intl/uconv/src/nsUnicodeToUTF8.cpp@b8a032363ba2 (annotated)
intl/uconv/src/nsUnicodeToUTF8.cpp
Thu, 22 Jan 2015 13:21:57 +0100
- author
- Michael Schloh von Bennewitz <michael@schloh.com>
- date
- Thu, 22 Jan 2015 13:21:57 +0100
- branch
- TOR_BUG_9701
- changeset 15
- b8a032363ba2
- permissions
- -rw-r--r--
Incorporate requested changes from Mozilla in review:
https://bugzilla.mozilla.org/show_bug.cgi?id=1123480#c6
| michael@0 | 1 | /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */ |
| michael@0 | 2 | /* This Source Code Form is subject to the terms of the Mozilla Public |
| michael@0 | 3 | * License, v. 2.0. If a copy of the MPL was not distributed with this |
| michael@0 | 4 | * file, You can obtain one at http://mozilla.org/MPL/2.0/. */ |
| michael@0 | 5 | |
| michael@0 | 6 | //---------------------------------------------------------------------- |
| michael@0 | 7 | // Global functions and data [declaration] |
| michael@0 | 8 | #include "nsUnicodeToUTF8.h" |
| michael@0 | 9 | |
| michael@0 | 10 | NS_IMPL_ISUPPORTS(nsUnicodeToUTF8, nsIUnicodeEncoder) |
| michael@0 | 11 | |
| michael@0 | 12 | //---------------------------------------------------------------------- |
| michael@0 | 13 | // nsUnicodeToUTF8 class [implementation] |
| michael@0 | 14 | |
| michael@0 | 15 | NS_IMETHODIMP nsUnicodeToUTF8::GetMaxLength(const char16_t * aSrc, |
| michael@0 | 16 | int32_t aSrcLength, |
| michael@0 | 17 | int32_t * aDestLength) |
| michael@0 | 18 | { |
| michael@0 | 19 | // aSrc is interpreted as UTF16, 3 is normally enough. |
| michael@0 | 20 | // But when previous buffer only contains part of the surrogate pair, we |
| michael@0 | 21 | // need to complete it here. If the first word in following buffer is not |
| michael@0 | 22 | // in valid surrogate range, we need to convert the remaining of last buffer |
| michael@0 | 23 | // to 3 bytes. |
| michael@0 | 24 | *aDestLength = 3*aSrcLength + 3; |
| michael@0 | 25 | return NS_OK; |
| michael@0 | 26 | } |
| michael@0 | 27 | |
| michael@0 | 28 | NS_IMETHODIMP nsUnicodeToUTF8::Convert(const char16_t * aSrc, |
| michael@0 | 29 | int32_t * aSrcLength, |
| michael@0 | 30 | char * aDest, |
| michael@0 | 31 | int32_t * aDestLength) |
| michael@0 | 32 | { |
| michael@0 | 33 | const char16_t * src = aSrc; |
| michael@0 | 34 | const char16_t * srcEnd = aSrc + *aSrcLength; |
| michael@0 | 35 | char * dest = aDest; |
| michael@0 | 36 | int32_t destLen = *aDestLength; |
| michael@0 | 37 | uint32_t n; |
| michael@0 | 38 | |
| michael@0 | 39 | //complete remaining of last conversion |
| michael@0 | 40 | if (mHighSurrogate) { |
| michael@0 | 41 | if (src < srcEnd) { |
| michael@0 | 42 | *aDestLength = 0; |
| michael@0 | 43 | return NS_OK_UENC_MOREINPUT; |
| michael@0 | 44 | } |
| michael@0 | 45 | if (*aDestLength < 4) { |
| michael@0 | 46 | *aSrcLength = 0; |
| michael@0 | 47 | *aDestLength = 0; |
| michael@0 | 48 | return NS_OK_UENC_MOREOUTPUT; |
| michael@0 | 49 | } |
| michael@0 | 50 | if (*src < (char16_t)0xdc00 || *src > (char16_t)0xdfff) { //not a pair |
| michael@0 | 51 | *dest++ = (char)0xef; //replacement character |
| michael@0 | 52 | *dest++ = (char)0xbf; |
| michael@0 | 53 | *dest++ = (char)0xbd; |
| michael@0 | 54 | destLen -= 3; |
| michael@0 | 55 | } else { |
| michael@0 | 56 | n = ((mHighSurrogate - (char16_t)0xd800) << 10) + |
| michael@0 | 57 | (*src - (char16_t)0xdc00) + 0x10000; |
| michael@0 | 58 | *dest++ = (char)0xf0 | (n >> 18); |
| michael@0 | 59 | *dest++ = (char)0x80 | ((n >> 12) & 0x3f); |
| michael@0 | 60 | *dest++ = (char)0x80 | ((n >> 6) & 0x3f); |
| michael@0 | 61 | *dest++ = (char)0x80 | (n & 0x3f); |
| michael@0 | 62 | ++src; |
| michael@0 | 63 | destLen -= 4; |
| michael@0 | 64 | } |
| michael@0 | 65 | mHighSurrogate = 0; |
| michael@0 | 66 | } |
| michael@0 | 67 | |
| michael@0 | 68 | while (src < srcEnd) { |
| michael@0 | 69 | if ( *src <= 0x007f) { |
| michael@0 | 70 | if (destLen < 1) |
| michael@0 | 71 | goto error_more_output; |
| michael@0 | 72 | *dest++ = (char)*src; |
| michael@0 | 73 | --destLen; |
| michael@0 | 74 | } else if (*src <= 0x07ff) { |
| michael@0 | 75 | if (destLen < 2) |
| michael@0 | 76 | goto error_more_output; |
| michael@0 | 77 | *dest++ = (char)0xc0 | (*src >> 6); |
| michael@0 | 78 | *dest++ = (char)0x80 | (*src & 0x003f); |
| michael@0 | 79 | destLen -= 2; |
| michael@0 | 80 | } else if (*src >= (char16_t)0xd800 && *src <= (char16_t)0xdfff) { |
| michael@0 | 81 | if (*src >= (char16_t)0xdc00) { //not a pair |
| michael@0 | 82 | if (destLen < 3) |
| michael@0 | 83 | goto error_more_output; |
| michael@0 | 84 | *dest++ = (char)0xef; //replacement character |
| michael@0 | 85 | *dest++ = (char)0xbf; |
| michael@0 | 86 | *dest++ = (char)0xbd; |
| michael@0 | 87 | destLen -= 3; |
| michael@0 | 88 | ++src; |
| michael@0 | 89 | continue; |
| michael@0 | 90 | } |
| michael@0 | 91 | if ((src+1) >= srcEnd) { |
| michael@0 | 92 | //we need another surrogate to complete this unicode char |
| michael@0 | 93 | mHighSurrogate = *src; |
| michael@0 | 94 | *aDestLength = dest - aDest; |
| michael@0 | 95 | return NS_OK_UENC_MOREINPUT; |
| michael@0 | 96 | } |
| michael@0 | 97 | //handle surrogate |
| michael@0 | 98 | if (destLen < 4) |
| michael@0 | 99 | goto error_more_output; |
| michael@0 | 100 | if (*(src+1) < (char16_t)0xdc00 || *(src+1) > 0xdfff) { //not a pair |
| michael@0 | 101 | *dest++ = (char)0xef; //replacement character |
| michael@0 | 102 | *dest++ = (char)0xbf; |
| michael@0 | 103 | *dest++ = (char)0xbd; |
| michael@0 | 104 | destLen -= 3; |
| michael@0 | 105 | } else { |
| michael@0 | 106 | n = ((*src - (char16_t)0xd800) << 10) + (*(src+1) - (char16_t)0xdc00) + (uint32_t)0x10000; |
| michael@0 | 107 | *dest++ = (char)0xf0 | (n >> 18); |
| michael@0 | 108 | *dest++ = (char)0x80 | ((n >> 12) & 0x3f); |
| michael@0 | 109 | *dest++ = (char)0x80 | ((n >> 6) & 0x3f); |
| michael@0 | 110 | *dest++ = (char)0x80 | (n & 0x3f); |
| michael@0 | 111 | destLen -= 4; |
| michael@0 | 112 | ++src; |
| michael@0 | 113 | } |
| michael@0 | 114 | } else { |
| michael@0 | 115 | if (destLen < 3) |
| michael@0 | 116 | goto error_more_output; |
| michael@0 | 117 | //treat rest of the character as BMP |
| michael@0 | 118 | *dest++ = (char)0xe0 | (*src >> 12); |
| michael@0 | 119 | *dest++ = (char)0x80 | ((*src >> 6) & 0x003f); |
| michael@0 | 120 | *dest++ = (char)0x80 | (*src & 0x003f); |
| michael@0 | 121 | destLen -= 3; |
| michael@0 | 122 | } |
| michael@0 | 123 | ++src; |
| michael@0 | 124 | } |
| michael@0 | 125 | |
| michael@0 | 126 | *aDestLength = dest - aDest; |
| michael@0 | 127 | return NS_OK; |
| michael@0 | 128 | |
| michael@0 | 129 | error_more_output: |
| michael@0 | 130 | *aSrcLength = src - aSrc; |
| michael@0 | 131 | *aDestLength = dest - aDest; |
| michael@0 | 132 | return NS_OK_UENC_MOREOUTPUT; |
| michael@0 | 133 | } |
| michael@0 | 134 | |
| michael@0 | 135 | NS_IMETHODIMP nsUnicodeToUTF8::Finish(char * aDest, int32_t * aDestLength) |
| michael@0 | 136 | { |
| michael@0 | 137 | char * dest = aDest; |
| michael@0 | 138 | |
| michael@0 | 139 | if (mHighSurrogate) { |
| michael@0 | 140 | if (*aDestLength < 3) { |
| michael@0 | 141 | *aDestLength = 0; |
| michael@0 | 142 | return NS_OK_UENC_MOREOUTPUT; |
| michael@0 | 143 | } |
| michael@0 | 144 | *dest++ = (char)0xef; //replacement character |
| michael@0 | 145 | *dest++ = (char)0xbf; |
| michael@0 | 146 | *dest++ = (char)0xbd; |
| michael@0 | 147 | mHighSurrogate = 0; |
| michael@0 | 148 | *aDestLength = 3; |
| michael@0 | 149 | return NS_OK; |
| michael@0 | 150 | } |
| michael@0 | 151 | |
| michael@0 | 152 | *aDestLength = 0; |
| michael@0 | 153 | return NS_OK; |
| michael@0 | 154 | } |
