split.c revision 1.11
1 /* $NetBSD: split.c,v 1.11 2003/06/10 16:57:05 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 1.11 2003/06/10 16:57:05 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 '-': /* 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 (strcmp(*argv, "-") == 0) 134 ifd = STDIN_FILENO; 135 else if ((ifd = open(*argv, O_RDONLY, 0)) < 0) 136 err(1, "%s", *argv); 137 ++argv; 138 } 139 140 if (*argv != NULL) { 141 if (strlen(*argv) + sfxlen > NAME_MAX) 142 errx(EXIT_FAILURE, "Output file name too long"); 143 (void)strcpy(fname, *argv++); /* File name prefix. */ 144 } else { 145 if (1 + sfxlen > NAME_MAX) 146 errx(EXIT_FAILURE, "Output file name too long"); 147 } 148 149 if (*argv != NULL) 150 usage(); 151 152 if (numlines == 0) 153 numlines = DEFLINE; 154 else if (bytecnt) 155 usage(); 156 157 if (ifd == -1) /* Stdin by default. */ 158 ifd = 0; 159 160 if (bytecnt) { 161 split1(bytecnt); 162 } else { 163 split2(numlines); 164 } 165 return 0; 166 } 167 168 /* 169 * split1 -- 170 * Split the input by bytes. 171 */ 172 static void 173 split1(unsigned long long bytecnt) 174 { 175 unsigned long long bcnt, dist; 176 ssize_t len; 177 char *C; 178 char bfr[MAXBSIZE]; 179 180 for (bcnt = 0;;) 181 switch (len = read(ifd, bfr, MAXBSIZE)) { 182 case 0: 183 exit(0); 184 case -1: 185 err(1, "read"); 186 /* NOTREACHED */ 187 default: 188 if (!file_open) { 189 newfile(); 190 file_open = 1; 191 } 192 if (bcnt + len >= bytecnt) { 193 dist = bytecnt - bcnt; 194 if (bigwrite(ofd, bfr, dist) != dist) 195 err(1, "write"); 196 len -= dist; 197 for (C = bfr + dist; len >= bytecnt; 198 len -= bytecnt, C += bytecnt) { 199 newfile(); 200 if (bigwrite(ofd, 201 C, (int)bytecnt) != bytecnt) 202 err(1, "write"); 203 } 204 if (len) { 205 newfile(); 206 if (bigwrite(ofd, C, len) != len) 207 err(1, "write"); 208 } else 209 file_open = 0; 210 bcnt = len; 211 } else { 212 bcnt += len; 213 if (bigwrite(ofd, bfr, len) != len) 214 err(1, "write"); 215 } 216 } 217 } 218 219 /* 220 * split2 -- 221 * Split the input by lines. 222 */ 223 static void 224 split2(unsigned long long numlines) 225 { 226 unsigned long long lcnt, bcnt; 227 ssize_t len; 228 char *Ce, *Cs; 229 char bfr[MAXBSIZE]; 230 231 for (lcnt = 0;;) 232 switch (len = read(ifd, bfr, MAXBSIZE)) { 233 case 0: 234 exit(0); 235 case -1: 236 err(1, "read"); 237 /* NOTREACHED */ 238 default: 239 if (!file_open) { 240 newfile(); 241 file_open = 1; 242 } 243 for (Cs = Ce = bfr; len--; Ce++) 244 if (*Ce == '\n' && ++lcnt == numlines) { 245 bcnt = Ce - Cs + 1; 246 if (bigwrite(ofd, Cs, bcnt) != bcnt) 247 err(1, "write"); 248 lcnt = 0; 249 Cs = Ce + 1; 250 if (len) 251 newfile(); 252 else 253 file_open = 0; 254 } 255 if (Cs < Ce) { 256 bcnt = Ce - Cs; 257 if (bigwrite(ofd, Cs, bcnt) != bcnt) 258 err(1, "write"); 259 } 260 } 261 } 262 263 /* 264 * newfile -- 265 * Open a new output file. 266 */ 267 static void 268 newfile(void) 269 { 270 static int fnum; 271 static int defname; 272 static char *fpnt; 273 int quot, i; 274 275 if (ofd == -1) { 276 if (fname[0] == '\0') { 277 fname[0] = 'x'; 278 fpnt = fname + 1; 279 defname = 1; 280 } else { 281 fpnt = fname + strlen(fname); 282 defname = 0; 283 } 284 ofd = fileno(stdout); 285 } 286 /* 287 * Hack to increase max files; original code wandered through 288 * magic characters. Maximum files is 3 * 26 * 26 == 2028 289 */ 290 fpnt[sfxlen] = '\0'; 291 quot = fnum; 292 for (i = sfxlen - 1; i >= 0; i--) { 293 fpnt[i] = quot % 26 + 'a'; 294 quot = quot / 26; 295 } 296 if (quot > 0) { 297 if (!defname || fname[0] == 'z') 298 errx(1, "too many files."); 299 ++fname[0]; 300 fnum = 0; 301 } 302 ++fnum; 303 if (!freopen(fname, "w", stdout)) 304 err(1, "%s", fname); 305 } 306 307 static unsigned long long 308 bigwrite(int fd, const void *buf, unsigned long long len) 309 { 310 const char *ptr = buf; 311 unsigned long long sofar = 0; 312 313 while (len != 0) { 314 ssize_t w, nw = (len > INT_MAX) ? INT_MAX : (ssize_t)len; 315 if ((w = write(fd, ptr, nw)) == -1) 316 return sofar; 317 len -= w; 318 ptr += w; 319 sofar += w; 320 } 321 return sofar; 322 } 323 324 325 static void 326 usage(void) 327 { 328 (void)fprintf(stderr, 329 "Usage: %s [-b byte_count] [-l line_count] [-a suffix_length] " 330 "[file [prefix]]\n", getprogname()); 331 exit(1); 332 } 333
Indexes created Tue Sep 23 21:09:50 GMT 2025