1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
|
#include <sys/types.h>
#include <ctype.h>
#include <err.h>
#include <errno.h>
#include <libgen.h>
#include <limits.h>
#include <stdarg.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <time.h>
#include "util.h"
static void
encodehex(unsigned char c, char *s)
{
static const char *table = "0123456789ABCDEF";
s[0] = table[((c - (c % 16)) / 16) % 16];
s[1] = table[c % 16];
}
int
parseuri(const char *s, struct uri *u, int rel)
{
const char *p = s;
size_t i;
memset(u, 0, sizeof(struct uri));
if (!*s)
return 0;
/* prefix is "//", don't read protocol, skip to domain parsing */
if (!strncmp(p, "//", 2)) {
p += 2; /* skip "//" */
} else {
/* protocol part */
for (p = s; *p && (isalpha((int)*p) || isdigit((int)*p) ||
*p == '+' || *p == '-' || *p == '.'); p++)
;
if (!strncmp(p, "://", 3)) {
if (p - s + 1 >= (ssize_t)sizeof(u->proto))
return -1; /* protocol too long */
memcpy(u->proto, s, p - s);
p += 3; /* skip "://" */
} else {
p = s; /* no protocol format, set to start */
/* relative url: read rest as path, else as domain */
if (rel)
goto readpath;
}
}
/* domain / host part, skip until "/" or end. */
i = strcspn(p, "/");
if (i + 1 >= sizeof(u->host))
return -1; /* host too long */
memcpy(u->host, p, i);
p = &p[i];
readpath:
if (u->host[0]) {
p = &p[strspn(p, "/")];
strlcpy(u->path, "/", sizeof(u->path));
} else {
/* having no host is an error in this case */
if (!rel)
return -1;
}
/* treat truncation as an error */
return strlcat(u->path, p, sizeof(u->path)) >= sizeof(u->path) ? -1 : 0;
}
/* Get absolute uri; if `link` is relative use `base` to make it absolute.
* the returned string in `buf` is uri encoded, see: encodeuri(). */
int
absuri(const char *link, const char *base, char *buf, size_t bufsiz)
{
struct uri ulink, ubase;
char tmp[4096] = "", *p;
int r = -1, c;
size_t i;
buf[0] = '\0';
if (parseuri(base, &ubase, 0) == -1 ||
parseuri(link, &ulink, 1) == -1)
return -1;
if (!ulink.host[0] && !ubase.host[0])
return -1;
r = snprintf(tmp, sizeof(tmp), "%s://%s",
ulink.proto[0] ?
ulink.proto :
(ubase.proto[0] ? ubase.proto : "http"),
!strncmp(link, "//", 2) ?
ulink.host :
(ulink.host[0] ? ulink.host : ubase.host));
if (r == -1 || (size_t)r >= sizeof(tmp))
return -1;
/* relative to root */
if (!ulink.host[0] && ulink.path[0] != '/') {
/* relative to base url path */
if (ulink.path[0]) {
if ((p = strrchr(ubase.path, '/'))) {
/* temporary null-terminate */
c = *(++p);
*p = '\0';
i = strlcat(tmp, ubase.path, sizeof(tmp));
*p = c; /* restore */
if (i >= sizeof(tmp))
return -1;
}
} else {
if (strlcat(tmp, ubase.path, sizeof(tmp)) >= sizeof(tmp))
return -1;
}
}
if (strlcat(tmp, ulink.path, sizeof(tmp)) >= sizeof(tmp))
return -1;
return encodeuri(tmp, buf, bufsiz);
}
int
encodeuri(const char *s, char *buf, size_t bufsiz)
{
size_t i, b;
if (!bufsiz)
return -1;
for (i = 0, b = 0; s[i]; i++) {
if ((int)s[i] == ' ' ||
(unsigned char)s[i] > 127 ||
iscntrl((int)s[i])) {
if (b + 3 >= bufsiz)
return -1;
buf[b++] = '%';
encodehex(s[i], &buf[b]);
b += 2;
} else {
if (b >= bufsiz)
return -1;
buf[b++] = s[i];
}
}
if (b >= bufsiz)
return -1;
buf[b] = '\0';
return 0;
}
/* Read a field-separated line from 'fp',
* separated by a character 'separator',
* 'fields' is a list of pointers with a size of FieldLast (must be >0).
* 'line' buffer is allocated using malloc, 'size' will contain the allocated
* buffer size.
* returns: amount of fields read (>0) or -1 on error. */
ssize_t
parseline(char **line, size_t *size, char *fields[FieldLast], FILE *fp)
{
char *prev, *s;
size_t i;
if (getline(line, size, fp) <= 0)
return -1;
for (prev = *line, i = 0;
(s = strchr(prev, '\t')) && i < FieldLast - 1;
i++) {
*s = '\0';
fields[i] = prev;
prev = s + 1;
}
fields[i++] = prev;
/* make non-parsed fields empty. */
for (; i < FieldLast; i++)
fields[i] = "";
return (ssize_t)i;
}
/* Parse time to time_t, assumes time_t is signed. */
int
strtotime(const char *s, time_t *t)
{
long l;
char *e;
errno = 0;
l = strtol(s, &e, 10);
if (*s == '\0' || *e != '\0')
return -1;
if (t)
*t = (time_t)l;
return 0;
}
/* Escape characters below as HTML 2.0 / XML 1.0. */
void
xmlencode(const char *s, FILE *fp)
{
for (; *s; s++) {
switch(*s) {
case '<': fputs("<", fp); break;
case '>': fputs(">", fp); break;
case '\'': fputs("'", fp); break;
case '&': fputs("&", fp); break;
case '"': fputs(""", fp); break;
default: fputc(*s, fp);
}
}
}
/* Some implementations of basename(3) return a pointer to a static
* internal buffer (OpenBSD). Others modify the contents of `path` (POSIX).
* This is a wrapper function that is compatible with both versions.
* The program will error out if basename(3) failed, this can only happen
* with the OpenBSD version. */
char *
xbasename(const char *path)
{
char *p, *b;
if (!(p = strdup(path)))
err(1, "strdup");
if (!(b = basename(p)))
err(1, "basename");
if (!(b = strdup(b)))
err(1, "strdup");
free(p);
return b;
}
|