1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/uconv/src/nsUnicodeToUTF8.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,154 @@
1.4 +/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
1.5 +/* This Source Code Form is subject to the terms of the Mozilla Public
1.6 + * License, v. 2.0. If a copy of the MPL was not distributed with this
1.7 + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
1.8 +
1.9 +//----------------------------------------------------------------------
1.10 +// Global functions and data [declaration]
1.11 +#include "nsUnicodeToUTF8.h"
1.12 +
1.13 +NS_IMPL_ISUPPORTS(nsUnicodeToUTF8, nsIUnicodeEncoder)
1.14 +
1.15 +//----------------------------------------------------------------------
1.16 +// nsUnicodeToUTF8 class [implementation]
1.17 +
1.18 +NS_IMETHODIMP nsUnicodeToUTF8::GetMaxLength(const char16_t * aSrc,
1.19 + int32_t aSrcLength,
1.20 + int32_t * aDestLength)
1.21 +{
1.22 + // aSrc is interpreted as UTF16, 3 is normally enough.
1.23 + // But when previous buffer only contains part of the surrogate pair, we
1.24 + // need to complete it here. If the first word in following buffer is not
1.25 + // in valid surrogate range, we need to convert the remaining of last buffer
1.26 + // to 3 bytes.
1.27 + *aDestLength = 3*aSrcLength + 3;
1.28 + return NS_OK;
1.29 +}
1.30 +
1.31 +NS_IMETHODIMP nsUnicodeToUTF8::Convert(const char16_t * aSrc,
1.32 + int32_t * aSrcLength,
1.33 + char * aDest,
1.34 + int32_t * aDestLength)
1.35 +{
1.36 + const char16_t * src = aSrc;
1.37 + const char16_t * srcEnd = aSrc + *aSrcLength;
1.38 + char * dest = aDest;
1.39 + int32_t destLen = *aDestLength;
1.40 + uint32_t n;
1.41 +
1.42 + //complete remaining of last conversion
1.43 + if (mHighSurrogate) {
1.44 + if (src < srcEnd) {
1.45 + *aDestLength = 0;
1.46 + return NS_OK_UENC_MOREINPUT;
1.47 + }
1.48 + if (*aDestLength < 4) {
1.49 + *aSrcLength = 0;
1.50 + *aDestLength = 0;
1.51 + return NS_OK_UENC_MOREOUTPUT;
1.52 + }
1.53 + if (*src < (char16_t)0xdc00 || *src > (char16_t)0xdfff) { //not a pair
1.54 + *dest++ = (char)0xef; //replacement character
1.55 + *dest++ = (char)0xbf;
1.56 + *dest++ = (char)0xbd;
1.57 + destLen -= 3;
1.58 + } else {
1.59 + n = ((mHighSurrogate - (char16_t)0xd800) << 10) +
1.60 + (*src - (char16_t)0xdc00) + 0x10000;
1.61 + *dest++ = (char)0xf0 | (n >> 18);
1.62 + *dest++ = (char)0x80 | ((n >> 12) & 0x3f);
1.63 + *dest++ = (char)0x80 | ((n >> 6) & 0x3f);
1.64 + *dest++ = (char)0x80 | (n & 0x3f);
1.65 + ++src;
1.66 + destLen -= 4;
1.67 + }
1.68 + mHighSurrogate = 0;
1.69 + }
1.70 +
1.71 + while (src < srcEnd) {
1.72 + if ( *src <= 0x007f) {
1.73 + if (destLen < 1)
1.74 + goto error_more_output;
1.75 + *dest++ = (char)*src;
1.76 + --destLen;
1.77 + } else if (*src <= 0x07ff) {
1.78 + if (destLen < 2)
1.79 + goto error_more_output;
1.80 + *dest++ = (char)0xc0 | (*src >> 6);
1.81 + *dest++ = (char)0x80 | (*src & 0x003f);
1.82 + destLen -= 2;
1.83 + } else if (*src >= (char16_t)0xd800 && *src <= (char16_t)0xdfff) {
1.84 + if (*src >= (char16_t)0xdc00) { //not a pair
1.85 + if (destLen < 3)
1.86 + goto error_more_output;
1.87 + *dest++ = (char)0xef; //replacement character
1.88 + *dest++ = (char)0xbf;
1.89 + *dest++ = (char)0xbd;
1.90 + destLen -= 3;
1.91 + ++src;
1.92 + continue;
1.93 + }
1.94 + if ((src+1) >= srcEnd) {
1.95 + //we need another surrogate to complete this unicode char
1.96 + mHighSurrogate = *src;
1.97 + *aDestLength = dest - aDest;
1.98 + return NS_OK_UENC_MOREINPUT;
1.99 + }
1.100 + //handle surrogate
1.101 + if (destLen < 4)
1.102 + goto error_more_output;
1.103 + if (*(src+1) < (char16_t)0xdc00 || *(src+1) > 0xdfff) { //not a pair
1.104 + *dest++ = (char)0xef; //replacement character
1.105 + *dest++ = (char)0xbf;
1.106 + *dest++ = (char)0xbd;
1.107 + destLen -= 3;
1.108 + } else {
1.109 + n = ((*src - (char16_t)0xd800) << 10) + (*(src+1) - (char16_t)0xdc00) + (uint32_t)0x10000;
1.110 + *dest++ = (char)0xf0 | (n >> 18);
1.111 + *dest++ = (char)0x80 | ((n >> 12) & 0x3f);
1.112 + *dest++ = (char)0x80 | ((n >> 6) & 0x3f);
1.113 + *dest++ = (char)0x80 | (n & 0x3f);
1.114 + destLen -= 4;
1.115 + ++src;
1.116 + }
1.117 + } else {
1.118 + if (destLen < 3)
1.119 + goto error_more_output;
1.120 + //treat rest of the character as BMP
1.121 + *dest++ = (char)0xe0 | (*src >> 12);
1.122 + *dest++ = (char)0x80 | ((*src >> 6) & 0x003f);
1.123 + *dest++ = (char)0x80 | (*src & 0x003f);
1.124 + destLen -= 3;
1.125 + }
1.126 + ++src;
1.127 + }
1.128 +
1.129 + *aDestLength = dest - aDest;
1.130 + return NS_OK;
1.131 +
1.132 +error_more_output:
1.133 + *aSrcLength = src - aSrc;
1.134 + *aDestLength = dest - aDest;
1.135 + return NS_OK_UENC_MOREOUTPUT;
1.136 +}
1.137 +
1.138 +NS_IMETHODIMP nsUnicodeToUTF8::Finish(char * aDest, int32_t * aDestLength)
1.139 +{
1.140 + char * dest = aDest;
1.141 +
1.142 + if (mHighSurrogate) {
1.143 + if (*aDestLength < 3) {
1.144 + *aDestLength = 0;
1.145 + return NS_OK_UENC_MOREOUTPUT;
1.146 + }
1.147 + *dest++ = (char)0xef; //replacement character
1.148 + *dest++ = (char)0xbf;
1.149 + *dest++ = (char)0xbd;
1.150 + mHighSurrogate = 0;
1.151 + *aDestLength = 3;
1.152 + return NS_OK;
1.153 + }
1.154 +
1.155 + *aDestLength = 0;
1.156 + return NS_OK;
1.157 +}
