split.c revision 1.10
1 /* $NetBSD: split.c,v 1.10 2003/06/10 16:32:54 bjh21 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. All advertising materials mentioning features or use of this software 16 * must display the following acknowledgement: 17 * This product includes software developed by the University of 18 * California, Berkeley and its contributors. 19 * 4. Neither the name of the University nor the names of its contributors 20 * may be used to endorse or promote products derived from this software 21 * without specific prior written permission. 22 * 23 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 24 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 25 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 26 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 27 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 28 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 29 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 30 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 31 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 32 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 33 * SUCH DAMAGE. 34 */ 35 36 #include <sys/cdefs.h> 37 #ifndef lint 38 __COPYRIGHT("@(#) Copyright (c) 1987, 1993, 1994\n\ 39 The Regents of the University of California. All rights reserved.\n"); 40 #endif /* not lint */ 41 42 #ifndef lint 43 #if 0 44 static char sccsid[] = "@(#)split.c 8.3 (Berkeley) 4/25/94"; 45 #endif 46 __RCSID("$NetBSD: split.c,v 1.10 2003/06/10 16:32:54 bjh21 Exp $"); 47 #endif /* not lint */ 48 49 #include <sys/param.h> 50 51 #include <ctype.h> 52 #include <err.h> 53 #include <fcntl.h> 54 #include <stdio.h> 55 #include <stdlib.h> 56 #include <string.h> 57 #include <unistd.h> 58 59 #define DEFLINE 1000 /* Default num lines per file. */ 60 61 static int file_open; /* If a file open. */ 62 static int ifd = -1, ofd = -1; /* Input/output file descriptors. */ 63 static char fname[MAXPATHLEN]; /* File name prefix. */ 64 static int sfxlen = 2; /* suffix length. */ 65 66 int main(int, char **); 67 static void newfile(void); 68 static void split1(unsigned long long); 69 static void split2(unsigned long long); 70 static void usage(void) __attribute__((__noreturn__)); 71 static unsigned long long bigwrite __P((int, const void *, unsigned long long)); 72 73 int 74 main(int argc, char *argv[]) 75 { 76 int ch; 77 char *ep, *p; 78 unsigned long long bytecnt = 0; /* Byte count to split on. */ 79 unsigned long long numlines = 0;/* Line count to split on. */ 80 81 while ((ch = getopt(argc, argv, "-0123456789b:l:a:")) != -1) 82 switch (ch) { 83 case '0': case '1': case '2': case '3': case '4': 84 case '5': case '6': case '7': case '8': case '9': 85 /* 86 * Undocumented kludge: split was originally designed 87 * to take a number after a dash. 88 */ 89 if (numlines == 0) { 90 p = argv[optind - 1]; 91 if (p[0] == '-' && p[1] == ch && !p[2]) 92 numlines = strtol(++p, &ep, 10); 93 else 94 numlines = 95 strtol(argv[optind] + 1, &ep, 10); 96 if (numlines <= 0 || *ep) 97 errx(1, 98 "%s: illegal line count.", optarg); 99 } 100 break; 101 case '-': /* Undocumented: historic stdin flag. */ 102 if (ifd != -1) 103 usage(); 104 ifd = 0; 105 break; 106 case 'b': /* Byte count. */ 107 if ((bytecnt = strtoull(optarg, &ep, 10)) <= 0 || 108 (*ep != '\0' && *ep != 'k' && *ep != 'm')) 109 errx(1, "%s: illegal byte count.", optarg); 110 if (*ep == 'k') 111 bytecnt *= 1024; 112 else if (*ep == 'm') 113 bytecnt *= 1024 * 1024; 114 break; 115 case 'l': /* Line count. */ 116 if (numlines != 0) 117 usage(); 118 if ((numlines = strtoull(optarg, &ep, 10)) <= 0 || *ep) 119 errx(1, "%s: illegal line count.", optarg); 120 break; 121 case 'a': /* Suffix length. */ 122 if ((sfxlen = strtol(optarg, &ep, 10)) <= 0 || *ep) 123 errx(1, "%s: illegal suffix length.", optarg); 124 break; 125 default: 126 usage(); 127 } 128 argv += optind; 129 argc -= optind; 130 131 if (*argv != NULL) 132 if (ifd == -1) { /* Input file. */ 133 if ((ifd = open(*argv, O_RDONLY, 0)) < 0) 134 err(1, "%s", *argv); 135 ++argv; 136 } 137 138 if (*argv != NULL) { 139 if (strlen(*argv) + sfxlen > NAME_MAX) 140 errx(EXIT_FAILURE, "Output file name too long"); 141 (void)strcpy(fname, *argv++); /* File name prefix. */ 142 } else { 143 if (1 + sfxlen > NAME_MAX) 144 errx(EXIT_FAILURE, "Output file name too long"); 145 } 146 147 if (*argv != NULL) 148 usage(); 149 150 if (numlines == 0) 151 numlines = DEFLINE; 152 else if (bytecnt) 153 usage(); 154 155 if (ifd == -1) /* Stdin by default. */ 156 ifd = 0; 157 158 if (bytecnt) { 159 split1(bytecnt); 160 } else { 161 split2(numlines); 162 } 163 return 0; 164 } 165 166 /* 167 * split1 -- 168 * Split the input by bytes. 169 */ 170 static void 171 split1(unsigned long long bytecnt) 172 { 173 unsigned long long bcnt, dist; 174 ssize_t len; 175 char *C; 176 char bfr[MAXBSIZE]; 177 178 for (bcnt = 0;;) 179 switch (len = read(ifd, bfr, MAXBSIZE)) { 180 case 0: 181 exit(0); 182 case -1: 183 err(1, "read"); 184 /* NOTREACHED */ 185 default: 186 if (!file_open) { 187 newfile(); 188 file_open = 1; 189 } 190 if (bcnt + len >= bytecnt) { 191 dist = bytecnt - bcnt; 192 if (bigwrite(ofd, bfr, dist) != dist) 193 err(1, "write"); 194 len -= dist; 195 for (C = bfr + dist; len >= bytecnt; 196 len -= bytecnt, C += bytecnt) { 197 newfile(); 198 if (bigwrite(ofd, 199 C, (int)bytecnt) != bytecnt) 200 err(1, "write"); 201 } 202 if (len) { 203 newfile(); 204 if (bigwrite(ofd, C, len) != len) 205 err(1, "write"); 206 } else 207 file_open = 0; 208 bcnt = len; 209 } else { 210 bcnt += len; 211 if (bigwrite(ofd, bfr, len) != len) 212 err(1, "write"); 213 } 214 } 215 } 216 217 /* 218 * split2 -- 219 * Split the input by lines. 220 */ 221 static void 222 split2(unsigned long long numlines) 223 { 224 unsigned long long lcnt, bcnt; 225 ssize_t len; 226 char *Ce, *Cs; 227 char bfr[MAXBSIZE]; 228 229 for (lcnt = 0;;) 230 switch (len = read(ifd, bfr, MAXBSIZE)) { 231 case 0: 232 exit(0); 233 case -1: 234 err(1, "read"); 235 /* NOTREACHED */ 236 default: 237 if (!file_open) { 238 newfile(); 239 file_open = 1; 240 } 241 for (Cs = Ce = bfr; len--; Ce++) 242 if (*Ce == '\n' && ++lcnt == numlines) { 243 bcnt = Ce - Cs + 1; 244 if (bigwrite(ofd, Cs, bcnt) != bcnt) 245 err(1, "write"); 246 lcnt = 0; 247 Cs = Ce + 1; 248 if (len) 249 newfile(); 250 else 251 file_open = 0; 252 } 253 if (Cs < Ce) { 254 bcnt = Ce - Cs; 255 if (bigwrite(ofd, Cs, bcnt) != bcnt) 256 err(1, "write"); 257 } 258 } 259 } 260 261 /* 262 * newfile -- 263 * Open a new output file. 264 */ 265 static void 266 newfile(void) 267 { 268 static int fnum; 269 static int defname; 270 static char *fpnt; 271 int quot, i; 272 273 if (ofd == -1) { 274 if (fname[0] == '\0') { 275 fname[0] = 'x'; 276 fpnt = fname + 1; 277 defname = 1; 278 } else { 279 fpnt = fname + strlen(fname); 280 defname = 0; 281 } 282 ofd = fileno(stdout); 283 } 284 /* 285 * Hack to increase max files; original code wandered through 286 * magic characters. Maximum files is 3 * 26 * 26 == 2028 287 */ 288 fpnt[sfxlen] = '\0'; 289 quot = fnum; 290 for (i = sfxlen - 1; i >= 0; i--) { 291 fpnt[i] = quot % 26 + 'a'; 292 quot = quot / 26; 293 } 294 if (quot > 0) { 295 if (!defname || fname[0] == 'z') 296 errx(1, "too many files."); 297 ++fname[0]; 298 fnum = 0; 299 } 300 ++fnum; 301 if (!freopen(fname, "w", stdout)) 302 err(1, "%s", fname); 303 } 304 305 static unsigned long long 306 bigwrite(int fd, const void *buf, unsigned long long len) 307 { 308 const char *ptr = buf; 309 unsigned long long sofar = 0; 310 311 while (len != 0) { 312 ssize_t w, nw = (len > INT_MAX) ? INT_MAX : (ssize_t)len; 313 if ((w = write(fd, ptr, nw)) == -1) 314 return sofar; 315 len -= w; 316 ptr += w; 317 sofar += w; 318 } 319 return sofar; 320 } 321 322 323 static void 324 usage(void) 325 { 326 (void)fprintf(stderr, 327 "Usage: %s [-b byte_count] [-l line_count] [-a suffix_length] " 328 "[file [prefix]]\n", getprogname()); 329 exit(1); 330 } 331
Indexes created Wed Sep 24 05:09:52 GMT 2025