| 1 |
/*************************************************************************** |
| 2 |
* _ _ ____ _ |
| 3 |
* Project ___| | | | _ \| | |
| 4 |
* / __| | | | |_) | | |
| 5 |
* | (__| |_| | _ <| |___ |
| 6 |
* \___|\___/|_| \_\_____| |
| 7 |
* |
| 8 |
* Copyright (C) 1998 - 2018, Daniel Stenberg, <[email protected]>, 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.haxx.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 alloc; |
| 83 |
char *ns; |
| 84 |
char *testing_ptr = NULL; |
| 85 |
size_t newlen; |
| 86 |
size_t strindex = 0; |
| 87 |
size_t length; |
| 88 |
CURLcode result; |
| 89 |
|
| 90 |
if(inlength < 0) |
| 91 |
return NULL; |
| 92 |
|
| 93 |
alloc = (inlength?(size_t)inlength:strlen(string)) + 1; |
| 94 |
newlen = alloc; |
| 95 |
|
| 96 |
ns = malloc(alloc); |
| 97 |
if(!ns) |
| 98 |
return NULL; |
| 99 |
|
| 100 |
length = alloc-1; |
| 101 |
while(length--) { |
| 102 |
unsigned char in = *string; /* we need to treat the characters unsigned */ |
| 103 |
|
| 104 |
if(Curl_isunreserved(in)) |
| 105 |
/* just copy this */ |
| 106 |
ns[strindex++] = in; |
| 107 |
else { |
| 108 |
/* encode it */ |
| 109 |
newlen += 2; /* the size grows with two, since this'll become a %XX */ |
| 110 |
if(newlen > alloc) { |
| 111 |
alloc *= 2; |
| 112 |
testing_ptr = Curl_saferealloc(ns, alloc); |
| 113 |
if(!testing_ptr) |
| 114 |
return NULL; |
| 115 |
ns = testing_ptr; |
| 116 |
} |
| 117 |
|
| 118 |
result = Curl_convert_to_network(data, (char *)&in, 1); |
| 119 |
if(result) { |
| 120 |
/* Curl_convert_to_network calls failf if unsuccessful */ |
| 121 |
free(ns); |
| 122 |
return NULL; |
| 123 |
} |
| 124 |
|
| 125 |
msnprintf(&ns[strindex], 4, "%%%02X", in); |
| 126 |
|
| 127 |
strindex += 3; |
| 128 |
} |
| 129 |
string++; |
| 130 |
} |
| 131 |
ns[strindex] = 0; /* terminate it */ |
| 132 |
return ns; |
| 133 |
} |
| 134 |
|
| 135 |
/* |
| 136 |
* Curl_urldecode() URL decodes the given string. |
| 137 |
* |
| 138 |
* Optionally detects control characters (byte codes lower than 32) in the |
| 139 |
* data and rejects such data. |
| 140 |
* |
| 141 |
* Returns a pointer to a malloced string in *ostring with length given in |
| 142 |
* *olen. If length == 0, the length is assumed to be strlen(string). |
| 143 |
* |
| 144 |
* 'data' can be set to NULL but then this function can't convert network |
| 145 |
* data to host for non-ascii. |
| 146 |
*/ |
| 147 |
CURLcode Curl_urldecode(struct Curl_easy *data, |
| 148 |
const char *string, size_t length, |
| 149 |
char **ostring, size_t *olen, |
| 150 |
bool reject_ctrl) |
| 151 |
{ |
| 152 |
size_t alloc = (length?length:strlen(string)) + 1; |
| 153 |
char *ns = malloc(alloc); |
| 154 |
size_t strindex = 0; |
| 155 |
unsigned long hex; |
| 156 |
CURLcode result = CURLE_OK; |
| 157 |
|
| 158 |
if(!ns) |
| 159 |
return CURLE_OUT_OF_MEMORY; |
| 160 |
|
| 161 |
while(--alloc > 0) { |
| 162 |
unsigned char in = *string; |
| 163 |
if(('%' == in) && (alloc > 2) && |
| 164 |
ISXDIGIT(string[1]) && ISXDIGIT(string[2])) { |
| 165 |
/* this is two hexadecimal digits following a '%' */ |
| 166 |
char hexstr[3]; |
| 167 |
char *ptr; |
| 168 |
hexstr[0] = string[1]; |
| 169 |
hexstr[1] = string[2]; |
| 170 |
hexstr[2] = 0; |
| 171 |
|
| 172 |
hex = strtoul(hexstr, &ptr, 16); |
| 173 |
|
| 174 |
in = curlx_ultouc(hex); /* this long is never bigger than 255 anyway */ |
| 175 |
|
| 176 |
if(data) { |
| 177 |
result = Curl_convert_from_network(data, (char *)&in, 1); |
| 178 |
if(result) { |
| 179 |
/* Curl_convert_from_network calls failf if unsuccessful */ |
| 180 |
free(ns); |
| 181 |
return result; |
| 182 |
} |
| 183 |
} |
| 184 |
|
| 185 |
string += 2; |
| 186 |
alloc -= 2; |
| 187 |
} |
| 188 |
|
| 189 |
if(reject_ctrl && (in < 0x20)) { |
| 190 |
free(ns); |
| 191 |
return CURLE_URL_MALFORMAT; |
| 192 |
} |
| 193 |
|
| 194 |
ns[strindex++] = in; |
| 195 |
string++; |
| 196 |
} |
| 197 |
ns[strindex] = 0; /* terminate it */ |
| 198 |
|
| 199 |
if(olen) |
| 200 |
/* store output size */ |
| 201 |
*olen = strindex; |
| 202 |
|
| 203 |
/* store output string */ |
| 204 |
*ostring = ns; |
| 205 |
|
| 206 |
return CURLE_OK; |
| 207 |
} |
| 208 |
|
| 209 |
/* |
| 210 |
* Unescapes the given URL escaped string of given length. Returns a |
| 211 |
* pointer to a malloced string with length given in *olen. |
| 212 |
* If length == 0, the length is assumed to be strlen(string). |
| 213 |
* If olen == NULL, no output length is stored. |
| 214 |
*/ |
| 215 |
char *curl_easy_unescape(struct Curl_easy *data, const char *string, |
| 216 |
int length, int *olen) |
| 217 |
{ |
| 218 |
char *str = NULL; |
| 219 |
if(length >= 0) { |
| 220 |
size_t inputlen = length; |
| 221 |
size_t outputlen; |
| 222 |
CURLcode res = Curl_urldecode(data, string, inputlen, &str, &outputlen, |
| 223 |
FALSE); |
| 224 |
if(res) |
| 225 |
return NULL; |
| 226 |
|
| 227 |
if(olen) { |
| 228 |
if(outputlen <= (size_t) INT_MAX) |
| 229 |
*olen = curlx_uztosi(outputlen); |
| 230 |
else |
| 231 |
/* too large to return in an int, fail! */ |
| 232 |
Curl_safefree(str); |
| 233 |
} |
| 234 |
} |
| 235 |
return str; |
| 236 |
} |
| 237 |
|
| 238 |
/* For operating systems/environments that use different malloc/free |
| 239 |
systems for the app and for this library, we provide a free that uses |
| 240 |
the library's memory system */ |
| 241 |
void curl_free(void *p) |
| 242 |
{ |
| 243 |
free(p); |
| 244 |
} |