1 /*
2 Inferno lib9/tokenize.c
3 http://code.google.com/p/inferno-os/source/browse/lib9/tokenize.c
4
5 Copyright © 1994-1999 Lucent Technologies Inc. All rights reserved.
6 Revisions Copyright © 2000-2007 Vita Nuova Holdings Limited (www.vitanuova.com). All rights reserved.
7
8 Permission is hereby granted, free of charge, to any person obtaining a copy
9 of this software and associated documentation files (the "Software"), to deal
10 in the Software without restriction, including without limitation the rights
11 to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
12 copies of the Software, and to permit persons to whom the Software is
13 furnished to do so, subject to the following conditions:
14
15 The above copyright notice and this permission notice shall be included in
16 all copies or substantial portions of the Software.
17
18 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
19 IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
20 FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
21 AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
22 LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
23 OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
24 THE SOFTWARE.
25 */
26
27 #include <u.h>
28 #include <libc.h>
29
30 static char qsep[] = " \t\r\n";
31
32 static char*
33 qtoken(char *s, char *sep)
34 {
35 int quoting;
36 char *t;
37
38 quoting = 0;
39 t = s; /* s is output string, t is input string */
40 while(*t!='\0' && (quoting || utfrune(sep, *t)==nil)){
41 if(*t != '\''){
42 *s++ = *t++;
43 continue;
44 }
45 /* *t is a quote */
46 if(!quoting){
47 quoting = 1;
48 t++;
49 continue;
50 }
51 /* quoting and we're on a quote */
52 if(t[1] != '\''){
53 /* end of quoted section; absorb closing quote */
54 t++;
55 quoting = 0;
56 continue;
57 }
58 /* doubled quote; fold one quote into two */
59 t++;
60 *s++ = *t++;
61 }
62 if(*s != '\0'){
63 *s = '\0';
64 if(t == s)
65 t++;
66 }
67 return t;
68 }
69
70 static char*
71 etoken(char *t, char *sep)
72 {
73 int quoting;
74
75 /* move to end of next token */
76 quoting = 0;
77 while(*t!='\0' && (quoting || utfrune(sep, *t)==nil)){
78 if(*t != '\''){
79 t++;
80 continue;
81 }
82 /* *t is a quote */
83 if(!quoting){
84 quoting = 1;
85 t++;
86 continue;
87 }
88 /* quoting and we're on a quote */
89 if(t[1] != '\''){
90 /* end of quoted section; absorb closing quote */
91 t++;
92 quoting = 0;
93 continue;
94 }
95 /* doubled quote; fold one quote into two */
96 t += 2;
97 }
98 return t;
99 }
100
101 int
102 gettokens(char *s, char **args, int maxargs, char *sep)
103 {
104 int nargs;
105
106 for(nargs=0; nargs<maxargs; nargs++){
107 while(*s!='\0' && utfrune(sep, *s)!=nil)
108 *s++ = '\0';
109 if(*s == '\0')
110 break;
111 args[nargs] = s;
112 s = etoken(s, sep);
113 }
114
115 return nargs;
116 }
117
118 int
119 tokenize(char *s, char **args, int maxargs)
120 {
121 int nargs;
122
123 for(nargs=0; nargs<maxargs; nargs++){
124 while(*s!='\0' && utfrune(qsep, *s)!=nil)
125 s++;
126 if(*s == '\0')
127 break;
128 args[nargs] = s;
129 s = qtoken(s, qsep);
130 }
131
132 return nargs;
133 }