_') // ignore embedded _ | |
| 2568 { p++; | |
| 2569 continue; | |
| 2570 } | |
| 2571 if (c == '.' && p[1] != '.') | |
| 2572 goto real_; | |
| 2573 else if (c == 'i' || c == 'f' || c == 'F' || | |
| 2574 c == 'e' || c == 'E') | |
| 2575 { | |
| 2576 real_: // It's a real number. Back up and rescan as a real | |
| 2577 p = start; | |
| 2578 return inreal(t); | |
| 2579 } | |
| 2580 else if (c == 'L' && p[1] == 'i') | |
| 2581 goto real_; | |
| 2582 goto done; | |
| 2583 } | |
| 2584 break; | |
| 2585 | |
| 2586 case STATE.STATE_hex0: // reading hex number | |
| 2587 case STATE.STATE_hex: | |
| 2588 if (! ishex(cast(ubyte)c)) | |
| 2589 { | |
| 2590 if (c == '_') // ignore embedded _ | |
| 2591 { p++; | |
| 2592 continue; | |
| 2593 } | |
| 2594 if (c == '.' && p[1] != '.') | |
| 2595 goto real_; | |
| 2596 if (c == 'P' || c == 'p' || c == 'i') | |
| 2597 goto real_; | |
| 2598 if (state == STATE.STATE_hex0) | |
| 2599 error("Hex digit expected, not '%c'", c); | |
| 2600 goto done; | |
| 2601 } | |
| 2602 state = STATE.STATE_hex; | |
| 2603 break; | |
| 2604 | |
| 2605 version (ZEROH) { | |
| 2606 hexh: | |
| 2607 state = STATE.STATE_hexh; | |
| 2608 case STATE.STATE_hexh: // parse numbers like 0FFh | |
| 2609 if (!ishex(c)) | |
| 2610 { | |
| 2611 if (c == 'H' || c == 'h') | |
| 2612 { | |
| 2613 p++; | |
| 2614 base = 16; | |
| 2615 goto done; | |
| 2616 } | |
| 2617 else | |
| 2618 { | |
| 2619 // Check for something like 1E3 or 0E24 | |
| 2620 if (memchr(cast(char*)stringbuffer.data, 'E', stringbuffer.offset) || | |
| 2621 memchr(cast(char*)stringbuffer.data, 'e', stringbuffer.offset)) | |
| 2622 goto real_; | |
| 2623 error("Hex digit expected, not '%c'", c); | |
| 2624 goto done; | |
| 2625 } | |
| 2626 } | |
| 2627 break; | |
| 2628 } | |
| 2629 | |
| 2630 case STATE.STATE_octal: // reading octal number | |
| 2631 case STATE.STATE_octale: // reading octal number with non-octal digits | |
| 2632 if (!isoctal(cast(ubyte)c)) | |
| 2633 { | |
| 2634 version (ZEROH) { | |
| 2635 if (ishex(c) | |
| 2636 || c == 'H' || c == 'h' | |
| 2637 ) | |
| 2638 goto hexh; | |
| 2639 } | |
| 2640 if (c == '_') // ignore embedded _ | |
| 2641 { p++; | |
| 2642 continue; | |
| 2643 } | |
| 2644 if (c == '.' && p[1] != '.') | |
| 2645 goto real_; | |
| 2646 if (c == 'i') | |
| 2647 goto real_; | |
| 2648 if (isdigit(c)) | |
| 2649 { | |
| 2650 state = STATE.STATE_octale; | |
| 2651 } | |
| 2652 else | |
| 2653 goto done; | |
| 2654 } | |
| 2655 break; | |
| 2656 | |
| 2657 case STATE.STATE_binary0: // starting binary number | |
| 2658 case STATE.STATE_binary: // reading binary number | |
| 2659 if (c != '0' && c != '1') | |
| 2660 { | |
| 2661 version (ZEROH) { | |
| 2662 if (ishex(c) | |
| 2663 || c == 'H' || c == 'h' | |
| 2664 ) | |
| 2665 goto hexh; | |
| 2666 } | |
| 2667 if (c == '_') // ignore embedded _ | |
| 2668 { p++; | |
| 2669 continue; | |
| 2670 } | |
| 2671 if (state == STATE.STATE_binary0) | |
| 2672 { error("binary digit expected"); | |
| 2673 state = STATE.STATE_error; | |
| 2674 break; | |
| 2675 } | |
| 2676 else | |
| 2677 goto done; | |
| 2678 } | |
| 2679 state = STATE.STATE_binary; | |
| 2680 break; | |
| 2681 | |
| 2682 case STATE.STATE_error: // for error recovery | |
| 2683 if (!isdigit(c)) // scan until non-digit | |
| 2684 goto done; | |
| 2685 break; | |
| 2686 | |
| 2687 default: | |
| 2688 assert(0); | |
| 2689 } | |
| 2690 stringbuffer.writeByte(c); | |
| 2691 p++; | |
| 2692 } | |
| 2693 done: | |
| 2694 stringbuffer.writeByte(0); // terminate string | |
| 2695 if (state == STATE.STATE_octale) | |
| 2696 error("Octal digit expected"); | |
| 2697 | |
| 2698 ulong n; // unsigned >=64 bit integer type | |
| 2699 | |
| 2700 if (stringbuffer.offset == 2 && (state == STATE.STATE_decimal || state == STATE.STATE_0)) | |
| 2701 n = stringbuffer.data[0] - '0'; | |
| 2702 else | |
| 2703 { | |
| 2704 // Convert string to integer | |
| 2705 version (__DMC__) { | |
| 2706 errno = 0; | |
| 2707 n = strtoull(cast(char*)stringbuffer.data,null,base); | |
| 2708 if (errno == ERANGE) | |
| 2709 error("integer overflow"); | |
| 2710 } else { | |
| 2711 // Not everybody implements strtoull() | |
| 2712 char* p = cast(char*)stringbuffer.data; | |
| 2713 int r = 10, d; | |
| 2714 | |
| 2715 if (*p == '0') | |
| 2716 { | |
| 2717 if (p[1] == 'x' || p[1] == 'X') | |
| 2718 p += 2, r = 16; | |
| 2719 else if (p[1] == 'b' || p[1] == 'B') | |
| 2720 p += 2, r = 2; | |
| 2721 else if (isdigit(p[1])) | |
| 2722 p += 1, r = 8; | |
| 2723 } | |
| 2724 | |
| 2725 n = 0; | |
| 2726 while (1) | |
| 2727 { | |
| 2728 if (*p >= '0' && *p <= '9') | |
| 2729 d = *p - '0'; | |
| 2730 else if (*p >= 'a' && *p <= 'z') | |
| 2731 d = *p - 'a' + 10; | |
| 2732 else if (*p >= 'A' && *p <= 'Z') | |
| 2733 d = *p - 'A' + 10; | |
| 2734 else | |
| 2735 break; | |
| 2736 if (d >= r) | |
| 2737 break; | |
| 2738 ulong n2 = n * r; | |
| 2739 //printf("n2 / r = %llx, n = %llx\n", n2/r, n); | |
| 2740 if (n2 / r != n || n2 + d < n) | |
| 2741 { | |
| 2742 error ("integer overflow"); | |
| 2743 break; | |
| 2744 } | |
| 2745 | |
| 2746 n = n2 + d; | |
| 2747 p++; | |
| 2748 } | |
| 2749 } | |
| 2750 if (n.sizeof > 8 && | |
| 2751 n > 0xFFFFFFFFFFFFFFFF) // if n needs more than 64 bits | |
| 2752 error("integer overflow"); | |
| 2753 } | |
| 2754 | |
| 2755 // Parse trailing 'u', 'U', 'l' or 'L' in any combination | |
| 2756 while (1) | |
| 2757 { FLAGS f; | |
| 2758 | |
| 2759 switch (*p) | |
| 2760 { case 'U': | |
| 2761 case 'u': | |
| 2762 f = FLAGS.FLAGS_unsigned; | |
| 2763 goto L1; | |
| 2764 | |
| 2765 case 'l': | |
| 2766 if (1 || !global.params.useDeprecated) | |
| 2767 error("'l' suffix is deprecated, use 'L' instead"); | |
| 2768 case 'L': | |
| 2769 f = FLAGS.FLAGS_long; | |
| 2770 L1: | |
| 2771 p++; | |
| 2772 if (flags & f) | |
| 2773 error("unrecognized token"); | |
| 2774 flags = (flags | f); | |
| 2775 continue; | |
| 2776 default: | |
| 2777 break; | |
| 2778 } | |
| 2779 break; | |
| 2780 } | |
| 2781 | |
| 2782 switch (flags) | |
| 2783 { | |
| 2784 case FLAGS.FLAGS_undefined: | |
| 2785 /* Octal or Hexadecimal constant. | |
| 2786 * First that fits: int, uint, long, ulong | |
| 2787 */ | |
| 2788 if (n & 0x8000000000000000) | |
| 2789 result = TOK.TOKuns64v; | |
| 2790 else if (n & 0xFFFFFFFF00000000) | |
| 2791 result = TOK.TOKint64v; | |
| 2792 else if (n & 0x80000000) | |
| 2793 result = TOK.TOKuns32v; | |
| 2794 else | |
| 2795 result = TOK.TOKint32v; | |
| 2796 break; | |
| 2797 | |
| 2798 case FLAGS.FLAGS_decimal: | |
| 2799 /* First that fits: int, long, long long | |
| 2800 */ | |
| 2801 if (n & 0x8000000000000000) | |
| 2802 { error("signed integer overflow"); | |
| 2803 result = TOK.TOKuns64v; | |
| 2804 } | |
| 2805 else if (n & 0xFFFFFFFF80000000) | |
| 2806 result = TOK.TOKint64v; | |
| 2807 else | |
| 2808 result = TOK.TOKint32v; | |
| 2809 break; | |
| 2810 | |
| 2811 case FLAGS.FLAGS_unsigned: | |
| 2812 case FLAGS.FLAGS_decimal | FLAGS.FLAGS_unsigned: | |
| 2813 /* First that fits: uint, ulong | |
| 2814 */ | |
| 2815 if (n & 0xFFFFFFFF00000000) | |
| 2816 result = TOK.TOKuns64v; | |
| 2817 else | |
| 2818 result = TOK.TOKuns32v; | |
| 2819 break; | |
| 2820 | |
| 2821 case FLAGS.FLAGS_decimal | FLAGS.FLAGS_long: | |
| 2822 if (n & 0x8000000000000000) | |
| 2823 { error("signed integer overflow"); | |
| 2824 result = TOK.TOKuns64v; | |
| 2825 } | |
| 2826 else | |
| 2827 result = TOK.TOKint64v; | |
| 2828 break; | |
| 2829 | |
| 2830 case FLAGS.FLAGS_long: | |
| 2831 if (n & 0x8000000000000000) | |
| 2832 result = TOK.TOKuns64v; | |
| 2833 else | |
| 2834 result = TOK.TOKint64v; | |
| 2835 break; | |
| 2836 | |
| 2837 case FLAGS.FLAGS_unsigned | FLAGS.FLAGS_long: | |
| 2838 case FLAGS.FLAGS_decimal | FLAGS.FLAGS_unsigned | FLAGS.FLAGS_long: | |
| 2839 result = TOK.TOKuns64v; | |
| 2840 break; | |
| 2841 | |
| 2842 default: | |
| 2843 debug { | |
| 2844 printf("%x\n",flags); | |
| 2845 } | |
| 2846 assert(0); | |
| 2847 } | |
| 2848 t.uns64value = n; | |
| 2849 return result; | |
| 2850 } | |
| 2851 | |
| 2852 /************************************** | |
| 2853 * Read in characters, converting them to real. | |
| 2854 * Bugs: | |
| 2855 * Exponent overflow not detected. | |
| 2856 * Too much requested precision is not detected. | |
| 2857 */ | |
| 2858 TOK inreal(Token* t) | |
| 2859 in | |
| 2860 { | |
| 2861 assert(*p == '.' || isdigit(*p)); | |
| 2862 } | |
| 2863 out (result) | |
| 2864 { | |
| 2865 switch (result) | |
| 2866 { | |
| 2867 case TOKfloat32v: | |
| 2868 case TOKfloat64v: | |
| 2869 case TOKfloat80v: | |
| 2870 case TOKimaginary32v: | |
| 2871 case TOKimaginary64v: | |
| 2872 case TOKimaginary80v: | |
| 2873 break; | |
| 2874 | |
| 2875 default: | |
| 2876 assert(0); | |
| 2877 } | |
| 2878 } | |
| 2879 body | |
| 2880 { | |
| 2881 int dblstate; | |
| 2882 uint c; | |
| 2883 char hex; // is this a hexadecimal-floating-constant? | |
| 2884 TOK result; | |
| 2885 | |
| 2886 //printf("Lexer.inreal()\n"); | |
| 2887 stringbuffer.reset(); | |
| 2888 dblstate = 0; | |
| 2889 hex = 0; | |
| 2890 Lnext: | |
| 2891 while (true) | |
| 2892 { | |
| 2893 // Get next char from input | |
| 2894 c = *p++; | |
| 2895 //printf("dblstate = %d, c = '%c'\n", dblstate, c); | |
| 2896 while (true) | |
| 2897 { | |
| 2898 switch (dblstate) | |
| 2899 { | |
| 2900 case 0: // opening state | |
| 2901 if (c == '0') | |
| 2902 dblstate = 9; | |
| 2903 else if (c == '.') | |
| 2904 dblstate = 3; | |
| 2905 else | |
| 2906 dblstate = 1; | |
| 2907 break; | |
| 2908 | |
| 2909 case 9: | |
| 2910 dblstate = 1; | |
| 2911 if (c == 'X' || c == 'x') | |
| 2912 { | |
| 2913 hex++; | |
| 2914 break; | |
| 2915 } | |
| 2916 case 1: // digits to left of . | |
| 2917 case 3: // digits to right of . | |
| 2918 case 7: // continuing exponent digits | |
| 2919 if (!isdigit(c) && !(hex && isxdigit(c))) | |
| 2920 { | |
| 2921 if (c == '_') | |
| 2922 goto Lnext; // ignore embedded '_' | |
| 2923 dblstate++; | |
| 2924 continue; | |
| 2925 } | |
| 2926 break; | |
| 2927 | |
| 2928 case 2: // no more digits to left of . | |
| 2929 if (c == '.') | |
| 2930 { | |
| 2931 dblstate++; | |
| 2932 break; | |
| 2933 } | |
| 2934 case 4: // no more digits to right of . | |
| 2935 if ((c == 'E' || c == 'e') || | |
| 2936 hex && (c == 'P' || c == 'p')) | |
| 2937 { | |
| 2938 dblstate = 5; | |
| 2939 hex = 0; // exponent is always decimal | |
| 2940 break; | |
| 2941 } | |
| 2942 if (hex) | |
| 2943 error("binary-exponent-part required"); | |
| 2944 goto done; | |
| 2945 | |
| 2946 case 5: // looking immediately to right of E | |
| 2947 dblstate++; | |
| 2948 if (c == '-' || c == '+') | |
| 2949 break; | |
| 2950 case 6: // 1st exponent digit expected | |
| 2951 if (!isdigit(c)) | |
| 2952 error("exponent expected"); | |
| 2953 dblstate++; | |
| 2954 break; | |
| 2955 | |
| 2956 case 8: // past end of exponent digits | |
| 2957 goto done; | |
| 79 | 2958 |
| 2959 default: | |
| 2960 assert(0, "inreal.dblstate has unexpected value"); | |
| 0 | 2961 } |
| 2962 break; | |
| 2963 } | |
| 2964 stringbuffer.writeByte(c); | |
| 2965 } | |
| 2966 done: | |
| 2967 p--; | |
| 2968 | |
| 2969 stringbuffer.writeByte(0); | |
| 2970 | |
| 114 | 2971 version (Windows) { /// && __DMC__ |
| 0 | 2972 char* save = __locale_decpoint; |
| 2973 __locale_decpoint = cast(char*)".".ptr; | |
| 2974 } | |
| 2975 t.float80value = strtold(cast(char*)stringbuffer.data, null); | |
| 2976 | |
| 2977 errno = 0; | |
| 2978 switch (*p) | |
| 2979 { | |
| 2980 case 'F': | |
| 2981 case 'f': | |
| 2982 strtof(cast(char*)stringbuffer.data, null); | |
| 2983 result = TOKfloat32v; | |
| 2984 p++; | |
| 2985 break; | |
| 2986 | |
| 2987 default: | |
| 2988 strtod(cast(char*)stringbuffer.data, null); | |
| 2989 result = TOKfloat64v; | |
| 2990 break; | |
| 2991 | |
| 2992 case 'l': | |
| 2993 if (!global.params.useDeprecated) | |
| 2994 error("'l' suffix is deprecated, use 'L' instead"); | |
| 2995 case 'L': | |
| 2996 result = TOKfloat80v; | |
| 2997 p++; | |
| 2998 break; | |
| 2999 } | |
| 3000 if (*p == 'i' || *p == 'I') | |
| 3001 { | |
| 3002 if (!global.params.useDeprecated && *p == 'I') | |
| 3003 error("'I' suffix is deprecated, use 'i' instead"); | |
| 3004 p++; | |
| 3005 switch (result) | |
| 3006 { | |
| 3007 case TOKfloat32v: | |
| 3008 result = TOKimaginary32v; | |
| 3009 break; | |
| 3010 case TOKfloat64v: | |
| 3011 result = TOKimaginary64v; | |
| 3012 break; | |
| 3013 case TOKfloat80v: | |
| 3014 result = TOKimaginary80v; | |
| 3015 break; | |
|
157
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3016 default: |
| 0 | 3017 } |
| 3018 } | |
| 3019 | |
| 114 | 3020 version (Windows) { ///&& __DMC__ |
| 0 | 3021 __locale_decpoint = save; |
| 3022 } | |
| 3023 if (errno == ERANGE) | |
| 3024 error("number is not representable"); | |
| 3025 | |
| 3026 return result; | |
| 3027 } | |
| 3028 | |
| 3029 void error(T...)(string format, T t) | |
| 3030 { | |
| 3031 error(this.loc, format, t); | |
| 3032 } | |
| 3033 | |
| 3034 void error(T...)(Loc loc, string format, T t) | |
| 3035 { | |
| 3036 if (mod && !global.gag) | |
| 3037 { | |
| 3038 string p = loc.toChars(); | |
| 3039 if (p.length != 0) | |
| 3040 writef("%s: ", p); | |
| 3041 | |
| 3042 writefln(format, t); | |
| 3043 | |
| 3044 if (global.errors >= 20) // moderate blizzard of cascading messages | |
| 3045 fatal(); | |
| 3046 } | |
| 3047 | |
| 3048 global.errors++; | |
| 3049 } | |
| 3050 | |
| 162 | 3051 /********************************************* |
| 3052 * Do pragma. | |
| 3053 * Currently, the only pragma supported is: | |
| 3054 * #line linnum [filespec] | |
| 3055 */ | |
| 0 | 3056 void pragma_() |
| 3057 { | |
| 162 | 3058 Token tok; |
| 3059 int linnum; | |
| 3060 string filespec = null; | |
| 3061 Loc loc = this.loc; | |
| 3062 | |
| 3063 scan(&tok); | |
| 3064 if (tok.value != TOKidentifier || tok.ident != Id.line) | |
| 3065 goto Lerr; | |
| 3066 | |
| 3067 scan(&tok); | |
| 3068 if (tok.value == TOKint32v || tok.value == TOKint64v) | |
| 3069 linnum = cast(int)(tok.uns64value - 1); /// | |
| 3070 else | |
| 3071 goto Lerr; | |
| 3072 | |
| 3073 while (1) | |
| 3074 { | |
| 3075 switch (*p) | |
| 3076 { | |
| 3077 case 0: | |
| 3078 case 0x1A: | |
| 3079 case '\n': | |
| 3080 Lnewline: | |
| 3081 this.loc.linnum = linnum; | |
| 3082 if (filespec != null) | |
| 3083 this.loc.filename = filespec; | |
| 3084 return; | |
| 3085 | |
| 3086 case '\r': | |
| 3087 p++; | |
| 3088 if (*p != '\n') | |
| 3089 { p--; | |
| 3090 goto Lnewline; | |
| 3091 } | |
| 3092 continue; | |
| 3093 | |
| 3094 case ' ': | |
| 3095 case '\t': | |
| 3096 case '\v': | |
| 3097 case '\f': | |
| 3098 p++; | |
| 3099 continue; // skip white space | |
| 3100 | |
| 3101 case '_': | |
| 3102 if (mod && memcmp(p, "__FILE__".ptr, 8) == 0) | |
| 3103 { | |
| 3104 p += 8; | |
| 3105 filespec = (loc.filename ? loc.filename : mod.ident.toChars()); | |
| 3106 } | |
| 3107 continue; | |
| 3108 | |
| 3109 case '"': | |
| 3110 if (filespec) | |
| 3111 goto Lerr; | |
| 3112 stringbuffer.reset(); | |
| 3113 p++; | |
| 3114 while (1) | |
| 3115 { | |
| 3116 uint c; | |
| 3117 | |
| 3118 c = *p; | |
| 3119 switch (c) | |
| 3120 { | |
| 3121 case '\n': | |
| 3122 case '\r': | |
| 3123 case 0: | |
| 3124 case 0x1A: | |
| 3125 goto Lerr; | |
| 3126 | |
| 3127 case '"': | |
| 3128 stringbuffer.writeByte(0); | |
| 3129 filespec = stringbuffer.extractString(); /// | |
| 3130 p++; | |
| 3131 break; | |
| 3132 | |
| 3133 default: | |
| 3134 if (c & 0x80) | |
| 3135 { | |
| 3136 uint u = decodeUTF(); | |
| 3137 if (u == PS || u == LS) | |
| 3138 goto Lerr; | |
| 3139 } | |
| 3140 stringbuffer.writeByte(c); | |
| 3141 p++; | |
| 3142 continue; | |
| 3143 } | |
| 3144 break; | |
| 3145 } | |
| 3146 continue; | |
| 3147 | |
| 3148 default: | |
| 3149 if (*p & 0x80) | |
| 3150 { | |
| 3151 uint u = decodeUTF(); | |
| 3152 if (u == PS || u == LS) | |
| 3153 goto Lnewline; | |
| 3154 } | |
| 3155 goto Lerr; | |
| 3156 } | |
| 3157 } | |
| 3158 | |
| 3159 Lerr: | |
| 3160 error(loc, "#line integer [\"filespec\"]\\n expected"); | |
| 0 | 3161 } |
| 3162 | |
| 49 | 3163 /******************************************** |
| 3164 * Decode UTF character. | |
| 3165 * Issue error messages for invalid sequences. | |
| 3166 * Return decoded character, advance p to last character in UTF sequence. | |
| 3167 */ | |
| 0 | 3168 uint decodeUTF() |
| 3169 { | |
| 49 | 3170 dchar u; |
| 3171 ubyte c; | |
| 3172 ubyte* s = p; | |
| 3173 size_t len; | |
| 3174 size_t idx; | |
| 3175 string msg; | |
| 3176 | |
| 3177 c = *s; | |
| 3178 assert(c & 0x80); | |
| 3179 | |
| 3180 // Check length of remaining string up to 6 UTF-8 characters | |
| 3181 for (len = 1; len < 6 && s[len]; len++) { | |
| 3182 ; | |
| 3183 } | |
| 3184 | |
| 3185 idx = 0; | |
| 3186 msg = utf_decodeChar(cast(string)s[0..len], &idx, &u); | |
| 3187 p += idx - 1; | |
| 3188 if (msg) | |
| 3189 { | |
| 3190 error("%s", msg); | |
| 3191 } | |
| 3192 return u; | |
| 0 | 3193 } |
| 3194 | |
|
157
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3195 /*************************************************** |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3196 * Parse doc comment embedded between t.ptr and p. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3197 * Remove trailing blanks and tabs from lines. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3198 * Replace all newlines with \n. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3199 * Remove leading comment character from each line. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3200 * Decide if it's a lineComment or a blockComment. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3201 * Append to previous one for this token. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3202 */ |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3203 void getDocComment(Token* t, uint lineComment) |
| 0 | 3204 { |
|
157
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3205 /* ct tells us which kind of comment it is: '!', '/', '*', or '+' |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3206 */ |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3207 ubyte ct = t.ptr[2]; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3208 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3209 /* Start of comment text skips over / * *, / + +, or / / / |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3210 */ |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3211 ubyte* q = t.ptr + 3; // start of comment text |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3212 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3213 ubyte* qend = p; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3214 if (ct == '*' || ct == '+') |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3215 qend -= 2; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3216 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3217 /* Scan over initial row of ****'s or ++++'s or ////'s |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3218 */ |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3219 for (; q < qend; q++) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3220 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3221 if (*q != ct) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3222 break; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3223 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3224 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3225 /* Remove trailing row of ****'s or ++++'s |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3226 */ |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3227 if (ct != '/' && ct != '!') |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3228 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3229 for (; q < qend; qend--) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3230 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3231 if (qend[-1] != ct) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3232 break; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3233 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3234 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3235 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3236 /* Comment is now [q .. qend]. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3237 * Canonicalize it into buf[]. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3238 */ |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3239 OutBuffer buf = new OutBuffer; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3240 int linestart = 0; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3241 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3242 for (; q < qend; q++) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3243 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3244 ubyte c = *q; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3245 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3246 switch (c) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3247 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3248 case '*': |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3249 case '+': |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3250 if (linestart && c == ct) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3251 { linestart = 0; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3252 /* Trim preceding whitespace up to preceding \n |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3253 */ |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3254 while (buf.offset && (buf.data[buf.offset - 1] == ' ' || buf.data[buf.offset - 1] == '\t')) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3255 buf.offset--; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3256 continue; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3257 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3258 break; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3259 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3260 case ' ': |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3261 case '\t': |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3262 break; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3263 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3264 case '\r': |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3265 if (q[1] == '\n') |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3266 continue; // skip the \r |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3267 goto Lnewline; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3268 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3269 default: |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3270 if (c == 226) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3271 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3272 // If LS or PS |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3273 if (q[1] == 128 && |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3274 (q[2] == 168 || q[2] == 169)) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3275 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3276 q += 2; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3277 goto Lnewline; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3278 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3279 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3280 linestart = 0; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3281 break; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3282 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3283 Lnewline: |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3284 c = '\n'; // replace all newlines with \n |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3285 case '\n': |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3286 linestart = 1; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3287 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3288 /* Trim trailing whitespace |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3289 */ |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3290 while (buf.offset && (buf.data[buf.offset - 1] == ' ' || buf.data[buf.offset - 1] == '\t')) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3291 buf.offset--; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3292 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3293 break; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3294 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3295 buf.writeByte(c); |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3296 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3297 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3298 // Always end with a newline |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3299 if (!buf.offset || buf.data[buf.offset - 1] != '\n') |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3300 buf.writeByte('\n'); |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3301 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3302 buf.writeByte(0); |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3303 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3304 // It's a line comment if the start of the doc comment comes |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3305 // after other non-whitespace on the same line. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3306 string* dc = (lineComment && anyToken) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3307 ? &t.lineComment |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3308 : &t.blockComment; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3309 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3310 // Combine with previous doc comment, if any |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3311 if (*dc) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3312 *dc = combineComments(*dc, cast(string) buf.data[0 .. buf.size]); // TODO: utf decode etc? |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3313 else |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3314 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3315 auto bufsize = buf.size; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3316 *dc = cast(string) buf.extractData()[0..bufsize]; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3317 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3318 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3319 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3320 /******************************************** |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3321 * Combine two document comments into one, |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3322 * separated by a newline. |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3323 */ |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3324 static string combineComments(string c1, string c2) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3325 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3326 //printf("Lexer::combineComments('%s', '%s')\n", c1, c2); |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3327 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3328 string c = c2; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3329 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3330 if (c1) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3331 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3332 c = c1; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3333 if (c2) |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3334 { |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3335 size_t len1 = c1.length; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3336 size_t len2 = c2.length; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3337 |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3338 c = c1.idup; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3339 if (len1 && c1[$-1] != '\n') |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3340 c ~= '\n'; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3341 c ~= c2; |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3342 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3343 } |
|
b7b61140701d
* added all missing default cases in switch statements
trass3r
parents:
135
diff
changeset
|
3344 return c; |
| 0 | 3345 } |
| 3346 | |
| 3347 static bool isValidIdentifier(string p) | |
| 3348 { | |
| 3349 if (p.length == 0) { | |
| 3350 return false; | |
| 3351 } | |
| 3352 | |
| 3353 if (p[0] >= '0' && p[0] <= '9') { // beware of isdigit() on signed chars | |
| 3354 return false; | |
| 3355 } | |
| 3356 | |
| 3357 size_t idx = 0; | |
| 3358 while (idx < p.length) | |
| 3359 { | |
| 3360 dchar dc; | |
| 3361 | |
| 3362 if (utf_decodeChar(p, &idx, &dc) !is null) { | |
| 3363 return false; | |
| 3364 } | |
| 3365 | |
| 3366 if (!((dc >= 0x80 && isUniAlpha(dc)) || isalnum(dc) || dc == '_')) { | |
| 3367 return false; | |
| 3368 } | |
| 3369 } | |
| 3370 | |
| 3371 return true; | |
| 3372 } | |
| 3373 | |
| 79 | 3374 /// TODO: use normal string append when GC works |
| 3375 static string combineComments(const(char)[] c1, const(char)[] c2) | |
| 0 | 3376 { |
| 79 | 3377 //writef("Lexer.combineComments('%s', '%s')\n", c1, c2); |
| 0 | 3378 |
| 79 | 3379 char[] c = cast(char[]) c2; |
| 3380 | |
| 3381 if (c1 !is null) | |
| 0 | 3382 { |
| 79 | 3383 c = cast(char[]) c1; |
| 3384 if (c2 !is null) | |
| 0 | 3385 { |
| 79 | 3386 c = cast(char[]) (GC.malloc(c1.length + 1 + c2.length)[0 .. c1.length + 1 + c2.length]); |
| 3387 size_t len1 = c1.length; | |
| 3388 c[0..len1] = c1[]; | |
| 3389 c[len1++] = '\n'; | |
| 3390 c[len1 .. len1 + c2.length] = c2[]; | |
| 0 | 3391 } |
| 3392 } | |
| 79 | 3393 |
| 3394 return cast(string)c; | |
| 0 | 3395 } |
| 3396 } |