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#include <ctype.h>
28#include <curl/curl.h>
29
30#include <stdio.h>
31#include <stdlib.h>
32#include <string.h>
33#include "curl_memory.h"
34#include "urldata.h"
35#include "warnless.h"
36#include "non-ascii.h"
37
38#define _MPRINTF_REPLACE /* use our functions only */
39#include <curl/mprintf.h>
40
41/* The last #include file should be: */
42#include "memdebug.h"
43
44/* Portable character check (remember EBCDIC). Do not use isalnum() because
45   its behavior is altered by the current locale.
46   See http://tools.ietf.org/html/rfc3986#section-2.3
47*/
48static bool Curl_isunreserved(unsigned char in)
49{
50  switch (in) {
51    case '0': case '1': case '2': case '3': case '4':
52    case '5': case '6': case '7': case '8': case '9':
53    case 'a': case 'b': case 'c': case 'd': case 'e':
54    case 'f': case 'g': case 'h': case 'i': case 'j':
55    case 'k': case 'l': case 'm': case 'n': case 'o':
56    case 'p': case 'q': case 'r': case 's': case 't':
57    case 'u': case 'v': case 'w': case 'x': case 'y': case 'z':
58    case 'A': case 'B': case 'C': case 'D': case 'E':
59    case 'F': case 'G': case 'H': case 'I': case 'J':
60    case 'K': case 'L': case 'M': case 'N': case 'O':
61    case 'P': case 'Q': case 'R': case 'S': case 'T':
62    case 'U': case 'V': case 'W': case 'X': case 'Y': case 'Z':
63    case '-': case '.': case '_': case '~':
64      return TRUE;
65    default:
66      break;
67  }
68  return FALSE;
69}
70
71/* for ABI-compatibility with previous versions */
72char *curl_escape(const char *string, int inlength)
73{
74  return curl_easy_escape(NULL, string, inlength);
75}
76
77/* for ABI-compatibility with previous versions */
78char *curl_unescape(const char *string, int length)
79{
80  return curl_easy_unescape(NULL, string, length, NULL);
81}
82
83char *curl_easy_escape(CURL *handle, const char *string, int inlength)
84{
85  size_t alloc = (inlength?(size_t)inlength:strlen(string))+1;
86  char *ns;
87  char *testing_ptr = NULL;
88  unsigned char in; /* we need to treat the characters unsigned */
89  size_t newlen = alloc;
90  int strindex=0;
91  size_t length;
92  CURLcode res;
93
94  ns = malloc(alloc);
95  if(!ns)
96    return NULL;
97
98  length = alloc-1;
99  while(length--) {
100    in = *string;
101
102    if(Curl_isunreserved(in))
103      /* just copy this */
104      ns[strindex++]=in;
105    else {
106      /* encode it */
107      newlen += 2; /* the size grows with two, since this'll become a %XX */
108      if(newlen > alloc) {
109        alloc *= 2;
110        testing_ptr = realloc(ns, alloc);
111        if(!testing_ptr) {
112          free( ns );
113          return NULL;
114        }
115        else {
116          ns = testing_ptr;
117        }
118      }
119
120      res = Curl_convert_to_network(handle, &in, 1);
121      if(res) {
122        /* Curl_convert_to_network calls failf if unsuccessful */
123        free(ns);
124        return NULL;
125      }
126
127      snprintf(&ns[strindex], 4, "%%%02X", in);
128
129      strindex+=3;
130    }
131    string++;
132  }
133  ns[strindex]=0; /* terminate it */
134  return ns;
135}
136
137/*
138 * Unescapes the given URL escaped string of given length. Returns a
139 * pointer to a malloced string with length given in *olen.
140 * If length == 0, the length is assumed to be strlen(string).
141 * If olen == NULL, no output length is stored.
142 */
143char *curl_easy_unescape(CURL *handle, const char *string, int length,
144                         int *olen)
145{
146  int alloc = (length?length:(int)strlen(string))+1;
147  char *ns = malloc(alloc);
148  unsigned char in;
149  int strindex=0;
150  unsigned long hex;
151  CURLcode res;
152
153  if(!ns)
154    return NULL;
155
156  while(--alloc > 0) {
157    in = *string;
158    if(('%' == in) && ISXDIGIT(string[1]) && ISXDIGIT(string[2])) {
159      /* this is two hexadecimal digits following a '%' */
160      char hexstr[3];
161      char *ptr;
162      hexstr[0] = string[1];
163      hexstr[1] = string[2];
164      hexstr[2] = 0;
165
166      hex = strtoul(hexstr, &ptr, 16);
167
168      in = curlx_ultouc(hex); /* this long is never bigger than 255 anyway */
169
170      res = Curl_convert_from_network(handle, &in, 1);
171      if(res) {
172        /* Curl_convert_from_network calls failf if unsuccessful */
173        free(ns);
174        return NULL;
175      }
176
177      string+=2;
178      alloc-=2;
179    }
180
181    ns[strindex++] = in;
182    string++;
183  }
184  ns[strindex]=0; /* terminate it */
185
186  if(olen)
187    /* store output size */
188    *olen = strindex;
189  return ns;
190}
191
192/* For operating systems/environments that use different malloc/free
193   systems for the app and for this library, we provide a free that uses
194   the library's memory system */
195void curl_free(void *p)
196{
197  if(p)
198    free(p);
199}
200