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 	p = request;
43 	pt = temp;
44 
45 	while (*p) {
46 		if (*p == '%') {
47 			if (p[1] && p[2]) {
48 				*pt++ = hex_to_int(p[1]) << 4 | hex_to_int(p[2]);
49 				p += 2;
50 			}
51 		} else if (*p == '+') {
52 			*pt++ = ' ';
53 		} else {
54 			*pt++ = *p;
55 		}
56 		p++;
57 	}
58 	*pt = '\0';
59 
60 	return temp;
61 }
62 
63 int
64 get_path_from_request(char *request, char *finalpath)
65 {
66 	char hostname[MAXREQLEN];
67 	char localpath[MAXREQLEN];
68 	char temp[MAXREQLEN];
69 	char *p, *decoded_request;
70 	int pos = 0, ret;
71 
72 	memset(hostname, 0, sizeof(hostname));
73 	memset(localpath, 0, sizeof(localpath));
74 	memset(temp, 0, sizeof(temp));
75 
76 	p = request;
77 
78 	if ((p = strchr(request, '\r')) == NULL) {
79 		log_info("\\r missing from request, abort processing");
80 		return -1;
81 	}
82 
83 	*p = '\0';	/* Strip \r\n */
84 	p = request;
85 
86 	if (strncmp(p, "gemini://", 9) != 0) {
87 		log_info("Gemini scheme missing, abort processing");
88 		return -1;
89 	}
90 	memmove(request, p + 9, strlen(request) + 1 - 9);
91 
92 	decoded_request = uridecode(request);
93 
94 	/* save hostname */
95 	if ((p = strchr(decoded_request, '/')) != NULL)
96 		snprintf(hostname, strlen(decoded_request) - strlen(p)+1, "%s",
97 			decoded_request);
98 	else
99 		snprintf(hostname, strlen(decoded_request)+1, "%s", decoded_request);
100 
101 	/* Strip possible port (e.g. :1965) from hostname */
102 	if ((p = strrchr(hostname, ':')) != NULL) {
103 		pos = strlen(hostname) - strlen(p);
104 		if (pos < 0 || pos > _POSIX_HOST_NAME_MAX)
105 			fatalx("pos while shorten hostname out of range");
106 		hostname[pos] = '\0';
107 	}
108 
109 	/* Remove ../ for security reasons */
110 	while ((p = strstr(decoded_request, "/..")) != NULL) {
111 		memmove(decoded_request, p + 3, strlen(p) + 1 - 3);
112 	}
113 
114 	if ((p = strchr(decoded_request, '/')) != NULL) {
115 		/* Save all after the first / in localpath */
116 		snprintf(localpath, strlen(decoded_request), "%s", p+1);
117 		if (strlen(localpath) == 0) {
118 			/*
119 			 * If the request is 'example.com/', localpart will be empty. In this case
120 			 * write the default to it.
121 			 */
122 			sprintf(localpath, "index.gmi");
123 		}
124 	} else {
125 		/* There is no slash in the request, so assume index.gmi */
126 		sprintf(localpath, "index.gmi");
127 	}
128 
129 	/*
130 	 *We do not need to take the base dir aka /var/db/gemini into account
131 	 * since we already chroot() to _PATH_TWIND_CHROOT .
132 	 *
133 	 * Here, a string truncation could happen.  This can be implemented
134 	 * better! XXX FIXME
135 	 */
136 	snprintf(finalpath, MAXREQLEN, "%s/%s", hostname, localpath);
137 
138 	/* Check if the wanted path exists and if it's a directory */
139 	ret = check_gemini_file(finalpath);
140 	if (ret < 0) {
141 		log_debug("%s not found", finalpath);
142 		free(decoded_request);
143 		return -2;
144 	} else if (ret == 1) {
145 		log_debug("%s is a directory", finalpath);
146 		/* Auto append index.gmi if destination is a directory */
147 		snprintf(temp, MAXREQLEN, "%s", finalpath);
148 		snprintf(finalpath, MAXREQLEN, "%s/index.gmi", temp);
149 	}
150 
151 	log_debug("Got request for %s on server %s -> %s",
152 		localpath, hostname, finalpath);
153 
154 	/* decoded_request is no longer used, so it can be freed */
155 	free(decoded_request);
156 
157 	return 0;
158 }
159