Annotation of src/usr.bin/wc/wc.c, Revision 1.19
1.19 ! deraadt 1: /* $OpenBSD: wc.c,v 1.18 2015/10/03 14:39:25 deraadt Exp $ */
1.2 deraadt 2:
1.1 deraadt 3: /*
1.3 millert 4: * Copyright (c) 1980, 1987, 1991, 1993
5: * The Regents of the University of California. All rights reserved.
1.1 deraadt 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.
1.9 millert 15: * 3. Neither the name of the University nor the names of its contributors
1.1 deraadt 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:
1.17 deraadt 32: #include <sys/param.h> /* MAXBSIZE */
33: #include <sys/stat.h>
34: #include <sys/file.h>
1.1 deraadt 35: #include <stdio.h>
36: #include <stdlib.h>
37: #include <string.h>
38: #include <locale.h>
39: #include <ctype.h>
1.3 millert 40: #include <err.h>
1.1 deraadt 41: #include <unistd.h>
1.11 espie 42: #include <util.h>
1.1 deraadt 43:
1.3 millert 44: int64_t tlinect, twordct, tcharct;
1.11 espie 45: int doline, doword, dochar, humanchar;
1.3 millert 46: int rval;
47: extern char *__progname;
48:
1.7 millert 49: void print_counts(int64_t, int64_t, int64_t, char *);
1.14 deraadt 50: void format_and_print(long long);
1.7 millert 51: void cnt(char *);
1.1 deraadt 52:
53: int
1.8 deraadt 54: main(int argc, char *argv[])
1.1 deraadt 55: {
1.6 mpech 56: int ch;
1.1 deraadt 57:
58: setlocale(LC_ALL, "");
1.18 deraadt 59:
1.19 ! deraadt 60: if (pledge("stdio rpath", NULL) == -1)
! 61: err(1, "pledge");
1.1 deraadt 62:
1.11 espie 63: while ((ch = getopt(argc, argv, "lwchm")) != -1)
1.16 okan 64: switch(ch) {
1.1 deraadt 65: case 'l':
66: doline = 1;
67: break;
68: case 'w':
69: doword = 1;
70: break;
71: case 'c':
72: case 'm':
73: dochar = 1;
74: break;
1.11 espie 75: case 'h':
76: humanchar = 1;
77: break;
1.1 deraadt 78: case '?':
79: default:
1.3 millert 80: (void)fprintf(stderr,
1.11 espie 81: "usage: %s [-c | -m] [-hlw] [file ...]\n",
1.3 millert 82: __progname);
1.1 deraadt 83: exit(1);
84: }
85: argv += optind;
86: argc -= optind;
87:
88: /*
89: * wc is unusual in that its flags are on by default, so,
90: * if you don't get any arguments, you have to turn them
91: * all on.
92: */
1.3 millert 93: if (!doline && !doword && !dochar)
1.1 deraadt 94: doline = doword = dochar = 1;
95:
96: if (!*argv) {
97: cnt((char *)NULL);
98: } else {
99: int dototal = (argc > 1);
100:
101: do {
102: cnt(*argv);
103: } while(*++argv);
104:
1.3 millert 105: if (dototal)
1.10 deraadt 106: print_counts(tlinect, twordct, tcharct, "total");
1.1 deraadt 107: }
108:
109: exit(rval);
110: }
111:
1.3 millert 112: void
1.8 deraadt 113: cnt(char *file)
1.1 deraadt 114: {
1.6 mpech 115: u_char *C;
116: short gotsp;
117: int len;
118: int64_t linect, wordct, charct;
1.1 deraadt 119: struct stat sbuf;
120: int fd;
121: u_char buf[MAXBSIZE];
122:
123: linect = wordct = charct = 0;
124: if (file) {
125: if ((fd = open(file, O_RDONLY, 0)) < 0) {
1.3 millert 126: warn("%s", file);
1.1 deraadt 127: rval = 1;
128: return;
129: }
130: } else {
131: fd = STDIN_FILENO;
132: }
1.10 deraadt 133:
1.1 deraadt 134: if (!doword) {
135: /*
1.3 millert 136: * Line counting is split out because it's a lot
1.1 deraadt 137: * faster to get lines than to get words, since
138: * the word count requires some logic.
139: */
140: if (doline) {
1.3 millert 141: while ((len = read(fd, buf, MAXBSIZE)) > 0) {
1.1 deraadt 142: charct += len;
143: for (C = buf; len--; ++C)
144: if (*C == '\n')
145: ++linect;
146: }
147: if (len == -1) {
1.3 millert 148: warn("%s", file);
1.1 deraadt 149: rval = 1;
150: }
151: }
152: /*
1.3 millert 153: * If all we need is the number of characters and
1.1 deraadt 154: * it's a directory or a regular or linked file, just
155: * stat the puppy. We avoid testing for it not being
156: * a special device in case someone adds a new type
157: * of inode.
158: */
159: else if (dochar) {
1.3 millert 160: mode_t ifmt;
1.1 deraadt 161:
162: if (fstat(fd, &sbuf)) {
1.3 millert 163: warn("%s", file);
1.1 deraadt 164: rval = 1;
165: } else {
166: ifmt = sbuf.st_mode & S_IFMT;
167: if (ifmt == S_IFREG || ifmt == S_IFLNK
1.3 millert 168: || ifmt == S_IFDIR) {
1.1 deraadt 169: charct = sbuf.st_size;
170: } else {
1.3 millert 171: while ((len = read(fd, buf, MAXBSIZE)) > 0)
1.1 deraadt 172: charct += len;
173: if (len == -1) {
1.3 millert 174: warn("%s", file);
1.1 deraadt 175: rval = 1;
176: }
177: }
178: }
179: }
1.3 millert 180: } else {
181: /* Do it the hard way... */
1.1 deraadt 182: gotsp = 1;
183: while ((len = read(fd, buf, MAXBSIZE)) > 0) {
1.3 millert 184: /*
185: * This loses in the presence of multi-byte characters.
186: * To do it right would require a function to return a
187: * character while knowing how many bytes it consumed.
188: */
1.1 deraadt 189: charct += len;
190: for (C = buf; len--; ++C) {
1.15 deraadt 191: if (isspace(*C)) {
1.1 deraadt 192: gotsp = 1;
1.3 millert 193: if (*C == '\n')
1.1 deraadt 194: ++linect;
195: } else {
196: /*
197: * This line implements the POSIX
198: * spec, i.e. a word is a "maximal
199: * string of characters delimited by
200: * whitespace." Notice nothing was
201: * said about a character being
202: * printing or non-printing.
203: */
204: if (gotsp) {
205: gotsp = 0;
206: ++wordct;
207: }
208: }
209: }
210: }
211: if (len == -1) {
1.3 millert 212: warn("%s", file);
1.1 deraadt 213: rval = 1;
214: }
215: }
216:
1.12 otto 217: print_counts(linect, wordct, charct, file);
1.1 deraadt 218:
1.3 millert 219: /*
220: * Don't bother checking doline, doword, or dochar -- speeds
1.10 deraadt 221: * up the common case
1.3 millert 222: */
1.1 deraadt 223: tlinect += linect;
224: twordct += wordct;
225: tcharct += charct;
226:
1.3 millert 227: if (close(fd) != 0) {
228: warn("%s", file);
1.1 deraadt 229: rval = 1;
230: }
231: }
232:
1.11 espie 233: void
234: format_and_print(long long v)
235: {
236: if (humanchar) {
237: char result[FMT_SCALED_STRSIZE];
238:
239: (void)fmt_scaled(v, result);
240: (void)printf("%7s", result);
241: } else {
242: (void)printf(" %7lld", v);
243: }
244: }
245:
1.1 deraadt 246: void
1.8 deraadt 247: print_counts(int64_t lines, int64_t words, int64_t chars, char *name)
1.1 deraadt 248: {
249: if (doline)
1.11 espie 250: format_and_print((long long)lines);
1.1 deraadt 251: if (doword)
1.11 espie 252: format_and_print((long long)words);
1.1 deraadt 253: if (dochar)
1.11 espie 254: format_and_print((long long)chars);
1.1 deraadt 255:
1.12 otto 256: if (name)
257: (void)printf(" %s\n", name);
258: else
259: (void)printf("\n");
1.1 deraadt 260: }