/src/gdal/curl/lib/escape.c
Line  | Count  | Source (jump to first uncovered line)  | 
1  |  | /***************************************************************************  | 
2  |  |  *                                  _   _ ____  _  | 
3  |  |  *  Project                     ___| | | |  _ \| |  | 
4  |  |  *                             / __| | | | |_) | |  | 
5  |  |  *                            | (__| |_| |  _ <| |___  | 
6  |  |  *                             \___|\___/|_| \_\_____|  | 
7  |  |  *  | 
8  |  |  * Copyright (C) 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  |  |  * SPDX-License-Identifier: curl  | 
22  |  |  *  | 
23  |  |  ***************************************************************************/  | 
24  |  |  | 
25  |  | /* Escape and unescape URL encoding in strings. The functions return a new  | 
26  |  |  * allocated string or NULL if an error occurred.  */  | 
27  |  |  | 
28  |  | #include "curl_setup.h"  | 
29  |  |  | 
30  |  | #include <curl/curl.h>  | 
31  |  |  | 
32  |  | struct Curl_easy;  | 
33  |  |  | 
34  |  | #include "urldata.h"  | 
35  |  | #include "curlx/warnless.h"  | 
36  |  | #include "escape.h"  | 
37  |  | #include "strdup.h"  | 
38  |  | #include "curlx/strparse.h"  | 
39  |  |  | 
40  |  | /* The last 3 #include files should be in this order */  | 
41  |  | #include "curl_printf.h"  | 
42  |  | #include "curl_memory.h"  | 
43  |  | #include "memdebug.h"  | 
44  |  |  | 
45  |  | /* for ABI-compatibility with previous versions */  | 
46  |  | char *curl_escape(const char *string, int inlength)  | 
47  | 0  | { | 
48  | 0  |   return curl_easy_escape(NULL, string, inlength);  | 
49  | 0  | }  | 
50  |  |  | 
51  |  | /* for ABI-compatibility with previous versions */  | 
52  |  | char *curl_unescape(const char *string, int length)  | 
53  | 0  | { | 
54  | 0  |   return curl_easy_unescape(NULL, string, length, NULL);  | 
55  | 0  | }  | 
56  |  |  | 
57  |  | /* Escapes for URL the given unescaped string of given length.  | 
58  |  |  * 'data' is ignored since 7.82.0.  | 
59  |  |  */  | 
60  |  | char *curl_easy_escape(CURL *data, const char *string,  | 
61  |  |                        int inlength)  | 
62  | 0  | { | 
63  | 0  |   size_t length;  | 
64  | 0  |   struct dynbuf d;  | 
65  | 0  |   (void)data;  | 
66  |  | 
  | 
67  | 0  |   if(!string || (inlength < 0))  | 
68  | 0  |     return NULL;  | 
69  |  |  | 
70  | 0  |   length = (inlength ? (size_t)inlength : strlen(string));  | 
71  | 0  |   if(!length)  | 
72  | 0  |     return strdup(""); | 
73  |  |  | 
74  | 0  |   curlx_dyn_init(&d, length * 3 + 1);  | 
75  |  | 
  | 
76  | 0  |   while(length--) { | 
77  |  |     /* treat the characters unsigned */  | 
78  | 0  |     unsigned char in = (unsigned char)*string++;  | 
79  |  | 
  | 
80  | 0  |     if(ISUNRESERVED(in)) { | 
81  |  |       /* append this */  | 
82  | 0  |       if(curlx_dyn_addn(&d, &in, 1))  | 
83  | 0  |         return NULL;  | 
84  | 0  |     }  | 
85  | 0  |     else { | 
86  |  |       /* encode it */  | 
87  | 0  |       unsigned char out[3]={'%'}; | 
88  | 0  |       Curl_hexbyte(&out[1], in, FALSE);  | 
89  | 0  |       if(curlx_dyn_addn(&d, out, 3))  | 
90  | 0  |         return NULL;  | 
91  | 0  |     }  | 
92  | 0  |   }  | 
93  |  |  | 
94  | 0  |   return curlx_dyn_ptr(&d);  | 
95  | 0  | }  | 
96  |  |  | 
97  |  | /*  | 
98  |  |  * Curl_urldecode() URL decodes the given string.  | 
99  |  |  *  | 
100  |  |  * Returns a pointer to a malloced string in *ostring with length given in  | 
101  |  |  * *olen. If length == 0, the length is assumed to be strlen(string).  | 
102  |  |  *  | 
103  |  |  * ctrl options:  | 
104  |  |  * - REJECT_NADA: accept everything  | 
105  |  |  * - REJECT_CTRL: rejects control characters (byte codes lower than 32) in  | 
106  |  |  *                the data  | 
107  |  |  * - REJECT_ZERO: rejects decoded zero bytes  | 
108  |  |  *  | 
109  |  |  * The values for the enum starts at 2, to make the assert detect legacy  | 
110  |  |  * invokes that used TRUE/FALSE (0 and 1).  | 
111  |  |  */  | 
112  |  |  | 
113  |  | CURLcode Curl_urldecode(const char *string, size_t length,  | 
114  |  |                         char **ostring, size_t *olen,  | 
115  |  |                         enum urlreject ctrl)  | 
116  | 355  | { | 
117  | 355  |   size_t alloc;  | 
118  | 355  |   char *ns;  | 
119  |  |  | 
120  | 355  |   DEBUGASSERT(string);  | 
121  | 355  |   DEBUGASSERT(ctrl >= REJECT_NADA); /* crash on TRUE/FALSE */  | 
122  |  |  | 
123  | 355  |   alloc = (length ? length : strlen(string));  | 
124  | 355  |   ns = malloc(alloc + 1);  | 
125  |  |  | 
126  | 355  |   if(!ns)  | 
127  | 0  |     return CURLE_OUT_OF_MEMORY;  | 
128  |  |  | 
129  |  |   /* store output string */  | 
130  | 355  |   *ostring = ns;  | 
131  |  |  | 
132  | 10.1k  |   while(alloc) { | 
133  | 9.76k  |     unsigned char in = (unsigned char)*string;  | 
134  | 9.76k  |     if(('%' == in) && (alloc > 2) && | 
135  | 9.76k  |        ISXDIGIT(string[1]) && ISXDIGIT(string[2])) { | 
136  |  |       /* this is two hexadecimal digits following a '%' */  | 
137  | 12  |       in = (unsigned char)((Curl_hexval(string[1]) << 4) |  | 
138  | 12  |                            Curl_hexval(string[2]));  | 
139  | 12  |       string += 3;  | 
140  | 12  |       alloc -= 3;  | 
141  | 12  |     }  | 
142  | 9.75k  |     else { | 
143  | 9.75k  |       string++;  | 
144  | 9.75k  |       alloc--;  | 
145  | 9.75k  |     }  | 
146  |  |  | 
147  | 9.76k  |     if(((ctrl == REJECT_CTRL) && (in < 0x20)) ||  | 
148  | 9.76k  |        ((ctrl == REJECT_ZERO) && (in == 0))) { | 
149  | 0  |       Curl_safefree(*ostring);  | 
150  | 0  |       return CURLE_URL_MALFORMAT;  | 
151  | 0  |     }  | 
152  |  |  | 
153  | 9.76k  |     *ns++ = (char)in;  | 
154  | 9.76k  |   }  | 
155  | 355  |   *ns = 0; /* terminate it */  | 
156  |  |  | 
157  | 355  |   if(olen)  | 
158  |  |     /* store output size */  | 
159  | 114  |     *olen = ns - *ostring;  | 
160  |  |  | 
161  | 355  |   return CURLE_OK;  | 
162  | 355  | }  | 
163  |  |  | 
164  |  | /*  | 
165  |  |  * Unescapes the given URL escaped string of given length. Returns a  | 
166  |  |  * pointer to a malloced string with length given in *olen.  | 
167  |  |  * If length == 0, the length is assumed to be strlen(string).  | 
168  |  |  * If olen == NULL, no output length is stored.  | 
169  |  |  * 'data' is ignored since 7.82.0.  | 
170  |  |  */  | 
171  |  | char *curl_easy_unescape(CURL *data, const char *string,  | 
172  |  |                          int length, int *olen)  | 
173  | 0  | { | 
174  | 0  |   char *str = NULL;  | 
175  | 0  |   (void)data;  | 
176  | 0  |   if(string && (length >= 0)) { | 
177  | 0  |     size_t inputlen = (size_t)length;  | 
178  | 0  |     size_t outputlen;  | 
179  | 0  |     CURLcode res = Curl_urldecode(string, inputlen, &str, &outputlen,  | 
180  | 0  |                                   REJECT_NADA);  | 
181  | 0  |     if(res)  | 
182  | 0  |       return NULL;  | 
183  |  |  | 
184  | 0  |     if(olen) { | 
185  | 0  |       if(outputlen <= (size_t) INT_MAX)  | 
186  | 0  |         *olen = curlx_uztosi(outputlen);  | 
187  | 0  |       else  | 
188  |  |         /* too large to return in an int, fail! */  | 
189  | 0  |         Curl_safefree(str);  | 
190  | 0  |     }  | 
191  | 0  |   }  | 
192  | 0  |   return str;  | 
193  | 0  | }  | 
194  |  |  | 
195  |  | /* For operating systems/environments that use different malloc/free  | 
196  |  |    systems for the app and for this library, we provide a free that uses  | 
197  |  |    the library's memory system */  | 
198  |  | void curl_free(void *p)  | 
199  | 0  | { | 
200  | 0  |   free(p);  | 
201  | 0  | }  | 
202  |  |  | 
203  |  | /*  | 
204  |  |  * Curl_hexencode()  | 
205  |  |  *  | 
206  |  |  * Converts binary input to lowercase hex-encoded ASCII output.  | 
207  |  |  * Null-terminated.  | 
208  |  |  */  | 
209  |  | void Curl_hexencode(const unsigned char *src, size_t len, /* input length */  | 
210  |  |                     unsigned char *out, size_t olen) /* output buffer size */  | 
211  | 0  | { | 
212  | 0  |   DEBUGASSERT(src && len && (olen >= 3));  | 
213  | 0  |   if(src && len && (olen >= 3)) { | 
214  | 0  |     while(len-- && (olen >= 3)) { | 
215  | 0  |       Curl_hexbyte(out, *src, TRUE);  | 
216  | 0  |       ++src;  | 
217  | 0  |       out += 2;  | 
218  | 0  |       olen -= 2;  | 
219  | 0  |     }  | 
220  | 0  |     *out = 0;  | 
221  | 0  |   }  | 
222  | 0  |   else if(olen)  | 
223  | 0  |     *out = 0;  | 
224  | 0  | }  | 
225  |  |  | 
226  |  | /* Curl_hexbyte  | 
227  |  |  *  | 
228  |  |  * Output a single unsigned char as a two-digit hex number, lowercase or  | 
229  |  |  * uppercase  | 
230  |  |  */  | 
231  |  | void Curl_hexbyte(unsigned char *dest, /* must fit two bytes */  | 
232  |  |                   unsigned char val,  | 
233  |  |                   bool lowercase)  | 
234  | 286k  | { | 
235  | 286k  |   const unsigned char *t = lowercase ? Curl_ldigits : Curl_udigits;  | 
236  | 286k  |   dest[0] = t[val >> 4];  | 
237  | 286k  |   dest[1] = t[val & 0x0F];  | 
238  | 286k  | }  |