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 34 #include <sys/cdefs.h> 35 __FBSDID("$FreeBSD$"); 36 37 #ifndef lint 38 static const char copyright[] = 39 "@(#) Copyright (c) 1987, 1993, 1994\n\ 40 The Regents of the University of California. All rights reserved.\n"; 41 #endif 42 43 #ifndef lint 44 static const char sccsid[] = "@(#)split.c 8.2 (Berkeley) 4/16/94"; 45 #endif 46 47 #include <sys/param.h> 48 49 #include <ctype.h> 50 #include <err.h> 51 #include <errno.h> 52 #include <fcntl.h> 53 #include <inttypes.h> 54 #include <limits.h> 55 #include <locale.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 off_t bytecnt; /* Byte count to split on. */ 67 long numlines; /* Line count to split on. */ 68 int file_open; /* If a file open. */ 69 int ifd = -1, ofd = -1; /* Input/output file descriptors. */ 70 char bfr[MAXBSIZE]; /* I/O buffer. */ 71 char fname[MAXPATHLEN]; /* File name prefix. */ 72 regex_t rgx; 73 int pflag; 74 long sufflen = 2; /* File name suffix length. */ 75 76 void newfile(void); 77 void split1(void); 78 void split2(void); 79 static void usage(void); 80 81 int 82 main(int argc, char **argv) 83 { 84 intmax_t bytecnti; 85 long scale; 86 int ch; 87 char *ep, *p; 88 89 setlocale(LC_ALL, ""); 90 91 while ((ch = getopt(argc, argv, "0123456789a:b:l:p:")) != -1) 92 switch (ch) { 93 case '0': case '1': case '2': case '3': case '4': 94 case '5': case '6': case '7': case '8': case '9': 95 /* 96 * Undocumented kludge: split was originally designed 97 * to take a number after a dash. 98 */ 99 if (numlines == 0) { 100 p = argv[optind - 1]; 101 if (p[0] == '-' && p[1] == ch && !p[2]) 102 numlines = strtol(++p, &ep, 10); 103 else 104 numlines = 105 strtol(argv[optind] + 1, &ep, 10); 106 if (numlines <= 0 || *ep) 107 errx(EX_USAGE, 108 "%s: illegal line count", optarg); 109 } 110 break; 111 case 'a': /* Suffix length */ 112 if ((sufflen = strtol(optarg, &ep, 10)) <= 0 || *ep) 113 errx(EX_USAGE, 114 "%s: illegal suffix length", optarg); 115 break; 116 case 'b': /* Byte count. */ 117 errno = 0; 118 if ((bytecnti = strtoimax(optarg, &ep, 10)) <= 0 || 119 (*ep != '\0' && *ep != 'k' && *ep != 'm') || 120 errno != 0) 121 errx(EX_USAGE, 122 "%s: illegal byte count", optarg); 123 if (*ep == 'k') 124 scale = 1024; 125 else if (*ep == 'm') 126 scale = 1024 * 1024; 127 else 128 scale = 1; 129 if (bytecnti > OFF_MAX / scale) 130 errx(EX_USAGE, "%s: offset too large", optarg); 131 bytecnt = (off_t)(bytecnti * scale); 132 break; 133 case 'l': /* Line count. */ 134 if (numlines != 0) 135 usage(); 136 if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep) 137 errx(EX_USAGE, 138 "%s: illegal line count", optarg); 139 break; 140 case 'p': /* pattern matching. */ 141 if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0) 142 errx(EX_USAGE, "%s: illegal regexp", optarg); 143 pflag = 1; 144 break; 145 default: 146 usage(); 147 } 148 argv += optind; 149 argc -= optind; 150 151 if (*argv != NULL) { /* Input file. */ 152 if (strcmp(*argv, "-") == 0) 153 ifd = STDIN_FILENO; 154 else if ((ifd = open(*argv, O_RDONLY, 0)) < 0) 155 err(EX_NOINPUT, "%s", *argv); 156 ++argv; 157 } 158 if (*argv != NULL) /* File name prefix. */ 159 if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname)) 160 errx(EX_USAGE, "file name prefix is too long"); 161 if (*argv != NULL) 162 usage(); 163 164 if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname)) 165 errx(EX_USAGE, "suffix is too long"); 166 if (pflag && (numlines != 0 || bytecnt != 0)) 167 usage(); 168 169 if (numlines == 0) 170 numlines = DEFLINE; 171 else if (bytecnt != 0) 172 usage(); 173 174 if (ifd == -1) /* Stdin by default. */ 175 ifd = 0; 176 177 if (bytecnt) { 178 split1(); 179 exit (0); 180 } 181 split2(); 182 if (pflag) 183 regfree(&rgx); 184 exit(0); 185 } 186 187 /* 188 * split1 -- 189 * Split the input by bytes. 190 */ 191 void 192 split1(void) 193 { 194 off_t bcnt; 195 char *C; 196 ssize_t dist, len; 197 198 for (bcnt = 0;;) 199 switch ((len = read(ifd, bfr, MAXBSIZE))) { 200 case 0: 201 exit(0); 202 case -1: 203 err(EX_IOERR, "read"); 204 /* NOTREACHED */ 205 default: 206 if (!file_open) 207 newfile(); 208 if (bcnt + len >= bytecnt) { 209 dist = bytecnt - bcnt; 210 if (write(ofd, bfr, dist) != dist) 211 err(EX_IOERR, "write"); 212 len -= dist; 213 for (C = bfr + dist; len >= bytecnt; 214 len -= bytecnt, C += bytecnt) { 215 newfile(); 216 if (write(ofd, 217 C, bytecnt) != bytecnt) 218 err(EX_IOERR, "write"); 219 } 220 if (len != 0) { 221 newfile(); 222 if (write(ofd, C, len) != len) 223 err(EX_IOERR, "write"); 224 } else 225 file_open = 0; 226 bcnt = len; 227 } else { 228 bcnt += len; 229 if (write(ofd, bfr, len) != len) 230 err(EX_IOERR, "write"); 231 } 232 } 233 } 234 235 /* 236 * split2 -- 237 * Split the input by lines. 238 */ 239 void 240 split2(void) 241 { 242 long lcnt = 0; 243 FILE *infp; 244 245 /* Stick a stream on top of input file descriptor */ 246 if ((infp = fdopen(ifd, "r")) == NULL) 247 err(EX_NOINPUT, "fdopen"); 248 249 /* Process input one line at a time */ 250 while (fgets(bfr, sizeof(bfr), infp) != NULL) { 251 const int len = strlen(bfr); 252 253 /* If line is too long to deal with, just write it out */ 254 if (bfr[len - 1] != '\n') 255 goto writeit; 256 257 /* Check if we need to start a new file */ 258 if (pflag) { 259 regmatch_t pmatch; 260 261 pmatch.rm_so = 0; 262 pmatch.rm_eo = len - 1; 263 if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0) 264 newfile(); 265 } else if (lcnt++ == numlines) { 266 newfile(); 267 lcnt = 1; 268 } 269 270 writeit: 271 /* Open output file if needed */ 272 if (!file_open) 273 newfile(); 274 275 /* Write out line */ 276 if (write(ofd, bfr, len) != len) 277 err(EX_IOERR, "write"); 278 } 279 280 /* EOF or error? */ 281 if (ferror(infp)) 282 err(EX_IOERR, "read"); 283 else 284 exit(0); 285 } 286 287 /* 288 * newfile -- 289 * Open a new output file. 290 */ 291 void 292 newfile(void) 293 { 294 long i, maxfiles, tfnum; 295 static long fnum; 296 static int defname; 297 static char *fpnt; 298 299 if (ofd == -1) { 300 if (fname[0] == '\0') { 301 fname[0] = 'x'; 302 fpnt = fname + 1; 303 defname = 1; 304 } else { 305 fpnt = fname + strlen(fname); 306 defname = 0; 307 } 308 ofd = fileno(stdout); 309 } 310 311 /* maxfiles = 26^sufflen, but don't use libm. */ 312 for (maxfiles = 1, i = 0; i < sufflen; i++) 313 if ((maxfiles *= 26) <= 0) 314 errx(EX_USAGE, "suffix is too long (max %ld)", i); 315 316 if (fnum == maxfiles) 317 errx(EX_DATAERR, "too many files"); 318 319 /* Generate suffix of sufflen letters */ 320 tfnum = fnum; 321 i = sufflen - 1; 322 do { 323 fpnt[i] = tfnum % 26 + 'a'; 324 tfnum /= 26; 325 } while (i-- > 0); 326 fpnt[sufflen] = '\0'; 327 328 ++fnum; 329 if (!freopen(fname, "w", stdout)) 330 err(EX_IOERR, "%s", fname); 331 file_open = 1; 332 } 333 334 static void 335 usage(void) 336 { 337 (void)fprintf(stderr, 338 "usage: split [-l line_count] [-a suffix_length] [file [prefix]]\n" 339 " split -b byte_count[k|m] [-a suffix_length] [file [prefix]]\n" 340 " split -p pattern [-a suffix_length] [file [prefix]]\n"); 341 exit(EX_USAGE); 342 } 343