1 /*
2  * Copyright (c) 2021 Matthias Schmidt <xhr@giessen.ccc.de>
3  *
4  * Permission to use, copy, modify, and distribute this software for any
5  * purpose with or without fee is hereby granted, provided that the above
6  * copyright notice and this permission notice appear in all copies.
7  *
8  * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
9  * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
10  * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
11  * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
12  * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
13  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
14  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
15  */
16 
17 #include <ctype.h>
18 #include <limits.h>
19 #include <string.h>
20 
21 #include "log.h"
22 #include "twind.h"
23 
24 char hex_to_int(char);
25 char* uridecode(const char *);
26 
27 /*
28  * The following two functions are from https://geekhideout.com/urlcode.shtml
29  * and provided without license restrictions
30  */
31 char hex_to_int(char ch) {
32 	return isdigit(ch) ? ch - '0' : tolower(ch) - 'a' + 10;
33 }
34 
35 char *
36 uridecode(const char *request)
37 {
38 	char *temp = xmalloc(strlen(request) + 1);
39 	const char *p;
40 	char *pt;
41 
42 	memset(temp, 0, strlen(request)+1);
43 
44 	p = request;
45 	pt = temp;
46 
47 	while (*p) {
48 		if (*p == '%') {
49 			if (p[1] && p[2]) {
50 				*pt++ = hex_to_int(p[1]) << 4 | hex_to_int(p[2]);
51 				p += 2;
52 			}
53 		} else if (*p == '+') {
54 			*pt++ = ' ';
55 		} else {
56 			*pt++ = *p;
57 		}
58 		p++;
59 	}
60 	*pt = '\0';
61 
62 	return temp;
63 }
64 
65 int
66 get_path_from_request(char *request, char *finalpath)
67 {
68 	char hostname[MAXREQLEN];
69 	char localpath[MAXREQLEN];
70 	char temp[MAXREQLEN];
71 	char *p, *decoded_request;
72 	int pos = 0, ret;
73 
74 	memset(hostname, 0, sizeof(hostname));
75 	memset(localpath, 0, sizeof(localpath));
76 	memset(temp, 0, sizeof(temp));
77 
78 	p = request;
79 
80 	if ((p = strchr(request, '\r')) == NULL) {
81 		log_info("\\r missing from request, abort processing");
82 		return -1;
83 	}
84 
85 	*p = '\0';	/* Strip \r\n */
86 	p = request;
87 
88 	if (strncmp(p, "gemini://", 9) != 0) {
89 		log_info("Gemini scheme missing, abort processing");
90 		return -1;
91 	}
92 	memmove(request, p + 9, strlen(request) + 1 - 9);
93 
94 	decoded_request = uridecode(request);
95 
96 	/* save hostname */
97 	if ((p = strchr(decoded_request, '/')) != NULL)
98 		snprintf(hostname, strlen(decoded_request) - strlen(p)+1, "%s",
99 			decoded_request);
100 	else
101 		snprintf(hostname, strlen(decoded_request)+1, "%s", decoded_request);
102 
103 	/* Strip possible port (e.g. :1965) from hostname */
104 	if ((p = strrchr(hostname, ':')) != NULL) {
105 		pos = strlen(hostname) - strlen(p);
106 		if (pos < 0 || pos > _POSIX_HOST_NAME_MAX)
107 			fatalx("pos while shorten hostname out of range");
108 		hostname[pos] = '\0';
109 	}
110 
111 	/* Remove ../ for security reasons */
112 	while ((p = strstr(decoded_request, "/..")) != NULL) {
113 		memmove(decoded_request, p + 3, strlen(p) + 1 - 3);
114 	}
115 
116 	if ((p = strchr(decoded_request, '/')) != NULL) {
117 		/* Save all after the first / in localpath */
118 		snprintf(localpath, strlen(decoded_request), "%s", p+1);
119 		if (strlen(localpath) == 0) {
120 			/*
121 			 * If the request is 'example.com/', localpart will be empty. In this case
122 			 * write the default to it.
123 			 */
124 			sprintf(localpath, "index.gmi");
125 		}
126 	} else {
127 		/* There is no slash in the request, so assume index.gmi */
128 		sprintf(localpath, "index.gmi");
129 	}
130 
131 	/*
132 	 *We do not need to take the base dir aka /var/db/gemini into account
133 	 * since we already chroot() to _PATH_TWIND_CHROOT .
134 	 *
135 	 * Here, a string truncation could happen.  This can be implemented
136 	 * better! XXX FIXME
137 	 */
138 	snprintf(finalpath, MAXREQLEN, "%s/%s", hostname, localpath);
139 
140 	/* Check if the wanted path exists and if it's a directory */
141 	ret = check_gemini_file(finalpath);
142 	if (ret < 0) {
143 		log_debug("%s not found", finalpath);
144 		free(decoded_request);
145 		return -2;
146 	} else if (ret == 1) {
147 		log_debug("%s is a directory", finalpath);
148 		/* Auto append index.gmi if destination is a directory */
149 		snprintf(temp, MAXREQLEN, "%s", finalpath);
150 		snprintf(finalpath, MAXREQLEN, "%s/index.gmi", temp);
151 	}
152 
153 	log_debug("Got request for %s on server %s -> %s",
154 		localpath, hostname, finalpath);
155 
156 	/* decoded_request is no longer used, so it can be freed */
157 	free(decoded_request);
158 
159 	return 0;
160 }
161