1 /* 2 * Copyright (c) 1987, 1993, 1994 3 * The Regents of the University of California. All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer. 10 * 2. Redistributions in binary form must reproduce the above copyright 11 * notice, this list of conditions and the following disclaimer in the 12 * documentation and/or other materials provided with the distribution. 13 * 4. Neither the name of the University nor the names of its contributors 14 * may be used to endorse or promote products derived from this software 15 * without specific prior written permission. 16 * 17 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 20 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 27 * SUCH DAMAGE. 28 */ 29 30 #include <sys/cdefs.h> 31 __FBSDID("$FreeBSD$"); 32 33 #ifndef lint 34 static const char copyright[] = 35 "@(#) Copyright (c) 1987, 1993, 1994\n\ 36 The Regents of the University of California. All rights reserved.\n"; 37 #endif 38 39 #ifndef lint 40 static const char sccsid[] = "@(#)split.c 8.2 (Berkeley) 4/16/94"; 41 #endif 42 43 #include <sys/param.h> 44 #include <sys/types.h> 45 #include <sys/stat.h> 46 47 #include <ctype.h> 48 #include <err.h> 49 #include <errno.h> 50 #include <fcntl.h> 51 #include <inttypes.h> 52 #include <limits.h> 53 #include <locale.h> 54 #include <stdbool.h> 55 #include <stdint.h> 56 #include <stdio.h> 57 #include <stdlib.h> 58 #include <string.h> 59 #include <unistd.h> 60 #include <regex.h> 61 #include <sysexits.h> 62 63 #define DEFLINE 1000 /* Default num lines per file. */ 64 65 static off_t bytecnt; /* Byte count to split on. */ 66 static off_t chunks = 0; /* Chunks count to split into. */ 67 static long numlines; /* Line count to split on. */ 68 static int file_open; /* If a file open. */ 69 static int ifd = -1, ofd = -1; /* Input/output file descriptors. */ 70 static char bfr[MAXBSIZE]; /* I/O buffer. */ 71 static char fname[MAXPATHLEN]; /* File name prefix. */ 72 static regex_t rgx; 73 static int pflag; 74 static bool dflag; 75 static long sufflen = 2; /* File name suffix length. */ 76 77 static void newfile(void); 78 static void split1(void); 79 static void split2(void); 80 static void split3(void); 81 static void usage(void); 82 83 int 84 main(int argc, char **argv) 85 { 86 intmax_t bytecnti; 87 long scale; 88 int ch; 89 char *ep, *p; 90 91 setlocale(LC_ALL, ""); 92 93 dflag = false; 94 while ((ch = getopt(argc, argv, "0123456789a:b:dl:n:p:")) != -1) 95 switch (ch) { 96 case '0': case '1': case '2': case '3': case '4': 97 case '5': case '6': case '7': case '8': case '9': 98 /* 99 * Undocumented kludge: split was originally designed 100 * to take a number after a dash. 101 */ 102 if (numlines == 0) { 103 p = argv[optind - 1]; 104 if (p[0] == '-' && p[1] == ch && !p[2]) 105 numlines = strtol(++p, &ep, 10); 106 else 107 numlines = 108 strtol(argv[optind] + 1, &ep, 10); 109 if (numlines <= 0 || *ep) 110 errx(EX_USAGE, 111 "%s: illegal line count", optarg); 112 } 113 break; 114 case 'a': /* Suffix length */ 115 if ((sufflen = strtol(optarg, &ep, 10)) <= 0 || *ep) 116 errx(EX_USAGE, 117 "%s: illegal suffix length", optarg); 118 break; 119 case 'b': /* Byte count. */ 120 errno = 0; 121 if ((bytecnti = strtoimax(optarg, &ep, 10)) <= 0 || 122 strchr("kKmMgG", *ep) == NULL || errno != 0) 123 errx(EX_USAGE, 124 "%s: illegal byte count", optarg); 125 if (*ep == 'k' || *ep == 'K') 126 scale = 1024; 127 else if (*ep == 'm' || *ep == 'M') 128 scale = 1024 * 1024; 129 else if (*ep == 'g' || *ep == 'G') 130 scale = 1024 * 1024 * 1024; 131 else 132 scale = 1; 133 if (bytecnti > OFF_MAX / scale) 134 errx(EX_USAGE, "%s: offset too large", optarg); 135 bytecnt = (off_t)(bytecnti * scale); 136 break; 137 case 'd': /* Decimal suffix */ 138 dflag = true; 139 break; 140 case 'l': /* Line count. */ 141 if (numlines != 0) 142 usage(); 143 if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep) 144 errx(EX_USAGE, 145 "%s: illegal line count", optarg); 146 break; 147 case 'n': /* Chunks. */ 148 if (!isdigit((unsigned char)optarg[0]) || 149 (chunks = (size_t)strtoul(optarg, &ep, 10)) == 0 || 150 *ep != '\0') { 151 errx(EX_USAGE, "%s: illegal number of chunks", 152 optarg); 153 } 154 break; 155 156 case 'p': /* pattern matching. */ 157 if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0) 158 errx(EX_USAGE, "%s: illegal regexp", optarg); 159 pflag = 1; 160 break; 161 default: 162 usage(); 163 } 164 argv += optind; 165 argc -= optind; 166 167 if (*argv != NULL) { /* Input file. */ 168 if (strcmp(*argv, "-") == 0) 169 ifd = STDIN_FILENO; 170 else if ((ifd = open(*argv, O_RDONLY, 0)) < 0) 171 err(EX_NOINPUT, "%s", *argv); 172 ++argv; 173 } 174 if (*argv != NULL) /* File name prefix. */ 175 if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname)) 176 errx(EX_USAGE, "file name prefix is too long"); 177 if (*argv != NULL) 178 usage(); 179 180 if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname)) 181 errx(EX_USAGE, "suffix is too long"); 182 if (pflag && (numlines != 0 || bytecnt != 0 || chunks != 0)) 183 usage(); 184 185 if (numlines == 0) 186 numlines = DEFLINE; 187 else if (bytecnt != 0 || chunks != 0) 188 usage(); 189 190 if (bytecnt && chunks) 191 usage(); 192 193 if (ifd == -1) /* Stdin by default. */ 194 ifd = 0; 195 196 if (bytecnt) { 197 split1(); 198 exit (0); 199 } else if (chunks) { 200 split3(); 201 exit (0); 202 } 203 split2(); 204 if (pflag) 205 regfree(&rgx); 206 exit(0); 207 } 208 209 /* 210 * split1 -- 211 * Split the input by bytes. 212 */ 213 static void 214 split1(void) 215 { 216 off_t bcnt; 217 char *C; 218 ssize_t dist, len; 219 int nfiles; 220 221 nfiles = 0; 222 223 for (bcnt = 0;;) 224 switch ((len = read(ifd, bfr, MAXBSIZE))) { 225 case 0: 226 exit(0); 227 case -1: 228 err(EX_IOERR, "read"); 229 /* NOTREACHED */ 230 default: 231 if (!file_open) { 232 if (!chunks || (nfiles < chunks)) { 233 newfile(); 234 nfiles++; 235 } 236 } 237 if (bcnt + len >= bytecnt) { 238 dist = bytecnt - bcnt; 239 if (write(ofd, bfr, dist) != dist) 240 err(EX_IOERR, "write"); 241 len -= dist; 242 for (C = bfr + dist; len >= bytecnt; 243 len -= bytecnt, C += bytecnt) { 244 if (!chunks || (nfiles < chunks)) { 245 newfile(); 246 nfiles++; 247 } 248 if (write(ofd, 249 C, bytecnt) != bytecnt) 250 err(EX_IOERR, "write"); 251 } 252 if (len != 0) { 253 if (!chunks || (nfiles < chunks)) { 254 newfile(); 255 nfiles++; 256 } 257 if (write(ofd, C, len) != len) 258 err(EX_IOERR, "write"); 259 } else 260 file_open = 0; 261 bcnt = len; 262 } else { 263 bcnt += len; 264 if (write(ofd, bfr, len) != len) 265 err(EX_IOERR, "write"); 266 } 267 } 268 } 269 270 /* 271 * split2 -- 272 * Split the input by lines. 273 */ 274 static void 275 split2(void) 276 { 277 long lcnt = 0; 278 FILE *infp; 279 280 /* Stick a stream on top of input file descriptor */ 281 if ((infp = fdopen(ifd, "r")) == NULL) 282 err(EX_NOINPUT, "fdopen"); 283 284 /* Process input one line at a time */ 285 while (fgets(bfr, sizeof(bfr), infp) != NULL) { 286 const int len = strlen(bfr); 287 288 /* If line is too long to deal with, just write it out */ 289 if (bfr[len - 1] != '\n') 290 goto writeit; 291 292 /* Check if we need to start a new file */ 293 if (pflag) { 294 regmatch_t pmatch; 295 296 pmatch.rm_so = 0; 297 pmatch.rm_eo = len - 1; 298 if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0) 299 newfile(); 300 } else if (lcnt++ == numlines) { 301 newfile(); 302 lcnt = 1; 303 } 304 305 writeit: 306 /* Open output file if needed */ 307 if (!file_open) 308 newfile(); 309 310 /* Write out line */ 311 if (write(ofd, bfr, len) != len) 312 err(EX_IOERR, "write"); 313 } 314 315 /* EOF or error? */ 316 if (ferror(infp)) 317 err(EX_IOERR, "read"); 318 else 319 exit(0); 320 } 321 322 /* 323 * split3 -- 324 * Split the input into specified number of chunks 325 */ 326 static void 327 split3(void) 328 { 329 struct stat sb; 330 331 if (fstat(ifd, &sb) == -1) { 332 err(1, "stat"); 333 /* NOTREACHED */ 334 } 335 336 if (chunks > sb.st_size) { 337 errx(1, "can't split into more than %d files", 338 (int)sb.st_size); 339 /* NOTREACHED */ 340 } 341 342 bytecnt = sb.st_size / chunks; 343 split1(); 344 } 345 346 347 /* 348 * newfile -- 349 * Open a new output file. 350 */ 351 static void 352 newfile(void) 353 { 354 long i, maxfiles, tfnum; 355 static long fnum; 356 static char *fpnt; 357 char beg, end; 358 int pattlen; 359 360 if (ofd == -1) { 361 if (fname[0] == '\0') { 362 fname[0] = 'x'; 363 fpnt = fname + 1; 364 } else { 365 fpnt = fname + strlen(fname); 366 } 367 ofd = fileno(stdout); 368 } 369 370 if (dflag) { 371 beg = '0'; 372 end = '9'; 373 } 374 else { 375 beg = 'a'; 376 end = 'z'; 377 } 378 pattlen = end - beg + 1; 379 380 /* maxfiles = pattlen^sufflen, but don't use libm. */ 381 for (maxfiles = 1, i = 0; i < sufflen; i++) 382 if (LONG_MAX / pattlen < maxfiles) 383 errx(EX_USAGE, "suffix is too long (max %ld)", i); 384 else 385 maxfiles *= pattlen; 386 387 if (fnum == maxfiles) 388 errx(EX_DATAERR, "too many files"); 389 390 /* Generate suffix of sufflen letters */ 391 tfnum = fnum; 392 i = sufflen - 1; 393 do { 394 fpnt[i] = tfnum % pattlen + beg; 395 tfnum /= pattlen; 396 } while (i-- > 0); 397 fpnt[sufflen] = '\0'; 398 399 ++fnum; 400 if (!freopen(fname, "w", stdout)) 401 err(EX_IOERR, "%s", fname); 402 file_open = 1; 403 } 404 405 static void 406 usage(void) 407 { 408 (void)fprintf(stderr, 409 "usage: split [-l line_count] [-a suffix_length] [file [prefix]]\n" 410 " split -b byte_count[K|k|M|m|G|g] [-a suffix_length] [file [prefix]]\n" 411 " split -n chunk_count [-a suffix_length] [file [prefix]]\n" 412 " split -p pattern [-a suffix_length] [file [prefix]]\n"); 413 exit(EX_USAGE); 414 } 415