1 /* $OpenBSD: main.c,v 1.84 2014/12/21 09:33:12 espie 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 81 struct keyblk { 82 char *knam; /* keyword name */ 83 int ktyp; /* keyword type */ 84 }; 85 86 struct keyblk keywrds[] = { /* m4 keywords to be installed */ 87 { "include", INCLTYPE }, 88 { "sinclude", SINCTYPE }, 89 { "define", DEFITYPE }, 90 { "defn", DEFNTYPE }, 91 { "divert", DIVRTYPE | NOARGS }, 92 { "expr", EXPRTYPE }, 93 { "eval", EXPRTYPE }, 94 { "substr", SUBSTYPE }, 95 { "ifelse", IFELTYPE }, 96 { "ifdef", IFDFTYPE }, 97 { "len", LENGTYPE }, 98 { "incr", INCRTYPE }, 99 { "decr", DECRTYPE }, 100 { "dnl", DNLNTYPE | NOARGS }, 101 { "changequote", CHNQTYPE | NOARGS }, 102 { "changecom", CHNCTYPE | NOARGS }, 103 { "index", INDXTYPE }, 104 #ifdef EXTENDED 105 { "paste", PASTTYPE }, 106 { "spaste", SPASTYPE }, 107 /* Newer extensions, needed to handle gnu-m4 scripts */ 108 { "indir", INDIRTYPE}, 109 { "builtin", BUILTINTYPE}, 110 { "patsubst", PATSTYPE}, 111 { "regexp", REGEXPTYPE}, 112 { "esyscmd", ESYSCMDTYPE}, 113 { "__file__", FILENAMETYPE | NOARGS}, 114 { "__line__", LINETYPE | NOARGS}, 115 #endif 116 { "popdef", POPDTYPE }, 117 { "pushdef", PUSDTYPE }, 118 { "dumpdef", DUMPTYPE | NOARGS }, 119 { "shift", SHIFTYPE | NOARGS }, 120 { "translit", TRNLTYPE }, 121 { "undefine", UNDFTYPE }, 122 { "undivert", UNDVTYPE | NOARGS }, 123 { "divnum", DIVNTYPE | NOARGS }, 124 { "maketemp", MKTMTYPE }, 125 { "mkstemp", MKTMTYPE }, 126 { "errprint", ERRPTYPE | NOARGS }, 127 { "m4wrap", M4WRTYPE | NOARGS }, 128 { "m4exit", EXITTYPE | NOARGS }, 129 { "syscmd", SYSCTYPE }, 130 { "sysval", SYSVTYPE | NOARGS }, 131 { "traceon", TRACEONTYPE | NOARGS }, 132 { "traceoff", TRACEOFFTYPE | NOARGS }, 133 134 #if defined(unix) || defined(__unix__) 135 { "unix", SELFTYPE | NOARGS }, 136 #else 137 #ifdef vms 138 { "vms", SELFTYPE | NOARGS }, 139 #endif 140 #endif 141 }; 142 143 #define MAXKEYS (sizeof(keywrds)/sizeof(struct keyblk)) 144 145 extern int optind; 146 extern char *optarg; 147 148 #define MAXRECORD 50 149 static struct position { 150 char *name; 151 unsigned long line; 152 } quotes[MAXRECORD], paren[MAXRECORD]; 153 154 static void record(struct position *, int); 155 static void dump_stack(struct position *, int); 156 157 static void macro(void); 158 static void initkwds(void); 159 static ndptr inspect(int, char *); 160 static int do_look_ahead(int, const char *); 161 static void reallyoutputstr(const char *); 162 static void reallyputchar(int); 163 164 static void enlarge_stack(void); 165 166 int main(int, char *[]); 167 168 int exit_code = 0; 169 170 int 171 main(int argc, char *argv[]) 172 { 173 int c; 174 int n; 175 char *p; 176 177 if (signal(SIGINT, SIG_IGN) != SIG_IGN) 178 signal(SIGINT, onintr); 179 180 init_macros(); 181 initspaces(); 182 STACKMAX = INITSTACKMAX; 183 184 mstack = xreallocarray(NULL, STACKMAX, sizeof(stae), NULL); 185 sstack = xalloc(STACKMAX, NULL); 186 187 maxout = 0; 188 outfile = NULL; 189 resizedivs(MAXOUT); 190 191 while ((c = getopt(argc, argv, "gst:d:D:U:o:I:P")) != -1) 192 switch(c) { 193 194 case 'D': /* define something..*/ 195 for (p = optarg; *p; p++) 196 if (*p == '=') 197 break; 198 if (*p) 199 *p++ = EOS; 200 dodefine(optarg, p); 201 break; 202 case 'I': 203 addtoincludepath(optarg); 204 break; 205 case 'P': 206 prefix_builtins = 1; 207 break; 208 case 'U': /* undefine... */ 209 macro_popdef(optarg); 210 break; 211 case 'g': 212 mimic_gnu = 1; 213 break; 214 case 'd': 215 set_trace_flags(optarg); 216 break; 217 case 's': 218 synch_lines = 1; 219 break; 220 case 't': 221 mark_traced(optarg, 1); 222 break; 223 case 'o': 224 trace_file(optarg); 225 break; 226 case '?': 227 usage(); 228 } 229 230 argc -= optind; 231 argv += optind; 232 233 initkwds(); 234 if (mimic_gnu) 235 setup_builtin("format", FORMATTYPE); 236 237 active = stdout; /* default active output */ 238 bbase[0] = bufbase; 239 if (!argc) { 240 sp = -1; /* stack pointer initialized */ 241 fp = 0; /* frame pointer initialized */ 242 set_input(infile+0, stdin, "stdin"); 243 /* default input (naturally) */ 244 macro(); 245 } else 246 for (; argc--; ++argv) { 247 p = *argv; 248 if (p[0] == '-' && p[1] == EOS) 249 set_input(infile, stdin, "stdin"); 250 else if (fopen_trypath(infile, p) == NULL) 251 err(1, "%s", p); 252 sp = -1; 253 fp = 0; 254 macro(); 255 release_input(infile); 256 } 257 258 if (wrapindex) { 259 int i; 260 261 ilevel = 0; /* in case m4wrap includes.. */ 262 bufbase = bp = buf; /* use the entire buffer */ 263 if (mimic_gnu) { 264 while (wrapindex != 0) { 265 for (i = 0; i < wrapindex; i++) 266 pbstr(m4wraps[i]); 267 wrapindex =0; 268 macro(); 269 } 270 } else { 271 for (i = 0; i < wrapindex; i++) { 272 pbstr(m4wraps[i]); 273 macro(); 274 } 275 } 276 } 277 278 if (active != stdout) 279 active = stdout; /* reset output just in case */ 280 for (n = 1; n < maxout; n++) /* default wrap-up: undivert */ 281 if (outfile[n] != NULL) 282 getdiv(n); 283 /* remove bitbucket if used */ 284 if (outfile[0] != NULL) { 285 (void) fclose(outfile[0]); 286 } 287 288 return exit_code; 289 } 290 291 /* 292 * Look ahead for `token'. 293 * (on input `t == token[0]') 294 * Used for comment and quoting delimiters. 295 * Returns 1 if `token' present; copied to output. 296 * 0 if `token' not found; all characters pushed back 297 */ 298 static int 299 do_look_ahead(int t, const char *token) 300 { 301 int i; 302 303 assert((unsigned char)t == (unsigned char)token[0]); 304 305 for (i = 1; *++token; i++) { 306 t = gpbc(); 307 if (t == EOF || (unsigned char)t != (unsigned char)*token) { 308 pushback(t); 309 while (--i) 310 pushback(*--token); 311 return 0; 312 } 313 } 314 return 1; 315 } 316 317 #define LOOK_AHEAD(t, token) (t != EOF && \ 318 (unsigned char)(t)==(unsigned char)(token)[0] && \ 319 do_look_ahead(t,token)) 320 321 /* 322 * macro - the work horse.. 323 */ 324 static void 325 macro(void) 326 { 327 char token[MAXTOK+1]; 328 int t, l; 329 ndptr p; 330 int nlpar; 331 332 cycle { 333 t = gpbc(); 334 335 if (LOOK_AHEAD(t,lquote)) { /* strip quotes */ 336 nlpar = 0; 337 record(quotes, nlpar++); 338 /* 339 * Opening quote: scan forward until matching 340 * closing quote has been found. 341 */ 342 do { 343 344 l = gpbc(); 345 if (LOOK_AHEAD(l,rquote)) { 346 if (--nlpar > 0) 347 outputstr(rquote); 348 } else if (LOOK_AHEAD(l,lquote)) { 349 record(quotes, nlpar++); 350 outputstr(lquote); 351 } else if (l == EOF) { 352 if (nlpar == 1) 353 warnx("unclosed quote:"); 354 else 355 warnx("%d unclosed quotes:", nlpar); 356 dump_stack(quotes, nlpar); 357 exit(1); 358 } else { 359 if (nlpar > 0) { 360 if (sp < 0) 361 reallyputchar(l); 362 else 363 CHRSAVE(l); 364 } 365 } 366 } 367 while (nlpar != 0); 368 } else if (sp < 0 && LOOK_AHEAD(t, scommt)) { 369 reallyoutputstr(scommt); 370 371 for(;;) { 372 t = gpbc(); 373 if (LOOK_AHEAD(t, ecommt)) { 374 reallyoutputstr(ecommt); 375 break; 376 } 377 if (t == EOF) 378 break; 379 reallyputchar(t); 380 } 381 } else if (t == '_' || isalpha(t)) { 382 p = inspect(t, token); 383 if (p != NULL) 384 pushback(l = gpbc()); 385 if (p == NULL || (l != LPAREN && 386 (macro_getdef(p)->type & NEEDARGS) != 0)) 387 outputstr(token); 388 else { 389 /* 390 * real thing.. First build a call frame: 391 */ 392 pushf(fp); /* previous call frm */ 393 pushf(macro_getdef(p)->type); /* type of the call */ 394 pushf(is_traced(p)); 395 pushf(0); /* parenthesis level */ 396 fp = sp; /* new frame pointer */ 397 /* 398 * now push the string arguments: 399 */ 400 pushdef(p); /* defn string */ 401 pushs1((char *)macro_name(p)); /* macro name */ 402 pushs(ep); /* start next..*/ 403 404 if (l != LPAREN && PARLEV == 0) { 405 /* no bracks */ 406 chrsave(EOS); 407 408 if (sp == STACKMAX) 409 errx(1, "internal stack overflow"); 410 eval((const char **) mstack+fp+1, 2, 411 CALTYP, TRACESTATUS); 412 413 ep = PREVEP; /* flush strspace */ 414 sp = PREVSP; /* previous sp.. */ 415 fp = PREVFP; /* rewind stack...*/ 416 } 417 } 418 } else if (t == EOF) { 419 if (!mimic_gnu /* you can puke right there */ 420 && sp > -1 && ilevel <= 0) { 421 warnx( "unexpected end of input, unclosed parenthesis:"); 422 dump_stack(paren, PARLEV); 423 exit(1); 424 } 425 if (ilevel <= 0) 426 break; /* all done thanks.. */ 427 release_input(infile+ilevel--); 428 emit_synchline(); 429 bufbase = bbase[ilevel]; 430 continue; 431 } else if (sp < 0) { /* not in a macro at all */ 432 reallyputchar(t); /* output directly.. */ 433 } 434 435 else switch(t) { 436 437 case LPAREN: 438 if (PARLEV > 0) 439 chrsave(t); 440 while (isspace(l = gpbc())) /* skip blank, tab, nl.. */ 441 if (PARLEV > 0) 442 chrsave(l); 443 pushback(l); 444 record(paren, PARLEV++); 445 break; 446 447 case RPAREN: 448 if (--PARLEV > 0) 449 chrsave(t); 450 else { /* end of argument list */ 451 chrsave(EOS); 452 453 if (sp == STACKMAX) 454 errx(1, "internal stack overflow"); 455 456 eval((const char **) mstack+fp+1, sp-fp, 457 CALTYP, TRACESTATUS); 458 459 ep = PREVEP; /* flush strspace */ 460 sp = PREVSP; /* previous sp.. */ 461 fp = PREVFP; /* rewind stack...*/ 462 } 463 break; 464 465 case COMMA: 466 if (PARLEV == 1) { 467 chrsave(EOS); /* new argument */ 468 while (isspace(l = gpbc())) 469 ; 470 pushback(l); 471 pushs(ep); 472 } else 473 chrsave(t); 474 break; 475 476 default: 477 if (LOOK_AHEAD(t, scommt)) { 478 char *p; 479 for (p = scommt; *p; p++) 480 chrsave(*p); 481 for(;;) { 482 t = gpbc(); 483 if (LOOK_AHEAD(t, ecommt)) { 484 for (p = ecommt; *p; p++) 485 chrsave(*p); 486 break; 487 } 488 if (t == EOF) 489 break; 490 CHRSAVE(t); 491 } 492 } else 493 CHRSAVE(t); /* stack the char */ 494 break; 495 } 496 } 497 } 498 499 /* 500 * output string directly, without pushing it for reparses. 501 */ 502 void 503 outputstr(const char *s) 504 { 505 if (sp < 0) 506 reallyoutputstr(s); 507 else 508 while (*s) 509 CHRSAVE(*s++); 510 } 511 512 void 513 reallyoutputstr(const char *s) 514 { 515 if (synch_lines) { 516 while (*s) { 517 fputc(*s, active); 518 if (*s++ == '\n') { 519 infile[ilevel].synch_lineno++; 520 if (infile[ilevel].synch_lineno != 521 infile[ilevel].lineno) 522 do_emit_synchline(); 523 } 524 } 525 } else 526 fputs(s, active); 527 } 528 529 void 530 reallyputchar(int c) 531 { 532 putc(c, active); 533 if (synch_lines && c == '\n') { 534 infile[ilevel].synch_lineno++; 535 if (infile[ilevel].synch_lineno != infile[ilevel].lineno) 536 do_emit_synchline(); 537 } 538 } 539 540 /* 541 * build an input token.. 542 * consider only those starting with _ or A-Za-z. 543 */ 544 static ndptr 545 inspect(int c, char *tp) 546 { 547 char *name = tp; 548 char *etp = tp+MAXTOK; 549 ndptr p; 550 551 *tp++ = c; 552 553 while ((isalnum(c = gpbc()) || c == '_') && tp < etp) 554 *tp++ = c; 555 if (c != EOF) 556 PUSHBACK(c); 557 *tp = EOS; 558 /* token is too long, it won't match anything, but it can still 559 * be output. */ 560 if (tp == ep) { 561 outputstr(name); 562 while (isalnum(c = gpbc()) || c == '_') { 563 if (sp < 0) 564 reallyputchar(c); 565 else 566 CHRSAVE(c); 567 } 568 *name = EOS; 569 return NULL; 570 } 571 572 p = ohash_find(¯os, ohash_qlookupi(¯os, name, (const char **)&tp)); 573 if (p == NULL) 574 return NULL; 575 if (macro_getdef(p) == NULL) 576 return NULL; 577 return p; 578 } 579 580 /* 581 * initkwds - initialise m4 keywords as fast as possible. 582 * This very similar to install, but without certain overheads, 583 * such as calling lookup. Malloc is not used for storing the 584 * keyword strings, since we simply use the static pointers 585 * within keywrds block. 586 */ 587 static void 588 initkwds(void) 589 { 590 unsigned int type; 591 int i; 592 593 for (i = 0; i < MAXKEYS; i++) { 594 type = keywrds[i].ktyp & TYPEMASK; 595 if ((keywrds[i].ktyp & NOARGS) == 0) 596 type |= NEEDARGS; 597 setup_builtin(keywrds[i].knam, type); 598 } 599 } 600 601 static void 602 record(struct position *t, int lev) 603 { 604 if (lev < MAXRECORD) { 605 t[lev].name = CURRENT_NAME; 606 t[lev].line = CURRENT_LINE; 607 } 608 } 609 610 static void 611 dump_stack(struct position *t, int lev) 612 { 613 int i; 614 615 for (i = 0; i < lev; i++) { 616 if (i == MAXRECORD) { 617 fprintf(stderr, " ...\n"); 618 break; 619 } 620 fprintf(stderr, " %s at line %lu\n", 621 t[i].name, t[i].line); 622 } 623 } 624 625 626 static void 627 enlarge_stack(void) 628 { 629 STACKMAX += STACKMAX/2; 630 mstack = xreallocarray(mstack, STACKMAX, sizeof(stae), 631 "Evaluation stack overflow (%lu)", 632 (unsigned long)STACKMAX); 633 sstack = xrealloc(sstack, STACKMAX, 634 "Evaluation stack overflow (%lu)", 635 (unsigned long)STACKMAX); 636 } 637