| 1 | /***************************************************************************
|
|---|
| 2 | * _ _ ____ _
|
|---|
| 3 | * Project ___| | | | _ \| |
|
|---|
| 4 | * / __| | | | |_) | |
|
|---|
| 5 | * | (__| |_| | _ <| |___
|
|---|
| 6 | * \___|\___/|_| \_\_____|
|
|---|
| 7 | *
|
|---|
| 8 | * Copyright (C) 1998 - 2022, 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 | /*
|
|---|
| 26 | * IDN conversions
|
|---|
| 27 | */
|
|---|
| 28 |
|
|---|
| 29 | #include "curl_setup.h"
|
|---|
| 30 | #include "urldata.h"
|
|---|
| 31 | #include "idn.h"
|
|---|
| 32 | #include "sendf.h"
|
|---|
| 33 | #include "curl_multibyte.h"
|
|---|
| 34 | #include "warnless.h"
|
|---|
| 35 |
|
|---|
| 36 | #ifdef USE_LIBIDN2
|
|---|
| 37 | #include <idn2.h>
|
|---|
| 38 |
|
|---|
| 39 | #if defined(WIN32) && defined(UNICODE)
|
|---|
| 40 | #define IDN2_LOOKUP(name, host, flags) \
|
|---|
| 41 | idn2_lookup_u8((const uint8_t *)name, (uint8_t **)host, flags)
|
|---|
| 42 | #else
|
|---|
| 43 | #define IDN2_LOOKUP(name, host, flags) \
|
|---|
| 44 | idn2_lookup_ul((const char *)name, (char **)host, flags)
|
|---|
| 45 | #endif
|
|---|
| 46 | #endif /* USE_LIBIDN2 */
|
|---|
| 47 |
|
|---|
| 48 | /* The last 3 #include files should be in this order */
|
|---|
| 49 | #include "curl_printf.h"
|
|---|
| 50 | #include "curl_memory.h"
|
|---|
| 51 | #include "memdebug.h"
|
|---|
| 52 |
|
|---|
| 53 | #ifdef USE_WIN32_IDN
|
|---|
| 54 | /* using Windows kernel32 and normaliz libraries. */
|
|---|
| 55 |
|
|---|
| 56 | #if !defined(_WIN32_WINNT) || _WIN32_WINNT < 0x600
|
|---|
| 57 | WINBASEAPI int WINAPI IdnToAscii(DWORD dwFlags,
|
|---|
| 58 | const WCHAR *lpUnicodeCharStr,
|
|---|
| 59 | int cchUnicodeChar,
|
|---|
| 60 | WCHAR *lpASCIICharStr,
|
|---|
| 61 | int cchASCIIChar);
|
|---|
| 62 | WINBASEAPI int WINAPI IdnToUnicode(DWORD dwFlags,
|
|---|
| 63 | const WCHAR *lpASCIICharStr,
|
|---|
| 64 | int cchASCIIChar,
|
|---|
| 65 | WCHAR *lpUnicodeCharStr,
|
|---|
| 66 | int cchUnicodeChar);
|
|---|
| 67 | #endif
|
|---|
| 68 |
|
|---|
| 69 | #define IDN_MAX_LENGTH 255
|
|---|
| 70 |
|
|---|
| 71 | bool Curl_win32_idn_to_ascii(const char *in, char **out)
|
|---|
| 72 | {
|
|---|
| 73 | bool success = FALSE;
|
|---|
| 74 |
|
|---|
| 75 | wchar_t *in_w = curlx_convert_UTF8_to_wchar(in);
|
|---|
| 76 | if(in_w) {
|
|---|
| 77 | wchar_t punycode[IDN_MAX_LENGTH];
|
|---|
| 78 | int chars = IdnToAscii(0, in_w, -1, punycode, IDN_MAX_LENGTH);
|
|---|
| 79 | curlx_unicodefree(in_w);
|
|---|
| 80 | if(chars) {
|
|---|
| 81 | char *mstr = curlx_convert_wchar_to_UTF8(punycode);
|
|---|
| 82 | if(mstr) {
|
|---|
| 83 | *out = strdup(mstr);
|
|---|
| 84 | curlx_unicodefree(mstr);
|
|---|
| 85 | if(*out)
|
|---|
| 86 | success = TRUE;
|
|---|
| 87 | }
|
|---|
| 88 | }
|
|---|
| 89 | }
|
|---|
| 90 |
|
|---|
| 91 | return success;
|
|---|
| 92 | }
|
|---|
| 93 |
|
|---|
| 94 | #endif /* USE_WIN32_IDN */
|
|---|
| 95 |
|
|---|
| 96 | /*
|
|---|
| 97 | * Helpers for IDNA conversions.
|
|---|
| 98 | */
|
|---|
| 99 | bool Curl_is_ASCII_name(const char *hostname)
|
|---|
| 100 | {
|
|---|
| 101 | /* get an UNSIGNED local version of the pointer */
|
|---|
| 102 | const unsigned char *ch = (const unsigned char *)hostname;
|
|---|
| 103 |
|
|---|
| 104 | if(!hostname) /* bad input, consider it ASCII! */
|
|---|
| 105 | return TRUE;
|
|---|
| 106 |
|
|---|
| 107 | while(*ch) {
|
|---|
| 108 | if(*ch++ & 0x80)
|
|---|
| 109 | return FALSE;
|
|---|
| 110 | }
|
|---|
| 111 | return TRUE;
|
|---|
| 112 | }
|
|---|
| 113 |
|
|---|
| 114 | #ifdef USE_IDN
|
|---|
| 115 | /*
|
|---|
| 116 | * Curl_idn_decode() returns an allocated IDN decoded string if it was
|
|---|
| 117 | * possible. NULL on error.
|
|---|
| 118 | */
|
|---|
| 119 | static char *Curl_idn_decode(const char *input)
|
|---|
| 120 | {
|
|---|
| 121 | char *decoded = NULL;
|
|---|
| 122 | #ifdef USE_LIBIDN2
|
|---|
| 123 | if(idn2_check_version(IDN2_VERSION)) {
|
|---|
| 124 | int flags = IDN2_NFC_INPUT
|
|---|
| 125 | #if IDN2_VERSION_NUMBER >= 0x00140000
|
|---|
| 126 | /* IDN2_NFC_INPUT: Normalize input string using normalization form C.
|
|---|
| 127 | IDN2_NONTRANSITIONAL: Perform Unicode TR46 non-transitional
|
|---|
| 128 | processing. */
|
|---|
| 129 | | IDN2_NONTRANSITIONAL
|
|---|
| 130 | #endif
|
|---|
| 131 | ;
|
|---|
| 132 | int rc = IDN2_LOOKUP(input, &decoded, flags);
|
|---|
| 133 | if(rc != IDN2_OK)
|
|---|
| 134 | /* fallback to TR46 Transitional mode for better IDNA2003
|
|---|
| 135 | compatibility */
|
|---|
| 136 | rc = IDN2_LOOKUP(input, &decoded, IDN2_TRANSITIONAL);
|
|---|
| 137 | if(rc != IDN2_OK)
|
|---|
| 138 | decoded = NULL;
|
|---|
| 139 | }
|
|---|
| 140 | #elif defined(USE_WIN32_IDN)
|
|---|
| 141 | if(!Curl_win32_idn_to_ascii(input, &decoded))
|
|---|
| 142 | decoded = NULL;
|
|---|
| 143 | #endif
|
|---|
| 144 | return decoded;
|
|---|
| 145 | }
|
|---|
| 146 |
|
|---|
| 147 | /*
|
|---|
| 148 | * Frees data allocated by idnconvert_hostname()
|
|---|
| 149 | */
|
|---|
| 150 | void Curl_free_idnconverted_hostname(struct hostname *host)
|
|---|
| 151 | {
|
|---|
| 152 | #if defined(USE_LIBIDN2)
|
|---|
| 153 | if(host->encalloc) {
|
|---|
| 154 | idn2_free(host->encalloc); /* must be freed with idn2_free() since this was
|
|---|
| 155 | allocated by libidn */
|
|---|
| 156 | host->encalloc = NULL;
|
|---|
| 157 | }
|
|---|
| 158 | #elif defined(USE_WIN32_IDN)
|
|---|
| 159 | free(host->encalloc); /* must be freed with free() since this was
|
|---|
| 160 | allocated by Curl_win32_idn_to_ascii */
|
|---|
| 161 | host->encalloc = NULL;
|
|---|
| 162 | #else
|
|---|
| 163 | (void)host;
|
|---|
| 164 | #endif
|
|---|
| 165 | }
|
|---|
| 166 |
|
|---|
| 167 | #endif /* USE_IDN */
|
|---|
| 168 |
|
|---|
| 169 | /*
|
|---|
| 170 | * Perform any necessary IDN conversion of hostname
|
|---|
| 171 | */
|
|---|
| 172 | CURLcode Curl_idnconvert_hostname(struct hostname *host)
|
|---|
| 173 | {
|
|---|
| 174 | /* set the name we use to display the host name */
|
|---|
| 175 | host->dispname = host->name;
|
|---|
| 176 |
|
|---|
| 177 | #ifdef USE_IDN
|
|---|
| 178 | /* Check name for non-ASCII and convert hostname if we can */
|
|---|
| 179 | if(!Curl_is_ASCII_name(host->name)) {
|
|---|
| 180 | char *decoded = Curl_idn_decode(host->name);
|
|---|
| 181 | if(decoded) {
|
|---|
| 182 | /* successful */
|
|---|
| 183 | host->encalloc = decoded;
|
|---|
| 184 | /* change the name pointer to point to the encoded hostname */
|
|---|
| 185 | host->name = host->encalloc;
|
|---|
| 186 | }
|
|---|
| 187 | else
|
|---|
| 188 | return CURLE_URL_MALFORMAT;
|
|---|
| 189 | }
|
|---|
| 190 | #endif
|
|---|
| 191 | return CURLE_OK;
|
|---|
| 192 | }
|
|---|
| 193 |
|
|---|