1 /* $OpenBSD: split.c,v 1.4 1999/02/04 03:53:48 millert Exp $ */ 2 /* $NetBSD: split.c,v 1.5 1995/08/31 22:22:05 jtc Exp $ */ 3 4 /* 5 * Copyright (c) 1987, 1993, 1994 6 * The Regents of the University of California. All rights reserved. 7 * 8 * Redistribution and use in source and binary forms, with or without 9 * modification, are permitted provided that the following conditions 10 * are met: 11 * 1. Redistributions of source code must retain the above copyright 12 * notice, this list of conditions and the following disclaimer. 13 * 2. Redistributions in binary form must reproduce the above copyright 14 * notice, this list of conditions and the following disclaimer in the 15 * documentation and/or other materials provided with the distribution. 16 * 3. All advertising materials mentioning features or use of this software 17 * must display the following acknowledgement: 18 * This product includes software developed by the University of 19 * California, Berkeley and its contributors. 20 * 4. Neither the name of the University nor the names of its contributors 21 * may be used to endorse or promote products derived from this software 22 * without specific prior written permission. 23 * 24 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 25 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 27 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 28 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 29 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 30 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 31 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 32 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 33 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 34 * SUCH DAMAGE. 35 */ 36 37 #ifndef lint 38 static char copyright[] = 39 "@(#) Copyright (c) 1987, 1993, 1994\n\ 40 The Regents of the University of California. All rights reserved.\n"; 41 #endif /* not lint */ 42 43 #ifndef lint 44 #if 0 45 static char sccsid[] = "@(#)split.c 8.3 (Berkeley) 4/25/94"; 46 #else 47 static char rcsid[] = "$OpenBSD: split.c,v 1.4 1999/02/04 03:53:48 millert Exp $"; 48 #endif 49 #endif /* not lint */ 50 51 #include <sys/param.h> 52 #include <sys/types.h> 53 54 #include <ctype.h> 55 #include <err.h> 56 #include <fcntl.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 long 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 75 void newfile __P((void)); 76 void split1 __P((void)); 77 void split2 __P((void)); 78 void usage __P((void)); 79 80 int 81 main(argc, argv) 82 int argc; 83 char *argv[]; 84 { 85 int ch; 86 char *ep, *p; 87 88 while ((ch = getopt(argc, argv, "-0123456789b: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 'b': /* Byte count. */ 114 if ((bytecnt = strtol(optarg, &ep, 10)) <= 0 || 115 (*ep != '\0' && *ep != 'k' && *ep != 'm')) 116 errx(EX_USAGE, 117 "%s: illegal byte count", optarg); 118 if (*ep == 'k') 119 bytecnt *= 1024; 120 else if (*ep == 'm') 121 bytecnt *= 1048576; 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 ((ifd = open(*argv, O_RDONLY, 0)) < 0) 144 err(EX_NOINPUT, "%s", *argv); 145 ++argv; 146 } 147 if (*argv != NULL) /* File name prefix. */ 148 (void)strcpy(fname, *argv++); 149 if (*argv != NULL) 150 usage(); 151 152 if (pflag && (numlines != 0 || bytecnt != 0)) 153 usage(); 154 155 if (numlines == 0) 156 numlines = DEFLINE; 157 else if (bytecnt != 0) 158 usage(); 159 160 if (ifd == -1) /* Stdin by default. */ 161 ifd = 0; 162 163 if (bytecnt) { 164 split1(); 165 exit (0); 166 } 167 split2(); 168 if (pflag) 169 regfree(&rgx); 170 exit(0); 171 } 172 173 /* 174 * split1 -- 175 * Split the input by bytes. 176 */ 177 void 178 split1() 179 { 180 long bcnt; 181 int dist, len; 182 char *C; 183 184 for (bcnt = 0;;) 185 switch ((len = read(ifd, bfr, MAXBSIZE))) { 186 case 0: 187 exit(0); 188 case -1: 189 err(EX_IOERR, "read"); 190 /* NOTREACHED */ 191 default: 192 if (!file_open) 193 newfile(); 194 if (bcnt + len >= bytecnt) { 195 dist = bytecnt - bcnt; 196 if (write(ofd, bfr, dist) != dist) 197 err(EX_IOERR, "write"); 198 len -= dist; 199 for (C = bfr + dist; len >= bytecnt; 200 len -= bytecnt, C += bytecnt) { 201 newfile(); 202 if (write(ofd, 203 C, (int)bytecnt) != bytecnt) 204 err(EX_IOERR, "write"); 205 } 206 if (len != 0) { 207 newfile(); 208 if (write(ofd, C, len) != len) 209 err(EX_IOERR, "write"); 210 } else 211 file_open = 0; 212 bcnt = len; 213 } else { 214 bcnt += len; 215 if (write(ofd, bfr, len) != len) 216 err(EX_IOERR, "write"); 217 } 218 } 219 } 220 221 /* 222 * split2 -- 223 * Split the input by lines. 224 */ 225 void 226 split2() 227 { 228 long lcnt = 0; 229 FILE *infp; 230 231 /* Stick a stream on top of input file descriptor */ 232 if ((infp = fdopen(ifd, "r")) == NULL) 233 err(EX_NOINPUT, "fdopen"); 234 235 /* Process input one line at a time */ 236 while (fgets(bfr, sizeof(bfr), infp) != NULL) { 237 const int len = strlen(bfr); 238 239 /* If line is too long to deal with, just write it out */ 240 if (bfr[len - 1] != '\n') 241 goto writeit; 242 243 /* Check if we need to start a new file */ 244 if (pflag) { 245 regmatch_t pmatch; 246 247 pmatch.rm_so = 0; 248 pmatch.rm_eo = len - 1; 249 if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0) 250 newfile(); 251 } else if (lcnt++ == numlines) { 252 newfile(); 253 lcnt = 1; 254 } 255 256 writeit: 257 /* Open output file if needed */ 258 if (!file_open) 259 newfile(); 260 261 /* Write out line */ 262 if (write(ofd, bfr, len) != len) 263 err(EX_IOERR, "write"); 264 } 265 266 /* EOF or error? */ 267 if (ferror(infp)) 268 err(EX_IOERR, "read"); 269 else 270 exit(0); 271 } 272 273 /* 274 * newfile -- 275 * Open a new output file. 276 */ 277 void 278 newfile() 279 { 280 static long fnum; 281 static int defname; 282 static char *fpnt; 283 284 if (ofd == -1) { 285 if (fname[0] == '\0') { 286 fname[0] = 'x'; 287 fpnt = fname + 1; 288 defname = 1; 289 } else { 290 fpnt = fname + strlen(fname); 291 defname = 0; 292 } 293 ofd = fileno(stdout); 294 } 295 /* 296 * Hack to increase max files; original code wandered through 297 * magic characters. Maximum files is 3 * 26 * 26 == 2028 298 */ 299 #define MAXFILES 676 300 if (fnum == MAXFILES) { 301 if (!defname || fname[0] == 'z') 302 errx(EX_DATAERR, "too many files"); 303 ++fname[0]; 304 fnum = 0; 305 } 306 fpnt[0] = fnum / 26 + 'a'; 307 fpnt[1] = fnum % 26 + 'a'; 308 ++fnum; 309 if (!freopen(fname, "w", stdout)) 310 err(EX_IOERR, "%s", fname); 311 file_open = 1; 312 } 313 314 void 315 usage() 316 { 317 extern char *__progname; 318 319 (void)fprintf(stderr, 320 "usage: %s [-b byte_count] [-l line_count] [-p pattern] [file [prefix]]\n", 321 __progname); 322 exit(EX_USAGE); 323 } 324