1 /* 2 * Copyright (c) 1985 Sun Microsystems, Inc. 3 * Copyright (c) 1980, 1993 4 * The Regents of the University of California. All rights reserved. 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 3. All advertising materials mentioning features or use of this software 16 * must display the following acknowledgement: 17 * This product includes software developed by the University of 18 * California, Berkeley and its contributors. 19 * 4. Neither the name of the University nor the names of its contributors 20 * may be used to endorse or promote products derived from this software 21 * without specific prior written permission. 22 * 23 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 24 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 25 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 26 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 27 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 28 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 29 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 30 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 31 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 32 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 33 * SUCH DAMAGE. 34 */ 35 36 #if 0 37 #ifndef lint 38 static char sccsid[] = "@(#)lexi.c 8.1 (Berkeley) 6/6/93"; 39 #endif /* not lint */ 40 #endif 41 #include <sys/cdefs.h> 42 __FBSDID("$FreeBSD$"); 43 44 /* 45 * Here we have the token scanner for indent. It scans off one token and puts 46 * it in the global variable "token". It returns a code, indicating the type 47 * of token scanned. 48 */ 49 50 #include <err.h> 51 #include <stdio.h> 52 #include <ctype.h> 53 #include <stdlib.h> 54 #include <string.h> 55 #include "indent_globs.h" 56 #include "indent_codes.h" 57 #include "indent.h" 58 59 #define alphanum 1 60 #define opchar 3 61 62 struct templ { 63 const char *rwd; 64 int rwcode; 65 }; 66 67 struct templ specials[1000] = 68 { 69 {"switch", 1}, 70 {"case", 2}, 71 {"break", 0}, 72 {"struct", 3}, 73 {"union", 3}, 74 {"enum", 3}, 75 {"default", 2}, 76 {"int", 4}, 77 {"char", 4}, 78 {"float", 4}, 79 {"double", 4}, 80 {"long", 4}, 81 {"short", 4}, 82 {"typdef", 4}, 83 {"unsigned", 4}, 84 {"register", 4}, 85 {"static", 4}, 86 {"global", 4}, 87 {"extern", 4}, 88 {"void", 4}, 89 {"const", 4}, 90 {"volatile", 4}, 91 {"goto", 0}, 92 {"return", 0}, 93 {"if", 5}, 94 {"while", 5}, 95 {"for", 5}, 96 {"else", 6}, 97 {"do", 6}, 98 {"sizeof", 7}, 99 {0, 0} 100 }; 101 102 char chartype[128] = 103 { /* this is used to facilitate the decision of 104 * what type (alphanumeric, operator) each 105 * character is */ 106 0, 0, 0, 0, 0, 0, 0, 0, 107 0, 0, 0, 0, 0, 0, 0, 0, 108 0, 0, 0, 0, 0, 0, 0, 0, 109 0, 0, 0, 0, 0, 0, 0, 0, 110 0, 3, 0, 0, 1, 3, 3, 0, 111 0, 0, 3, 3, 0, 3, 0, 3, 112 1, 1, 1, 1, 1, 1, 1, 1, 113 1, 1, 0, 0, 3, 3, 3, 3, 114 0, 1, 1, 1, 1, 1, 1, 1, 115 1, 1, 1, 1, 1, 1, 1, 1, 116 1, 1, 1, 1, 1, 1, 1, 1, 117 1, 1, 1, 0, 0, 0, 3, 1, 118 0, 1, 1, 1, 1, 1, 1, 1, 119 1, 1, 1, 1, 1, 1, 1, 1, 120 1, 1, 1, 1, 1, 1, 1, 1, 121 1, 1, 1, 0, 3, 0, 3, 0 122 }; 123 124 int 125 lexi(void) 126 { 127 int unary_delim; /* this is set to 1 if the current token 128 * forces a following operator to be unary */ 129 static int last_code; /* the last token type returned */ 130 static int l_struct; /* set to 1 if the last token was 'struct' */ 131 int code; /* internal code to be returned */ 132 char qchar; /* the delimiter character for a string */ 133 134 e_token = s_token; /* point to start of place to save token */ 135 unary_delim = false; 136 ps.col_1 = ps.last_nl; /* tell world that this token started in 137 * column 1 iff the last thing scanned was nl */ 138 ps.last_nl = false; 139 140 while (*buf_ptr == ' ' || *buf_ptr == '\t') { /* get rid of blanks */ 141 ps.col_1 = false; /* leading blanks imply token is not in column 142 * 1 */ 143 if (++buf_ptr >= buf_end) 144 fill_buffer(); 145 } 146 147 /* Scan an alphanumeric token */ 148 if (chartype[(int)*buf_ptr] == alphanum || (buf_ptr[0] == '.' && isdigit(buf_ptr[1]))) { 149 /* 150 * we have a character or number 151 */ 152 const char *j; /* used for searching thru list of 153 * 154 * reserved words */ 155 struct templ *p; 156 157 if (isdigit(*buf_ptr) || (buf_ptr[0] == '.' && isdigit(buf_ptr[1]))) { 158 int seendot = 0, 159 seenexp = 0, 160 seensfx = 0; 161 if (*buf_ptr == '0' && 162 (buf_ptr[1] == 'x' || buf_ptr[1] == 'X')) { 163 *e_token++ = *buf_ptr++; 164 *e_token++ = *buf_ptr++; 165 while (isxdigit(*buf_ptr)) { 166 CHECK_SIZE_TOKEN; 167 *e_token++ = *buf_ptr++; 168 } 169 } 170 else 171 while (1) { 172 if (*buf_ptr == '.') { 173 if (seendot) 174 break; 175 else 176 seendot++; 177 } 178 CHECK_SIZE_TOKEN; 179 *e_token++ = *buf_ptr++; 180 if (!isdigit(*buf_ptr) && *buf_ptr != '.') { 181 if ((*buf_ptr != 'E' && *buf_ptr != 'e') || seenexp) 182 break; 183 else { 184 seenexp++; 185 seendot++; 186 CHECK_SIZE_TOKEN; 187 *e_token++ = *buf_ptr++; 188 if (*buf_ptr == '+' || *buf_ptr == '-') 189 *e_token++ = *buf_ptr++; 190 } 191 } 192 } 193 while (1) { 194 if (!(seensfx & 1) && 195 (*buf_ptr == 'U' || *buf_ptr == 'u')) { 196 CHECK_SIZE_TOKEN; 197 *e_token++ = *buf_ptr++; 198 seensfx |= 1; 199 continue; 200 } 201 if (!(seensfx & 2) && 202 (*buf_ptr == 'L' || *buf_ptr == 'l')) { 203 CHECK_SIZE_TOKEN; 204 if (buf_ptr[1] == buf_ptr[0]) 205 *e_token++ = *buf_ptr++; 206 *e_token++ = *buf_ptr++; 207 seensfx |= 2; 208 continue; 209 } 210 break; 211 } 212 } 213 else 214 while (chartype[(int)*buf_ptr] == alphanum || *buf_ptr == BACKSLASH) { 215 /* fill_buffer() terminates buffer with newline */ 216 if (*buf_ptr == BACKSLASH) { 217 if (*(buf_ptr + 1) == '\n') { 218 buf_ptr += 2; 219 if (buf_ptr >= buf_end) 220 fill_buffer(); 221 } else 222 break; 223 } 224 CHECK_SIZE_TOKEN; 225 /* copy it over */ 226 *e_token++ = *buf_ptr++; 227 if (buf_ptr >= buf_end) 228 fill_buffer(); 229 } 230 *e_token++ = '\0'; 231 while (*buf_ptr == ' ' || *buf_ptr == '\t') { /* get rid of blanks */ 232 if (++buf_ptr >= buf_end) 233 fill_buffer(); 234 } 235 ps.its_a_keyword = false; 236 ps.sizeof_keyword = false; 237 if (l_struct && !ps.p_l_follow) { 238 /* if last token was 'struct' and we're not 239 * in parentheses, then this token 240 * should be treated as a declaration */ 241 l_struct = false; 242 last_code = ident; 243 ps.last_u_d = true; 244 return (decl); 245 } 246 ps.last_u_d = l_struct; /* Operator after identifier is binary 247 * unless last token was 'struct' */ 248 l_struct = false; 249 last_code = ident; /* Remember that this is the code we will 250 * return */ 251 252 if (auto_typedefs) { 253 const char *q = s_token; 254 /* Check if we have an "_t" in the end */ 255 if (q[0] && q[1] && 256 (strcmp(q + strlen(q) - 2, "_t") == 0)) { 257 ps.its_a_keyword = true; 258 ps.last_u_d = true; 259 goto found_auto_typedef; 260 } 261 } 262 263 /* 264 * This loop will check if the token is a keyword. 265 */ 266 for (p = specials; (j = p->rwd) != 0; p++) { 267 const char *q = s_token; /* point at scanned token */ 268 if (*j++ != *q++ || *j++ != *q++) 269 continue; /* This test depends on the fact that 270 * identifiers are always at least 1 character 271 * long (ie. the first two bytes of the 272 * identifier are always meaningful) */ 273 if (q[-1] == 0) 274 break; /* If its a one-character identifier */ 275 while (*q++ == *j) 276 if (*j++ == 0) 277 goto found_keyword; /* I wish that C had a multi-level 278 * break... */ 279 } 280 if (p->rwd) { /* we have a keyword */ 281 found_keyword: 282 ps.its_a_keyword = true; 283 ps.last_u_d = true; 284 switch (p->rwcode) { 285 case 1: /* it is a switch */ 286 return (swstmt); 287 case 2: /* a case or default */ 288 return (casestmt); 289 290 case 3: /* a "struct" */ 291 /* 292 * Next time around, we will want to know that we have had a 293 * 'struct' 294 */ 295 l_struct = true; 296 /* FALLTHROUGH */ 297 298 case 4: /* one of the declaration keywords */ 299 found_auto_typedef: 300 if (ps.p_l_follow) { 301 ps.cast_mask |= (1 << ps.p_l_follow) & ~ps.sizeof_mask; 302 break; /* inside parens: cast, param list or sizeof */ 303 } 304 last_code = decl; 305 return (decl); 306 307 case 5: /* if, while, for */ 308 return (sp_paren); 309 310 case 6: /* do, else */ 311 return (sp_nparen); 312 313 case 7: 314 ps.sizeof_keyword = true; 315 default: /* all others are treated like any other 316 * identifier */ 317 return (ident); 318 } /* end of switch */ 319 } /* end of if (found_it) */ 320 if (*buf_ptr == '(' && ps.tos <= 1 && ps.ind_level == 0) { 321 char *tp = buf_ptr; 322 while (tp < buf_end) 323 if (*tp++ == ')' && (*tp == ';' || *tp == ',')) 324 goto not_proc; 325 strncpy(ps.procname, token, sizeof ps.procname - 1); 326 ps.in_parameter_declaration = 1; 327 rparen_count = 1; 328 not_proc:; 329 } 330 /* 331 * The following hack attempts to guess whether or not the current 332 * token is in fact a declaration keyword -- one that has been 333 * typedefd 334 */ 335 if (((*buf_ptr == '*' && buf_ptr[1] != '=') || isalpha(*buf_ptr) || *buf_ptr == '_') 336 && !ps.p_l_follow 337 && !ps.block_init 338 && (ps.last_token == rparen || ps.last_token == semicolon || 339 ps.last_token == decl || 340 ps.last_token == lbrace || ps.last_token == rbrace)) { 341 ps.its_a_keyword = true; 342 ps.last_u_d = true; 343 last_code = decl; 344 return decl; 345 } 346 if (last_code == decl) /* if this is a declared variable, then 347 * following sign is unary */ 348 ps.last_u_d = true; /* will make "int a -1" work */ 349 last_code = ident; 350 return (ident); /* the ident is not in the list */ 351 } /* end of procesing for alpanum character */ 352 353 /* Scan a non-alphanumeric token */ 354 355 *e_token++ = *buf_ptr; /* if it is only a one-character token, it is 356 * moved here */ 357 *e_token = '\0'; 358 if (++buf_ptr >= buf_end) 359 fill_buffer(); 360 361 switch (*token) { 362 case '\n': 363 unary_delim = ps.last_u_d; 364 ps.last_nl = true; /* remember that we just had a newline */ 365 code = (had_eof ? 0 : newline); 366 367 /* 368 * if data has been exhausted, the newline is a dummy, and we should 369 * return code to stop 370 */ 371 break; 372 373 case '\'': /* start of quoted character */ 374 case '"': /* start of string */ 375 qchar = *token; 376 if (troff) { 377 e_token[-1] = '`'; 378 if (qchar == '"') 379 *e_token++ = '`'; 380 e_token = chfont(&bodyf, &stringf, e_token); 381 } 382 do { /* copy the string */ 383 while (1) { /* move one character or [/<char>]<char> */ 384 if (*buf_ptr == '\n') { 385 diag2(1, "Unterminated literal"); 386 goto stop_lit; 387 } 388 CHECK_SIZE_TOKEN; /* Only have to do this once in this loop, 389 * since CHECK_SIZE guarantees that there 390 * are at least 5 entries left */ 391 *e_token = *buf_ptr++; 392 if (buf_ptr >= buf_end) 393 fill_buffer(); 394 if (*e_token == BACKSLASH) { /* if escape, copy extra char */ 395 if (*buf_ptr == '\n') /* check for escaped newline */ 396 ++line_no; 397 if (troff) { 398 *++e_token = BACKSLASH; 399 if (*buf_ptr == BACKSLASH) 400 *++e_token = BACKSLASH; 401 } 402 *++e_token = *buf_ptr++; 403 ++e_token; /* we must increment this again because we 404 * copied two chars */ 405 if (buf_ptr >= buf_end) 406 fill_buffer(); 407 } 408 else 409 break; /* we copied one character */ 410 } /* end of while (1) */ 411 } while (*e_token++ != qchar); 412 if (troff) { 413 e_token = chfont(&stringf, &bodyf, e_token - 1); 414 if (qchar == '"') 415 *e_token++ = '\''; 416 } 417 stop_lit: 418 code = ident; 419 break; 420 421 case ('('): 422 case ('['): 423 unary_delim = true; 424 code = lparen; 425 break; 426 427 case (')'): 428 case (']'): 429 code = rparen; 430 break; 431 432 case '#': 433 unary_delim = ps.last_u_d; 434 code = preesc; 435 break; 436 437 case '?': 438 unary_delim = true; 439 code = question; 440 break; 441 442 case (':'): 443 code = colon; 444 unary_delim = true; 445 break; 446 447 case (';'): 448 unary_delim = true; 449 code = semicolon; 450 break; 451 452 case ('{'): 453 unary_delim = true; 454 455 /* 456 * if (ps.in_or_st) ps.block_init = 1; 457 */ 458 /* ? code = ps.block_init ? lparen : lbrace; */ 459 code = lbrace; 460 break; 461 462 case ('}'): 463 unary_delim = true; 464 /* ? code = ps.block_init ? rparen : rbrace; */ 465 code = rbrace; 466 break; 467 468 case 014: /* a form feed */ 469 unary_delim = ps.last_u_d; 470 ps.last_nl = true; /* remember this so we can set 'ps.col_1' 471 * right */ 472 code = form_feed; 473 break; 474 475 case (','): 476 unary_delim = true; 477 code = comma; 478 break; 479 480 case '.': 481 unary_delim = false; 482 code = period; 483 break; 484 485 case '-': 486 case '+': /* check for -, +, --, ++ */ 487 code = (ps.last_u_d ? unary_op : binary_op); 488 unary_delim = true; 489 490 if (*buf_ptr == token[0]) { 491 /* check for doubled character */ 492 *e_token++ = *buf_ptr++; 493 /* buffer overflow will be checked at end of loop */ 494 if (last_code == ident || last_code == rparen) { 495 code = (ps.last_u_d ? unary_op : postop); 496 /* check for following ++ or -- */ 497 unary_delim = false; 498 } 499 } 500 else if (*buf_ptr == '=') 501 /* check for operator += */ 502 *e_token++ = *buf_ptr++; 503 else if (*buf_ptr == '>') { 504 /* check for operator -> */ 505 *e_token++ = *buf_ptr++; 506 if (!pointer_as_binop) { 507 unary_delim = false; 508 code = unary_op; 509 ps.want_blank = false; 510 } 511 } 512 break; /* buffer overflow will be checked at end of 513 * switch */ 514 515 case '=': 516 if (ps.in_or_st) 517 ps.block_init = 1; 518 #ifdef undef 519 if (chartype[*buf_ptr] == opchar) { /* we have two char assignment */ 520 e_token[-1] = *buf_ptr++; 521 if ((e_token[-1] == '<' || e_token[-1] == '>') && e_token[-1] == *buf_ptr) 522 *e_token++ = *buf_ptr++; 523 *e_token++ = '='; /* Flip =+ to += */ 524 *e_token = 0; 525 } 526 #else 527 if (*buf_ptr == '=') {/* == */ 528 *e_token++ = '='; /* Flip =+ to += */ 529 buf_ptr++; 530 *e_token = 0; 531 } 532 #endif 533 code = binary_op; 534 unary_delim = true; 535 break; 536 /* can drop thru!!! */ 537 538 case '>': 539 case '<': 540 case '!': /* ops like <, <<, <=, !=, etc */ 541 if (*buf_ptr == '>' || *buf_ptr == '<' || *buf_ptr == '=') { 542 *e_token++ = *buf_ptr; 543 if (++buf_ptr >= buf_end) 544 fill_buffer(); 545 } 546 if (*buf_ptr == '=') 547 *e_token++ = *buf_ptr++; 548 code = (ps.last_u_d ? unary_op : binary_op); 549 unary_delim = true; 550 break; 551 552 default: 553 if (token[0] == '/' && *buf_ptr == '*') { 554 /* it is start of comment */ 555 *e_token++ = '*'; 556 557 if (++buf_ptr >= buf_end) 558 fill_buffer(); 559 560 code = comment; 561 unary_delim = ps.last_u_d; 562 break; 563 } 564 while (*(e_token - 1) == *buf_ptr || *buf_ptr == '=') { 565 /* 566 * handle ||, &&, etc, and also things as in int *****i 567 */ 568 *e_token++ = *buf_ptr; 569 if (++buf_ptr >= buf_end) 570 fill_buffer(); 571 } 572 code = (ps.last_u_d ? unary_op : binary_op); 573 unary_delim = true; 574 575 576 } /* end of switch */ 577 if (code != newline) { 578 l_struct = false; 579 last_code = code; 580 } 581 if (buf_ptr >= buf_end) /* check for input buffer empty */ 582 fill_buffer(); 583 ps.last_u_d = unary_delim; 584 *e_token = '\0'; /* null terminate the token */ 585 return (code); 586 } 587 588 /* 589 * Add the given keyword to the keyword table, using val as the keyword type 590 */ 591 void 592 addkey(char *key, int val) 593 { 594 struct templ *p = specials; 595 while (p->rwd) 596 if (p->rwd[0] == key[0] && strcmp(p->rwd, key) == 0) 597 return; 598 else 599 p++; 600 if (p >= specials + sizeof specials / sizeof specials[0]) 601 return; /* For now, table overflows are silently 602 * ignored */ 603 p->rwd = key; 604 p->rwcode = val; 605 p[1].rwd = 0; 606 p[1].rwcode = 0; 607 } 608