The Tor Browser / file comparison
comparison: intl/uconv/src/nsUnicodeToUTF8.cpp
intl/uconv/src/nsUnicodeToUTF8.cpp
- branch
- TOR_BUG_9701
- changeset 8
- 97036ab72558
equal
deleted
inserted
replaced
| -1:000000000000 | 0:16370bf68d92 |
|---|---|
| 1 /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */ | |
| 2 /* This Source Code Form is subject to the terms of the Mozilla Public | |
| 3 * License, v. 2.0. If a copy of the MPL was not distributed with this | |
| 4 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */ | |
| 5 | |
| 6 //---------------------------------------------------------------------- | |
| 7 // Global functions and data [declaration] | |
| 8 #include "nsUnicodeToUTF8.h" | |
| 9 | |
| 10 NS_IMPL_ISUPPORTS(nsUnicodeToUTF8, nsIUnicodeEncoder) | |
| 11 | |
| 12 //---------------------------------------------------------------------- | |
| 13 // nsUnicodeToUTF8 class [implementation] | |
| 14 | |
| 15 NS_IMETHODIMP nsUnicodeToUTF8::GetMaxLength(const char16_t * aSrc, | |
| 16 int32_t aSrcLength, | |
| 17 int32_t * aDestLength) | |
| 18 { | |
| 19 // aSrc is interpreted as UTF16, 3 is normally enough. | |
| 20 // But when previous buffer only contains part of the surrogate pair, we | |
| 21 // need to complete it here. If the first word in following buffer is not | |
| 22 // in valid surrogate range, we need to convert the remaining of last buffer | |
| 23 // to 3 bytes. | |
| 24 *aDestLength = 3*aSrcLength + 3; | |
| 25 return NS_OK; | |
| 26 } | |
| 27 | |
| 28 NS_IMETHODIMP nsUnicodeToUTF8::Convert(const char16_t * aSrc, | |
| 29 int32_t * aSrcLength, | |
| 30 char * aDest, | |
| 31 int32_t * aDestLength) | |
| 32 { | |
| 33 const char16_t * src = aSrc; | |
| 34 const char16_t * srcEnd = aSrc + *aSrcLength; | |
| 35 char * dest = aDest; | |
| 36 int32_t destLen = *aDestLength; | |
| 37 uint32_t n; | |
| 38 | |
| 39 //complete remaining of last conversion | |
| 40 if (mHighSurrogate) { | |
| 41 if (src < srcEnd) { | |
| 42 *aDestLength = 0; | |
| 43 return NS_OK_UENC_MOREINPUT; | |
| 44 } | |
| 45 if (*aDestLength < 4) { | |
| 46 *aSrcLength = 0; | |
| 47 *aDestLength = 0; | |
| 48 return NS_OK_UENC_MOREOUTPUT; | |
| 49 } | |
| 50 if (*src < (char16_t)0xdc00 || *src > (char16_t)0xdfff) { //not a pair | |
| 51 *dest++ = (char)0xef; //replacement character | |
| 52 *dest++ = (char)0xbf; | |
| 53 *dest++ = (char)0xbd; | |
| 54 destLen -= 3; | |
| 55 } else { | |
| 56 n = ((mHighSurrogate - (char16_t)0xd800) << 10) + | |
| 57 (*src - (char16_t)0xdc00) + 0x10000; | |
| 58 *dest++ = (char)0xf0 | (n >> 18); | |
| 59 *dest++ = (char)0x80 | ((n >> 12) & 0x3f); | |
| 60 *dest++ = (char)0x80 | ((n >> 6) & 0x3f); | |
| 61 *dest++ = (char)0x80 | (n & 0x3f); | |
| 62 ++src; | |
| 63 destLen -= 4; | |
| 64 } | |
| 65 mHighSurrogate = 0; | |
| 66 } | |
| 67 | |
| 68 while (src < srcEnd) { | |
| 69 if ( *src <= 0x007f) { | |
| 70 if (destLen < 1) | |
| 71 goto error_more_output; | |
| 72 *dest++ = (char)*src; | |
| 73 --destLen; | |
| 74 } else if (*src <= 0x07ff) { | |
| 75 if (destLen < 2) | |
| 76 goto error_more_output; | |
| 77 *dest++ = (char)0xc0 | (*src >> 6); | |
| 78 *dest++ = (char)0x80 | (*src & 0x003f); | |
| 79 destLen -= 2; | |
| 80 } else if (*src >= (char16_t)0xd800 && *src <= (char16_t)0xdfff) { | |
| 81 if (*src >= (char16_t)0xdc00) { //not a pair | |
| 82 if (destLen < 3) | |
| 83 goto error_more_output; | |
| 84 *dest++ = (char)0xef; //replacement character | |
| 85 *dest++ = (char)0xbf; | |
| 86 *dest++ = (char)0xbd; | |
| 87 destLen -= 3; | |
| 88 ++src; | |
| 89 continue; | |
| 90 } | |
| 91 if ((src+1) >= srcEnd) { | |
| 92 //we need another surrogate to complete this unicode char | |
| 93 mHighSurrogate = *src; | |
| 94 *aDestLength = dest - aDest; | |
| 95 return NS_OK_UENC_MOREINPUT; | |
| 96 } | |
| 97 //handle surrogate | |
| 98 if (destLen < 4) | |
| 99 goto error_more_output; | |
| 100 if (*(src+1) < (char16_t)0xdc00 || *(src+1) > 0xdfff) { //not a pair | |
| 101 *dest++ = (char)0xef; //replacement character | |
| 102 *dest++ = (char)0xbf; | |
| 103 *dest++ = (char)0xbd; | |
| 104 destLen -= 3; | |
| 105 } else { | |
| 106 n = ((*src - (char16_t)0xd800) << 10) + (*(src+1) - (char16_t)0xdc00) + (uint32_t)0x10000; | |
| 107 *dest++ = (char)0xf0 | (n >> 18); | |
| 108 *dest++ = (char)0x80 | ((n >> 12) & 0x3f); | |
| 109 *dest++ = (char)0x80 | ((n >> 6) & 0x3f); | |
| 110 *dest++ = (char)0x80 | (n & 0x3f); | |
| 111 destLen -= 4; | |
| 112 ++src; | |
| 113 } | |
| 114 } else { | |
| 115 if (destLen < 3) | |
| 116 goto error_more_output; | |
| 117 //treat rest of the character as BMP | |
| 118 *dest++ = (char)0xe0 | (*src >> 12); | |
| 119 *dest++ = (char)0x80 | ((*src >> 6) & 0x003f); | |
| 120 *dest++ = (char)0x80 | (*src & 0x003f); | |
| 121 destLen -= 3; | |
| 122 } | |
| 123 ++src; | |
| 124 } | |
| 125 | |
| 126 *aDestLength = dest - aDest; | |
| 127 return NS_OK; | |
| 128 | |
| 129 error_more_output: | |
| 130 *aSrcLength = src - aSrc; | |
| 131 *aDestLength = dest - aDest; | |
| 132 return NS_OK_UENC_MOREOUTPUT; | |
| 133 } | |
| 134 | |
| 135 NS_IMETHODIMP nsUnicodeToUTF8::Finish(char * aDest, int32_t * aDestLength) | |
| 136 { | |
| 137 char * dest = aDest; | |
| 138 | |
| 139 if (mHighSurrogate) { | |
| 140 if (*aDestLength < 3) { | |
| 141 *aDestLength = 0; | |
| 142 return NS_OK_UENC_MOREOUTPUT; | |
| 143 } | |
| 144 *dest++ = (char)0xef; //replacement character | |
| 145 *dest++ = (char)0xbf; | |
| 146 *dest++ = (char)0xbd; | |
| 147 mHighSurrogate = 0; | |
| 148 *aDestLength = 3; | |
| 149 return NS_OK; | |
| 150 } | |
| 151 | |
| 152 *aDestLength = 0; | |
| 153 return NS_OK; | |
| 154 } |
