1 /*************************************************************************** 2 * _ _ ____ _ 3 * Project ___| | | | _ \| | 4 * / __| | | | |_) | | 5 * | (__| |_| | _ <| |___ 6 * \___|\___/|_| \_\_____| 7 * 8 * Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel (at) 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.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 static 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 unsigned char in; /* we need to treat the characters unsigned */ 86 size_t newlen; 87 size_t strindex = 0; 88 size_t length; 89 CURLcode result; 90 91 if(inlength < 0) 92 return NULL; 93 94 alloc = (inlength?(size_t)inlength:strlen(string)) + 1; 95 newlen = alloc; 96 97 ns = malloc(alloc); 98 if(!ns) 99 return NULL; 100 101 length = alloc-1; 102 while(length--) { 103 in = *string; 104 105 if(Curl_isunreserved(in)) 106 /* just copy this */ 107 ns[strindex++] = in; 108 else { 109 /* encode it */ 110 newlen += 2; /* the size grows with two, since this'll become a %XX */ 111 if(newlen > alloc) { 112 alloc *= 2; 113 testing_ptr = Curl_saferealloc(ns, alloc); 114 if(!testing_ptr) 115 return NULL; 116 ns = testing_ptr; 117 } 118 119 result = Curl_convert_to_network(data, (char *)&in, 1); 120 if(result) { 121 /* Curl_convert_to_network calls failf if unsuccessful */ 122 free(ns); 123 return NULL; 124 } 125 126 snprintf(&ns[strindex], 4, "%%%02X", in); 127 128 strindex += 3; 129 } 130 string++; 131 } 132 ns[strindex] = 0; /* terminate it */ 133 return ns; 134 } 135 136 /* 137 * Curl_urldecode() URL decodes the given string. 138 * 139 * Optionally detects control characters (byte codes lower than 32) in the 140 * data and rejects such data. 141 * 142 * Returns a pointer to a malloced string in *ostring with length given in 143 * *olen. If length == 0, the length is assumed to be strlen(string). 144 * 145 */ 146 CURLcode Curl_urldecode(struct Curl_easy *data, 147 const char *string, size_t length, 148 char **ostring, size_t *olen, 149 bool reject_ctrl) 150 { 151 size_t alloc = (length?length:strlen(string)) + 1; 152 char *ns = malloc(alloc); 153 unsigned char in; 154 size_t strindex = 0; 155 unsigned long hex; 156 CURLcode result; 157 158 if(!ns) 159 return CURLE_OUT_OF_MEMORY; 160 161 while(--alloc > 0) { 162 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 result = Curl_convert_from_network(data, (char *)&in, 1); 177 if(result) { 178 /* Curl_convert_from_network calls failf if unsuccessful */ 179 free(ns); 180 return result; 181 } 182 183 string += 2; 184 alloc -= 2; 185 } 186 187 if(reject_ctrl && (in < 0x20)) { 188 free(ns); 189 return CURLE_URL_MALFORMAT; 190 } 191 192 ns[strindex++] = in; 193 string++; 194 } 195 ns[strindex] = 0; /* terminate it */ 196 197 if(olen) 198 /* store output size */ 199 *olen = strindex; 200 201 /* store output string */ 202 *ostring = ns; 203 204 return CURLE_OK; 205 } 206 207 /* 208 * Unescapes the given URL escaped string of given length. Returns a 209 * pointer to a malloced string with length given in *olen. 210 * If length == 0, the length is assumed to be strlen(string). 211 * If olen == NULL, no output length is stored. 212 */ 213 char *curl_easy_unescape(struct Curl_easy *data, const char *string, 214 int length, int *olen) 215 { 216 char *str = NULL; 217 if(length >= 0) { 218 size_t inputlen = length; 219 size_t outputlen; 220 CURLcode res = Curl_urldecode(data, string, inputlen, &str, &outputlen, 221 FALSE); 222 if(res) 223 return NULL; 224 225 if(olen) { 226 if(outputlen <= (size_t) INT_MAX) 227 *olen = curlx_uztosi(outputlen); 228 else 229 /* too large to return in an int, fail! */ 230 Curl_safefree(str); 231 } 232 } 233 return str; 234 } 235 236 /* For operating systems/environments that use different malloc/free 237 systems for the app and for this library, we provide a free that uses 238 the library's memory system */ 239 void curl_free(void *p) 240 { 241 free(p); 242 } 243
