1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/uconv/ucvko/nsISO2022KRToUnicode.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,189 @@
1.4 +/* -*- Mode: C; tab-width: 4; 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 +#include "nsISO2022KRToUnicode.h"
1.9 +#include "nsUCSupport.h"
1.10 +#include "nsICharsetConverterManager.h"
1.11 +#include "nsServiceManagerUtils.h"
1.12 +
1.13 +NS_IMETHODIMP nsISO2022KRToUnicode::Convert(const char * aSrc, int32_t * aSrcLen, char16_t * aDest, int32_t * aDestLen)
1.14 +{
1.15 + static NS_DEFINE_CID(kCharsetConverterManagerCID, NS_ICHARSETCONVERTERMANAGER_CID);
1.16 +
1.17 + const unsigned char* srcEnd = (unsigned char*)aSrc + *aSrcLen;
1.18 + const unsigned char* src =(unsigned char*) aSrc;
1.19 + char16_t* destEnd = aDest + *aDestLen;
1.20 + char16_t* dest = aDest;
1.21 + while((src < srcEnd))
1.22 + {
1.23 + // if LF/CR, return to US-ASCII unconditionally.
1.24 + if ( *src == 0x0a || *src == 0x0d )
1.25 + mState = mState_Init;
1.26 +
1.27 + switch(mState)
1.28 + {
1.29 + case mState_Init:
1.30 + if(0x1b == *src) {
1.31 + mLastLegalState = mState_ASCII;
1.32 + mState = mState_ESC;
1.33 + break;
1.34 + }
1.35 + mState = mState_ASCII;
1.36 + // fall through
1.37 +
1.38 + case mState_ASCII:
1.39 + if(0x0e == *src) { // Shift-Out
1.40 + mState = mState_KSX1001_1992;
1.41 + mRunLength = 0;
1.42 + }
1.43 + else if(*src & 0x80) {
1.44 + if (CHECK_OVERRUN(dest, destEnd, 1))
1.45 + goto error1;
1.46 + *dest++ = 0xFFFD;
1.47 + }
1.48 + else {
1.49 + if (CHECK_OVERRUN(dest, destEnd, 1))
1.50 + goto error1;
1.51 + *dest++ = (char16_t) *src;
1.52 + }
1.53 + break;
1.54 +
1.55 + case mState_ESC:
1.56 + if('$' == *src) {
1.57 + mState = mState_ESC_24;
1.58 + }
1.59 + else {
1.60 + if (CHECK_OVERRUN(dest, destEnd, 2))
1.61 + goto error1;
1.62 + *dest++ = (char16_t) 0x1b;
1.63 + *dest++ = (0x80 & *src) ? 0xFFFD : (char16_t) *src;
1.64 + mState = mLastLegalState;
1.65 + }
1.66 + break;
1.67 +
1.68 + case mState_ESC_24: // ESC $
1.69 + if(')' == *src) {
1.70 + mState = mState_ESC_24_29;
1.71 + }
1.72 + else {
1.73 + if (CHECK_OVERRUN(dest, destEnd, 3))
1.74 + goto error1;
1.75 + *dest++ = (char16_t) 0x1b;
1.76 + *dest++ = (char16_t) '$';
1.77 + *dest++ = (0x80 & *src) ? 0xFFFD : (char16_t) *src;
1.78 + mState = mLastLegalState;
1.79 + }
1.80 + break;
1.81 +
1.82 + case mState_ESC_24_29: // ESC $ )
1.83 + mState = mLastLegalState;
1.84 + if('C' == *src) {
1.85 + mState = mState_ASCII;
1.86 + mRunLength = 0;
1.87 + }
1.88 + else {
1.89 + if (CHECK_OVERRUN(dest, destEnd, 4))
1.90 + goto error1;
1.91 + *dest++ = (char16_t) 0x1b;
1.92 + *dest++ = (char16_t) '$';
1.93 + *dest++ = (char16_t) ')';
1.94 + *dest++ = (0x80 & *src) ? 0xFFFD : (char16_t) *src;
1.95 + mState = mLastLegalState;
1.96 + }
1.97 + break;
1.98 +
1.99 + case mState_KSX1001_1992:
1.100 + if (0x20 < (uint8_t) *src && (uint8_t) *src < 0x7f) {
1.101 + mData = (uint8_t) *src;
1.102 + mState = mState_KSX1001_1992_2ndbyte;
1.103 + }
1.104 + else if (0x0f == *src) { // Shift-In (SI)
1.105 + mState = mState_ASCII;
1.106 + if (mRunLength == 0) {
1.107 + if (CHECK_OVERRUN(dest, destEnd, 1))
1.108 + goto error1;
1.109 + *dest++ = 0xFFFD;
1.110 + }
1.111 + mRunLength = 0;
1.112 + }
1.113 + else if ((uint8_t) *src == 0x20 || (uint8_t) *src == 0x09) {
1.114 + // Allow space and tab between SO and SI (i.e. in Hangul segment)
1.115 + if (CHECK_OVERRUN(dest, destEnd, 1))
1.116 + goto error1;
1.117 + mState = mState_KSX1001_1992;
1.118 + *dest++ = (char16_t) *src;
1.119 + ++mRunLength;
1.120 + }
1.121 + else { // Everything else is invalid.
1.122 + if (CHECK_OVERRUN(dest, destEnd, 1))
1.123 + goto error1;
1.124 + *dest++ = 0xFFFD;
1.125 + }
1.126 + break;
1.127 +
1.128 + case mState_KSX1001_1992_2ndbyte:
1.129 + if ( 0x20 < (uint8_t) *src && (uint8_t) *src < 0x7f ) {
1.130 + if (!mEUCKRDecoder) {
1.131 + // creating a delegate converter (EUC-KR)
1.132 + nsresult rv;
1.133 + nsCOMPtr<nsICharsetConverterManager> ccm =
1.134 + do_GetService(kCharsetConverterManagerCID, &rv);
1.135 + if (NS_SUCCEEDED(rv)) {
1.136 + rv = ccm->GetUnicodeDecoderRaw("EUC-KR", &mEUCKRDecoder);
1.137 + }
1.138 + }
1.139 +
1.140 + if (!mEUCKRDecoder) {// failed creating a delegate converter
1.141 + *dest++ = 0xFFFD;
1.142 + }
1.143 + else {
1.144 + if (CHECK_OVERRUN(dest, destEnd, 1))
1.145 + goto error1;
1.146 + unsigned char ksx[2];
1.147 + char16_t uni;
1.148 + int32_t ksxLen = 2, uniLen = 1;
1.149 + // mData is the original 1st byte.
1.150 + // *src is the present 2nd byte.
1.151 + // Put 2 bytes (one character) to ksx[] with EUC-KR encoding.
1.152 + ksx[0] = mData | 0x80;
1.153 + ksx[1] = *src | 0x80;
1.154 + // Convert EUC-KR to unicode.
1.155 + mEUCKRDecoder->Convert((const char *)ksx, &ksxLen, &uni, &uniLen);
1.156 + *dest++ = uni;
1.157 + ++mRunLength;
1.158 + }
1.159 + mState = mState_KSX1001_1992;
1.160 + }
1.161 + else { // Invalid
1.162 + if ( 0x0f == *src ) { // Shift-In (SI)
1.163 + mState = mState_ASCII;
1.164 + }
1.165 + else {
1.166 + mState = mState_KSX1001_1992;
1.167 + }
1.168 + if (CHECK_OVERRUN(dest, destEnd, 1))
1.169 + goto error1;
1.170 + *dest++ = 0xFFFD;
1.171 + }
1.172 + break;
1.173 +
1.174 + case mState_ERROR:
1.175 + mState = mLastLegalState;
1.176 + if (CHECK_OVERRUN(dest, destEnd, 1))
1.177 + goto error1;
1.178 + *dest++ = 0xFFFD;
1.179 + break;
1.180 +
1.181 + } // switch
1.182 + src++;
1.183 + }
1.184 + *aDestLen = dest - aDest;
1.185 + return NS_OK;
1.186 +
1.187 +error1:
1.188 + *aDestLen = dest-aDest;
1.189 + *aSrcLen = src-(unsigned char*)aSrc;
1.190 + return NS_OK_UDEC_MOREOUTPUT;
1.191 +}
1.192 +
