Tera TermRevision: 7462
http://sourceforge.jp/projects/ttssh2/scm/svn/commits/7462
Author: zmatsuo
Date: 2019-03-10 02:32:42 +0900 (Sun, 10 Mar 2019)
Log Message:
-----------
unicode関連をcodeconv.cに集めた
CP932へ/からの変換を関数に分離(UTF32ToCP932(), UTF32ToCP932())
unicodeからDEC特殊文字変換を関数に分離(UTF32ToDecSp())
SJIS2UTF8() (内部コードからUTF-8へ出力)を language.c から ttcmn.c に移動
_WideCharToMultiByte(), _MultiByteToWideChar() の変換した文字数の戻り値修正
UTF32ToMBCP()追加
UTF32_CP932()追加
MBCPToUTF32()追加
WideCharToUTF8(), WideCharToCP932() の仕様を変更
変換テーブルとWindows APIをつかったコード変換の優先順位を変更できるようにした
Modified Paths:
--------------
trunk/teraterm/common/codeconv.cpp
trunk/teraterm/common/codeconv.h
trunk/teraterm/teraterm/CMakeLists.txt
trunk/teraterm/teraterm/vtterm.c
trunk/teraterm/ttpcmn/language.c
trunk/teraterm/ttpcmn/language.h
trunk/teraterm/ttpcmn/ttcmn.c
-------------- next part --------------
Modified: trunk/teraterm/common/codeconv.cpp
===================================================================
--- trunk/teraterm/common/codeconv.cpp 2019-03-05 16:22:05 UTC (rev 7461)
+++ trunk/teraterm/common/codeconv.cpp 2019-03-09 17:32:42 UTC (rev 7462)
@@ -26,14 +26,21 @@
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
+/* unicode\x8A֘A\x82̕\xB6\x8E\x9A\x83R\x81[\x83h\x95ϊ\xB7 */
+
#include <windows.h>
#include <string.h>
+#include <assert.h>
#include <crtdbg.h>
#if (defined(_MSC_VER) && (_MSC_VER >= 1600)) || !defined(_MSC_VER)
#include <stdint.h>
#endif
+#include "codemap.h"
#include "codeconv.h"
+// cp932\x95ϊ\xB7\x8E\x9E\x81AWindows API \x82\xE6\x82\xE8 Tera Term \x82̕ϊ\xB7\x83e\x81[\x83u\x83\x8B\x82\xF0\x97D\x90悷\x82\xE9
+//#define PRIORITY_CP932_TABLE
+
#if defined(_MSC_VER) && (_MSC_VER < 1600)
typedef unsigned char uint8_t;
typedef unsigned short uint16_t;
@@ -47,12 +54,158 @@
#define _wcsdup(s) _wcsdup_dbg((s), _NORMAL_BLOCK, __FILE__, __LINE__)
#endif
+/*
+ * \x8C\xA9\x82\xA9\x82\xE7\x82Ȃ\xA2\x8Fꍇ\x82\xCD 0 \x82\xF0\x95Ԃ\xB7
+ */
+static unsigned short _ConvertUnicode(unsigned short code, const codemap_t *table, int tmax)
+{
+ int low, mid, high;
+ unsigned short result;
+
+ low = 0;
+ high = tmax - 1;
+ result = 0; // convert error
+
+ // binary search
+ while (low < high) {
+ mid = (low + high) / 2;
+ if (table[mid].from_code < code) {
+ low = mid + 1;
+ } else {
+ high = mid;
+ }
+ }
+
+ if (table[low].from_code == code) {
+ result = table[low].to_code;
+ }
+
+ return (result);
+}
+
+static int IsHighSurrogate(wchar_t u16)
+{
+ return 0xd800 <= u16 && u16 < 0xdc00;
+}
+
+static int IsLowSurrogate(wchar_t u16)
+{
+ return 0xdc00 <= u16 && u16 < 0xe000;
+}
+
/**
- * UTF-32 \x82\xA9\x82\xE7 UTF-8 \x82֕ϊ\xB7\x82\xB7\x82\xE9
+ * 1\x95\xB6\x8E\x9A\x82Ƃ\xB5\x82Ĉ\xB5\x82\xA4\x82\xBD\x82߂ɁA\x95K\x97v\x82ȃL\x83\x83\x83\x89\x83N\x83^\x90\x94\x82\xE9
+ * @retval 0 \x95\xB6\x8E\x9A\x82Ƃ\xB5\x82Ĉ\xB5\x82\xA6\x82Ȃ\xA2(\x95\xB6\x8E\x9A\x83R\x81[\x83h\x82\xAA\x82\xA8\x82\xA9\x82\xB5\x82\xA2)
+ * @retval 1 1\x83L\x83\x83\x83\x89\x83N\x83^\x82\xC51\x95\xB6\x8E\x9A\x82Ƃ\xB5\x82Ĉ\xB5\x82\xA6\x82\xE9
+ * @retval 2 2\x83L\x83\x83\x83\x89\x83N\x83^\x82\xC51\x95\xB6\x8E\x9A\x82Ƃ\xB5\x82Ĉ\xB5\x82\xA6\x82\xE9
+ */
+#if 0
+static size_t UTF16GetCharCount(const wchar_t *wstr_ptr, size_t wstr_len)
+{
+ wchar_t u16;
+ assert(wstr_ptr != NULL);
+ if (wstr_len == 0) {
+ return 0;
+ }
+ u16 = *wstr_ptr++;
+ if (IsHighSurrogate(u16)) {
+ if (wstr_len >= 2) {
+ const wchar_t u16_lo = *wstr_ptr++;
+ if (IsLowSurrogate(u16_lo)) {
+ return 2;
+ } else {
+ return 0;
+ }
+ } else {
+ return 0;
+ }
+ } else if (IsLowSurrogate(u16)) {
+ return 0;
+ }
+ return 1;
+}
+#endif
+
+/**
+ * code page \x82\xCC mulit byte \x95\xB6\x8E\x9A\x82\xF0 UTF-32\x82֕ϊ\xB7\x82\xB7\x82\xE9
+ * @param KCode \x83}\x83\x8B\x83`\x83o\x83C\x83g\x82̕\xB6\x8E\x9A\x83R\x81[\x83h(0x0000-0xffff)
+ * @param CoePage \x83}\x83\x8B\x83`\x83o\x83C\x83g\x82̃R\x81[\x83h\x83y\x81[\x83W
+ * @retval unicode(UTF-32\x95\xB6\x8E\x9A\x83R\x81[\x83h)
+ */
+unsigned int MBCPToUTF32(unsigned short KCode, int CodePage)
+{
+ unsigned int c;
+
+ if (CodePage == 932) {
+ c = CP932ToUTF32(KCode);
+ } else {
+ char buf[3];
+ wchar_t wchar;
+ int ret;
+ int len = 0;
+ if (KCode < 0x100) {
+ buf[0] = KCode & 0xff;
+ len = 1;
+ } else {
+ buf[0] = KCode >> 8;
+ buf[1] = KCode & 0xff;
+ len = 2;
+ }
+ ret = MultiByteToWideChar(CodePage, MB_ERR_INVALID_CHARS, buf, len, &wchar, 1);
+ if (ret <= 0) {
+ c = 0;
+ } else {
+ c = (unsigned int)wchar;
+ }
+ }
+ return c;
+}
+
+/**
+ * wchar_t\x95\xB6\x8E\x9A\x97\xE7unicode(UTF-32)\x82\xF01\x95\xB6\x8E\x9A\x8E\xE6\x82\xE8\x8Fo\x82\xB7
+ * @retval 0 \x95\xB6\x8E\x9A\x82Ƃ\xB5\x82Ĉ\xB5\x82\xA6\x82Ȃ\xA2(\x95\xB6\x8E\x9A\x83R\x81[\x83h\x82\xAA\x82\xA8\x82\xA9\x82\xB5\x82\xA2)
+ * @retval 1 1\x83L\x83\x83\x83\x89\x83N\x83^\x82\xC51\x95\xB6\x8E\x9A\x82Ƃ\xB5\x82Ĉ\xB5\x82\xA6\x82\xE9
+ * @retval 2 2\x83L\x83\x83\x83\x89\x83N\x83^\x82\xC51\x95\xB6\x8E\x9A\x82Ƃ\xB5\x82Ĉ\xB5\x82\xA6\x82\xE9
+ */
+size_t UTF16ToUTF32(const wchar_t *wstr_ptr, size_t wstr_len, unsigned int *u32)
+{
+ assert(wstr_ptr != NULL);
+ if (wstr_len == 0) {
+ *u32 = 0;
+ return 0;
+ }
+ const wchar_t u16 = *wstr_ptr++;
+ // \x83T\x83\x8D\x83Q\x81[\x83g high?
+ if (IsHighSurrogate(u16)) {
+ if (wstr_len >= 2) {
+ const wchar_t u16_lo = *wstr_ptr++;
+ if (IsLowSurrogate(u16_lo)) {
+ // \x83T\x83\x8D\x83Q\x81[\x83g\x83y\x83A \x83f\x83R\x81[\x83h
+ *u32 = 0x10000 + (u16 - 0xd800) * 0x400 + (u16_lo - 0xdc00);
+ return 2;
+ } else {
+ *u32 = 0;
+ return 0;
+ }
+ } else {
+ *u32 = 0;
+ return 0;
+ }
+ } else if (IsLowSurrogate(u16)) {
+ *u32 = 0;
+ return 0;
+ } else {
+ *u32 = u16;
+ return 1;
+ }
+}
+
+/**
+ * UTF-32\x95\xB6\x8E\x9A \x82\xA9\x82\xE7 UTF-8 \x82֕ϊ\xB7\x82\xB7\x82\xE9
* @param[in] u32 \x95ϊ\xB7\x82\xB7\x82\xE9UTF-32
* @param[in,out] u8_ptr \x95ϊ\xB7\x8C\xE3UTF-8\x95\xB6\x8E\x9A\x97\xF1\x8Fo\x97͐\xE6(NULL\x82̂Ƃ\xAB\x8Fo\x97͂\xB5\x82Ȃ\xA2)
* @param[in] u8_len UTF-8\x8Fo\x97͐敶\x8E\x9A\x90\x94(\x83o\x83b\x83t\x83@\x92\xB7,byte\x90\x94)
- * @retval \x8Eg\x97p\x82\xB5\x82\xBDutf8\x95\xB6\x8E\x9A\x90\x94(byte\x90\x94\x81j
+ * @retval \x8Fo\x97͂\xB5\x82\xBDutf8\x95\xB6\x8E\x9A\x90\x94(byte\x90\x94\x81j
* 0=\x83G\x83\x89\x81[
*/
size_t UTF32ToUTF8(uint32_t u32, char *u8_ptr_, size_t u8_len)
@@ -113,7 +266,7 @@
* @param[in] u8_len UTF-8\x95\xB6\x8E\x9A\x97\xB3
* @param[out] u32 \x95ϊ\xB7\x82\xB5\x82\xBDUTF-32\x95\xB6\x8E\x9A
* @retval \x8Eg\x97p\x82\xB5\x82\xBDUTF-8\x95\xB6\x8E\x9A\x90\x94(byte\x90\x94\x81j
- * 0=\x83G\x83\x89\x81[
+ * 0=\x83G\x83\x89\x81[(\x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD)
*/
size_t UTF8ToUTF32(const char *u8_ptr_, size_t u8_len, uint32_t *u32_)
{
@@ -191,33 +344,286 @@
return u8_in;
}
-// WideCharToMultiByte\x82\xCCUTF8\x93\xC1\x89\xBB\x94\xC5
-int WideCharToUTF8(const wchar_t *wstr_ptr, int wstr_len, char *u8_ptr, int u8_len)
+/**
+ * UTF-32 \x82\xA9\x82\xE7 UTF-16 \x82֕ϊ\xB7\x82\xB7\x82\xE9
+ * @param[in] u32 \x95ϊ\xB7\x82\xB7\x82\xE9UTF-32
+ * @param[in,out] wstr_ptr \x95ϊ\xB7\x8C\xE3UTF-16\x95\xB6\x8E\x9A\x97\xF1\x8Fo\x97͐\xE6(NULL\x82̂Ƃ\xAB\x8Fo\x97͂\xB5\x82Ȃ\xA2)
+ * @param[in] wstr_len UTF-16\x8Fo\x97͐敶\x8E\x9A\x90\x94(\x95\xB6\x8E\x9A\x90\x94,sizeof(wchar_t)*wstr_len bytes)
+ * @retval \x8Fo\x97͂\xB5\x82\xBDUTF-16\x95\xB6\x8E\x9A\x90\x94(sizeof(wchar_t)\x94{\x82\xB7\x82\xE9\x82\xC6byte\x90\x94)
+ * 0=\x83G\x83\x89\x81[(\x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD)
+ */
+size_t UTF32ToUTF16(uint32_t u32, wchar_t *wstr_ptr, size_t wstr_len)
{
- int u8_out_sum = 0;
- if (u8_ptr == NULL) {
- u8_len = 4;
+ size_t u16_out;
+ if (u32 < 0x10000) {
+ if (wstr_len >= 1) {
+ if (wstr_ptr != NULL) {
+ *wstr_ptr++ = (uint16_t)u32;
+ }
+ u16_out = 1;
+ } else {
+ u16_out = 0;
+ }
+ } else if (u32 <= 0x10ffff) {
+ if (wstr_len >= 2) {
+ if (wstr_ptr != NULL) {
+ // \x83T\x83\x8D\x83Q\x81[\x83g \x83G\x83\x93\x83R\x81[\x83h
+ *wstr_ptr++ = uint16_t((u32 - 0x10000) / 0x400) + 0xd800;
+ *wstr_ptr++ = uint16_t((u32 - 0x10000) % 0x400) + 0xdc00;
+ }
+ u16_out = 2;
+ } else {
+ u16_out = 0;
+ }
} else {
- if (u8_len == 0) {
- return 0;
+ u16_out = 0;
+ }
+ return u16_out;
+}
+
+/**
+ * UTF-32\x95\xB6\x8E\x9A\x82\xF0CP932\x95\xB6\x8E\x9A(Shift_JIS) 1\x95\xB6\x8E\x9A\x82֕ϊ\xB7\x82\xB7\x82\xE9
+ * @retval \x8Eg\x97p\x82\xB5\x82\xBDCP932\x95\xB6\x8E\x9A
+ * 0=\x83G\x83\x89\x81[(\x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD)
+ */
+unsigned short UTF32_CP932(unsigned int u32)
+{
+#include "../teraterm/uni2sjis.map" // mapUnicodeToSJIS[]
+ char mbstr[2];
+ unsigned short mb;
+ DWORD mblen;
+ wchar_t u16_str[2];
+ size_t u16_len;
+
+ if (u32 < 0x80) {
+ return (unsigned short)u32;
+ }
+
+#if defined(PRIORITY_CP932_TABLE)
+ if (u32 < 0x10000) {
+ wchar_t u16 = (wchar_t)u32;
+ // Tera Term\x82̕ϊ\xB7\x83e\x81[\x83u\x83\x8B\x82\xC5 Unicode -> Shift_JIS\x82֕ϊ\xB7
+ mb = _ConvertUnicode(u16, mapUnicodeToSJIS, _countof(mapUnicodeToSJIS));
+ if (mb != 0) {
+ // \x95ϊ\xB7\x82ł\xAB\x82\xBD
+ return mb;
}
}
- if (wstr_len < 0) {
+#endif
+ u16_len = UTF32ToUTF16(u32, u16_str, 2);
+ if (u16_len == 0) {
+ return 0;
+ }
+ mblen = WideCharToMultiByte(932, 0, u16_str, (int)u16_len, mbstr, 2, NULL, NULL);
+ switch (mblen) {
+ case 0:
+ case 1:
+ default:
+ if (mblen == 0 || mbstr[0] == '?') {
+ goto next_convert;
+ } else {
+ mb = (unsigned char)mbstr[0];
+ return mb;
+ }
+ case 2:
+ if (mbstr[0] == '?' && mbstr[1] == '?') {
+ // 2byte\x8Fo\x97\xCD && "??" \x82̏ꍇ\x82͕ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD
+ goto next_convert;
+ }
+ mb = (((unsigned char)mbstr[0]) << 8) | (unsigned char)mbstr[1];
+ return mb;
+ }
+
+next_convert:
+#if !defined(PRIORITY_CP932_TABLE)
+ if (u32 < 0x10000) {
+ wchar_t u16 = (wchar_t)u32;
+ // Tera Term\x82̕ϊ\xB7\x83e\x81[\x83u\x83\x8B\x82\xC5 Unicode -> Shift_JIS\x82֕ϊ\xB7
+ mb = _ConvertUnicode(u16, mapUnicodeToSJIS, _countof(mapUnicodeToSJIS));
+ if (mb != 0) {
+ // \x95ϊ\xB7\x82ł\xAB\x82\xBD
+ return mb;
+ }
+ }
+#endif
+ return 0;
+}
+
+/**
+ * CP932\x95\xB6\x8E\x9A(Shift_JIS) 1\x95\xB6\x8E\x9A\x82\xA9\x82\xE7UTF-32\x82֕ϊ\xB7\x82\xB7\x82\xE9
+ * @param[in] cp932 CP932\x95\xB6\x8E\x9A
+ * @retval \x95ϊ\xB7\x82\xB5\x82\xBDUTF-32\x95\xB6\x8E\x9A\x90\x94
+ * 0=\x83G\x83\x89\x81[(\x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD)
+ */
+unsigned int CP932ToUTF32(unsigned short cp932)
+{
+#include "../ttpcmn/sjis2uni.map" // mapSJISToUnicode[]
+ wchar_t wchar;
+ int ret;
+ unsigned int u32;
+ unsigned char buf[2];
+ int len = 0;
+
+#if defined(PRIORITY_CP932_TABLE)
+ u32 = _ConvertUnicode(cp932, mapSJISToUnicode, sizeof(mapSJISToUnicode)/sizeof(mapSJISToUnicode[0]));
+ if (u32 != 0) {
+ return u32;
+ }
+#endif
+ if (cp932 < 0x100) {
+ buf[0] = cp932 & 0xff;
+ len = 1;
+ } else {
+ buf[0] = cp932 >> 8;
+ buf[1] = cp932 & 0xff;
+ len = 2;
+ }
+ ret = MultiByteToWideChar(932, MB_ERR_INVALID_CHARS, (char *)buf, len, &wchar, 1);
+ if (ret <= 0) {
+ // MultiByteToWideChar()\x82\xAA\x95ϊ\xB7\x8E\xB8\x94s
+#if !defined(PRIORITY_CP932_TABLE)
+ u32 = _ConvertUnicode(cp932, mapSJISToUnicode, sizeof(mapSJISToUnicode)/sizeof(mapSJISToUnicode[0]));
+ // \x83e\x81[\x83u\x83\x8B\x82ɂ\xE0\x82Ȃ\xA9\x82\xC1\x82\xBD\x8Fꍇ c = 0(\x95ϊ\xB7\x8E\xB8\x94s\x8E\x9E)
+#else
+ u32 = 0;
+#endif
+ } else {
+ u32 = (unsigned int)wchar;
+ }
+
+ return u32;
+}
+
+/**
+ * Unicode\x82\xA9\x82\xE7DEC\x93\xC1\x8Eꕶ\x8E\x9A\x82֕ϊ\xB7
+ * @param u32 UTF-32\x95\xB6\x8E\x9A\x83R\x81[\x83h
+ * @return \x89\xBA\x88\xCA8bit DEC\x93\xC1\x8Eꕶ\x8E\x9A\x83R\x81[\x83h
+ * \x8F\xE3\x88\xCA8bit \x95\xB6\x8E\x9A\x83R\x81[\x83h\x8E\xED\x95\xCA (1,2,4)
+ * file://../../doc/ja/html/setup/teraterm-term.html \x8EQ\x8F\xC6
+ * 0 \x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD
+ */
+unsigned short UTF32ToDecSp(unsigned int u32)
+{
+#include "../teraterm/unisym2decsp.map" // mapUnicodeSymbolToDecSp[]
+ unsigned short cset;
+ if (u32 > 0x10000) {
+ cset = 0;
+ } else {
+ const unsigned short u16 = (unsigned short)u32;
+ cset = _ConvertUnicode(u16, mapUnicodeSymbolToDecSp, _countof(mapUnicodeSymbolToDecSp));
+ }
+ return cset;
+}
+
+/**
+ * UTF-32 \x82\xA9\x82\xE7 CP932 \x82֕ϊ\xB7\x82\xB7\x82\xE9
+ * @param[in] u32 \x95ϊ\xB7\x82\xB7\x82\xE9UTF-32
+ * @param[in,out] mb_ptr \x95ϊ\xB7\x8C\xE3CP932\x95\xB6\x8E\x9A\x97\xF1\x8Fo\x97͐\xE6(NULL\x82̂Ƃ\xAB\x8Fo\x97͂\xB5\x82Ȃ\xA2)
+ * @param[in] mb_len CP932\x8Fo\x97͐敶\x8E\x9A\x90\x94(\x95\xB6\x8E\x9A\x90\x94,sizeof(wchar_t)*wstr_len bytes)
+ * @retval \x8Fo\x97͂\xB5\x82\xBDCP932\x95\xB6\x8E\x9A\x90\x94(byte\x90\x94)
+ * 0=\x83G\x83\x89\x81[(\x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD)
+ */
+size_t UTF32ToCP932(uint32_t u32, char *mb_ptr, size_t mb_len)
+{
+ size_t cp932_out;
+ const uint16_t cp932 = UTF32_CP932(u32);
+ if (cp932 == 0 && u32 != 0) {
+ return 0;
+ }
+ if (mb_ptr == NULL) {
+ mb_len = 2;
+ }
+ if (cp932 < 0x100) {
+ if (mb_len >= 1) {
+ if (mb_ptr != NULL) {
+ *mb_ptr = cp932 & 0xff;
+ }
+ cp932_out = 1;
+ } else {
+ cp932_out = 0;
+ }
+ } else {
+ if (mb_len >= 2) {
+ if (mb_ptr != NULL) {
+ mb_ptr[0] = (cp932 >> 8) & 0xff;
+ mb_ptr[1] = cp932 & 0xff;
+ }
+ cp932_out = 2;
+ } else {
+ cp932_out = 0;
+ }
+ }
+ return cp932_out;
+}
+
+size_t UTF32ToMBCP(unsigned int u32, int code_page, char *mb_ptr, size_t mb_len)
+{
+ wchar_t u16_str[2];
+ size_t u16_len;
+ u16_len = UTF32ToUTF16(u32, u16_str, 2);
+ if (u16_len == 0) {
+ return 0;
+ }
+ mb_len = WideCharToMultiByte(code_page, 0, u16_str, u16_len, mb_ptr, mb_len, NULL, NULL);
+ if (mb_len == 1 && mb_ptr[0] == '?' && u32 != '?') {
+ // \x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD\x82Ƃ\xAB\x81A\x96߂\xE8\x92l=1, \x95\xB6\x8E\x9A[0]='?' \x82\xF0\x95Ԃ\xB5\x82Ă\xAD\x82\xE9
+ mb_len = 0;
+ }
+ return mb_len;
+}
+
+/**
+ * wchar_t(UTF-16)\x95\xB6\x8E\x9A\x97\xF1\x82\xF0\x83}\x83\x8B\x83`\x83o\x83C\x83g\x82ɕϊ\xB7\x82\xB7\x82\xE9
+ *
+ * @param[in] *wstr_ptr wchar_t\x95\xB6\x8E\x9A\x97\xF1
+ * @param[in,out] *wstr_len wchar_t\x95\xB6\x8E\x9A\x97\xF1\x92\xB7
+ * NULL\x82܂\xBD\x82\xCD0\x82̂Ƃ\xAB\x8E\xA9\x93\xAE\x81AL'\0'\x82Ń^\x81[\x83~\x83l\x81[\x83g\x82\xB7\x82邱\x82\xC6)
+ * NULL\x88ȊO\x82̂Ƃ\xAB\x93\xFC\x97͂\xB5\x82\xBD\x95\xB6\x8E\x9A\x90\x94\x82\xF0\x95Ԃ\xB7
+ * @param[in] *mb_ptr \x95ϊ\xB7\x82\xB5\x82\xBD\x95\xB6\x8E\x9A\x97\xF1\x82\xF0\x8E\xFB\x94[\x82\xB7\x82\xE9\x83|\x83C\x83\x93\x83^
+ * (NULL\x82̂Ƃ\xAB\x95ϊ\xB7\x82\xB9\x82\xB8\x82ɕ\xB6\x8E\x9A\x90\x94\x82\xF0\x83J\x83E\x83\x93\x83g\x82\xB7\x82\xE9)
+ * @param[in,out] *mb_len \x95ϊ\xB7\x82\xB5\x82\xBD\x95\xB6\x8E\x9A\x97\xF1\x82\xF0\x8E\xFB\x94[\x82ł\xAB\x82\xE9\x83T\x83C\x83Y,byte\x90\x94,
+ * mb_ptr\x82\xAANULL\x82̂Ƃ\xAB\x8Fo\x97͉\\x83T\x83C\x83Y\x82͕s\x97v
+ * \x95ϊ\xB7\x82\xB5\x82\xBD\x83}\x83\x8B\x83`\x83o\x83C\x83g\x95\xB6\x8E\x9A\x97\xF1\x82̒\xB7\x82\xB3\x82\xF0\x95Ԃ\xB7
+ * L'\0'\x82\xF0\x95ϊ\xB7\x82\xB5\x82\xBD\x82\xE7'\0'\x82\xE0\x8A܂\xDE
+ * mb_ptr\x82\xAANULL\x82̂Ƃ\xAB\x82ł\xE0\x92\xB7\x82\xB3\x82͕Ԃ\xB7
+ * @param[in] UTF32ToMB UTF32\x82\xF0\x83}\x83\x8B\x83`\x83o\x83C\x83g\x82ɕϊ\xB7\x82\xB7\x82\xE9\x8A\x94\x82ւ̃|\x83C\x83\x93\x83^
+ */
+static void WideCharToMB(const wchar_t *wstr_ptr, size_t *wstr_len_,
+ char *mb_ptr, size_t *mb_len_,
+ size_t (*UTF32ToMB)(uint32_t u32, char *mb_ptr, size_t mb_len))
+{
+ size_t wstr_len;
+ size_t mb_len;
+ size_t mb_out_sum = 0;
+ size_t wstr_in = 0;
+
+ assert(wstr_ptr != NULL);
+ if (mb_ptr == NULL) {
+ // \x95ϊ\xB7\x95\xB6\x8E\x9A\x97\xF1\x82\xF0\x8F\x91\x82\xAB\x8Fo\x82\xB3\x82Ȃ\xA2
+ mb_len = 4; // 1\x95\xB6\x8E\x9A4byte\x82ɂ͎\xFB\x82܂\xE9\x82͂\xB8
+ } else {
+ mb_len = *mb_len_;
+ }
+ if (wstr_len_ == NULL || *wstr_len_ == 0) {
wstr_len = (int)wcslen(wstr_ptr) + 1;
+ } else {
+ wstr_len = *wstr_len_;
}
- while(u8_len > 0 && wstr_len > 0) {
+ while(mb_len > 0 && wstr_len > 0) {
const wchar_t u16 = *wstr_ptr++;
uint32_t u32 = u16;
- size_t u8_out;
+ size_t mb_out;
wstr_len--;
+ wstr_in++;
// \x83T\x83\x8D\x83Q\x81[\x83g high?
- if (0xd800 <= u16 && u16 < 0xdc00) {
+ if (IsHighSurrogate(u16)) {
if (wstr_len >= 1) {
const wchar_t u16_lo = *wstr_ptr++;
wstr_len--;
+ wstr_in++;
// \x83T\x83\x8D\x83Q\x81[\x83g low?
- if (0xdc00 <= u16_lo && u16_lo < 0xe000) {
+ if (IsLowSurrogate(u16_lo)) {
// \x83T\x83\x8D\x83Q\x81[\x83g\x83y\x83A \x83f\x83R\x81[\x83h
u32 = 0x10000 + (u16 - 0xd800) * 0x400 + (u16_lo - 0xdc00);
} else {
@@ -224,34 +630,75 @@
goto unknown_code;
}
} else {
- unknown_code:
- if (u8_ptr != NULL) {
- *u8_ptr++ = '?';
- }
- u8_out = 1;
- goto loop_next;
+ goto unknown_code;
}
}
- u8_out = UTF32ToUTF8(u32, u8_ptr, u8_len);
- if (u8_out == 0) {
- goto unknown_code;
+ mb_out = UTF32ToMB(u32, mb_ptr, mb_len);
+ if (mb_out == 0) {
+ unknown_code:
+ if (mb_ptr != NULL) {
+ // \x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD\x8Fꍇ
+ *mb_ptr++ = '?';
+ }
+ mb_out = 1;
}
- loop_next:
- u8_out_sum += u8_out;
- if (u8_ptr != NULL) {
- u8_ptr += u8_out;
- u8_len -= u8_out;
+ mb_out_sum += mb_out;
+ if (mb_ptr != NULL) {
+ mb_ptr += mb_out;
+ mb_len -= mb_out;
}
}
- return u8_out_sum;
+
+ if (wstr_len_ != NULL) {
+ *wstr_len_ = wstr_in;
+ }
+ *mb_len_ = mb_out_sum;
}
+// WideCharToMultiByte\x82\xCCUTF8\x93\xC1\x89\xBB\x94\xC5
+void WideCharToUTF8(const wchar_t *wstr_ptr, size_t *wstr_len, char *u8_ptr, size_t *u8_len)
+{
+ WideCharToMB(wstr_ptr, wstr_len, u8_ptr, u8_len, UTF32ToUTF8);
+}
+
+void WideCharToCP932(const wchar_t *wstr_ptr, size_t *wstr_len, char *cp932_ptr, size_t *cp932_len)
+{
+ WideCharToMB(wstr_ptr, wstr_len,
+ cp932_ptr, cp932_len,
+ UTF32ToCP932);
+}
+
+void WideCharToMBCP(const wchar_t *wstr_ptr, size_t *wstr_len, char *mb_ptr, size_t *mb_len,
+ int code_page)
+{
+ size_t (*utf32_to_mb)(uint32_t u32, char *mb_ptr, size_t mb_len);
+ switch (code_page) {
+ case CP_UTF8:
+ utf32_to_mb = UTF32ToUTF8;
+ break;
+ case 932:
+ utf32_to_mb = UTF32ToCP932;
+ break;
+ default:
+ *mb_len = 0;
+ return;
+ }
+
+ WideCharToMB(wstr_ptr, wstr_len,
+ mb_ptr, mb_len,
+ utf32_to_mb);
+}
+
// MultiByteToWideChar\x82\xCCUTF8\x93\xC1\x89\xBB\x94\xC5
-int UTF8ToWideChar(const char *u8_ptr, int u8_len, wchar_t *wstr_ptr, int wstr_len)
+int UTF8ToWideChar(const char *u8_ptr, int u8_len_, wchar_t *wstr_ptr, int wstr_len_)
{
+ size_t u8_len;
+ size_t wstr_len = wstr_len_;
size_t u16_out_sum = 0;
- if (u8_len < 0) {
+ if (u8_len_ < 0) {
u8_len = strlen(u8_ptr) + 1;
+ } else {
+ u8_len = u8_len_;
}
if (wstr_ptr == NULL) {
wstr_len = 1;
@@ -295,16 +742,18 @@
}
u16_out_sum += u16_out;
}
- return u16_out_sum;
+ return (int)u16_out_sum;
}
/**
* wchar_t\x95\xB6\x8E\x9A\x97\xF1\x82\xF0\x83}\x83\x8B\x83`\x83o\x83C\x83g\x95\xB6\x8E\x9A\x97\xF1\x82֕ϊ\xB7
* @param[in] *wstr_ptr wchar_t\x95\xB6\x8E\x9A\x97\xF1
- * @param[in] wstr_len wchar_t\x95\xB6\x8E\x9A\x97\xF1\x92\xB7(0\x82̂Ƃ\xAB\x8E\xA9\x93\xAE)
+ * @param[in] wstr_len wchar_t\x95\xB6\x8E\x9A\x97\xF1\x92\xB7(0\x82̂Ƃ\xAB\x8E\xA9\x93\xAE\x81A\x8E\xA9\x93\xAE\x82̂Ƃ\xAB\x82\xCDL'\0'\x82Ń^\x81[\x83~\x83l\x81[\x83g\x82\xB7\x82邱\x82\xC6)
* @param[in] code_page \x95ϊ\xB7\x90\xE6\x83R\x81[\x83h\x83y\x81[\x83W
- * @param[out] *mb_len_ mb\x95\xB6\x8E\x9A\x97\xF1\x92\xB7(NULL\x82̂Ƃ\xAB\x93\xE0\x95\x94\x83G\x83\x89\x81[)
+ * @param[out] *mb_len_ \x95ϊ\xB7\x82\xB5\x82\xBD\x95\xB6\x8E\x9A\x97\xF1\x92\xB7,byte\x90\x94,L'\0'\x82\xF0\x95ϊ\xB7\x82\xB5\x82\xBD\x82\xE7'\0'\x82\xE0\x8A܂\xDE
+ * (NULL\x82̂Ƃ\xAB\x95\xB6\x8E\x9A\x97\xF0\x95Ԃ\xB3\x82Ȃ\xA2)
* @retval mb\x95\xB6\x8E\x9A\x97\xF1\x82ւ̃|\x83C\x83\x93\x83^(NULL\x82̎\x9E\x95ϊ\xB7\x83G\x83\x89\x81[)
+ * \x8Eg\x97p\x8C\xE3 free() \x82\xB7\x82邱\x82\xC6
*/
char *_WideCharToMultiByte(const wchar_t *wstr_ptr, size_t wstr_len, int code_page, size_t *mb_len_)
{
@@ -317,9 +766,11 @@
wstr_len = wcslen(wstr_ptr) + 1;
}
int len;
- if (code_page == CP_UTF8) {
- len = WideCharToUTF8(wstr_ptr, (DWORD)wstr_len,
- NULL, 0);
+ if (code_page == CP_UTF8 || code_page == 932) {
+ size_t wl = wstr_len;
+ size_t ml;
+ WideCharToMBCP(wstr_ptr, &wl, NULL, &ml, code_page);
+ len = ml;
} else {
len = ::WideCharToMultiByte(code_page, flags,
wstr_ptr, (DWORD)wstr_len,
@@ -333,9 +784,11 @@
if (mb_ptr == NULL) {
return NULL;
}
- if (code_page == CP_UTF8) {
- len = WideCharToUTF8(wstr_ptr, (DWORD)wstr_len,
- mb_ptr, len);
+ if (code_page == CP_UTF8 || code_page == 932) {
+ size_t wl = wstr_len;
+ size_t ml = len;
+ WideCharToMBCP(wstr_ptr, &wl, mb_ptr, &ml, code_page);
+ len = ml;
} else {
len = ::WideCharToMultiByte(code_page, flags,
wstr_ptr, (DWORD)wstr_len,
@@ -347,7 +800,8 @@
return NULL;
}
if (mb_len_ != NULL) {
- *mb_len_ = len - 1;
+ // \x95ϊ\xB7\x82\xB5\x82\xBD\x95\xB6\x8E\x9A\x97\xF1\x90\x94(byte\x90\x94)\x82\xF0\x95Ԃ\xB7
+ *mb_len_ = len;
}
return mb_ptr;
}
@@ -355,10 +809,12 @@
/**
* \x83}\x83\x8B\x83`\x83o\x83C\x83g\x95\xB6\x8E\x9A\x97\xF1\x82\xF0wchar_t\x95\xB6\x8E\x9A\x97\xF1\x82֕ϊ\xB7
* @param[in] *str_ptr mb(char)\x95\xB6\x8E\x9A\x97\xF1
- * @param[in] str_len mb(char)\x95\xB6\x8E\x9A\x97\xF1\x92\xB7(0\x82̂Ƃ\xAB\x8E\xA9\x93\xAE)
+ * @param[in] str_len mb(char)\x95\xB6\x8E\x9A\x97\xF1\x92\xB7(0\x82̂Ƃ\xAB\x8E\xA9\x93\xAE\x81A\x8E\xA9\x93\xAE\x82̂Ƃ\xAB\x82\xCD'\0'\x82Ń^\x81[\x83~\x83l\x81[\x83g\x82\xB7\x82邱\x82\xC6)
* @param[in] code_page \x95ϊ\xB7\x8C\xB3\x83R\x81[\x83h\x83y\x81[\x83W
- * @param[out] *w_len_ wchar_t\x95\xB6\x8E\x9A\x97\xF1\x92\xB7
- * @retval mb\x95\xB6\x8E\x9A\x97\xF1\x82ւ̃|\x83C\x83\x93\x83^(NULL\x82̎\x9E\x95ϊ\xB7\x83G\x83\x89\x81[)
+ * @param[out] *w_len_ wchar_t\x95\xB6\x8E\x9A\x97\xF1\x92\xB7,wchar_t\x90\x94,'\0'\x82\xF0\x95ϊ\xB7\x82\xB5\x82\xBD\x82\xE7L'\0'\x82\xE0\x8A܂\xDE
+ * (NULL\x82̂Ƃ\xAB\x95\xB6\x8E\x9A\x97\xF0\x95Ԃ\xB3\x82Ȃ\xA2)
+ * @retval wchar_t\x95\xB6\x8E\x9A\x97\xF1\x82ւ̃|\x83C\x83\x93\x83^(NULL\x82̎\x9E\x95ϊ\xB7\x83G\x83\x89\x81[)
+ * \x8Eg\x97p\x8C\xE3 free() \x82\xB7\x82邱\x82\xC6
*/
wchar_t *_MultiByteToWideChar(const char *str_ptr, size_t str_len, int code_page, size_t *w_len_)
{
@@ -402,7 +858,8 @@
return NULL;
}
if (w_len_ != NULL) {
- *w_len_ = len - 1;
+ // \x95ϊ\xB7\x82\xB5\x82\xBD\x95\xB6\x8E\x9A\x97\xF1\x90\x94(wchar_t\x90\x94)\x82\xF0\x95Ԃ\xB7
+ *w_len_ = len;
}
return wstr_ptr;
}
Modified: trunk/teraterm/common/codeconv.h
===================================================================
--- trunk/teraterm/common/codeconv.h 2019-03-05 16:22:05 UTC (rev 7461)
+++ trunk/teraterm/common/codeconv.h 2019-03-09 17:32:42 UTC (rev 7462)
@@ -35,12 +35,24 @@
#endif
-// 1char
+// simple code convert
+unsigned int CP932ToUTF32(unsigned short cp932);
+unsigned short UTF32ToDecSp(unsigned int u32);
+unsigned int MBCPToUTF32(unsigned short KCode, int CodePage);
+unsigned short UTF32_CP932(unsigned int u32);
+
+// 1char ToUTF32
+size_t UTF8ToUTF32(const char *u8_ptr_, size_t u8_len, unsigned int *u32_);
+size_t UTF16ToUTF32(const wchar_t *wstr_ptr, size_t wstr_len, unsigned int *u32);
+// 1char UTF32To
+size_t UTF32ToUTF16(unsigned int u32, wchar_t *wstr_ptr, size_t wstr_len);
size_t UTF32ToUTF8(unsigned int u32, char *u8_ptr, size_t u8_len);
-size_t UTF8ToUTF32(const char *u8_ptr_, size_t u8_len, unsigned int *u32_);
+size_t UTF32ToCP932(unsigned int u32, char *mb_ptr, size_t mb_len);
+size_t UTF32ToMBCP(unsigned int u32, int code_page, char *mb_ptr, size_t mb_len);
// MultiByteToWideChar() wrappers
-int WideCharToUTF8(const wchar_t *wstr_ptr, int wstr_len, char *u8_ptr, int u8_len);
+void WideCharToUTF8(const wchar_t *wstr_ptr, size_t *wstr_len, char *u8_ptr, size_t *u8_len);
+void WideCharToCP932(const wchar_t *wstr_ptr, size_t *wstr_len, char *cp932_ptr, size_t *cp932_len);
int UTF8ToWideChar(const char *u8_ptr, int u8_len, wchar_t *wstr_ptr, int wstr_len);
// API wrappers
Modified: trunk/teraterm/teraterm/CMakeLists.txt
===================================================================
--- trunk/teraterm/teraterm/CMakeLists.txt 2019-03-05 16:22:05 UTC (rev 7461)
+++ trunk/teraterm/teraterm/CMakeLists.txt 2019-03-09 17:32:42 UTC (rev 7462)
@@ -31,7 +31,12 @@
../common/i18n.h
../common/dllutil.cpp
../common/dllutil.h
- ../ttpcmn/language.h
+ ../common/codeconv.h
+ ../common/codeconv.cpp
+ #
+ ../teraterm/unisym2decsp.map
+ ../teraterm/uni2sjis.map
+ ../ttpcmn/sjis2uni.map
)
source_group(
@@ -102,8 +107,6 @@
WSAAsyncGetAddrInfo.c
WSAAsyncGetAddrInfo.h
#
- uni2sjis.map
- unisym2decsp.map
uni_combining.map
#
teraterm.manifest
Modified: trunk/teraterm/teraterm/vtterm.c
===================================================================
--- trunk/teraterm/teraterm/vtterm.c 2019-03-05 16:22:05 UTC (rev 7461)
+++ trunk/teraterm/teraterm/vtterm.c 2019-03-09 17:32:42 UTC (rev 7462)
@@ -52,7 +52,8 @@
#include "telnet.h"
#include "ttime.h"
#include "clipboar.h"
-#include "../ttpcmn/language.h"
+#include "codeconv.h"
+#include "codeconv.h"
#include "vtterm.h"
@@ -5408,13 +5409,6 @@
}
//
-// UTF-8
-//
-#include "uni2sjis.map"
-#include "unisym2decsp.map"
-
-
-//
// Unicode Combining Character Support
//
#include "uni_combining.map"
@@ -5467,17 +5461,22 @@
return (index);
}
-// unicode(UTF-16,wchar_t)\x82\xF0\x83o\x83b\x83t\x83@\x82֏\x91\x82\xAB\x8D\x9E\x82\xDE
+// unicode(UTF-32,wchar_t)\x82\xF0\x83o\x83b\x83t\x83@\x82֏\x91\x82\xAB\x8D\x9E\x82\xDE
static void UnicodeToCP932(unsigned int code)
{
- wchar_t wchar = (wchar_t)code;
+ wchar_t wchar;
int ret;
char mbchar[2];
unsigned short cset;
+ if (code >= 0x10000) {
+ goto unknown;
+ }
+ wchar = (wchar_t)code;
+
// Unicode\x82\xA9\x82\xE7DEC\x93\xC1\x8Eꕶ\x8E\x9A\x82ւ̃}\x83b\x83s\x83\x93\x83O
if (ts.UnicodeDecSpMapping) {
- cset = ConvertUnicode(wchar, mapUnicodeSymbolToDecSp, MAPSIZE(mapUnicodeSymbolToDecSp));
+ cset = UTF32ToDecSp(wchar);
if (((cset >> 8) & ts.UnicodeDecSpMapping) != 0) {
PutDecSp(cset & 0xff);
return;
@@ -5485,7 +5484,24 @@
}
// Unicode -> \x93\xE0\x95\x94\x83R\x81[\x83h(ts.CodePage)\x82֕ϊ\xB7\x82\xB5\x82ďo\x97\xCD
- ret = WideCharToMultiByte(ts.CodePage, 0, &wchar, 1, mbchar, 2, NULL, NULL);
+ if (ts.CodePage == 932) {
+ ret = (int)UTF16ToCP932(&wchar, 1, &cset);
+ if (ret == 0) {
+ // \x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD
+ ;
+ } else if (cset < 0x100) {
+ // 1byte\x95\xB6\x8E\x9A
+ mbchar[0] = (char)cset;
+ ret = 1;
+ } else {
+ // 2byte\x95\xB6\x8E\x9A
+ mbchar[0] = (char)(cset >> 8);
+ mbchar[1] = (char)(cset & 0xff);
+ ret = 2;
+ }
+ } else {
+ ret = WideCharToMultiByte(ts.CodePage, 0, &wchar, 1, mbchar, 2, NULL, NULL);
+ }
if (ret == 1 && mbchar[0] == '?' && code != '?') {
// \x95ϊ\xB7\x82ł\xAB\x82Ȃ\xA9\x82\xC1\x82\xBD\x82Ƃ\xAB\x81Aret=1, '?' \x82\xF0\x95Ԃ\xB5\x82Ă\xAD\x82\xE9
ret = 0;
@@ -5492,17 +5508,7 @@
}
switch (ret) {
case 0:
- if (ts.CodePage == 932) {
- // CP932
- // U+301C\x82Ȃǂ͕ϊ\xB7\x82ł\xAB\x82Ȃ\xA2\x81BUnicode -> Shift_JIS\x82֕ϊ\xB7\x82\xB5\x82Ă݂\xE9\x81B
- cset = ConvertUnicode(code, mapUnicodeToSJIS, MAPSIZE(mapUnicodeToSJIS));
- if (cset != 0) {
- Kanji = cset & 0xff00;
- PutKanji(cset & 0x00ff);
- return;
- }
- }
-
+ unknown:
PutChar('?');
if (ts.UnknownUnicodeCharaAsWide) {
PutChar('?');
Modified: trunk/teraterm/ttpcmn/language.c
===================================================================
--- trunk/teraterm/ttpcmn/language.c 2019-03-05 16:22:05 UTC (rev 7461)
+++ trunk/teraterm/ttpcmn/language.c 2019-03-09 17:32:42 UTC (rev 7462)
@@ -35,9 +35,8 @@
#include <locale.h>
#include "language.h"
-#include "codeconv.h"
-#include "sjis2uni.map"
+// export\x82\xB3\x82\xEA\x82Ă\xA2\x82\xE9
unsigned short ConvertUnicode(unsigned short code, const codemap_t *table, int tmax)
{
int low, mid, high;
@@ -64,69 +63,6 @@
return (result);
}
-// \x93\xE0\x95\x94\x83R\x81[\x83h(CodePage)\x82\xF0UTF8\x82֕ϊ\xB7\x82\xB7\x82\xE9
-unsigned int PASCAL SJIS2UTF8(WORD KCode, int *byte, int CodePage)
-{
- wchar_t wchar;
- int ret;
- unsigned int code;
- unsigned int c, c1, c2, c3;
- unsigned char buf[3];
- unsigned char KCode_h;
- int len = 0;
-
- // \x93\xE0\x95\x94\x83R\x81[\x83h(CodePage)\x82\xA9\x82\xE7UTF-16LE\x82֕ϊ\xB7\x82\xB7\x82\xE9
- KCode_h = (unsigned char)(KCode >> 8);
- if (KCode_h != 0) {
- buf[len++] = KCode_h;
- }
- buf[len++] = KCode & 0xff;
- ret = MultiByteToWideChar(CodePage, MB_ERR_INVALID_CHARS, buf, len, &wchar, 1);
- if (ret <= 0) {
- // \x95ϊ\xB7\x8E\xB8\x94s
- unsigned short cset = 0;
- if (CodePage == 932) {
- // CP932
- cset = ConvertUnicode(KCode, mapSJISToUnicode, sizeof(mapSJISToUnicode)/sizeof(mapSJISToUnicode[0]));
- }
- if (cset == 0) {
- c = 0xfffd; // U+FFFD: Replacement Character
- } else {
- c = cset;
- }
- } else {
- c = (unsigned int)wchar;
- }
-
- // UTF-16LE\x82\xA9\x82\xE7UTF-8\x82֕ϊ\xB7\x82\xB7\x82\xE9
- if (c <= 0x0000007f) {
- // 0x00000000 <= c <= 0x0000007f
- code = (c & 0xff);
- *byte = 1;
-
- } else if (c <= 0x000007ff) {
- // 0x00000080 <= c <= 0x000007ff
- c1 = ((c >> 6) & 0x1f) | 0xc0;
- c2 = (c & 0x3f) | 0x80;
- code = (c1 << 8) | c2;
- *byte = 2;
-
- } else if (c <= 0x0000ffff) {
- // 0x00000800 <= c <= 0x0000ffff
- c1 = ((c >> 12) & 0xf) | 0xe0;
- c2 = ((c >> 6) & 0x3f) | 0x80;
- c3 = ((c) & 0x3f) | 0x80;
- code = (c1 << 16) | (c2 << 8) | c3;
- *byte = 3;
- } else {
- code = KCode;
- *byte = 2;
- }
-
- return (code);
-}
-
-
// Japanese SJIS -> JIS
WORD PASCAL SJIS2JIS(WORD KCode)
{
Modified: trunk/teraterm/ttpcmn/language.h
===================================================================
--- trunk/teraterm/ttpcmn/language.h 2019-03-05 16:22:05 UTC (rev 7461)
+++ trunk/teraterm/ttpcmn/language.h 2019-03-09 17:32:42 UTC (rev 7462)
@@ -35,13 +35,11 @@
#endif
/* proto types */
-unsigned int PASCAL SJIS2UTF8(WORD KCode, int *byte, int CodePage);
WORD PASCAL SJIS2JIS(WORD KCode);
WORD PASCAL SJIS2EUC(WORD KCode);
WORD PASCAL JIS2SJIS(WORD KCode);
BYTE PASCAL RussConv(int cin, int cout, BYTE b);
void PASCAL RussConvStr(int cin, int cout, PCHAR Str, int count);
-unsigned short ConvertUnicode(unsigned short code, const codemap_t *table, int tmax);
#ifdef __cplusplus
}
Modified: trunk/teraterm/ttpcmn/ttcmn.c
===================================================================
--- trunk/teraterm/ttpcmn/ttcmn.c 2019-03-05 16:22:05 UTC (rev 7461)
+++ trunk/teraterm/ttpcmn/ttcmn.c 2019-03-09 17:32:42 UTC (rev 7462)
@@ -1582,25 +1582,52 @@
return i;
}
+// \x93\xE0\x95\x94\x83R\x81[\x83h(CodePage)\x82\xF0UTF-32(UTF-16LE)\x82֕ϊ\xB7\x82\xB7\x82\xE9
+static unsigned int SJIS2UTF32(WORD KCode, int CodePage)
+{
+ unsigned int c;
+
+ // \x93\xE0\x95\x94\x83R\x81[\x83h(CodePage)\x82\xA9\x82\xE7UTF-16LE\x82֕ϊ\xB7\x82\xB7\x82\xE9
+ if (CodePage == 932) {
+ c = CP932ToUTF32(KCode);
+ } else {
+ unsigned char buf[3];
+ wchar_t wchar;
+ int ret;
+ int len = 0;
+ if (KCode < 0x100) {
+ buf[0] = KCode & 0xff;
+ len = 1;
+ } else {
+ buf[0] = KCode >> 8;
+ buf[1] = KCode & 0xff;
+ len = 2;
+ }
+ ret = MultiByteToWideChar(CodePage, MB_ERR_INVALID_CHARS, buf, len, &wchar, 1);
+ if (ret <= 0) {
+ c = 0;
+ } else {
+ c = (unsigned int)wchar;
+ }
+ }
+ if (c <= 0) {
+ // \x95ϊ\xB7\x8E\xB8\x94s
+ c = 0xfffd; // U+FFFD: Replacement Character
+ }
+
+ return c;
+}
+
+// \x93\xE0\x95\x94\x83R\x81[\x83h(CodePage)\x82\xF0UTF-8\x82֏o\x97͂\xB7\x82\xE9
static int OutputTextUTF8(WORD K, char *TempStr, PComVar cv)
{
+ int CodePage = *cv->CodePage;
unsigned int code;
int outlen;
- int TempLen = 0;
- code = SJIS2UTF8(K, &outlen, *cv->CodePage);
- switch (outlen) {
- case 4:
- TempStr[TempLen++] = (code >> 24) & 0xff;
- case 3:
- TempStr[TempLen++] = (code >> 16) & 0xff;
- case 2:
- TempStr[TempLen++] = (code >> 8) & 0xff;
- case 1:
- TempStr[TempLen++] = code & 0xff;
- }
-
- return TempLen;
+ code = SJIS2UTF32(K, CodePage);
+ outlen = UTF32ToUTF8(code, TempStr, 4);
+ return outlen;
}
//