split.c revision 1.20
1 /* $NetBSD: split.c,v 1.20 2003/08/07 11:15:56 agc Exp $ */ 2 3 /* 4 * Copyright (c) 1987, 1993, 1994 5 * The Regents of the University of California. All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 3. Neither the name of the University nor the names of its contributors 16 * may be used to endorse or promote products derived from this software 17 * without specific prior written permission. 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 22 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 29 * SUCH DAMAGE. 30 */ 31 32 #include <sys/cdefs.h> 33 #ifndef lint 34 __COPYRIGHT("@(#) Copyright (c) 1987, 1993, 1994\n\ 35 The Regents of the University of California. All rights reserved.\n"); 36 #endif /* not lint */ 37 38 #ifndef lint 39 #if 0 40 static char sccsid[] = "@(#)split.c 8.3 (Berkeley) 4/25/94"; 41 #endif 42 __RCSID("$NetBSD: split.c,v 1.20 2003/08/07 11:15:56 agc Exp $"); 43 #endif /* not lint */ 44 45 #include <sys/param.h> 46 47 #include <ctype.h> 48 #include <err.h> 49 #include <errno.h> 50 #include <fcntl.h> 51 #include <stdio.h> 52 #include <stdlib.h> 53 #include <string.h> 54 #include <unistd.h> 55 56 #define DEFLINE 1000 /* Default num lines per file. */ 57 58 static int file_open; /* If a file open. */ 59 static int ifd = STDIN_FILENO, ofd = -1; /* Input/output file descriptors. */ 60 static char *fname; /* File name prefix. */ 61 static size_t sfxlen = 2; /* suffix length. */ 62 63 int main(int, char **); 64 static void newfile(void); 65 static void split1(off_t); 66 static void split2(off_t); 67 static void usage(void) __attribute__((__noreturn__)); 68 static size_t bigwrite(int, void const *, size_t); 69 70 int 71 main(int argc, char *argv[]) 72 { 73 int ch; 74 char *ep, *p; 75 char const *base; 76 off_t bytecnt = 0; /* Byte count to split on. */ 77 off_t numlines = 0; /* Line count to split on. */ 78 79 while ((ch = getopt(argc, argv, "0123456789b:l:a:")) != -1) 80 switch (ch) { 81 case '0': case '1': case '2': case '3': case '4': 82 case '5': case '6': case '7': case '8': case '9': 83 /* 84 * Undocumented kludge: split was originally designed 85 * to take a number after a dash. 86 */ 87 if (numlines == 0) { 88 p = argv[optind - 1]; 89 if (p[0] == '-' && p[1] == ch && !p[2]) 90 p++; 91 else 92 p = argv[optind] + 1; 93 numlines = strtoull(p, &ep, 10); 94 if (numlines == 0 || *ep != '\0') 95 errx(1, "%s: illegal line count.", p); 96 } 97 break; 98 case 'b': /* Byte count. */ 99 if (!isdigit((unsigned char)optarg[0]) || 100 (bytecnt = strtoull(optarg, &ep, 10)) == 0 || 101 (*ep != '\0' && *ep != 'k' && *ep != 'm')) 102 errx(1, "%s: illegal byte count.", optarg); 103 if (*ep == 'k') 104 bytecnt *= 1024; 105 else if (*ep == 'm') 106 bytecnt *= 1024 * 1024; 107 break; 108 case 'l': /* Line count. */ 109 if (numlines != 0) 110 usage(); 111 if (!isdigit((unsigned char)optarg[0]) || 112 (numlines = strtoull(optarg, &ep, 10)) == 0 || 113 *ep != '\0') 114 errx(1, "%s: illegal line count.", optarg); 115 break; 116 case 'a': /* Suffix length. */ 117 if (!isdigit((unsigned char)optarg[0]) || 118 (sfxlen = (size_t)strtoul(optarg, &ep, 10)) == 0 || 119 *ep != '\0') 120 errx(1, "%s: illegal suffix length.", optarg); 121 break; 122 default: 123 usage(); 124 } 125 argv += optind; 126 argc -= optind; 127 128 if (*argv != NULL) { 129 if (strcmp(*argv, "-") != 0 && 130 (ifd = open(*argv, O_RDONLY, 0)) < 0) 131 err(1, "%s", *argv); 132 ++argv; 133 } 134 135 136 base = (*argv != NULL) ? *argv++ : "x"; 137 if ((fname = malloc(strlen(base) + sfxlen + 1)) == NULL) 138 err(EXIT_FAILURE, NULL); 139 (void)strcpy(fname, base); /* File name prefix. */ 140 141 if (*argv != NULL) 142 usage(); 143 144 if (numlines == 0) 145 numlines = DEFLINE; 146 else if (bytecnt) 147 usage(); 148 149 if (bytecnt) 150 split1(bytecnt); 151 else 152 split2(numlines); 153 154 return 0; 155 } 156 157 /* 158 * split1 -- 159 * Split the input by bytes. 160 */ 161 static void 162 split1(off_t bytecnt) 163 { 164 off_t bcnt; 165 ssize_t dist, len; 166 char *C; 167 char bfr[MAXBSIZE]; 168 169 for (bcnt = 0;;) 170 switch (len = read(ifd, bfr, MAXBSIZE)) { 171 case 0: 172 exit(0); 173 /* NOTREACHED */ 174 case -1: 175 err(1, "read"); 176 /* NOTREACHED */ 177 default: 178 if (!file_open) { 179 newfile(); 180 file_open = 1; 181 } 182 if (bcnt + len >= bytecnt) { 183 /* LINTED: bytecnt - bcnt <= len */ 184 dist = bytecnt - bcnt; 185 if (bigwrite(ofd, bfr, dist) != dist) 186 err(1, "write"); 187 len -= dist; 188 for (C = bfr + dist; len >= bytecnt; 189 /* LINTED: bytecnt <= len */ 190 len -= bytecnt, C += bytecnt) { 191 newfile(); 192 /* LINTED: as above */ 193 if (bigwrite(ofd, 194 C, bytecnt) != bytecnt) 195 err(1, "write"); 196 } 197 if (len) { 198 newfile(); 199 /* LINTED: len >= 0 */ 200 if (bigwrite(ofd, C, len) != len) 201 err(1, "write"); 202 } else 203 file_open = 0; 204 bcnt = len; 205 } else { 206 bcnt += len; 207 /* LINTED: len >= 0 */ 208 if (bigwrite(ofd, bfr, len) != len) 209 err(1, "write"); 210 } 211 } 212 } 213 214 /* 215 * split2 -- 216 * Split the input by lines. 217 */ 218 static void 219 split2(off_t numlines) 220 { 221 off_t lcnt; 222 size_t bcnt; 223 ssize_t len; 224 char *Ce, *Cs; 225 char bfr[MAXBSIZE]; 226 227 for (lcnt = 0;;) 228 switch (len = read(ifd, bfr, MAXBSIZE)) { 229 case 0: 230 exit(0); 231 /* NOTREACHED */ 232 case -1: 233 err(1, "read"); 234 /* NOTREACHED */ 235 default: 236 if (!file_open) { 237 newfile(); 238 file_open = 1; 239 } 240 for (Cs = Ce = bfr; len--; Ce++) 241 if (*Ce == '\n' && ++lcnt == numlines) { 242 bcnt = Ce - Cs + 1; 243 if (bigwrite(ofd, Cs, bcnt) != bcnt) 244 err(1, "write"); 245 lcnt = 0; 246 Cs = Ce + 1; 247 if (len) 248 newfile(); 249 else 250 file_open = 0; 251 } 252 if (Cs < Ce) { 253 bcnt = Ce - Cs; 254 if (bigwrite(ofd, Cs, bcnt) != bcnt) 255 err(1, "write"); 256 } 257 } 258 } 259 260 /* 261 * newfile -- 262 * Open a new output file. 263 */ 264 static void 265 newfile(void) 266 { 267 static int fnum; 268 static char *fpnt; 269 int quot, i; 270 271 if (ofd == -1) { 272 fpnt = fname + strlen(fname); 273 fpnt[sfxlen] = '\0'; 274 } else if (close(ofd) != 0) 275 err(1, "%s", fname); 276 277 quot = fnum; 278 for (i = sfxlen - 1; i >= 0; i--) { 279 fpnt[i] = quot % 26 + 'a'; 280 quot = quot / 26; 281 } 282 if (quot > 0) 283 errx(1, "too many files."); 284 ++fnum; 285 if ((ofd = open(fname, O_WRONLY | O_CREAT | O_TRUNC, DEFFILEMODE)) < 0) 286 err(1, "%s", fname); 287 } 288 289 static size_t 290 bigwrite(int fd, const void *buf, size_t len) 291 { 292 const char *ptr = buf; 293 size_t sofar = 0; 294 ssize_t w; 295 296 while (len != 0) { 297 if ((w = write(fd, ptr, len)) == -1) 298 return sofar; 299 len -= w; 300 ptr += w; 301 sofar += w; 302 } 303 return sofar; 304 } 305 306 307 static void 308 usage(void) 309 { 310 (void)fprintf(stderr, 311 "Usage: %s [-b byte_count] [-l line_count] [-a suffix_length] " 312 "[file [prefix]]\n", getprogname()); 313 exit(1); 314 } 315
Indexes created Wed Oct 01 12:09:54 GMT 2025