From 1bf5a906655ababdf843ddc0dd0f7e7bcd1a6ce0 Mon Sep 17 00:00:00 2001 From: Markus Scherer Date: Fri, 29 Oct 1999 20:55:42 +0000 Subject: [PATCH] ICU-86 new converter names UTF16_PlatformEndian & UTF16_OppositeEndian to have correct convrtrs.txt on all platforms X-SVN-Rev: 145 --- icu4c/data/convrtrs.txt | 7 +++---- icu4c/source/common/ucnv_bld.c | 16 +++++++++++++++- icu4c/source/common/ucnv_cnv.c | 12 ++++++------ icu4c/source/data/mappings/convrtrs.txt | 7 +++---- 4 files changed, 27 insertions(+), 15 deletions(-) diff --git a/icu4c/data/convrtrs.txt b/icu4c/data/convrtrs.txt index 4e1dc49fad2..bf91bca87da 100644 --- a/icu4c/data/convrtrs.txt +++ b/icu4c/data/convrtrs.txt @@ -5,11 +5,10 @@ # Algorithmic: DO NOT CHANGE LEFT COLUMN OF ALGORITHMIC CONVERTERS UTF8 utf8 utf-8 ibm-1208 utf_8 ibm1208 cp1208 -# If ICU is run on the intel platforms, NT, Linux etc., be sure to update the UTF16 entry with the following two lines: -# UTF16_BigEndian utf-16le utf-16 ibm-1200 ibm1200 cp1200 iso-10646-ucs-2 ucs-2 -# UTF16_LittleEndian utf-16be -UTF16_BigEndian utf-16be utf-16 ibm-1200 ibm1200 cp1200 iso-10646-ucs-2 ucs-2 +UTF16_BigEndian utf-16be UTF16_LittleEndian utf-16le +UTF16_PlatformEndian utf16 utf-16 ibm-1200 ibm1200 cp1200 iso-10646-ucs-2 ucs-2 +UTF16_OppositeEndian LATIN_1 iso-8859-1 iso_8859-1 ibm-819 ibm819 cp819 latin1 latin-1 ascii ascii-7 us-ascii 8859-1 csisolatin1 iso-ir-100 iso_8859-1:1978 ISO_2022 iso-2022 2022 cp2022 iso2022 iso_2022 diff --git a/icu4c/source/common/ucnv_bld.c b/icu4c/source/common/ucnv_bld.c index 8e8b911fc31..14c2740e50c 100644 --- a/icu4c/source/common/ucnv_bld.c +++ b/icu4c/source/common/ucnv_bld.c @@ -40,11 +40,12 @@ static const char *algorithmicConverterNames[] = { "UTF8", "UTF16_BigEndian", "UTF16_LittleEndian", + "UTF16_PlatformEndian", + "UTF16_OppositeEndian", "ISO_2022", "JIS", "EUC", "GB", - "ISO_2022", "" }; @@ -373,6 +374,18 @@ UConverterType return UCNV_GB; else if (icu_strcmp (realName, "ISO_2022") == 0) return UCNV_ISO_2022; + else if (icu_strcmp (realName, "UTF16_PlatformEndian") == 0) +# if U_IS_BIG_ENDIAN + return UCNV_UTF16_BigEndian; +# else + return UCNV_UTF16_LittleEndian; +# endif + else if (icu_strcmp (realName, "UTF16_OppositeEndian") == 0) +# if U_IS_BIG_ENDIAN + return UCNV_UTF16_LittleEndian; +# else + return UCNV_UTF16_BigEndian; +# endif else return UCNV_UNSUPPORTED_CONVERTER; } @@ -767,6 +780,7 @@ void break; } + case UCNV_UTF16_LittleEndian: { myConverter->sharedData->minBytesPerChar = 2; diff --git a/icu4c/source/common/ucnv_cnv.c b/icu4c/source/common/ucnv_cnv.c index 64e717a4e45..ff38ad20dd9 100644 --- a/icu4c/source/common/ucnv_cnv.c +++ b/icu4c/source/common/ucnv_cnv.c @@ -207,8 +207,8 @@ const char* escSeqStateTable_Result_2022[MAX_STATES_2022] = { NULL ,NULL ,NULL ,NULL ,"latin1" ,"latin1" ,"latin1" ,"ibm-865" ,"ibm-865" ,"ibm-865" ,"ibm-865" ,"ibm-865" ,"ibm-865" ,"ibm-895" ,"ibm-943" ,"latin1" ,"latin1" ,NULL ,"ibm-955" ,"ibm-367" ,"ibm-952" ,NULL ,"UTF8" ,NULL ,"ibm-955" ,"bm-367" ,"ibm-952" ,"ibm-949" ,"ibm-953" ,"ibm-1383" - ,"ibm-952" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"UTF16" ,"UTF16" ,"UTF16" - ,"UTF16" ,"UTF16" ,"UTF16" ,NULL ,"latin1" ,"ibm-912" ,"ibm-913" ,"ibm-914" ,"ibm-813" ,"ibm-1089" + ,"ibm-952" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"ibm-964" ,"UTF16_PlatformEndian" ,"UTF16_PlatformEndian" ,"UTF16_PlatformEndian" + ,"UTF16_PlatformEndian" ,"UTF16_PlatformEndian" ,"UTF16_PlatformEndian" ,NULL ,"latin1" ,"ibm-912" ,"ibm-913" ,"ibm-914" ,"ibm-813" ,"ibm-1089" ,"ibm-920" ,"ibm-915" ,"ibm-915" ,"latin1"}; UCNV_TableStates_2022 escSeqStateTable_Value_2022[MAX_STATES_2022] = { @@ -2110,8 +2110,8 @@ void T_UConverter_fromUnicode_UTF8 (UConverter * _this, int8_t targetCharByteNum = 0; UChar mySourceChar = 0x0000; uint32_t ch; - int16_t bytesToWrite = 0; - uint32_t ch2, i; + int16_t i, bytesToWrite = 0; + uint32_t ch2; char temp[4]; if (_this->fromUnicodeStatus) @@ -2222,8 +2222,8 @@ void T_UConverter_fromUnicode_UTF8_OFFSETS_LOGIC (UConverter * _this, int8_t targetCharByteNum = 0; UChar mySourceChar = 0x0000; uint32_t ch; - int16_t bytesToWrite = 0; - uint32_t ch2, i; + int16_t i, bytesToWrite = 0; + uint32_t ch2; char temp[4]; if (_this->fromUnicodeStatus) diff --git a/icu4c/source/data/mappings/convrtrs.txt b/icu4c/source/data/mappings/convrtrs.txt index 4e1dc49fad2..bf91bca87da 100644 --- a/icu4c/source/data/mappings/convrtrs.txt +++ b/icu4c/source/data/mappings/convrtrs.txt @@ -5,11 +5,10 @@ # Algorithmic: DO NOT CHANGE LEFT COLUMN OF ALGORITHMIC CONVERTERS UTF8 utf8 utf-8 ibm-1208 utf_8 ibm1208 cp1208 -# If ICU is run on the intel platforms, NT, Linux etc., be sure to update the UTF16 entry with the following two lines: -# UTF16_BigEndian utf-16le utf-16 ibm-1200 ibm1200 cp1200 iso-10646-ucs-2 ucs-2 -# UTF16_LittleEndian utf-16be -UTF16_BigEndian utf-16be utf-16 ibm-1200 ibm1200 cp1200 iso-10646-ucs-2 ucs-2 +UTF16_BigEndian utf-16be UTF16_LittleEndian utf-16le +UTF16_PlatformEndian utf16 utf-16 ibm-1200 ibm1200 cp1200 iso-10646-ucs-2 ucs-2 +UTF16_OppositeEndian LATIN_1 iso-8859-1 iso_8859-1 ibm-819 ibm819 cp819 latin1 latin-1 ascii ascii-7 us-ascii 8859-1 csisolatin1 iso-ir-100 iso_8859-1:1978 ISO_2022 iso-2022 2022 cp2022 iso2022 iso_2022