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 * 3. 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 <libutil.h> 53 #include <limits.h> 54 #include <locale.h> 55 #include <stdbool.h> 56 #include <stdint.h> 57 #include <stdio.h> 58 #include <stdlib.h> 59 #include <string.h> 60 #include <unistd.h> 61 #include <regex.h> 62 #include <sysexits.h> 63 64 #define DEFLINE 1000 /* Default num lines per file. */ 65 66 static off_t bytecnt; /* Byte count to split on. */ 67 static off_t chunks = 0; /* Chunks count to split into. */ 68 static long numlines; /* Line count to split on. */ 69 static int file_open; /* If a file open. */ 70 static int ifd = -1, ofd = -1; /* Input/output file descriptors. */ 71 static char bfr[MAXBSIZE]; /* I/O buffer. */ 72 static char fname[MAXPATHLEN]; /* File name prefix. */ 73 static regex_t rgx; 74 static int pflag; 75 static bool dflag; 76 static long sufflen = 2; /* File name suffix length. */ 77 78 static void newfile(void); 79 static void split1(void); 80 static void split2(void); 81 static void split3(void); 82 static void usage(void); 83 84 int 85 main(int argc, char **argv) 86 { 87 int ch; 88 int error; 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 error = expand_number(optarg, &bytecnt); 122 if (error == -1) 123 errx(EX_USAGE, "%s: offset too large", optarg); 124 break; 125 case 'd': /* Decimal suffix */ 126 dflag = true; 127 break; 128 case 'l': /* Line count. */ 129 if (numlines != 0) 130 usage(); 131 if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep) 132 errx(EX_USAGE, 133 "%s: illegal line count", optarg); 134 break; 135 case 'n': /* Chunks. */ 136 if (!isdigit((unsigned char)optarg[0]) || 137 (chunks = (size_t)strtoul(optarg, &ep, 10)) == 0 || 138 *ep != '\0') { 139 errx(EX_USAGE, "%s: illegal number of chunks", 140 optarg); 141 } 142 break; 143 144 case 'p': /* pattern matching. */ 145 if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0) 146 errx(EX_USAGE, "%s: illegal regexp", optarg); 147 pflag = 1; 148 break; 149 default: 150 usage(); 151 } 152 argv += optind; 153 argc -= optind; 154 155 if (*argv != NULL) { /* Input file. */ 156 if (strcmp(*argv, "-") == 0) 157 ifd = STDIN_FILENO; 158 else if ((ifd = open(*argv, O_RDONLY, 0)) < 0) 159 err(EX_NOINPUT, "%s", *argv); 160 ++argv; 161 } 162 if (*argv != NULL) /* File name prefix. */ 163 if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname)) 164 errx(EX_USAGE, "file name prefix is too long"); 165 if (*argv != NULL) 166 usage(); 167 168 if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname)) 169 errx(EX_USAGE, "suffix is too long"); 170 if (pflag && (numlines != 0 || bytecnt != 0 || chunks != 0)) 171 usage(); 172 173 if (numlines == 0) 174 numlines = DEFLINE; 175 else if (bytecnt != 0 || chunks != 0) 176 usage(); 177 178 if (bytecnt && chunks) 179 usage(); 180 181 if (ifd == -1) /* Stdin by default. */ 182 ifd = 0; 183 184 if (bytecnt) { 185 split1(); 186 exit (0); 187 } else if (chunks) { 188 split3(); 189 exit (0); 190 } 191 split2(); 192 if (pflag) 193 regfree(&rgx); 194 exit(0); 195 } 196 197 /* 198 * split1 -- 199 * Split the input by bytes. 200 */ 201 static void 202 split1(void) 203 { 204 off_t bcnt; 205 char *C; 206 ssize_t dist, len; 207 int nfiles; 208 209 nfiles = 0; 210 211 for (bcnt = 0;;) 212 switch ((len = read(ifd, bfr, MAXBSIZE))) { 213 case 0: 214 exit(0); 215 case -1: 216 err(EX_IOERR, "read"); 217 /* NOTREACHED */ 218 default: 219 if (!file_open) { 220 if (!chunks || (nfiles < chunks)) { 221 newfile(); 222 nfiles++; 223 } 224 } 225 if (bcnt + len >= bytecnt) { 226 dist = bytecnt - bcnt; 227 if (write(ofd, bfr, dist) != dist) 228 err(EX_IOERR, "write"); 229 len -= dist; 230 for (C = bfr + dist; len >= bytecnt; 231 len -= bytecnt, C += bytecnt) { 232 if (!chunks || (nfiles < chunks)) { 233 newfile(); 234 nfiles++; 235 } 236 if (write(ofd, 237 C, bytecnt) != bytecnt) 238 err(EX_IOERR, "write"); 239 } 240 if (len != 0) { 241 if (!chunks || (nfiles < chunks)) { 242 newfile(); 243 nfiles++; 244 } 245 if (write(ofd, C, len) != len) 246 err(EX_IOERR, "write"); 247 } else 248 file_open = 0; 249 bcnt = len; 250 } else { 251 bcnt += len; 252 if (write(ofd, bfr, len) != len) 253 err(EX_IOERR, "write"); 254 } 255 } 256 } 257 258 /* 259 * split2 -- 260 * Split the input by lines. 261 */ 262 static void 263 split2(void) 264 { 265 long lcnt = 0; 266 FILE *infp; 267 268 /* Stick a stream on top of input file descriptor */ 269 if ((infp = fdopen(ifd, "r")) == NULL) 270 err(EX_NOINPUT, "fdopen"); 271 272 /* Process input one line at a time */ 273 while (fgets(bfr, sizeof(bfr), infp) != NULL) { 274 const int len = strlen(bfr); 275 276 /* If line is too long to deal with, just write it out */ 277 if (bfr[len - 1] != '\n') 278 goto writeit; 279 280 /* Check if we need to start a new file */ 281 if (pflag) { 282 regmatch_t pmatch; 283 284 pmatch.rm_so = 0; 285 pmatch.rm_eo = len - 1; 286 if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0) 287 newfile(); 288 } else if (lcnt++ == numlines) { 289 newfile(); 290 lcnt = 1; 291 } 292 293 writeit: 294 /* Open output file if needed */ 295 if (!file_open) 296 newfile(); 297 298 /* Write out line */ 299 if (write(ofd, bfr, len) != len) 300 err(EX_IOERR, "write"); 301 } 302 303 /* EOF or error? */ 304 if (ferror(infp)) 305 err(EX_IOERR, "read"); 306 else 307 exit(0); 308 } 309 310 /* 311 * split3 -- 312 * Split the input into specified number of chunks 313 */ 314 static void 315 split3(void) 316 { 317 struct stat sb; 318 319 if (fstat(ifd, &sb) == -1) { 320 err(1, "stat"); 321 /* NOTREACHED */ 322 } 323 324 if (chunks > sb.st_size) { 325 errx(1, "can't split into more than %d files", 326 (int)sb.st_size); 327 /* NOTREACHED */ 328 } 329 330 bytecnt = sb.st_size / chunks; 331 split1(); 332 } 333 334 335 /* 336 * newfile -- 337 * Open a new output file. 338 */ 339 static void 340 newfile(void) 341 { 342 long i, maxfiles, tfnum; 343 static long fnum; 344 static char *fpnt; 345 char beg, end; 346 int pattlen; 347 348 if (ofd == -1) { 349 if (fname[0] == '\0') { 350 fname[0] = 'x'; 351 fpnt = fname + 1; 352 } else { 353 fpnt = fname + strlen(fname); 354 } 355 ofd = fileno(stdout); 356 } 357 358 if (dflag) { 359 beg = '0'; 360 end = '9'; 361 } 362 else { 363 beg = 'a'; 364 end = 'z'; 365 } 366 pattlen = end - beg + 1; 367 368 /* maxfiles = pattlen^sufflen, but don't use libm. */ 369 for (maxfiles = 1, i = 0; i < sufflen; i++) 370 if (LONG_MAX / pattlen < maxfiles) 371 errx(EX_USAGE, "suffix is too long (max %ld)", i); 372 else 373 maxfiles *= pattlen; 374 375 if (fnum == maxfiles) 376 errx(EX_DATAERR, "too many files"); 377 378 /* Generate suffix of sufflen letters */ 379 tfnum = fnum; 380 i = sufflen - 1; 381 do { 382 fpnt[i] = tfnum % pattlen + beg; 383 tfnum /= pattlen; 384 } while (i-- > 0); 385 fpnt[sufflen] = '\0'; 386 387 ++fnum; 388 if (!freopen(fname, "w", stdout)) 389 err(EX_IOERR, "%s", fname); 390 file_open = 1; 391 } 392 393 static void 394 usage(void) 395 { 396 (void)fprintf(stderr, 397 "usage: split [-l line_count] [-a suffix_length] [file [prefix]]\n" 398 " split -b byte_count[K|k|M|m|G|g] [-a suffix_length] [file [prefix]]\n" 399 " split -n chunk_count [-a suffix_length] [file [prefix]]\n" 400 " split -p pattern [-a suffix_length] [file [prefix]]\n"); 401 exit(EX_USAGE); 402 } 403