1/*************************************************************************** 2 * _ _ ____ _ 3 * Project ___| | | | _ \| | 4 * / __| | | | |_) | | 5 * | (__| |_| | _ <| |___ 6 * \___|\___/|_| \_\_____| 7 * 8 * Copyright (C) 1998 - 2011, 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 http://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 "setup.h" 27 28#include <curl/curl.h> 29 30#include "curl_memory.h" 31#include "urldata.h" 32#include "warnless.h" 33#include "non-ascii.h" 34 35#define _MPRINTF_REPLACE /* use our functions only */ 36#include <curl/mprintf.h> 37 38/* The last #include file should be: */ 39#include "memdebug.h" 40 41/* Portable character check (remember EBCDIC). Do not use isalnum() because 42 its behavior is altered by the current locale. 43 See http://tools.ietf.org/html/rfc3986#section-2.3 44*/ 45static bool Curl_isunreserved(unsigned char in) 46{ 47 switch (in) { 48 case '0': case '1': case '2': case '3': case '4': 49 case '5': case '6': case '7': case '8': case '9': 50 case 'a': case 'b': case 'c': case 'd': case 'e': 51 case 'f': case 'g': case 'h': case 'i': case 'j': 52 case 'k': case 'l': case 'm': case 'n': case 'o': 53 case 'p': case 'q': case 'r': case 's': case 't': 54 case 'u': case 'v': case 'w': case 'x': case 'y': case 'z': 55 case 'A': case 'B': case 'C': case 'D': case 'E': 56 case 'F': case 'G': case 'H': case 'I': case 'J': 57 case 'K': case 'L': case 'M': case 'N': case 'O': 58 case 'P': case 'Q': case 'R': case 'S': case 'T': 59 case 'U': case 'V': case 'W': case 'X': case 'Y': case 'Z': 60 case '-': case '.': case '_': case '~': 61 return TRUE; 62 default: 63 break; 64 } 65 return FALSE; 66} 67 68/* for ABI-compatibility with previous versions */ 69char *curl_escape(const char *string, int inlength) 70{ 71 return curl_easy_escape(NULL, string, inlength); 72} 73 74/* for ABI-compatibility with previous versions */ 75char *curl_unescape(const char *string, int length) 76{ 77 return curl_easy_unescape(NULL, string, length, NULL); 78} 79 80char *curl_easy_escape(CURL *handle, const char *string, int inlength) 81{ 82 size_t alloc = (inlength?(size_t)inlength:strlen(string))+1; 83 char *ns; 84 char *testing_ptr = NULL; 85 unsigned char in; /* we need to treat the characters unsigned */ 86 size_t newlen = alloc; 87 int strindex=0; 88 size_t length; 89 CURLcode res; 90 91 ns = malloc(alloc); 92 if(!ns) 93 return NULL; 94 95 length = alloc-1; 96 while(length--) { 97 in = *string; 98 99 if(Curl_isunreserved(in)) 100 /* just copy this */ 101 ns[strindex++]=in; 102 else { 103 /* encode it */ 104 newlen += 2; /* the size grows with two, since this'll become a %XX */ 105 if(newlen > alloc) { 106 alloc *= 2; 107 testing_ptr = realloc(ns, alloc); 108 if(!testing_ptr) { 109 free( ns ); 110 return NULL; 111 } 112 else { 113 ns = testing_ptr; 114 } 115 } 116 117 res = Curl_convert_to_network(handle, &in, 1); 118 if(res) { 119 /* Curl_convert_to_network calls failf if unsuccessful */ 120 free(ns); 121 return NULL; 122 } 123 124 snprintf(&ns[strindex], 4, "%%%02X", in); 125 126 strindex+=3; 127 } 128 string++; 129 } 130 ns[strindex]=0; /* terminate it */ 131 return ns; 132} 133 134/* 135 * Unescapes the given URL escaped string of given length. Returns a 136 * pointer to a malloced string with length given in *olen. 137 * If length == 0, the length is assumed to be strlen(string). 138 * If olen == NULL, no output length is stored. 139 */ 140char *curl_easy_unescape(CURL *handle, const char *string, int length, 141 int *olen) 142{ 143 int alloc = (length?length:(int)strlen(string))+1; 144 char *ns = malloc(alloc); 145 unsigned char in; 146 int strindex=0; 147 unsigned long hex; 148 CURLcode res; 149 150 if(!ns) 151 return NULL; 152 153 while(--alloc > 0) { 154 in = *string; 155 if(('%' == in) && ISXDIGIT(string[1]) && ISXDIGIT(string[2])) { 156 /* this is two hexadecimal digits following a '%' */ 157 char hexstr[3]; 158 char *ptr; 159 hexstr[0] = string[1]; 160 hexstr[1] = string[2]; 161 hexstr[2] = 0; 162 163 hex = strtoul(hexstr, &ptr, 16); 164 165 in = curlx_ultouc(hex); /* this long is never bigger than 255 anyway */ 166 167 res = Curl_convert_from_network(handle, &in, 1); 168 if(res) { 169 /* Curl_convert_from_network calls failf if unsuccessful */ 170 free(ns); 171 return NULL; 172 } 173 174 string+=2; 175 alloc-=2; 176 } 177 178 ns[strindex++] = in; 179 string++; 180 } 181 ns[strindex]=0; /* terminate it */ 182 183 if(olen) 184 /* store output size */ 185 *olen = strindex; 186 return ns; 187} 188 189/* For operating systems/environments that use different malloc/free 190 systems for the app and for this library, we provide a free that uses 191 the library's memory system */ 192void curl_free(void *p) 193{ 194 if(p) 195 free(p); 196} 197