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 <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 off_t bytecnt; /* Byte count to split on. */ 66 long numlines; /* Line count to split on. */ 67 int file_open; /* If a file open. */ 68 int ifd = -1, ofd = -1; /* Input/output file descriptors. */ 69 char bfr[MAXBSIZE]; /* I/O buffer. */ 70 char fname[MAXPATHLEN]; /* File name prefix. */ 71 regex_t rgx; 72 int pflag; 73 long sufflen = 2; /* File name suffix length. */ 74 75 void newfile(void); 76 void split1(void); 77 void split2(void); 78 static void usage(void); 79 80 int 81 main(int argc, char **argv) 82 { 83 intmax_t bytecnti; 84 long scale; 85 int ch; 86 char *ep, *p; 87 88 while ((ch = getopt(argc, argv, "-0123456789a:b:l:p:")) != -1) 89 switch (ch) { 90 case '0': case '1': case '2': case '3': case '4': 91 case '5': case '6': case '7': case '8': case '9': 92 /* 93 * Undocumented kludge: split was originally designed 94 * to take a number after a dash. 95 */ 96 if (numlines == 0) { 97 p = argv[optind - 1]; 98 if (p[0] == '-' && p[1] == ch && !p[2]) 99 numlines = strtol(++p, &ep, 10); 100 else 101 numlines = 102 strtol(argv[optind] + 1, &ep, 10); 103 if (numlines <= 0 || *ep) 104 errx(EX_USAGE, 105 "%s: illegal line count", optarg); 106 } 107 break; 108 case '-': /* Undocumented: historic stdin flag. */ 109 if (ifd != -1) 110 usage(); 111 ifd = 0; 112 break; 113 case 'a': /* Suffix length */ 114 if ((sufflen = strtol(optarg, &ep, 10)) <= 0 || *ep) 115 errx(EX_USAGE, 116 "%s: illegal suffix length", optarg); 117 break; 118 case 'b': /* Byte count. */ 119 errno = 0; 120 if ((bytecnti = strtoimax(optarg, &ep, 10)) <= 0 || 121 (*ep != '\0' && *ep != 'k' && *ep != 'm') || 122 errno != 0) 123 errx(EX_USAGE, 124 "%s: illegal byte count", optarg); 125 if (*ep == 'k') 126 scale = 1024; 127 else if (*ep == 'm') 128 scale = 1024 * 1024; 129 else 130 scale = 1; 131 if (bytecnti > OFF_MAX / scale) 132 errx(EX_USAGE, "%s: offset too large", optarg); 133 bytecnt = (off_t)(bytecnti * scale); 134 break; 135 case 'p' : /* pattern matching. */ 136 if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0) 137 errx(EX_USAGE, "%s: illegal regexp", optarg); 138 pflag = 1; 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 default: 148 usage(); 149 } 150 argv += optind; 151 argc -= optind; 152 153 if (*argv != NULL) 154 if (ifd == -1) { /* Input file. */ 155 if (strcmp(*argv, "-") == 0) 156 ifd = STDIN_FILENO; 157 else if ((ifd = open(*argv, O_RDONLY, 0)) < 0) 158 err(EX_NOINPUT, "%s", *argv); 159 ++argv; 160 } 161 if (*argv != NULL) /* File name prefix. */ 162 if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname)) 163 errx(EX_USAGE, "file name prefix is too long"); 164 if (*argv != NULL) 165 usage(); 166 167 if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname)) 168 errx(EX_USAGE, "suffix is too long"); 169 if (pflag && (numlines != 0 || bytecnt != 0)) 170 usage(); 171 172 if (numlines == 0) 173 numlines = DEFLINE; 174 else if (bytecnt != 0) 175 usage(); 176 177 if (ifd == -1) /* Stdin by default. */ 178 ifd = 0; 179 180 if (bytecnt) { 181 split1(); 182 exit (0); 183 } 184 split2(); 185 if (pflag) 186 regfree(&rgx); 187 exit(0); 188 } 189 190 /* 191 * split1 -- 192 * Split the input by bytes. 193 */ 194 void 195 split1(void) 196 { 197 off_t bcnt; 198 char *C; 199 ssize_t dist, len; 200 201 for (bcnt = 0;;) 202 switch ((len = read(ifd, bfr, MAXBSIZE))) { 203 case 0: 204 exit(0); 205 case -1: 206 err(EX_IOERR, "read"); 207 /* NOTREACHED */ 208 default: 209 if (!file_open) 210 newfile(); 211 if (bcnt + len >= bytecnt) { 212 dist = bytecnt - bcnt; 213 if (write(ofd, bfr, dist) != dist) 214 err(EX_IOERR, "write"); 215 len -= dist; 216 for (C = bfr + dist; len >= bytecnt; 217 len -= bytecnt, C += bytecnt) { 218 newfile(); 219 if (write(ofd, 220 C, bytecnt) != bytecnt) 221 err(EX_IOERR, "write"); 222 } 223 if (len != 0) { 224 newfile(); 225 if (write(ofd, C, len) != len) 226 err(EX_IOERR, "write"); 227 } else 228 file_open = 0; 229 bcnt = len; 230 } else { 231 bcnt += len; 232 if (write(ofd, bfr, len) != len) 233 err(EX_IOERR, "write"); 234 } 235 } 236 } 237 238 /* 239 * split2 -- 240 * Split the input by lines. 241 */ 242 void 243 split2(void) 244 { 245 long lcnt = 0; 246 FILE *infp; 247 248 /* Stick a stream on top of input file descriptor */ 249 if ((infp = fdopen(ifd, "r")) == NULL) 250 err(EX_NOINPUT, "fdopen"); 251 252 /* Process input one line at a time */ 253 while (fgets(bfr, sizeof(bfr), infp) != NULL) { 254 const int len = strlen(bfr); 255 256 /* If line is too long to deal with, just write it out */ 257 if (bfr[len - 1] != '\n') 258 goto writeit; 259 260 /* Check if we need to start a new file */ 261 if (pflag) { 262 regmatch_t pmatch; 263 264 pmatch.rm_so = 0; 265 pmatch.rm_eo = len - 1; 266 if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0) 267 newfile(); 268 } else if (lcnt++ == numlines) { 269 newfile(); 270 lcnt = 1; 271 } 272 273 writeit: 274 /* Open output file if needed */ 275 if (!file_open) 276 newfile(); 277 278 /* Write out line */ 279 if (write(ofd, bfr, len) != len) 280 err(EX_IOERR, "write"); 281 } 282 283 /* EOF or error? */ 284 if (ferror(infp)) 285 err(EX_IOERR, "read"); 286 else 287 exit(0); 288 } 289 290 /* 291 * newfile -- 292 * Open a new output file. 293 */ 294 void 295 newfile(void) 296 { 297 long i, maxfiles, tfnum; 298 static long fnum; 299 static int defname; 300 static char *fpnt; 301 302 if (ofd == -1) { 303 if (fname[0] == '\0') { 304 fname[0] = 'x'; 305 fpnt = fname + 1; 306 defname = 1; 307 } else { 308 fpnt = fname + strlen(fname); 309 defname = 0; 310 } 311 ofd = fileno(stdout); 312 } 313 314 /* maxfiles = 26^sufflen, but don't use libm. */ 315 for (maxfiles = 1, i = 0; i < sufflen; i++) 316 if ((maxfiles *= 26) <= 0) 317 errx(EX_USAGE, "suffix is too long (max %ld)", i); 318 319 /* 320 * Hack to increase max files; original code wandered through 321 * magic characters. 322 */ 323 if (fnum == maxfiles) { 324 if (!defname || fname[0] == 'z') 325 errx(EX_DATAERR, "too many files"); 326 ++fname[0]; 327 fnum = 0; 328 } 329 330 /* Generate suffix of sufflen letters */ 331 tfnum = fnum; 332 i = sufflen - 1; 333 do { 334 fpnt[i] = tfnum % 26 + 'a'; 335 tfnum /= 26; 336 } while (i-- > 0); 337 fpnt[sufflen] = '\0'; 338 339 ++fnum; 340 if (!freopen(fname, "w", stdout)) 341 err(EX_IOERR, "%s", fname); 342 file_open = 1; 343 } 344 345 static void 346 usage(void) 347 { 348 (void)fprintf(stderr, 349 "usage: split [-a sufflen] [-b byte_count] [-l line_count] [-p pattern]\n"); 350 (void)fprintf(stderr, 351 " [file [prefix]]\n"); 352 exit(EX_USAGE); 353 } 354