Mercurial > hg > index.cgi
annotate lwcc/lex.c @ 299:856caf91ffaa ccdev
Added token list structure and switched some stuff to use it
Swithced to using a token list structure instead of manually fiddling
pointers throughout the macro expansion code. Also fixed up some problematic
things related to stringification and concatenation.
| author | William Astle <lost@l-w.ca> |
|---|---|
| date | Sun, 15 Sep 2013 13:06:00 -0600 |
| parents | 6112c67728ba |
| children | 8d6c47395653 |
| rev | line source |
|---|---|
| 295 | 1 /* |
| 2 lwcc/lex.c | |
| 3 | |
| 4 Copyright © 2013 William Astle | |
| 5 | |
| 6 This file is part of LWTOOLS. | |
| 7 | |
| 8 LWTOOLS is free software: you can redistribute it and/or modify it under the | |
| 9 terms of the GNU General Public License as published by the Free Software | |
| 10 Foundation, either version 3 of the License, or (at your option) any later | |
| 11 version. | |
| 12 | |
| 13 This program is distributed in the hope that it will be useful, but WITHOUT | |
| 14 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or | |
| 15 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for | |
| 16 more details. | |
| 17 | |
| 18 You should have received a copy of the GNU General Public License along with | |
| 19 this program. If not, see <http://www.gnu.org/licenses/>. | |
| 20 */ | |
| 21 | |
| 22 #include <ctype.h> | |
| 23 #include <stdio.h> | |
| 24 | |
| 25 #include <lw_alloc.h> | |
| 26 | |
| 27 #include "cpp.h" | |
| 28 #include "strbuf.h" | |
| 29 #include "token.h" | |
| 30 | |
| 31 /* fetch a raw input byte from the current file. Will return CPP_EOF if | |
| 32 EOF is encountered and CPP_EOL if an end of line sequence is encountered. | |
| 33 End of line is defined as either CR, CRLF, LF, or LFCR. CPP_EOL is | |
| 34 returned on the first CR or LF encountered. The complementary CR or LF | |
| 35 is munched, if present, when the *next* character is read. This always | |
| 36 operates on file_stack. | |
| 37 | |
| 38 This function also accounts for line numbers in input files and also | |
| 39 character columns. | |
| 40 */ | |
| 41 static int fetch_byte_ll(struct preproc_info *pp) | |
| 42 { | |
| 43 int c; | |
| 44 | |
| 45 if (pp -> eolstate != 0) | |
| 46 { | |
| 47 pp -> lineno++; | |
| 48 pp -> column = 0; | |
| 49 } | |
| 50 c = getc(pp -> fp); | |
| 51 pp -> column++; | |
| 52 if (pp -> eolstate == 1) | |
| 53 { | |
| 54 // just saw CR, munch LF | |
| 55 if (c == 10) | |
| 56 c = getc(pp -> fp); | |
| 57 pp -> eolstate = 0; | |
| 58 } | |
| 59 else if (pp -> eolstate == 2) | |
| 60 { | |
| 61 // just saw LF, much CR | |
| 62 if (c == 13) | |
| 63 c = getc(pp -> fp); | |
| 64 pp -> eolstate = 0; | |
| 65 } | |
| 66 | |
| 67 if (c == 10) | |
| 68 { | |
| 69 // we have LF - end of line, flag to munch CR | |
| 70 pp -> eolstate = 2; | |
| 71 c = CPP_EOL; | |
| 72 } | |
| 73 else if (c == 13) | |
| 74 { | |
| 75 // we have CR - end of line, flag to munch LF | |
| 76 pp -> eolstate = 1; | |
| 77 c = CPP_EOL; | |
| 78 } | |
| 79 else if (c == EOF) | |
| 80 { | |
| 81 c = CPP_EOF; | |
| 82 } | |
| 83 return c; | |
| 84 } | |
| 85 | |
| 86 /* This function takes a sequence of bytes from the _ll function above | |
| 87 and does trigraph interpretation on it, but only if the global | |
| 88 trigraphs is nonzero. */ | |
| 89 static int fetch_byte_tg(struct preproc_info *pp) | |
| 90 { | |
| 91 int c; | |
| 92 | |
| 93 if (!pp -> trigraphs) | |
| 94 { | |
| 95 c = fetch_byte_ll(pp); | |
| 96 } | |
| 97 else | |
| 98 { | |
| 99 /* we have to do the trigraph shit here */ | |
| 100 if (pp -> ra != CPP_NOUNG) | |
| 101 { | |
| 102 if (pp -> qseen > 0) | |
| 103 { | |
| 104 c = '?'; | |
| 105 pp -> qseen -= 1; | |
| 106 return c; | |
| 107 } | |
| 108 else | |
| 109 { | |
| 110 c = pp -> ra; | |
| 111 pp -> ra = CPP_NOUNG; | |
| 112 return c; | |
| 113 } | |
| 114 } | |
| 115 | |
| 116 c = fetch_byte_ll(pp); | |
| 117 while (c == '?') | |
| 118 { | |
| 119 pp -> qseen++; | |
| 120 c = fetch_byte_ll(pp); | |
| 121 } | |
| 122 | |
| 123 if (pp -> qseen >= 2) | |
| 124 { | |
| 125 // we have a trigraph | |
| 126 switch (c) | |
| 127 { | |
| 128 case '=': | |
| 129 c = '#'; | |
| 130 pp -> qseen -= 2; | |
| 131 break; | |
| 132 | |
| 133 case '/': | |
| 134 c = '\\'; | |
| 135 pp -> qseen -= 2; | |
| 136 break; | |
| 137 | |
| 138 case '\'': | |
| 139 c = '^'; | |
| 140 pp -> qseen -= 2; | |
| 141 break; | |
| 142 | |
| 143 case '(': | |
| 144 c = '['; | |
| 145 pp -> qseen -= 2; | |
| 146 break; | |
| 147 | |
| 148 case ')': | |
| 149 c = ']'; | |
| 150 pp -> qseen -= 2; | |
| 151 break; | |
| 152 | |
| 153 case '!': | |
| 154 c = '|'; | |
| 155 pp -> qseen -= 2; | |
| 156 break; | |
| 157 | |
| 158 case '<': | |
| 159 c = '{'; | |
| 160 pp -> qseen -= 2; | |
| 161 break; | |
| 162 | |
| 163 case '>': | |
| 164 c = '}'; | |
| 165 pp -> qseen -= 2; | |
| 166 break; | |
| 167 | |
| 168 case '-': | |
| 169 c = '~'; | |
| 170 pp -> qseen -= 2; | |
| 171 break; | |
| 172 } | |
| 173 if (pp -> qseen > 0) | |
| 174 { | |
| 175 pp -> ra = c; | |
| 176 c = '?'; | |
| 177 pp -> qseen--; | |
| 178 } | |
| 179 } | |
| 180 else if (pp -> qseen > 0) | |
| 181 { | |
| 182 pp -> ra = c; | |
| 183 c = '?'; | |
| 184 pp -> qseen--; | |
| 185 } | |
| 186 } | |
| 187 return c; | |
| 188 } | |
| 189 | |
| 190 /* This function puts a byte back onto the front of the input stream used | |
| 191 by fetch_byte(). Theoretically, an unlimited number of characters can | |
| 192 be unfetched. Line and column counting may be incorrect if unfetched | |
| 193 characters cross a token boundary. */ | |
| 194 static void preproc_lex_unfetch_byte(struct preproc_info *pp, int c) | |
| 195 { | |
| 196 if (pp -> ungetbufl >= pp -> ungetbufs) | |
| 197 { | |
| 198 pp -> ungetbufs += 100; | |
| 199 pp -> ungetbuf = lw_realloc(pp -> ungetbuf, pp -> ungetbufs); | |
| 200 } | |
| 201 pp -> ungetbuf[pp -> ungetbufl++] = c; | |
| 202 } | |
| 203 | |
| 204 /* This function retrieves a byte from the input stream. It performs | |
| 205 backslash-newline splicing on the returned bytes. Any character | |
| 206 retrieved from the unfetch buffer is presumed to have already passed | |
| 207 the backslash-newline filter. */ | |
| 208 static int fetch_byte(struct preproc_info *pp) | |
| 209 { | |
| 210 int c; | |
| 211 | |
|
298
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
212 if (pp -> lexstr) |
|
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
213 { |
|
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
214 if (pp -> lexstr[pp -> lexstrloc]) |
|
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
215 return pp -> lexstr[pp -> lexstrloc++]; |
|
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
216 else |
|
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
217 return CPP_EOL; |
|
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
218 } |
|
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
219 |
| 295 | 220 if (pp -> ungetbufl > 0) |
| 221 { | |
| 222 pp -> ungetbufl--; | |
| 223 c = pp -> ungetbuf[pp -> ungetbufl]; | |
| 224 if (pp -> ungetbufl == 0) | |
| 225 { | |
| 226 lw_free(pp -> ungetbuf); | |
| 227 pp -> ungetbuf = NULL; | |
| 228 pp -> ungetbufs = 0; | |
| 229 } | |
| 230 return c; | |
| 231 } | |
| 232 | |
| 233 again: | |
| 234 if (pp -> unget != CPP_NOUNG) | |
| 235 { | |
| 236 c = pp -> unget; | |
| 237 pp -> unget = CPP_NOUNG; | |
| 238 } | |
| 239 else | |
| 240 { | |
| 241 c = fetch_byte_tg(pp); | |
| 242 } | |
| 243 if (c == '\\') | |
| 244 { | |
| 245 int c2; | |
| 246 c2 = fetch_byte_tg(pp); | |
| 247 if (c2 == CPP_EOL) | |
| 248 goto again; | |
| 249 else | |
| 250 pp -> unget = c2; | |
| 251 } | |
| 252 return c; | |
| 253 } | |
| 254 | |
| 255 | |
| 256 | |
| 257 /* | |
| 258 Lex a token off the current input file. | |
| 259 | |
| 260 Returned tokens are as follows: | |
| 261 | |
| 262 * all words starting with [a-zA-Z_] are returned as TOK_IDENT | |
| 263 * numbers are returned as their appropriate type | |
| 264 * all whitespace in a sequence, including comments, is returned as | |
| 265 a single instance of TOK_WSPACE | |
| 266 * TOK_EOL is returned in the case of the end of a line | |
| 267 * TOK_EOF is returned when the end of the file is reached | |
| 268 * If no TOK_EOL appears before TOK_EOF, a TOK_EOL will be synthesised | |
| 269 * Any symbolic operator, etc., recognized by C will be returned as such | |
| 270 a token | |
| 271 * TOK_HASH will be returned for a # | |
| 272 * trigraphs will be interpreted | |
| 273 * backslash-newline will be interpreted | |
| 274 * any instance of CR, LF, CRLF, or LFCR will be interpreted as TOK_EOL | |
| 275 */ | |
| 276 | |
| 277 | |
| 278 static int preproc_lex_fetch_byte(struct preproc_info *pp) | |
| 279 { | |
| 280 int c; | |
| 281 c = fetch_byte(pp); | |
| 282 if (c == CPP_EOF && pp -> eolseen == 0) | |
| 283 { | |
| 284 preproc_throw_warning(pp, "No newline at end of file"); | |
| 285 pp -> eolseen = 1; | |
| 286 return CPP_EOL; | |
| 287 } | |
| 288 | |
| 289 if (c == CPP_EOL) | |
| 290 { | |
| 291 pp -> eolseen = 1; | |
| 292 return c; | |
| 293 } | |
| 294 | |
| 295 pp -> eolseen = 0; | |
| 296 | |
| 297 /* convert comments to a single space here */ | |
| 298 if (c == '/') | |
| 299 { | |
| 300 int c2; | |
| 301 c2 = fetch_byte(pp); | |
| 302 if (c2 == '/') | |
| 303 { | |
| 304 /* single line comment */ | |
| 305 c = ' '; | |
| 306 for (;;) | |
| 307 { | |
| 308 c2 = fetch_byte(pp); | |
| 309 if (c2 == CPP_EOF || c2 == CPP_EOL) | |
| 310 break; | |
| 311 } | |
| 312 preproc_lex_unfetch_byte(pp, c2); | |
| 313 } | |
| 314 else if (c2 == '*') | |
| 315 { | |
| 316 /* block comment */ | |
| 317 c = ' '; | |
| 318 for (;;) | |
| 319 { | |
| 320 c2 = fetch_byte(pp); | |
| 321 if (c2 == CPP_EOL || c2 == CPP_EOF) | |
| 322 { | |
| 323 preproc_lex_unfetch_byte(pp, c); | |
| 324 break; | |
| 325 } | |
| 326 if (c2 == '*') | |
| 327 { | |
| 328 /* maybe end of comment */ | |
| 329 c2 = preproc_lex_fetch_byte(pp); | |
| 330 if (c2 == '/') | |
| 331 break; | |
| 332 } | |
| 333 } | |
| 334 } | |
| 335 else | |
| 336 { | |
| 337 /* not a comment - restore lookahead character */ | |
| 338 preproc_lex_unfetch_byte(pp, c2); | |
| 339 } | |
| 340 } | |
| 341 return c; | |
| 342 } | |
| 343 | |
| 344 struct token *preproc_lex_next_token(struct preproc_info *pp) | |
| 345 { | |
| 346 int sline = pp -> lineno; | |
| 347 int scol = pp -> column; | |
| 348 char *strval = NULL; | |
| 349 int ttype = TOK_NONE; | |
| 350 int c, c2; | |
| 351 int cl; | |
| 352 struct strbuf *strbuf; | |
| 353 struct token *t; | |
| 354 | |
| 355 c = preproc_lex_fetch_byte(pp); | |
| 356 if (c == CPP_EOF) | |
| 357 { | |
| 358 if (pp -> nlseen == 0) | |
| 359 { | |
| 360 c = CPP_EOL; | |
| 361 } | |
| 362 } | |
| 363 | |
| 364 if (c == CPP_EOF) | |
| 365 { | |
| 366 ttype = TOK_EOF; | |
| 367 goto out; | |
| 368 } | |
| 369 if (c == CPP_EOL) | |
| 370 { | |
| 371 pp -> nlseen = 1; | |
| 372 ttype = TOK_EOL; | |
| 373 goto out; | |
| 374 } | |
| 375 | |
| 376 pp -> nlseen = 0; | |
| 377 if (isspace(c)) | |
| 378 { | |
| 379 while (isspace(c)) | |
| 380 c = preproc_lex_fetch_byte(pp); | |
| 381 preproc_lex_unfetch_byte(pp, c); | |
| 382 ttype = TOK_WSPACE; | |
| 383 goto out; | |
| 384 } | |
| 385 | |
| 386 switch (c) | |
| 387 { | |
| 388 case '?': | |
| 389 ttype = TOK_QMARK; | |
| 390 goto out; | |
| 391 | |
| 392 case ':': | |
| 393 ttype = TOK_COLON; | |
| 394 goto out; | |
| 395 | |
| 396 case ',': | |
| 397 ttype = TOK_COMMA; | |
| 398 goto out; | |
| 399 | |
| 400 case '(': | |
| 401 ttype = TOK_OPAREN; | |
| 402 goto out; | |
| 403 | |
| 404 case ')': | |
| 405 ttype = TOK_CPAREN; | |
| 406 goto out; | |
| 407 | |
| 408 case '{': | |
| 409 ttype = TOK_OBRACE; | |
| 410 goto out; | |
| 411 | |
| 412 case '}': | |
| 413 ttype = TOK_CBRACE; | |
| 414 goto out; | |
| 415 | |
| 416 case '[': | |
| 417 ttype = TOK_OSQUARE; | |
| 418 goto out; | |
| 419 | |
| 420 case ']': | |
| 421 ttype = TOK_CSQUARE; | |
| 422 goto out; | |
| 423 | |
| 424 case '~': | |
| 425 ttype = TOK_COM; | |
| 426 goto out; | |
| 427 | |
| 428 case ';': | |
| 429 ttype = TOK_EOS; | |
| 430 goto out; | |
| 431 | |
| 432 /* and now for the possible multi character tokens */ | |
| 433 case '#': | |
| 434 ttype = TOK_HASH; | |
| 435 c = preproc_lex_fetch_byte(pp); | |
| 436 if (c == '#') | |
| 437 ttype = TOK_DBLHASH; | |
| 438 else | |
| 439 preproc_lex_unfetch_byte(pp, c); | |
| 440 goto out; | |
| 441 | |
| 442 case '^': | |
| 443 ttype = TOK_XOR; | |
| 444 c = preproc_lex_fetch_byte(pp); | |
| 445 if (c == '=') | |
| 446 ttype = TOK_XORASS; | |
| 447 else | |
| 448 preproc_lex_unfetch_byte(pp, c); | |
| 449 goto out; | |
| 450 | |
| 451 case '!': | |
| 452 ttype = TOK_BNOT; | |
| 453 c = preproc_lex_fetch_byte(pp); | |
| 454 if (c == '=') | |
| 455 ttype = TOK_NE; | |
| 456 else | |
| 457 preproc_lex_unfetch_byte(pp, c); | |
| 458 goto out; | |
| 459 | |
| 460 case '*': | |
| 461 ttype = TOK_STAR; | |
| 462 c = preproc_lex_fetch_byte(pp); | |
| 463 if (c == '=') | |
| 464 ttype = TOK_MULASS; | |
| 465 else | |
| 466 preproc_lex_unfetch_byte(pp, c); | |
| 467 goto out; | |
| 468 | |
| 469 case '/': | |
| 470 ttype = TOK_DIV; | |
| 471 c = preproc_lex_fetch_byte(pp); | |
| 472 if (c == '=') | |
| 473 ttype = TOK_DIVASS; | |
| 474 else | |
| 475 preproc_lex_unfetch_byte(pp, c); | |
| 476 goto out; | |
| 477 | |
| 478 case '=': | |
| 479 ttype = TOK_ASS; | |
| 480 c = preproc_lex_fetch_byte(pp); | |
| 481 if (c == '=') | |
| 482 ttype = TOK_EQ; | |
| 483 else | |
| 484 preproc_lex_unfetch_byte(pp, c); | |
| 485 goto out; | |
| 486 | |
| 487 case '%': | |
| 488 ttype = TOK_MOD; | |
| 489 c = preproc_lex_fetch_byte(pp); | |
| 490 if (c == '=') | |
| 491 ttype = TOK_MODASS; | |
| 492 else | |
| 493 preproc_lex_unfetch_byte(pp, c); | |
| 494 goto out; | |
| 495 | |
| 496 case '-': | |
| 497 ttype = TOK_SUB; | |
| 498 c = preproc_lex_fetch_byte(pp); | |
| 499 if (c == '=') | |
| 500 ttype = TOK_SUBASS; | |
| 501 else if (c == '-') | |
| 502 ttype = TOK_DBLSUB; | |
| 503 else if (c == '>') | |
| 504 ttype = TOK_ARROW; | |
| 505 else | |
| 506 preproc_lex_unfetch_byte(pp, c); | |
| 507 goto out; | |
| 508 | |
| 509 case '+': | |
| 510 ttype = TOK_ADD; | |
| 511 c = preproc_lex_fetch_byte(pp); | |
| 512 if (c == '=') | |
| 513 ttype = TOK_ADDASS; | |
| 514 else if (c == '+') | |
| 515 ttype = TOK_DBLADD; | |
| 516 else | |
| 517 preproc_lex_unfetch_byte(pp, c); | |
| 518 goto out; | |
| 519 | |
| 520 | |
| 521 case '&': | |
| 522 ttype = TOK_BWAND; | |
| 523 c = preproc_lex_fetch_byte(pp); | |
| 524 if (c == '=') | |
| 525 ttype = TOK_BWANDASS; | |
| 526 else if (c == '&') | |
| 527 ttype = TOK_BAND; | |
| 528 else | |
| 529 preproc_lex_unfetch_byte(pp, c); | |
| 530 goto out; | |
| 531 | |
| 532 case '|': | |
| 533 ttype = TOK_BWOR; | |
| 534 c = preproc_lex_fetch_byte(pp); | |
| 535 if (c == '=') | |
| 536 ttype = TOK_BWORASS; | |
| 537 else if (c == '|') | |
| 538 ttype = TOK_BOR; | |
| 539 else | |
| 540 preproc_lex_unfetch_byte(pp, c); | |
| 541 goto out; | |
| 542 | |
| 543 case '<': | |
| 544 ttype = TOK_LT; | |
| 545 c = preproc_lex_fetch_byte(pp); | |
| 546 if (c == '=') | |
| 547 ttype = TOK_LE; | |
| 548 else if (c == '<') | |
| 549 { | |
| 550 ttype = TOK_LSH; | |
| 551 c = preproc_lex_fetch_byte(pp); | |
| 552 if (c == '=') | |
| 553 ttype = TOK_LSHASS; | |
| 554 else | |
| 555 preproc_lex_unfetch_byte(pp, c); | |
| 556 } | |
| 557 else | |
| 558 preproc_lex_unfetch_byte(pp, c); | |
| 559 goto out; | |
| 560 | |
| 561 | |
| 562 case '>': | |
| 563 ttype = TOK_GT; | |
| 564 c = preproc_lex_fetch_byte(pp); | |
| 565 if (c == '=') | |
| 566 ttype = TOK_GE; | |
| 567 else if (c == '>') | |
| 568 { | |
| 569 ttype = TOK_RSH; | |
| 570 c = preproc_lex_fetch_byte(pp); | |
| 571 if (c == '=') | |
| 572 ttype = TOK_RSHASS; | |
| 573 else | |
| 574 preproc_lex_unfetch_byte(pp, c); | |
| 575 } | |
| 576 else | |
| 577 preproc_lex_unfetch_byte(pp, c); | |
| 578 goto out; | |
| 579 | |
| 580 case '\'': | |
| 581 /* character constant - turns into a uint */ | |
| 582 chrlit: | |
| 583 cl = 0; | |
| 584 strbuf = strbuf_new(); | |
| 585 for (;;) | |
| 586 { | |
| 587 c = preproc_lex_fetch_byte(pp); | |
| 588 if (c == CPP_EOF || c == CPP_EOL || c == '\'') | |
| 589 break; | |
| 590 cl++; | |
| 591 if (c == '\\') | |
| 592 { | |
| 593 strbuf_add(strbuf, '\\'); | |
| 594 c = preproc_lex_fetch_byte(pp); | |
| 595 if (c == CPP_EOF || c == CPP_EOL) | |
| 596 { | |
|
298
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
597 if (!pp -> lexstr) |
|
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
598 preproc_throw_error(pp, "Invalid character constant"); |
|
299
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
599 ttype = TOK_ERROR; |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
600 strval = strbuf_end(strbuf); |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
601 goto out; |
| 295 | 602 } |
| 603 cl++; | |
| 604 strbuf_add(strbuf, c); | |
| 605 continue; | |
| 606 } | |
| 607 strbuf_add(strbuf, c); | |
| 608 } | |
| 609 strval = strbuf_end(strbuf); | |
|
299
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
610 if (cl == 0) |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
611 { |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
612 ttype = TOK_ERROR; |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
613 if (!pp -> lexstr) |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
614 preproc_throw_error(pp, "Invalid character constant"); |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
615 } |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
616 else |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
617 ttype = TOK_CHR_LIT; |
| 295 | 618 goto out; |
| 619 | |
| 620 case '"': | |
| 621 strlit: | |
| 622 /* string literal */ | |
| 623 strbuf = strbuf_new(); | |
| 624 for (;;) | |
| 625 { | |
| 626 c = preproc_lex_fetch_byte(pp); | |
|
299
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
627 if (c == CPP_EOF || c == CPP_EOL) |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
628 { |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
629 ttype = TOK_ERROR; |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
630 strval = strbuf_end(strbuf); |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
631 if (!pp -> lexstr) |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
632 preproc_throw_error(pp, "Invalid string constant"); |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
633 goto out; |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
634 } |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
635 if (c == '"') |
| 295 | 636 break; |
| 637 if (c == '\\') | |
| 638 { | |
| 639 strbuf_add(strbuf, '\\'); | |
| 640 c = preproc_lex_fetch_byte(pp); | |
| 641 if (c == CPP_EOF || c == CPP_EOL) | |
| 642 { | |
|
299
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
643 ttype = TOK_ERROR; |
|
298
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
644 if (!pp -> lexstr) |
|
6112c67728ba
Add stringification and token concatenation
William Astle <lost@l-w.ca>
parents:
296
diff
changeset
|
645 preproc_throw_error(pp, "Invalid string constant"); |
|
299
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
646 strval = strbuf_end(strbuf); |
|
856caf91ffaa
Added token list structure and switched some stuff to use it
William Astle <lost@l-w.ca>
parents:
298
diff
changeset
|
647 goto out; |
| 295 | 648 } |
| 649 cl++; | |
| 650 strbuf_add(strbuf, c); | |
| 651 continue; | |
| 652 } | |
| 653 strbuf_add(strbuf, c); | |
| 654 } | |
| 655 strval = strbuf_end(strbuf); | |
| 656 ttype = TOK_STR_LIT; | |
| 657 goto out; | |
| 658 | |
| 659 case 'L': | |
| 660 /* check for wide string or wide char const */ | |
| 661 c2 = preproc_lex_fetch_byte(pp); | |
| 662 if (c2 == '\'') | |
| 663 { | |
| 664 goto chrlit; | |
| 665 } | |
| 666 else if (c2 == '"') | |
| 667 { | |
| 668 goto strlit; | |
| 669 } | |
| 670 preproc_lex_unfetch_byte(pp, c2); | |
| 671 /* fall through for identifier */ | |
| 672 case '_': | |
| 673 case 'a': case 'b': case 'c': case 'd': case 'e': case 'f': | |
| 674 case 'g': case 'h': case 'i': case 'j': case 'k': case 'l': | |
| 675 case 'm': case 'n': case 'o': case 'p': case 'q': case 'r': | |
| 676 case 's': case 't': case 'u': case 'v': case 'w': case 'x': | |
| 677 case 'y': case 'z': | |
| 678 case 'A': case 'B': case 'C': case 'D': case 'E': case 'F': | |
| 679 case 'G': case 'H': case 'I': case 'J': case 'K': | |
| 680 case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R': | |
| 681 case 'S': case 'T': case 'U': case 'V': case 'W': case 'X': | |
| 682 case 'Y': case 'Z': | |
| 683 /* we have an identifier here */ | |
| 684 strbuf = strbuf_new(); | |
| 685 strbuf_add(strbuf, c); | |
| 686 for (;;) | |
| 687 { | |
| 688 c = preproc_lex_fetch_byte(pp); | |
| 689 if ((c == '_') || (c >= '0' && c <= '9') || (c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')) | |
| 690 { | |
| 691 strbuf_add(strbuf, c); | |
| 692 continue; | |
| 693 } | |
| 694 else | |
| 695 { | |
| 696 strbuf_add(strbuf, 0); | |
| 697 strval = strbuf_end(strbuf); | |
| 698 break; | |
| 699 } | |
| 700 } | |
| 701 preproc_lex_unfetch_byte(pp, c); | |
| 702 ttype = TOK_IDENT; | |
| 703 goto out; | |
| 704 | |
| 705 case '.': | |
| 706 c = preproc_lex_fetch_byte(pp); | |
| 707 if (c >= '0' && c <= '9') | |
| 708 { | |
| 709 strbuf = strbuf_new(); | |
| 710 strbuf_add(strbuf, '.'); | |
| 711 goto numlit; | |
| 712 } | |
| 713 else if (c == '.') | |
| 714 { | |
| 715 c = preproc_lex_fetch_byte(pp); | |
| 716 if (c == '.') | |
| 717 { | |
| 718 ttype = TOK_ELLIPSIS; | |
| 719 goto out; | |
| 720 } | |
| 721 preproc_lex_unfetch_byte(pp, c); | |
| 722 } | |
| 723 preproc_lex_unfetch_byte(pp, c); | |
| 724 ttype = TOK_DOT; | |
| 725 goto out; | |
| 726 | |
| 727 case '0': case '1': case '2': case '3': case '4': | |
| 728 case '5': case '6': case '7': case '8': case '9': | |
| 729 strbuf = strbuf_new(); | |
| 730 numlit: | |
| 296 | 731 ttype = TOK_NUMBER; |
| 295 | 732 strbuf_add(strbuf, c); |
| 733 for (;;) | |
| 734 { | |
| 735 c = preproc_lex_fetch_byte(pp); | |
| 736 if (!((c == '_') || (c >= '0' && c <= '9') || (c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z'))) | |
| 737 break; | |
| 738 strbuf_add(strbuf, c); | |
| 739 if (c == 'e' || c == 'E' || c == 'p' || c == 'P') | |
| 740 { | |
| 741 c = preproc_lex_fetch_byte(pp); | |
| 742 if (c == '+' || c == '-') | |
| 743 { | |
| 744 strbuf_add(strbuf, c); | |
| 745 continue; | |
| 746 } | |
| 747 preproc_lex_unfetch_byte(pp, c); | |
| 748 } | |
| 749 } | |
| 750 strval = strbuf_end(strbuf); | |
| 751 preproc_lex_unfetch_byte(pp, c); | |
| 752 goto out; | |
| 753 | |
| 754 default: | |
| 755 ttype = TOK_CHAR; | |
| 756 strval = lw_alloc(2); | |
| 757 strval[0] = c; | |
| 758 strval[1] = 0; | |
| 759 break; | |
| 760 } | |
| 761 out: | |
| 762 t = token_create(ttype, strval, sline, scol, pp -> fn); | |
| 763 lw_free(strval); | |
| 764 return t; | |
| 765 } |