1 /* $OpenBSD: main.c,v 1.88 2022/12/04 23:50:48 cheloha Exp $ */ 2 /* $NetBSD: main.c,v 1.12 1997/02/08 23:54:49 cgd Exp $ */ 3 4 /*- 5 * Copyright (c) 1989, 1993 6 * The Regents of the University of California. All rights reserved. 7 * 8 * This code is derived from software contributed to Berkeley by 9 * Ozan Yigit at York University. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 14 * 1. Redistributions of source code must retain the above copyright 15 * notice, this list of conditions and the following disclaimer. 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 3. 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 /* 37 * main.c 38 * Facility: m4 macro processor 39 * by: oz 40 */ 41 42 #include <assert.h> 43 #include <signal.h> 44 #include <err.h> 45 #include <errno.h> 46 #include <unistd.h> 47 #include <stdio.h> 48 #include <ctype.h> 49 #include <string.h> 50 #include <stddef.h> 51 #include <stdint.h> 52 #include <stdlib.h> 53 #include <ohash.h> 54 #include "mdef.h" 55 #include "stdd.h" 56 #include "extern.h" 57 #include "pathnames.h" 58 59 stae *mstack; /* stack of m4 machine */ 60 char *sstack; /* shadow stack, for string space extension */ 61 static size_t STACKMAX; /* current maximum size of stack */ 62 int sp; /* current m4 stack pointer */ 63 int fp; /* m4 call frame pointer */ 64 struct input_file infile[MAXINP];/* input file stack (0=stdin) */ 65 FILE **outfile; /* diversion array(0=bitbucket)*/ 66 int maxout; 67 FILE *active; /* active output file pointer */ 68 int ilevel = 0; /* input file stack pointer */ 69 int oindex = 0; /* diversion index.. */ 70 char *null = ""; /* as it says.. just a null.. */ 71 char **m4wraps = NULL; /* m4wraps array. */ 72 int maxwraps = 0; /* size of m4wraps array */ 73 int wrapindex = 0; /* current offset in m4wraps */ 74 char lquote[MAXCCHARS+1] = {LQUOTE}; /* left quote character (`) */ 75 char rquote[MAXCCHARS+1] = {RQUOTE}; /* right quote character (') */ 76 char scommt[MAXCCHARS+1] = {SCOMMT}; /* start character for comment */ 77 char ecommt[MAXCCHARS+1] = {ECOMMT}; /* end character for comment */ 78 int synch_lines = 0; /* line synchronisation for C preprocessor */ 79 int prefix_builtins = 0; /* -P option to prefix builtin keywords */ 80 int error_warns = 0; /* -E option to make warnings exit_code = 1 */ 81 int fatal_warns = 0; /* -E -E option to make warnings fatal */ 82 83 struct keyblk { 84 char *knam; /* keyword name */ 85 int ktyp; /* keyword type */ 86 }; 87 88 struct keyblk keywrds[] = { /* m4 keywords to be installed */ 89 { "include", INCLTYPE }, 90 { "sinclude", SINCTYPE }, 91 { "define", DEFITYPE }, 92 { "defn", DEFNTYPE }, 93 { "divert", DIVRTYPE | NOARGS }, 94 { "expr", EXPRTYPE }, 95 { "eval", EXPRTYPE }, 96 { "substr", SUBSTYPE }, 97 { "ifelse", IFELTYPE }, 98 { "ifdef", IFDFTYPE }, 99 { "len", LENGTYPE }, 100 { "incr", INCRTYPE }, 101 { "decr", DECRTYPE }, 102 { "dnl", DNLNTYPE | NOARGS }, 103 { "changequote", CHNQTYPE | NOARGS }, 104 { "changecom", CHNCTYPE | NOARGS }, 105 { "index", INDXTYPE }, 106 #ifdef EXTENDED 107 { "paste", PASTTYPE }, 108 { "spaste", SPASTYPE }, 109 /* Newer extensions, needed to handle gnu-m4 scripts */ 110 { "indir", INDIRTYPE}, 111 { "builtin", BUILTINTYPE}, 112 { "patsubst", PATSTYPE}, 113 { "regexp", REGEXPTYPE}, 114 { "esyscmd", ESYSCMDTYPE}, 115 { "__file__", FILENAMETYPE | NOARGS}, 116 { "__line__", LINETYPE | NOARGS}, 117 #endif 118 { "popdef", POPDTYPE }, 119 { "pushdef", PUSDTYPE }, 120 { "dumpdef", DUMPTYPE | NOARGS }, 121 { "shift", SHIFTYPE | NOARGS }, 122 { "translit", TRNLTYPE }, 123 { "undefine", UNDFTYPE }, 124 { "undivert", UNDVTYPE | NOARGS }, 125 { "divnum", DIVNTYPE | NOARGS }, 126 { "maketemp", MKTMTYPE }, 127 { "mkstemp", MKTMTYPE }, 128 { "errprint", ERRPTYPE | NOARGS }, 129 { "m4wrap", M4WRTYPE | NOARGS }, 130 { "m4exit", EXITTYPE | NOARGS }, 131 { "syscmd", SYSCTYPE }, 132 { "sysval", SYSVTYPE | NOARGS }, 133 { "traceon", TRACEONTYPE | NOARGS }, 134 { "traceoff", TRACEOFFTYPE | NOARGS }, 135 136 { "unix", SELFTYPE | NOARGS }, 137 }; 138 139 #define MAXKEYS (sizeof(keywrds)/sizeof(struct keyblk)) 140 141 extern int optind; 142 extern char *optarg; 143 144 #define MAXRECORD 50 145 static struct position { 146 char *name; 147 unsigned long line; 148 } quotes[MAXRECORD], paren[MAXRECORD]; 149 150 static void record(struct position *, int); 151 static void dump_stack(struct position *, int); 152 153 static void macro(void); 154 static void initkwds(void); 155 static ndptr inspect(int, char *); 156 static int do_look_ahead(int, const char *); 157 static void reallyoutputstr(const char *); 158 static void reallyputchar(int); 159 160 static void enlarge_stack(void); 161 162 int main(int, char *[]); 163 164 int exit_code = 0; 165 166 int 167 main(int argc, char *argv[]) 168 { 169 int c; 170 int n; 171 char *p; 172 173 if (pledge("stdio rpath wpath cpath tmppath proc exec", NULL) == -1) 174 err(1, "pledge"); 175 176 if (signal(SIGINT, SIG_IGN) != SIG_IGN) 177 signal(SIGINT, onintr); 178 179 init_macros(); 180 initspaces(); 181 STACKMAX = INITSTACKMAX; 182 183 mstack = xreallocarray(NULL, STACKMAX, sizeof(stae), NULL); 184 sstack = xalloc(STACKMAX, NULL); 185 186 maxout = 0; 187 outfile = NULL; 188 resizedivs(MAXOUT); 189 190 while ((c = getopt(argc, argv, "gst:d:D:EU:o:I:P")) != -1) 191 switch(c) { 192 193 case 'D': /* define something..*/ 194 for (p = optarg; *p; p++) 195 if (*p == '=') 196 break; 197 if (*p) 198 *p++ = EOS; 199 dodefine(optarg, p); 200 break; 201 case 'E': /* like GNU m4 1.4.9+ */ 202 if (error_warns == 0) 203 error_warns = 1; 204 else 205 fatal_warns = 1; 206 break; 207 case 'I': 208 addtoincludepath(optarg); 209 break; 210 case 'P': 211 prefix_builtins = 1; 212 break; 213 case 'U': /* undefine... */ 214 macro_popdef(optarg); 215 break; 216 case 'g': 217 mimic_gnu = 1; 218 break; 219 case 'd': 220 set_trace_flags(optarg); 221 break; 222 case 's': 223 synch_lines = 1; 224 break; 225 case 't': 226 mark_traced(optarg, 1); 227 break; 228 case 'o': 229 trace_file(optarg); 230 break; 231 default: 232 usage(); 233 } 234 235 argc -= optind; 236 argv += optind; 237 238 initkwds(); 239 if (mimic_gnu) 240 setup_builtin("format", FORMATTYPE); 241 242 active = stdout; /* default active output */ 243 bbase[0] = bufbase; 244 if (!argc) { 245 sp = -1; /* stack pointer initialized */ 246 fp = 0; /* frame pointer initialized */ 247 set_input(infile+0, stdin, "stdin"); 248 /* default input (naturally) */ 249 macro(); 250 } else 251 for (; argc--; ++argv) { 252 p = *argv; 253 if (p[0] == '-' && p[1] == EOS) 254 set_input(infile, stdin, "stdin"); 255 else if (fopen_trypath(infile, p) == NULL) 256 err(1, "%s", p); 257 sp = -1; 258 fp = 0; 259 macro(); 260 release_input(infile); 261 } 262 263 if (wrapindex) { 264 int i; 265 266 ilevel = 0; /* in case m4wrap includes.. */ 267 bufbase = bp = buf; /* use the entire buffer */ 268 if (mimic_gnu) { 269 while (wrapindex != 0) { 270 for (i = 0; i < wrapindex; i++) 271 pbstr(m4wraps[i]); 272 wrapindex =0; 273 macro(); 274 } 275 } else { 276 for (i = 0; i < wrapindex; i++) { 277 pbstr(m4wraps[i]); 278 macro(); 279 } 280 } 281 } 282 283 if (active != stdout) 284 active = stdout; /* reset output just in case */ 285 for (n = 1; n < maxout; n++) /* default wrap-up: undivert */ 286 if (outfile[n] != NULL) 287 getdiv(n); 288 /* remove bitbucket if used */ 289 if (outfile[0] != NULL) { 290 (void) fclose(outfile[0]); 291 } 292 293 return exit_code; 294 } 295 296 /* 297 * Look ahead for `token'. 298 * (on input `t == token[0]') 299 * Used for comment and quoting delimiters. 300 * Returns 1 if `token' present; copied to output. 301 * 0 if `token' not found; all characters pushed back 302 */ 303 static int 304 do_look_ahead(int t, const char *token) 305 { 306 int i; 307 308 assert((unsigned char)t == (unsigned char)token[0]); 309 310 for (i = 1; *++token; i++) { 311 t = gpbc(); 312 if (t == EOF || (unsigned char)t != (unsigned char)*token) { 313 pushback(t); 314 while (--i) 315 pushback(*--token); 316 return 0; 317 } 318 } 319 return 1; 320 } 321 322 #define LOOK_AHEAD(t, token) (t != EOF && \ 323 (unsigned char)(t)==(unsigned char)(token)[0] && \ 324 do_look_ahead(t,token)) 325 326 /* 327 * macro - the work horse.. 328 */ 329 static void 330 macro(void) 331 { 332 char token[MAXTOK+1]; 333 int t, l; 334 ndptr p; 335 int nlpar; 336 337 cycle { 338 t = gpbc(); 339 340 if (LOOK_AHEAD(t,lquote)) { /* strip quotes */ 341 nlpar = 0; 342 record(quotes, nlpar++); 343 /* 344 * Opening quote: scan forward until matching 345 * closing quote has been found. 346 */ 347 do { 348 349 l = gpbc(); 350 if (LOOK_AHEAD(l,rquote)) { 351 if (--nlpar > 0) 352 outputstr(rquote); 353 } else if (LOOK_AHEAD(l,lquote)) { 354 record(quotes, nlpar++); 355 outputstr(lquote); 356 } else if (l == EOF) { 357 if (nlpar == 1) 358 warnx("unclosed quote:"); 359 else 360 warnx("%d unclosed quotes:", nlpar); 361 dump_stack(quotes, nlpar); 362 exit(1); 363 } else { 364 if (nlpar > 0) { 365 if (sp < 0) 366 reallyputchar(l); 367 else 368 CHRSAVE(l); 369 } 370 } 371 } 372 while (nlpar != 0); 373 } else if (sp < 0 && LOOK_AHEAD(t, scommt)) { 374 reallyoutputstr(scommt); 375 376 for(;;) { 377 t = gpbc(); 378 if (LOOK_AHEAD(t, ecommt)) { 379 reallyoutputstr(ecommt); 380 break; 381 } 382 if (t == EOF) 383 break; 384 reallyputchar(t); 385 } 386 } else if (t == '_' || isalpha(t)) { 387 p = inspect(t, token); 388 if (p != NULL) 389 pushback(l = gpbc()); 390 if (p == NULL || (l != LPAREN && 391 (macro_getdef(p)->type & NEEDARGS) != 0)) 392 outputstr(token); 393 else { 394 /* 395 * real thing.. First build a call frame: 396 */ 397 pushf(fp); /* previous call frm */ 398 pushf(macro_getdef(p)->type); /* type of the call */ 399 pushf(is_traced(p)); 400 pushf(0); /* parenthesis level */ 401 fp = sp; /* new frame pointer */ 402 /* 403 * now push the string arguments: 404 */ 405 pushdef(p); /* defn string */ 406 pushs1((char *)macro_name(p)); /* macro name */ 407 pushs(ep); /* start next..*/ 408 409 if (l != LPAREN && PARLEV == 0) { 410 /* no bracks */ 411 chrsave(EOS); 412 413 if (sp == STACKMAX) 414 errx(1, "internal stack overflow"); 415 eval((const char **) mstack+fp+1, 2, 416 CALTYP, TRACESTATUS); 417 418 ep = PREVEP; /* flush strspace */ 419 sp = PREVSP; /* previous sp.. */ 420 fp = PREVFP; /* rewind stack...*/ 421 } 422 } 423 } else if (t == EOF) { 424 if (!mimic_gnu /* you can puke right there */ 425 && sp > -1 && ilevel <= 0) { 426 warnx( "unexpected end of input, unclosed parenthesis:"); 427 dump_stack(paren, PARLEV); 428 exit(1); 429 } 430 if (ilevel <= 0) 431 break; /* all done thanks.. */ 432 release_input(infile+ilevel--); 433 emit_synchline(); 434 bufbase = bbase[ilevel]; 435 continue; 436 } else if (sp < 0) { /* not in a macro at all */ 437 reallyputchar(t); /* output directly.. */ 438 } 439 440 else switch(t) { 441 442 case LPAREN: 443 if (PARLEV > 0) 444 chrsave(t); 445 while (isspace(l = gpbc())) /* skip blank, tab, nl.. */ 446 if (PARLEV > 0) 447 chrsave(l); 448 pushback(l); 449 record(paren, PARLEV++); 450 break; 451 452 case RPAREN: 453 if (--PARLEV > 0) 454 chrsave(t); 455 else { /* end of argument list */ 456 chrsave(EOS); 457 458 if (sp == STACKMAX) 459 errx(1, "internal stack overflow"); 460 461 eval((const char **) mstack+fp+1, sp-fp, 462 CALTYP, TRACESTATUS); 463 464 ep = PREVEP; /* flush strspace */ 465 sp = PREVSP; /* previous sp.. */ 466 fp = PREVFP; /* rewind stack...*/ 467 } 468 break; 469 470 case COMMA: 471 if (PARLEV == 1) { 472 chrsave(EOS); /* new argument */ 473 while (isspace(l = gpbc())) 474 ; 475 pushback(l); 476 pushs(ep); 477 } else 478 chrsave(t); 479 break; 480 481 default: 482 if (LOOK_AHEAD(t, scommt)) { 483 char *p; 484 for (p = scommt; *p; p++) 485 chrsave(*p); 486 for(;;) { 487 t = gpbc(); 488 if (LOOK_AHEAD(t, ecommt)) { 489 for (p = ecommt; *p; p++) 490 chrsave(*p); 491 break; 492 } 493 if (t == EOF) 494 break; 495 CHRSAVE(t); 496 } 497 } else 498 CHRSAVE(t); /* stack the char */ 499 break; 500 } 501 } 502 } 503 504 /* 505 * output string directly, without pushing it for reparses. 506 */ 507 void 508 outputstr(const char *s) 509 { 510 if (sp < 0) 511 reallyoutputstr(s); 512 else 513 while (*s) 514 CHRSAVE(*s++); 515 } 516 517 void 518 reallyoutputstr(const char *s) 519 { 520 if (synch_lines) { 521 while (*s) { 522 fputc(*s, active); 523 if (*s++ == '\n') { 524 infile[ilevel].synch_lineno++; 525 if (infile[ilevel].synch_lineno != 526 infile[ilevel].lineno) 527 do_emit_synchline(); 528 } 529 } 530 } else 531 fputs(s, active); 532 } 533 534 void 535 reallyputchar(int c) 536 { 537 putc(c, active); 538 if (synch_lines && c == '\n') { 539 infile[ilevel].synch_lineno++; 540 if (infile[ilevel].synch_lineno != infile[ilevel].lineno) 541 do_emit_synchline(); 542 } 543 } 544 545 /* 546 * build an input token.. 547 * consider only those starting with _ or A-Za-z. 548 */ 549 static ndptr 550 inspect(int c, char *tp) 551 { 552 char *name = tp; 553 char *etp = tp+MAXTOK; 554 ndptr p; 555 556 *tp++ = c; 557 558 while ((isalnum(c = gpbc()) || c == '_') && tp < etp) 559 *tp++ = c; 560 if (c != EOF) 561 PUSHBACK(c); 562 *tp = EOS; 563 /* token is too long, it won't match anything, but it can still 564 * be output. */ 565 if (tp == ep) { 566 outputstr(name); 567 while (isalnum(c = gpbc()) || c == '_') { 568 if (sp < 0) 569 reallyputchar(c); 570 else 571 CHRSAVE(c); 572 } 573 *name = EOS; 574 return NULL; 575 } 576 577 p = ohash_find(¯os, ohash_qlookupi(¯os, name, (const char **)&tp)); 578 if (p == NULL) 579 return NULL; 580 if (macro_getdef(p) == NULL) 581 return NULL; 582 return p; 583 } 584 585 /* 586 * initkwds - initialise m4 keywords as fast as possible. 587 * This very similar to install, but without certain overheads, 588 * such as calling lookup. Malloc is not used for storing the 589 * keyword strings, since we simply use the static pointers 590 * within keywrds block. 591 */ 592 static void 593 initkwds(void) 594 { 595 unsigned int type; 596 int i; 597 598 for (i = 0; i < MAXKEYS; i++) { 599 type = keywrds[i].ktyp & TYPEMASK; 600 if ((keywrds[i].ktyp & NOARGS) == 0) 601 type |= NEEDARGS; 602 setup_builtin(keywrds[i].knam, type); 603 } 604 } 605 606 static void 607 record(struct position *t, int lev) 608 { 609 if (lev < MAXRECORD) { 610 t[lev].name = CURRENT_NAME; 611 t[lev].line = CURRENT_LINE; 612 } 613 } 614 615 static void 616 dump_stack(struct position *t, int lev) 617 { 618 int i; 619 620 for (i = 0; i < lev; i++) { 621 if (i == MAXRECORD) { 622 fprintf(stderr, " ...\n"); 623 break; 624 } 625 fprintf(stderr, " %s at line %lu\n", 626 t[i].name, t[i].line); 627 } 628 } 629 630 631 static void 632 enlarge_stack(void) 633 { 634 STACKMAX += STACKMAX/2; 635 mstack = xreallocarray(mstack, STACKMAX, sizeof(stae), 636 "Evaluation stack overflow (%lu)", 637 (unsigned long)STACKMAX); 638 sstack = xrealloc(sstack, STACKMAX, 639 "Evaluation stack overflow (%lu)", 640 (unsigned long)STACKMAX); 641 } 642