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. All advertising materials mentioning features or use of this software 14 * must display the following acknowledgement: 15 * This product includes software developed by the University of 16 * California, Berkeley and its contributors. 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) 1987, 1993, 1994 The Regents of the University of California. All rights reserved. 34 * @(#)split.c 8.2 (Berkeley) 4/16/94 35 * $FreeBSD: src/usr.bin/split/split.c,v 1.6.2.2 2002/07/25 12:46:36 tjr Exp $ 36 * $DragonFly: src/usr.bin/split/split.c,v 1.4 2003/10/25 14:14:52 hmp Exp $ 37 */ 38 39 #include <sys/param.h> 40 41 #include <ctype.h> 42 #include <err.h> 43 #include <errno.h> 44 #include <fcntl.h> 45 #include <limits.h> 46 #include <stdio.h> 47 #include <stdlib.h> 48 #include <string.h> 49 #include <unistd.h> 50 #include <regex.h> 51 #include <sysexits.h> 52 53 #define DEFLINE 1000 /* Default num lines per file. */ 54 55 off_t bytecnt; /* Byte count to split on. */ 56 long numlines; /* Line count to split on. */ 57 int file_open; /* If a file open. */ 58 int ifd = -1, ofd = -1; /* Input/output file descriptors. */ 59 char *bfr; /* I/O buffer. */ 60 char fname[MAXPATHLEN]; /* File name prefix. */ 61 regex_t rgx; 62 int pflag; 63 long sufflen = 2; /* File name suffix length. */ 64 65 void newfile(void); 66 void split1(void); 67 void split2(void); 68 static void usage(void); 69 70 int 71 main(int argc, char **argv) 72 { 73 long long bytecnti; 74 long scale; 75 int ch; 76 char *ep, *p; 77 78 while ((ch = getopt(argc, argv, "-0123456789a:b:l:p:")) != -1) 79 switch (ch) { 80 case '0': case '1': case '2': case '3': case '4': 81 case '5': case '6': case '7': case '8': case '9': 82 /* 83 * Undocumented kludge: split was originally designed 84 * to take a number after a dash. 85 */ 86 if (numlines == 0) { 87 p = argv[optind - 1]; 88 if (p[0] == '-' && p[1] == ch && !p[2]) 89 numlines = strtol(++p, &ep, 10); 90 else 91 numlines = 92 strtol(argv[optind] + 1, &ep, 10); 93 if (numlines <= 0 || *ep) 94 errx(EX_USAGE, 95 "%s: illegal line count", optarg); 96 } 97 break; 98 case '-': /* Undocumented: historic stdin flag. */ 99 if (ifd != -1) 100 usage(); 101 ifd = 0; 102 break; 103 case 'a': /* Suffix length */ 104 if ((sufflen = strtol(optarg, &ep, 10)) <= 0 || *ep) 105 errx(EX_USAGE, 106 "%s: illegal suffix length", optarg); 107 break; 108 case 'b': /* Byte count. */ 109 errno = 0; 110 if ((bytecnti = strtoll(optarg, &ep, 10)) <= 0 || 111 (*ep != '\0' && *ep != 'k' && *ep != 'm') || 112 errno != 0) 113 errx(EX_USAGE, 114 "%s: illegal byte count", optarg); 115 if (*ep == 'k') 116 scale = 1024; 117 else if (*ep == 'm') 118 scale = 1024 * 1024; 119 else 120 scale = 1; 121 bytecnt = (off_t)(bytecnti * scale); 122 break; 123 case 'p' : /* pattern matching. */ 124 if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0) 125 errx(EX_USAGE, "%s: illegal regexp", optarg); 126 pflag = 1; 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 default: 136 usage(); 137 } 138 argv += optind; 139 argc -= optind; 140 141 if (*argv != NULL) 142 if (ifd == -1) { /* Input file. */ 143 if (strcmp(*argv, "-") == 0) 144 ifd = STDIN_FILENO; 145 else if ((ifd = open(*argv, O_RDONLY, 0)) < 0) 146 err(EX_NOINPUT, "%s", *argv); 147 ++argv; 148 } 149 if (*argv != NULL) /* File name prefix. */ 150 if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname)) 151 errx(EX_USAGE, "file name prefix is too long"); 152 if (*argv != NULL) 153 usage(); 154 155 if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname)) 156 errx(EX_USAGE, "suffix is too long"); 157 if (pflag && (numlines != 0 || bytecnt != 0)) 158 usage(); 159 160 if (numlines == 0) 161 numlines = DEFLINE; 162 else if (bytecnt != 0) 163 usage(); 164 165 if (ifd == -1) /* Stdin by default. */ 166 ifd = 0; 167 168 if (bytecnt) { 169 split1(); 170 exit (0); 171 } 172 split2(); 173 if (pflag) 174 regfree(&rgx); 175 exit(0); 176 } 177 178 /* 179 * split1 -- 180 * Split the input by bytes. 181 */ 182 void 183 split1(void) 184 { 185 off_t bcnt; 186 char *C; 187 ssize_t dist, len; 188 189 if((bfr = (char *)malloc(bytecnt)) == NULL) 190 err(EX_OSERR, "malloc"); 191 192 for (bcnt = 0;;) 193 switch ((len = read(ifd, bfr, bytecnt))) { 194 case 0: 195 free(bfr); 196 exit(0); 197 case -1: 198 free(bfr); 199 err(EX_IOERR, "read"); 200 /* NOTREACHED */ 201 default: 202 if (!file_open) 203 newfile(); 204 if (bcnt + len >= bytecnt) { 205 dist = bytecnt - bcnt; 206 if (write(ofd, bfr, dist) != dist) 207 err(EX_IOERR, "write"); 208 len -= dist; 209 for (C = bfr + dist; len >= bytecnt; 210 len -= bytecnt, C += bytecnt) { 211 newfile(); 212 if (write(ofd, 213 C, (int)bytecnt) != bytecnt) { 214 free(bfr); 215 err(EX_IOERR, "write"); 216 } 217 } 218 if (len != 0) { 219 newfile(); 220 if (write(ofd, C, len) != len) { 221 free(bfr); 222 err(EX_IOERR, "write"); 223 } 224 } else 225 file_open = 0; 226 bcnt = len; 227 } else { 228 bcnt += len; 229 if (write(ofd, bfr, len) != len) { 230 free(bfr); 231 err(EX_IOERR, "write"); 232 } 233 } 234 } 235 free(bfr); 236 } 237 238 /* 239 * split2 -- 240 * Split the input by lines. 241 */ 242 void 243 split2(void) 244 { 245 int startofline = 1; 246 long lcnt = 0; 247 FILE *infp; 248 249 /* Stick a stream on top of input file descriptor */ 250 if ((infp = fdopen(ifd, "r")) == NULL) 251 err(EX_NOINPUT, "fdopen"); 252 253 if((bfr = (char *)malloc(MAXBSIZE)) == NULL) 254 err(EX_OSERR, "malloc"); 255 256 /* Process input one line at a time */ 257 while (fgets(bfr, MAXBSIZE, infp) != NULL) { 258 const int len = strlen(bfr); 259 260 /* Consider starting a new file only when at beginning of a line */ 261 if (startofline) { 262 /* Check if we need to start a new file */ 263 if (pflag) { 264 regmatch_t pmatch; 265 266 pmatch.rm_so = 0; 267 pmatch.rm_eo = len - 1; 268 if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0) 269 newfile(); 270 } else if (lcnt++ == numlines) { 271 newfile(); 272 lcnt = 1; 273 } 274 } 275 276 if (bfr[len - 1] != '\n') 277 startofline = 0; 278 else 279 startofline = 1; 280 281 /* Open output file if needed */ 282 if (!file_open) 283 newfile(); 284 285 /* Write out line */ 286 if (write(ofd, bfr, len) != len) { 287 free(bfr); 288 err(EX_IOERR, "write"); 289 } 290 } 291 292 free(bfr); 293 294 /* EOF or error? */ 295 if (ferror(infp)) 296 err(EX_IOERR, "read"); 297 else 298 exit(0); 299 } 300 301 /* 302 * newfile -- 303 * Open a new output file. 304 */ 305 void 306 newfile(void) 307 { 308 long i, maxfiles, tfnum; 309 static long fnum; 310 static int defname; 311 static char *fpnt; 312 313 if (ofd == -1) { 314 if (fname[0] == '\0') { 315 fname[0] = 'x'; 316 fpnt = fname + 1; 317 defname = 1; 318 } else { 319 fpnt = fname + strlen(fname); 320 defname = 0; 321 } 322 ofd = fileno(stdout); 323 } 324 325 /* maxfiles = 26^sufflen, but don't use libm. */ 326 for (maxfiles = 1, i = 0; i < sufflen; i++) 327 if ((maxfiles *= 26) <= 0) 328 errx(EX_USAGE, "suffix is too long (max %ld)", i); 329 330 /* 331 * Hack to increase max files; original code wandered through 332 * magic characters. 333 */ 334 if (fnum == maxfiles) { 335 if (!defname || fname[0] == 'z') 336 errx(EX_DATAERR, "too many files"); 337 ++fname[0]; 338 fnum = 0; 339 } 340 341 /* Generate suffix of sufflen letters */ 342 tfnum = fnum; 343 i = sufflen - 1; 344 do { 345 fpnt[i] = tfnum % 26 + 'a'; 346 tfnum /= 26; 347 } while (i-- > 0); 348 fpnt[sufflen] = '\0'; 349 350 ++fnum; 351 if (!freopen(fname, "w", stdout)) 352 err(EX_IOERR, "%s", fname); 353 file_open = 1; 354 } 355 356 static void 357 usage(void) 358 { 359 (void)fprintf(stderr, 360 "usage: split [-a sufflen] [-b byte_count] [-l line_count] [-p pattern]\n"); 361 (void)fprintf(stderr, 362 " [file [prefix]]\n"); 363 exit(EX_USAGE); 364 } 365