usr.bin/sed/main.c

   1 /*-
   2  * Copyright (c) 1992 Diomidis Spinellis.
   3  * Copyright (c) 1992, 1993
   4  *      The Regents of the University of California.  All rights reserved.
   5  *
   6  * This code is derived from software contributed to Berkeley by
   7  * Diomidis Spinellis of Imperial College, University of London.
   8  *
   9  * Redistribution and use in source and binary forms, with or without
  10  * modification, are permitted provided that the following conditions
  11  * are met:
  12  * 1. Redistributions of source code must retain the above copyright
  13  *    notice, this list of conditions and the following disclaimer.
  14  * 2. Redistributions in binary form must reproduce the above copyright
  15  *    notice, this list of conditions and the following disclaimer in the
  16  *    documentation and/or other materials provided with the distribution.
  17  * 4. Neither the name of the University nor the names of its contributors
  18  *    may be used to endorse or promote products derived from this software
  19  *    without specific prior written permission.
  20  *
  21  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  22  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  23  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  24  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  25  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  26  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  27  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  28  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  29  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  30  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  31  * SUCH DAMAGE.
  32  *
  33  * @(#) Copyright (c) 1992, 1993 The Regents of the University of California.  All rights reserved.
  34  * @(#)main.c   8.2 (Berkeley) 1/3/94
  35  * $FreeBSD: src/usr.bin/sed/main.c,v 1.41 2008/02/09 09:12:02 dwmalone Exp $
  36  * $DragonFly: src/usr.bin/sed/main.c,v 1.4 2008/04/08 13:23:38 swildner Exp $
  37  */
  38
  39 #include <sys/types.h>
  40 #include <sys/mman.h>
  41 #include <sys/param.h>
  42 #include <sys/stat.h>
  43
  44 #include <err.h>
  45 #include <errno.h>
  46 #include <fcntl.h>
  47 #include <libgen.h>
  48 #include <limits.h>
  49 #include <locale.h>
  50 #include <regex.h>
  51 #include <stddef.h>
  52 #include <stdio.h>
  53 #include <stdlib.h>
  54 #include <string.h>
  55 #include <unistd.h>
  56
  57 #include "defs.h"
  58 #include "extern.h"
  59
  60 /*
  61  * Linked list of units (strings and files) to be compiled
  62  */
  63 struct s_compunit {
  64         struct s_compunit *next;
  65         enum e_cut {CU_FILE, CU_STRING} type;
  66         char *s;                        /* Pointer to string or fname */
  67 };
  68
  69 /*
  70  * Linked list pointer to compilation units and pointer to current
  71  * next pointer.
  72  */
  73 static struct s_compunit *script, **cu_nextp = &script;
  74
  75 /*
  76  * Linked list of files to be processed
  77  */
  78 struct s_flist {
  79         char *fname;
  80         struct s_flist *next;
  81 };
  82
  83 /*
  84  * Linked list pointer to files and pointer to current
  85  * next pointer.
  86  */
  87 static struct s_flist *files, **fl_nextp = &files;
  88
  89 FILE *infile;                   /* Current input file */
  90 FILE *outfile;                  /* Current output file */
  91
  92 int aflag, eflag, nflag;
  93 int rflags = 0;
  94 static int rval;                /* Exit status */
  95
  96 static int ispan;               /* Whether inplace editing spans across files */
  97
  98 /*
  99  * Current file and line number; line numbers restart across compilation
 100  * units, but span across input files.  The latter is optional if editing
 101  * in place.
 102  */
 103 const char *fname;              /* File name. */
 104 const char *outfname;           /* Output file name */
 105 static char oldfname[PATH_MAX]; /* Old file name (for in-place editing) */
 106 static char tmpfname[PATH_MAX]; /* Temporary file name (for in-place editing) */
 107 static const char *inplace;     /* Inplace edit file extension. */
 108 u_long linenum;
 109
 110 static void add_compunit(enum e_cut, char *);
 111 static void add_file(char *);
 112 static void usage(void);
 113
 114 int
 115 main(int argc, char *argv[])
 116 {
 117         int c, fflag;
 118         char *temp_arg;
 119
 120         (void) setlocale(LC_ALL, "");
 121
 122         fflag = 0;
 123         inplace = NULL;
 124
 125         while ((c = getopt(argc, argv, "EI:ae:f:i:ln")) != -1)
 126                 switch (c) {
 127                 case 'E':
 128                         rflags = REG_EXTENDED;
 129                         break;
 130                 case 'I':
 131                         inplace = optarg;
 132                         ispan = 1;      /* span across input files */
 133                         break;
 134                 case 'a':
 135                         aflag = 1;
 136                         break;
 137                 case 'e':
 138                         eflag = 1;
 139                         if ((temp_arg = malloc(strlen(optarg) + 2)) == NULL)
 140                                 err(1, "malloc");
 141                         strcpy(temp_arg, optarg);
 142                         strcat(temp_arg, "\n");
 143                         add_compunit(CU_STRING, temp_arg);
 144                         break;
 145                 case 'f':
 146                         fflag = 1;
 147                         add_compunit(CU_FILE, optarg);
 148                         break;
 149                 case 'i':
 150                         inplace = optarg;
 151                         ispan = 0;      /* don't span across input files */
 152                         break;
 153                 case 'l':
 154                         if(setlinebuf(stdout) != 0)
 155                                 warnx("setlinebuf() failed");
 156                         break;
 157                 case 'n':
 158                         nflag = 1;
 159                         break;
 160                 default:
 161                 case '?':
 162                         usage();
 163                 }
 164         argc -= optind;
 165         argv += optind;
 166
 167         /* First usage case; script is the first arg */
 168         if (!eflag && !fflag && *argv) {
 169                 add_compunit(CU_STRING, *argv);
 170                 argv++;
 171         }
 172
 173         compile();
 174
 175         /* Continue with first and start second usage */
 176         if (*argv)
 177                 for (; *argv; argv++)
 178                         add_file(*argv);
 179         else
 180                 add_file(NULL);
 181         process();
 182         cfclose(prog, NULL);
 183         if (fclose(stdout))
 184                 err(1, "stdout");
 185         exit(rval);
 186 }
 187
 188 static void
 189 usage(void)
 190 {
 191         (void)fprintf(stderr, "%s\n%s\n",
 192                 "usage: sed script [-Ealn] [-i extension] [file ...]",
 193                 "       sed [-Ealn] [-i extension] [-e script] ... [-f script_file] ... [file ...]");
 194         exit(1);
 195 }
 196
 197 /*
 198  * Like fgets, but go through the chain of compilation units chaining them
 199  * together.  Empty strings and files are ignored.
 200  */
 201 char *
 202 cu_fgets(char *buf, int n, int *more)
 203 {
 204         static enum {ST_EOF, ST_FILE, ST_STRING} state = ST_EOF;
 205         static FILE *f;         /* Current open file */
 206         static char *s;         /* Current pointer inside string */
 207         static char string_ident[30];
 208         char *p;
 209
 210 again:
 211         switch (state) {
 212         case ST_EOF:
 213                 if (script == NULL) {
 214                         if (more != NULL)
 215                                 *more = 0;
 216                         return (NULL);
 217                 }
 218                 linenum = 0;
 219                 switch (script->type) {
 220                 case CU_FILE:
 221                         if ((f = fopen(script->s, "r")) == NULL)
 222                                 err(1, "%s", script->s);
 223                         fname = script->s;
 224                         state = ST_FILE;
 225                         goto again;
 226                 case CU_STRING:
 227                         if (((size_t)snprintf(string_ident,
 228                             sizeof(string_ident), "\"%s\"", script->s)) >=
 229                             sizeof(string_ident) - 1)
 230                                 (void)strcpy(string_ident +
 231                                     sizeof(string_ident) - 6, " ...\"");
 232                         fname = string_ident;
 233                         s = script->s;
 234                         state = ST_STRING;
 235                         goto again;
 236                 }
 237         case ST_FILE:
 238                 if ((p = fgets(buf, n, f)) != NULL) {
 239                         linenum++;
 240                         if (linenum == 1 && buf[0] == '#' && buf[1] == 'n')
 241                                 nflag = 1;
 242                         if (more != NULL)
 243                                 *more = !feof(f);
 244                         return (p);
 245                 }
 246                 script = script->next;
 247                 (void)fclose(f);
 248                 state = ST_EOF;
 249                 goto again;
 250         case ST_STRING:
 251                 if (linenum == 0 && s[0] == '#' && s[1] == 'n')
 252                         nflag = 1;
 253                 p = buf;
 254                 for (;;) {
 255                         if (n-- <= 1) {
 256                                 *p = '\0';
 257                                 linenum++;
 258                                 if (more != NULL)
 259                                         *more = 1;
 260                                 return (buf);
 261                         }
 262                         switch (*s) {
 263                         case '\0':
 264                                 state = ST_EOF;
 265                                 if (s == script->s) {
 266                                         script = script->next;
 267                                         goto again;
 268                                 } else {
 269                                         script = script->next;
 270                                         *p = '\0';
 271                                         linenum++;
 272                                         if (more != NULL)
 273                                                 *more = 0;
 274                                         return (buf);
 275                                 }
 276                         case '\n':
 277                                 *p++ = '\n';
 278                                 *p = '\0';
 279                                 s++;
 280                                 linenum++;
 281                                 if (more != NULL)
 282                                         *more = 0;
 283                                 return (buf);
 284                         default:
 285                                 *p++ = *s++;
 286                         }
 287                 }
 288         }
 289         /* NOTREACHED */
 290         return (NULL);
 291 }
 292
 293 /*
 294  * Like fgets, but go through the list of files chaining them together.
 295  * Set len to the length of the line.
 296  */
 297 int
 298 mf_fgets(SPACE *sp, enum e_spflag spflag)
 299 {
 300         struct stat sb;
 301         size_t len;
 302         char *p;
 303         int c;
 304         static int firstfile;
 305
 306         if (infile == NULL) {
 307                 /* stdin? */
 308                 if (files->fname == NULL) {
 309                         if (inplace != NULL)
 310                                 errx(1, "-I or -i may not be used with stdin");
 311                         infile = stdin;
 312                         fname = "stdin";
 313                         outfile = stdout;
 314                         outfname = "stdout";
 315                 }
 316                 firstfile = 1;
 317         }
 318
 319         for (;;) {
 320                 if (infile != NULL && (c = getc(infile)) != EOF) {
 321                         (void)ungetc(c, infile);
 322                         break;
 323                 }
 324                 /* If we are here then either eof or no files are open yet */
 325                 if (infile == stdin) {
 326                         sp->len = 0;
 327                         return (0);
 328                 }
 329                 if (infile != NULL) {
 330                         fclose(infile);
 331                         if (*oldfname != '\0') {
 332                                 if (rename(fname, oldfname) != 0) {
 333                                         warn("rename()");
 334                                         unlink(tmpfname);
 335                                         exit(1);
 336                                 }
 337                                 *oldfname = '\0';
 338                         }
 339                         if (*tmpfname != '\0') {
 340                                 if (outfile != NULL && outfile != stdout)
 341                                         fclose(outfile);
 342                                 outfile = NULL;
 343                                 rename(tmpfname, fname);
 344                                 *tmpfname = '\0';
 345                         }
 346                         outfname = NULL;
 347                 }
 348                 if (firstfile == 0)
 349                         files = files->next;
 350                 else
 351                         firstfile = 0;
 352                 if (files == NULL) {
 353                         sp->len = 0;
 354                         return (0);
 355                 }
 356                 fname = files->fname;
 357                 if (inplace != NULL) {
 358                         if (lstat(fname, &sb) != 0)
 359                                 err(1, "%s", fname);
 360                         if (!(sb.st_mode & S_IFREG))
 361                                 errx(1, "%s: %s %s", fname,
 362                                     "in-place editing only",
 363                                     "works for regular files");
 364                         if (*inplace != '\0') {
 365                                 strlcpy(oldfname, fname,
 366                                     sizeof(oldfname));
 367                                 len = strlcat(oldfname, inplace,
 368                                     sizeof(oldfname));
 369                                 if (len > sizeof(oldfname))
 370                                         errx(1, "%s: name too long", fname);
 371                         }
 372                         len = snprintf(tmpfname, sizeof(tmpfname),
 373                             "%s/.!%ld!%s", dirname(fname), (long)getpid(),
 374                             basename(fname));
 375                         if (len >= sizeof(tmpfname))
 376                                 errx(1, "%s: name too long", fname);
 377                         unlink(tmpfname);
 378                         if ((outfile = fopen(tmpfname, "w")) == NULL)
 379                                 err(1, "%s", fname);
 380                         fchown(fileno(outfile), sb.st_uid, sb.st_gid);
 381                         fchmod(fileno(outfile), sb.st_mode & ALLPERMS);
 382                         outfname = tmpfname;
 383                         if (!ispan) {
 384                                 linenum = 0;
 385                                 resetstate();
 386                         }
 387                 } else {
 388                         outfile = stdout;
 389                         outfname = "stdout";
 390                 }
 391                 if ((infile = fopen(fname, "r")) == NULL) {
 392                         warn("%s", fname);
 393                         rval = 1;
 394                         continue;
 395                 }
 396         }
 397         /*
 398          * We are here only when infile is open and we still have something
 399          * to read from it.
 400          *
 401          * Use fgetln so that we can handle essentially infinite input data.
 402          * Can't use the pointer into the stdio buffer as the process space
 403          * because the ungetc() can cause it to move.
 404          */
 405         p = fgetln(infile, &len);
 406         if (ferror(infile))
 407                 errx(1, "%s: %s", fname, strerror(errno ? errno : EIO));
 408         if (len != 0 && p[len - 1] == '\n')
 409                 len--;
 410         cspace(sp, p, len, spflag);
 411
 412         linenum++;
 413
 414         return (1);
 415 }
 416
 417 /*
 418  * Add a compilation unit to the linked list
 419  */
 420 static void
 421 add_compunit(enum e_cut type, char *s)
 422 {
 423         struct s_compunit *cu;
 424
 425         if ((cu = malloc(sizeof(struct s_compunit))) == NULL)
 426                 err(1, "malloc");
 427         cu->type = type;
 428         cu->s = s;
 429         cu->next = NULL;
 430         *cu_nextp = cu;
 431         cu_nextp = &cu->next;
 432 }
 433
 434 /*
 435  * Add a file to the linked list
 436  */
 437 static void
 438 add_file(char *s)
 439 {
 440         struct s_flist *fp;
 441
 442         if ((fp = malloc(sizeof(struct s_flist))) == NULL)
 443                 err(1, "malloc");
 444         fp->next = NULL;
 445         *fl_nextp = fp;
 446         fp->fname = s;
 447         fl_nextp = &fp->next;
 448 }
 449
 450 int
 451 lastline(void)
 452 {
 453         int ch;
 454
 455         if (files->next != NULL && (inplace == NULL || ispan))
 456                 return (0);
 457         if ((ch = getc(infile)) == EOF)
 458                 return (1);
 459         ungetc(ch, infile);
 460         return (0);
 461 }