1 /***********************************************************************
2 * *
3 * This software is part of the ast package *
4 * Copyright (c) 1992-2010 AT&T Intellectual Property *
5 * and is licensed under the *
6 * Common Public License, Version 1.0 *
7 * by AT&T Intellectual Property *
8 * *
9 * A copy of the License is available at *
10 * http://www.opensource.org/licenses/cpl1.0.txt *
11 * (with md5 checksum 059e8cd6165cb4c31e351f2b69388fd9) *
12 * *
13 * Information and Software Systems Research *
14 * AT&T Research *
15 * Florham Park NJ *
16 * *
17 * Glenn Fowler <gsf@research.att.com> *
18 * David Korn <dgk@research.att.com> *
19 * *
20 ***********************************************************************/
21 #pragma prototyped
22 /*
23 * David Korn
24 * AT&T Bell Laboratories
25 *
26 * count the number of bytes, words, and lines in a file
27 */
28
29 static const char usage[] =
30 "[-?\n@(#)$Id: wc (AT&T Research) 2009-11-28 $\n]"
31 USAGE_LICENSE
32 "[+NAME?wc - print the number of bytes, words, and lines in files]"
33 "[+DESCRIPTION?\bwc\b reads one or more input files and, by default, "
34 "for each file writes a line containing the number of newlines, "
35 "\aword\as, and bytes contained in each file followed by the "
36 "file name to standard output in that order. A \aword\a is "
37 "defined to be a non-zero length string delimited by \bisspace\b(3) "
38 "characters.]"
39 "[+?If more than one file is specified, \bwc\b writes a total count "
40 "for all of the named files with \btotal\b written instead "
41 "of the file name.]"
42 "[+?By default, \bwc\b writes all three counts. Options can specified "
43 "so that only certain counts are written. The options \b-c\b "
44 "and \b-m\b are mutually exclusive.]"
45 "[+?If no \afile\a is given, or if the \afile\a is \b-\b, \bwc\b "
46 "reads from standard input and no filename is written to standard "
47 "output. The start of the file is defined as the current offset.]"
48 "[l:lines?List the line counts.]"
49 "[w:words?List the word counts.]"
50 "[c:bytes|chars:chars?List the byte counts.]"
51 "[m|C:multibyte-chars?List the character counts.]"
52 "[q:quiet?Suppress invalid multibyte character warnings.]"
53 "[L:longest-line|max-line-length?List the longest line length; the newline,"
54 "if any, is not counted in the length.]"
55 "[N!:utf8?For \bUTF-8\b locales \b--noutf8\b disables \bUTF-8\b "
56 "optimzations and relies on the native \bmbtowc\b(3).]"
57 "\n"
58 "\n[file ...]\n"
59 "\n"
60 "[+EXIT STATUS?]{"
61 "[+0?All files processed successfully.]"
62 "[+>0?One or more files failed to open or could not be read.]"
63 "}"
64 "[+SEE ALSO?\bcat\b(1), \bisspace\b(3)]"
65 ;
66
67
68 #include <cmd.h>
69 #include <wc.h>
70 #include <ls.h>
71
72 #define ERRORMAX 125
73
printout(register Wc_t * wp,register char * name,register int mode)74 static void printout(register Wc_t *wp, register char *name,register int mode)
75 {
76 if (mode&WC_LINES)
77 sfprintf(sfstdout," %7I*d",sizeof(wp->lines),wp->lines);
78 if (mode&WC_WORDS)
79 sfprintf(sfstdout," %7I*d",sizeof(wp->words),wp->words);
80 if (mode&WC_CHARS)
81 sfprintf(sfstdout," %7I*d",sizeof(wp->chars),wp->chars);
82 if (mode&WC_LONGEST)
83 sfprintf(sfstdout," %7I*d",sizeof(wp->chars),wp->longest);
84 if (name)
85 sfprintf(sfstdout," %s",name);
86 sfputc(sfstdout,'\n');
87 }
88
89 int
b_wc(int argc,register char ** argv,void * context)90 b_wc(int argc,register char **argv, void* context)
91 {
92 register char *cp;
93 register int mode=0, n;
94 register Wc_t *wp;
95 Sfio_t *fp;
96 Sfoff_t tlines=0, twords=0, tchars=0;
97 struct stat statb;
98
99 cmdinit(argc, argv, context, ERROR_CATALOG, 0);
100 for (;;)
101 {
102 switch (optget(argv, usage))
103 {
104 case 'c':
105 mode |= WC_CHARS;
106 continue;
107 case 'l':
108 mode |= WC_LINES;
109 continue;
110 case 'L':
111 mode |= WC_LONGEST;
112 continue;
113 case 'N':
114 if (!opt_info.num)
115 mode |= WC_NOUTF8;
116 continue;
117 case 'm':
118 case 'C':
119 mode |= WC_MBYTE;
120 continue;
121 case 'q':
122 mode |= WC_QUIET;
123 continue;
124 case 'w':
125 mode |= WC_WORDS;
126 continue;
127 case ':':
128 error(2, "%s", opt_info.arg);
129 break;
130 case '?':
131 error(ERROR_usage(2), "%s", opt_info.arg);
132 break;
133 }
134 break;
135 }
136 argv += opt_info.index;
137 if (error_info.errors)
138 error(ERROR_usage(2), "%s", optusage(NiL));
139 if (mode&WC_MBYTE)
140 {
141 if (mode&WC_CHARS)
142 error(2, "-c and -C are mutually exclusive");
143 if (!mbwide())
144 mode &= ~WC_MBYTE;
145 mode |= WC_CHARS;
146 }
147 if (!(mode&(WC_WORDS|WC_CHARS|WC_LINES|WC_MBYTE|WC_LONGEST)))
148 mode |= (WC_WORDS|WC_CHARS|WC_LINES);
149 if (!(wp = wc_init(mode)))
150 error(3,"internal error");
151 if (cp = *argv)
152 argv++;
153 n = 0;
154 do
155 {
156 if (!cp || streq(cp,"-"))
157 fp = sfstdin;
158 else if (!(fp = sfopen(NiL,cp,"r")))
159 {
160 error(ERROR_system(0),"%s: cannot open",cp);
161 continue;
162 }
163 if (cp)
164 n++;
165 if (!(mode&(WC_WORDS|WC_LINES|WC_MBYTE|WC_LONGEST)) && fstat(sffileno(fp),&statb)>=0
166 && S_ISREG(statb.st_mode))
167 {
168 wp->chars = statb.st_size - lseek(sffileno(fp),0L,1);
169 lseek(sffileno(fp),0L,2);
170 }
171 else
172 wc_count(wp, fp, cp);
173 if (fp!=sfstdin)
174 sfclose(fp);
175 tchars += wp->chars;
176 twords += wp->words;
177 tlines += wp->lines;
178 printout(wp,cp,mode);
179 } while (cp= *argv++);
180 if (n > 1)
181 {
182 wp->lines = tlines;
183 wp->chars = tchars;
184 wp->words = twords;
185 printout(wp,"total",mode);
186 }
187 return error_info.errors<ERRORMAX?error_info.errors:ERRORMAX;
188 }
189