1 | /*************************************************************************** |
2 | * _ _ ____ _ |
3 | * Project ___| | | | _ \| | |
4 | * / __| | | | |_) | | |
5 | * | (__| |_| | _ <| |___ |
6 | * \___|\___/|_| \_\_____| |
7 | * |
8 | * Copyright (C) 1998 - 2020, Daniel Stenberg, <daniel@haxx.se>, et al. |
9 | * |
10 | * This software is licensed as described in the file COPYING, which |
11 | * you should have received as part of this distribution. The terms |
12 | * are also available at https://curl.se/docs/copyright.html. |
13 | * |
14 | * You may opt to use, copy, modify, merge, publish, distribute and/or sell |
15 | * copies of the Software, and permit persons to whom the Software is |
16 | * furnished to do so, under the terms of the COPYING file. |
17 | * |
18 | * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY |
19 | * KIND, either express or implied. |
20 | * |
21 | ***************************************************************************/ |
22 | |
23 | /* Escape and unescape URL encoding in strings. The functions return a new |
24 | * allocated string or NULL if an error occurred. */ |
25 | |
26 | #include "curl_setup.h" |
27 | |
28 | #include <curl/curl.h> |
29 | |
30 | #include "urldata.h" |
31 | #include "warnless.h" |
32 | #include "non-ascii.h" |
33 | #include "escape.h" |
34 | #include "strdup.h" |
35 | /* The last 3 #include files should be in this order */ |
36 | #include "curl_printf.h" |
37 | #include "curl_memory.h" |
38 | #include "memdebug.h" |
39 | |
40 | /* Portable character check (remember EBCDIC). Do not use isalnum() because |
41 | its behavior is altered by the current locale. |
42 | See https://tools.ietf.org/html/rfc3986#section-2.3 |
43 | */ |
44 | bool Curl_isunreserved(unsigned char in) |
45 | { |
46 | switch(in) { |
47 | case '0': case '1': case '2': case '3': case '4': |
48 | case '5': case '6': case '7': case '8': case '9': |
49 | case 'a': case 'b': case 'c': case 'd': case 'e': |
50 | case 'f': case 'g': case 'h': case 'i': case 'j': |
51 | case 'k': case 'l': case 'm': case 'n': case 'o': |
52 | case 'p': case 'q': case 'r': case 's': case 't': |
53 | case 'u': case 'v': case 'w': case 'x': case 'y': case 'z': |
54 | case 'A': case 'B': case 'C': case 'D': case 'E': |
55 | case 'F': case 'G': case 'H': case 'I': case 'J': |
56 | case 'K': case 'L': case 'M': case 'N': case 'O': |
57 | case 'P': case 'Q': case 'R': case 'S': case 'T': |
58 | case 'U': case 'V': case 'W': case 'X': case 'Y': case 'Z': |
59 | case '-': case '.': case '_': case '~': |
60 | return TRUE; |
61 | default: |
62 | break; |
63 | } |
64 | return FALSE; |
65 | } |
66 | |
67 | /* for ABI-compatibility with previous versions */ |
68 | char *curl_escape(const char *string, int inlength) |
69 | { |
70 | return curl_easy_escape(NULL, string, inlength); |
71 | } |
72 | |
73 | /* for ABI-compatibility with previous versions */ |
74 | char *curl_unescape(const char *string, int length) |
75 | { |
76 | return curl_easy_unescape(NULL, string, length, NULL); |
77 | } |
78 | |
79 | char *curl_easy_escape(struct Curl_easy *data, const char *string, |
80 | int inlength) |
81 | { |
82 | size_t length; |
83 | CURLcode result; |
84 | struct dynbuf d; |
85 | |
86 | if(inlength < 0) |
87 | return NULL; |
88 | |
89 | Curl_dyn_init(&d, CURL_MAX_INPUT_LENGTH * 3); |
90 | |
91 | length = (inlength?(size_t)inlength:strlen(string)); |
92 | if(!length) |
93 | return strdup("" ); |
94 | |
95 | while(length--) { |
96 | unsigned char in = *string; /* we need to treat the characters unsigned */ |
97 | |
98 | if(Curl_isunreserved(in)) { |
99 | /* append this */ |
100 | if(Curl_dyn_addn(&d, &in, 1)) |
101 | return NULL; |
102 | } |
103 | else { |
104 | /* encode it */ |
105 | char encoded[4]; |
106 | result = Curl_convert_to_network(data, (char *)&in, 1); |
107 | if(result) { |
108 | /* Curl_convert_to_network calls failf if unsuccessful */ |
109 | Curl_dyn_free(&d); |
110 | return NULL; |
111 | } |
112 | |
113 | msnprintf(encoded, sizeof(encoded), "%%%02X" , in); |
114 | if(Curl_dyn_add(&d, encoded)) |
115 | return NULL; |
116 | } |
117 | string++; |
118 | } |
119 | |
120 | return Curl_dyn_ptr(&d); |
121 | } |
122 | |
123 | /* |
124 | * Curl_urldecode() URL decodes the given string. |
125 | * |
126 | * Returns a pointer to a malloced string in *ostring with length given in |
127 | * *olen. If length == 0, the length is assumed to be strlen(string). |
128 | * |
129 | * 'data' can be set to NULL but then this function can't convert network |
130 | * data to host for non-ascii. |
131 | * |
132 | * ctrl options: |
133 | * - REJECT_NADA: accept everything |
134 | * - REJECT_CTRL: rejects control characters (byte codes lower than 32) in |
135 | * the data |
136 | * - REJECT_ZERO: rejects decoded zero bytes |
137 | * |
138 | * The values for the enum starts at 2, to make the assert detect legacy |
139 | * invokes that used TRUE/FALSE (0 and 1). |
140 | */ |
141 | |
142 | CURLcode Curl_urldecode(struct Curl_easy *data, |
143 | const char *string, size_t length, |
144 | char **ostring, size_t *olen, |
145 | enum urlreject ctrl) |
146 | { |
147 | size_t alloc; |
148 | char *ns; |
149 | size_t strindex = 0; |
150 | unsigned long hex; |
151 | CURLcode result = CURLE_OK; |
152 | |
153 | DEBUGASSERT(string); |
154 | DEBUGASSERT(ctrl >= REJECT_NADA); /* crash on TRUE/FALSE */ |
155 | |
156 | alloc = (length?length:strlen(string)) + 1; |
157 | ns = malloc(alloc); |
158 | |
159 | if(!ns) |
160 | return CURLE_OUT_OF_MEMORY; |
161 | |
162 | while(--alloc > 0) { |
163 | unsigned char in = *string; |
164 | if(('%' == in) && (alloc > 2) && |
165 | ISXDIGIT(string[1]) && ISXDIGIT(string[2])) { |
166 | /* this is two hexadecimal digits following a '%' */ |
167 | char hexstr[3]; |
168 | char *ptr; |
169 | hexstr[0] = string[1]; |
170 | hexstr[1] = string[2]; |
171 | hexstr[2] = 0; |
172 | |
173 | hex = strtoul(hexstr, &ptr, 16); |
174 | |
175 | in = curlx_ultouc(hex); /* this long is never bigger than 255 anyway */ |
176 | |
177 | if(data) { |
178 | result = Curl_convert_from_network(data, (char *)&in, 1); |
179 | if(result) { |
180 | /* Curl_convert_from_network calls failf if unsuccessful */ |
181 | free(ns); |
182 | return result; |
183 | } |
184 | } |
185 | |
186 | string += 2; |
187 | alloc -= 2; |
188 | } |
189 | |
190 | if(((ctrl == REJECT_CTRL) && (in < 0x20)) || |
191 | ((ctrl == REJECT_ZERO) && (in == 0))) { |
192 | free(ns); |
193 | return CURLE_URL_MALFORMAT; |
194 | } |
195 | |
196 | ns[strindex++] = in; |
197 | string++; |
198 | } |
199 | ns[strindex] = 0; /* terminate it */ |
200 | |
201 | if(olen) |
202 | /* store output size */ |
203 | *olen = strindex; |
204 | |
205 | /* store output string */ |
206 | *ostring = ns; |
207 | |
208 | return CURLE_OK; |
209 | } |
210 | |
211 | /* |
212 | * Unescapes the given URL escaped string of given length. Returns a |
213 | * pointer to a malloced string with length given in *olen. |
214 | * If length == 0, the length is assumed to be strlen(string). |
215 | * If olen == NULL, no output length is stored. |
216 | */ |
217 | char *curl_easy_unescape(struct Curl_easy *data, const char *string, |
218 | int length, int *olen) |
219 | { |
220 | char *str = NULL; |
221 | if(length >= 0) { |
222 | size_t inputlen = length; |
223 | size_t outputlen; |
224 | CURLcode res = Curl_urldecode(data, string, inputlen, &str, &outputlen, |
225 | REJECT_NADA); |
226 | if(res) |
227 | return NULL; |
228 | |
229 | if(olen) { |
230 | if(outputlen <= (size_t) INT_MAX) |
231 | *olen = curlx_uztosi(outputlen); |
232 | else |
233 | /* too large to return in an int, fail! */ |
234 | Curl_safefree(str); |
235 | } |
236 | } |
237 | return str; |
238 | } |
239 | |
240 | /* For operating systems/environments that use different malloc/free |
241 | systems for the app and for this library, we provide a free that uses |
242 | the library's memory system */ |
243 | void curl_free(void *p) |
244 | { |
245 | free(p); |
246 | } |
247 | |