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