Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 1 | /*************************************************************************** |
| 2 | * _ _ ____ _ |
| 3 | * Project ___| | | | _ \| | |
| 4 | * / __| | | | |_) | | |
| 5 | * | (__| |_| | _ <| |___ |
| 6 | * \___|\___/|_| \_\_____| |
| 7 | * |
Elliott Hughes | 82be86d | 2017-09-20 17:00:17 -0700 | [diff] [blame] | 8 | * Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel@haxx.se>, et al. |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 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 |
Alex Deymo | 8f1a214 | 2016-06-28 14:49:26 -0700 | [diff] [blame] | 12 | * are also available at https://curl.haxx.se/docs/copyright.html. |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 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 | #include "curl_setup.h" |
| 24 | |
Alex Deymo | 8f1a214 | 2016-06-28 14:49:26 -0700 | [diff] [blame] | 25 | #include <curl/curl.h> |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 26 | |
Alex Deymo | 8f1a214 | 2016-06-28 14:49:26 -0700 | [diff] [blame] | 27 | #include "dotdot.h" |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 28 | #include "curl_memory.h" |
Alex Deymo | 8f1a214 | 2016-06-28 14:49:26 -0700 | [diff] [blame] | 29 | |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 30 | /* The last #include file should be: */ |
| 31 | #include "memdebug.h" |
| 32 | |
| 33 | /* |
| 34 | * "Remove Dot Segments" |
Alex Deymo | 8f1a214 | 2016-06-28 14:49:26 -0700 | [diff] [blame] | 35 | * https://tools.ietf.org/html/rfc3986#section-5.2.4 |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 36 | */ |
| 37 | |
| 38 | /* |
| 39 | * Curl_dedotdotify() |
Alex Deymo | 8f1a214 | 2016-06-28 14:49:26 -0700 | [diff] [blame] | 40 | * @unittest: 1395 |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 41 | * |
| 42 | * This function gets a zero-terminated path with dot and dotdot sequences |
| 43 | * passed in and strips them off according to the rules in RFC 3986 section |
| 44 | * 5.2.4. |
| 45 | * |
| 46 | * The function handles a query part ('?' + stuff) appended but it expects |
| 47 | * that fragments ('#' + stuff) have already been cut off. |
| 48 | * |
| 49 | * RETURNS |
| 50 | * |
| 51 | * an allocated dedotdotified output string |
| 52 | */ |
| 53 | char *Curl_dedotdotify(const char *input) |
| 54 | { |
| 55 | size_t inlen = strlen(input); |
| 56 | char *clone; |
| 57 | size_t clen = inlen; /* the length of the cloned input */ |
Alex Deymo | 486467e | 2017-12-19 19:04:07 +0100 | [diff] [blame] | 58 | char *out = malloc(inlen + 1); |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 59 | char *outptr; |
| 60 | char *orgclone; |
| 61 | char *queryp; |
| 62 | if(!out) |
| 63 | return NULL; /* out of memory */ |
| 64 | |
| 65 | /* get a cloned copy of the input */ |
| 66 | clone = strdup(input); |
| 67 | if(!clone) { |
| 68 | free(out); |
| 69 | return NULL; |
| 70 | } |
| 71 | orgclone = clone; |
| 72 | outptr = out; |
| 73 | |
Alex Deymo | 8f1a214 | 2016-06-28 14:49:26 -0700 | [diff] [blame] | 74 | if(!*clone) { |
| 75 | /* zero length string, return that */ |
| 76 | free(out); |
| 77 | return clone; |
| 78 | } |
| 79 | |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 80 | /* |
| 81 | * To handle query-parts properly, we must find it and remove it during the |
| 82 | * dotdot-operation and then append it again at the end to the output |
| 83 | * string. |
| 84 | */ |
| 85 | queryp = strchr(clone, '?'); |
| 86 | if(queryp) |
| 87 | *queryp = 0; |
| 88 | |
| 89 | do { |
| 90 | |
| 91 | /* A. If the input buffer begins with a prefix of "../" or "./", then |
| 92 | remove that prefix from the input buffer; otherwise, */ |
| 93 | |
| 94 | if(!strncmp("./", clone, 2)) { |
Alex Deymo | 486467e | 2017-12-19 19:04:07 +0100 | [diff] [blame] | 95 | clone += 2; |
| 96 | clen -= 2; |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 97 | } |
| 98 | else if(!strncmp("../", clone, 3)) { |
Alex Deymo | 486467e | 2017-12-19 19:04:07 +0100 | [diff] [blame] | 99 | clone += 3; |
| 100 | clen -= 3; |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 101 | } |
| 102 | |
| 103 | /* B. if the input buffer begins with a prefix of "/./" or "/.", where |
| 104 | "." is a complete path segment, then replace that prefix with "/" in |
| 105 | the input buffer; otherwise, */ |
| 106 | else if(!strncmp("/./", clone, 3)) { |
Alex Deymo | 486467e | 2017-12-19 19:04:07 +0100 | [diff] [blame] | 107 | clone += 2; |
| 108 | clen -= 2; |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 109 | } |
| 110 | else if(!strcmp("/.", clone)) { |
| 111 | clone[1]='/'; |
| 112 | clone++; |
Alex Deymo | 486467e | 2017-12-19 19:04:07 +0100 | [diff] [blame] | 113 | clen -= 1; |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 114 | } |
| 115 | |
| 116 | /* C. if the input buffer begins with a prefix of "/../" or "/..", where |
| 117 | ".." is a complete path segment, then replace that prefix with "/" in |
| 118 | the input buffer and remove the last segment and its preceding "/" (if |
| 119 | any) from the output buffer; otherwise, */ |
| 120 | |
| 121 | else if(!strncmp("/../", clone, 4)) { |
Alex Deymo | 486467e | 2017-12-19 19:04:07 +0100 | [diff] [blame] | 122 | clone += 3; |
| 123 | clen -= 3; |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 124 | /* remove the last segment from the output buffer */ |
| 125 | while(outptr > out) { |
| 126 | outptr--; |
| 127 | if(*outptr == '/') |
| 128 | break; |
| 129 | } |
| 130 | *outptr = 0; /* zero-terminate where it stops */ |
| 131 | } |
| 132 | else if(!strcmp("/..", clone)) { |
| 133 | clone[2]='/'; |
Alex Deymo | 486467e | 2017-12-19 19:04:07 +0100 | [diff] [blame] | 134 | clone += 2; |
| 135 | clen -= 2; |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 136 | /* remove the last segment from the output buffer */ |
| 137 | while(outptr > out) { |
| 138 | outptr--; |
| 139 | if(*outptr == '/') |
| 140 | break; |
| 141 | } |
| 142 | *outptr = 0; /* zero-terminate where it stops */ |
| 143 | } |
| 144 | |
| 145 | /* D. if the input buffer consists only of "." or "..", then remove |
| 146 | that from the input buffer; otherwise, */ |
| 147 | |
| 148 | else if(!strcmp(".", clone) || !strcmp("..", clone)) { |
Alex Deymo | 486467e | 2017-12-19 19:04:07 +0100 | [diff] [blame] | 149 | *clone = 0; |
| 150 | *out = 0; |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 151 | } |
| 152 | |
| 153 | else { |
| 154 | /* E. move the first path segment in the input buffer to the end of |
| 155 | the output buffer, including the initial "/" character (if any) and |
| 156 | any subsequent characters up to, but not including, the next "/" |
| 157 | character or the end of the input buffer. */ |
| 158 | |
| 159 | do { |
| 160 | *outptr++ = *clone++; |
| 161 | clen--; |
| 162 | } while(*clone && (*clone != '/')); |
| 163 | *outptr = 0; |
| 164 | } |
| 165 | |
| 166 | } while(*clone); |
| 167 | |
| 168 | if(queryp) { |
| 169 | size_t qlen; |
| 170 | /* There was a query part, append that to the output. The 'clone' string |
| 171 | may now have been altered so we copy from the original input string |
| 172 | from the correct index. */ |
| 173 | size_t oindex = queryp - orgclone; |
| 174 | qlen = strlen(&input[oindex]); |
Alex Deymo | 486467e | 2017-12-19 19:04:07 +0100 | [diff] [blame] | 175 | memcpy(outptr, &input[oindex], qlen + 1); /* include the end zero byte */ |
Bertrand SIMONNET | e6cd738 | 2015-07-01 15:39:44 -0700 | [diff] [blame] | 176 | } |
| 177 | |
| 178 | free(orgclone); |
| 179 | return out; |
| 180 | } |