add codecnv

- add ucs4(UTF32) function
- OPENFILENAMEW function

codecnv/codecnv.h

@@ -21,8 +21,19 @@ UINT codecnv_euctoucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UI
 UINT codecnv_utf8toucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput);
 UINT codecnv_utf8tosjis(char *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput);
 
+UINT codecnv_ucs2len(const UINT16 *lpString);
 UINT codecnv_ucs2tosjis(char *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput);
 UINT codecnv_ucs2toutf8(char *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput);
+UINT codecnv_utf8toucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput);
+
+UINT codecnv_utf8toucs4_1(UINT32 *lpOutput, const char *lpInput, UINT cchInput);
+UINT codecnv_utf8toucs4(UINT32 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput);
+UINT codecnv_ucs4toutf8(char *lpOutput, UINT cchOutput, const UINT32 *lpInput, UINT cchInput);
+
+UINT codecnv_ucs4len(const UINT32 *lpString);
+UINT codecnv_ucs4toucs2_1(UINT16 *lpOutput, UINT cchOutput, const UINT32 *lpInput);
+UINT codecnv_ucs2toucs4(UINT32 *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput);
+UINT codecnv_ucs4toucs2(UINT16 *lpOutput, UINT cchOutput, const UINT32 *lpInput, UINT cchInput);
 
 #ifdef __cplusplus
 }

codecnv/eucsjis.c

@@ -3,7 +3,11 @@
  * @brief	Implementation of converting EUC to S-JIS
  */
 
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
 #include "compiler.h"
+#endif
 #include "codecnv.h"
 
 static UINT euctosjis(char *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput);

codecnv/eucucs2.c

@@ -3,7 +3,11 @@
  * @brief	Implementation of converting EUC to UTF-16
  */
 
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
 #include "compiler.h"
+#endif
 #include "codecnv.h"
 
 static UINT euctoucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput);

codecnv/sjiseuc.c

@@ -3,7 +3,11 @@
  * @brief	Implementation of converting S-JIS to EUC
  */
 
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
 #include "compiler.h"
+#endif
 #include "codecnv.h"
 
 static UINT sjistoeuc(char *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput);

codecnv/sjisucs2.c

@@ -3,7 +3,11 @@
  * @brief	Implementation of converting S-JIS to UCS2
  */
 
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
 #include "compiler.h"
+#endif
 #include "codecnv.h"
 
 //! undefined code

codecnv/ucs2sjis.c

@@ -3,7 +3,11 @@
  * @brief	Implementation of converting UCS2 to S-JIS
  */
 
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
 #include "compiler.h"
+#endif
 #include "codecnv.h"
 
 //! undefined code

codecnv/ucs2ucs4.c

@@ -0,0 +1,115 @@
+/**
+ * @file	ucs2ucs4.c
+ * @brief	Implementation of converting UTF-16 to UTF-32
+ */
+
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
+#include "compiler.h"
+#endif
+#include "codecnv.h"
+
+static BOOL isucs2highsurrogate(const UINT16 c);
+static BOOL isucs2lowsurrogate(const UINT16 c);
+static UINT ucs2toucs4(UINT32 *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput);
+
+/**
+ * Maps a UTF-16 string to a UTF-8 string
+ * @param[out] lpOutput Pointer to a buffer that receives the converted string
+ * @param[in] cchOutput Size, in characters, of the buffer indicated by lpOutput
+ * @param[in] lpInput Pointer to the character string to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters written to the buffer indicated by lpOutput
+ */
+UINT codecnv_ucs2toucs4(UINT32 *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput) {
+	UINT n = 0;
+	UINT nLength;
+
+	if(lpOutput != NULL && lpInput != NULL) {
+		if(cchOutput == 0) {
+			lpOutput = NULL;
+			cchOutput = (UINT)-1;
+		}
+
+		if(cchInput != (UINT)-1) {
+			// Binary mode
+			n = ucs2toucs4(lpOutput, cchOutput, lpInput, cchInput);
+		} else {
+			// String mode
+			nLength = ucs2toucs4(lpOutput, cchOutput - 1, lpInput, codecnv_ucs2len(lpInput));
+			lpOutput[nLength] = '\0';
+			n = nLength + 1;
+		}
+	}
+
+	return n;
+}
+
+static BOOL isucs2highsurrogate(const UINT16 c) {
+	return 0xD800 <= c && c < 0xDC00;
+}
+
+static BOOL isucs2lowsurrogate(const UINT16 c) {
+	return 0xDC00 <= c && c < 0xE000;
+}
+
+/**
+ * Maps a UTF-16 char to a UTF-32 character
+ * @param[out] lpOutput Pointer to a buffer that receives the converted character
+ * @param[in] lpInput Pointer to the character to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters read to the buffer indicated by lpInput
+ */
+UINT codecnv_ucs2toucs4_1(UINT32 *lpOutput, const UINT16 *lpInput, UINT cchInput) {
+	int n = 0;
+
+	if(isucs2highsurrogate(lpInput[0])) {
+		if(isucs2lowsurrogate(lpInput[1]) && cchInput >= 2) {
+			lpOutput[0]  = 0x10000;
+			lpOutput[0] += ((UINT32)lpInput[0] - 0xD800) * 0x400;
+			lpOutput[0] +=  (UINT32)lpInput[1] - 0xDC00         ;
+			n = 2;
+		} else if(lpInput[1] == 0 && cchInput >= 1) {
+			lpOutput[0] = lpInput[0];
+			n = 1;
+		}
+	} else if(isucs2lowsurrogate(lpInput[0])) {
+		if(lpInput[1] == 0 && cchInput >= 1) {
+			lpOutput[0] = lpInput[0];
+			n = 1;
+		}
+	} else {
+		if(cchInput >= 1) {
+			lpOutput[0] = lpInput[0];
+			n = 1;
+		}
+	}
+
+	return n;
+}
+
+/**
+ * Maps a UTF-16 string to a UTF-32 string (inner)
+ * @param[out] lpOutput Pointer to a buffer that receives the converted string
+ * @param[in] cchOutput Size, in characters, of the buffer indicated by lpOutput
+ * @param[in] lpInput Pointer to the character string to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters written to the buffer indicated by lpOutput
+ */
+static UINT ucs2toucs4(UINT32 *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput)
+{
+	UINT nRemain;
+	int n = 1;
+
+	nRemain = cchOutput;
+	while ((cchInput > 0) && (nRemain > 0) && (n > 0)) {
+		n = codecnv_ucs2toucs4_1(lpOutput, lpInput, nRemain);
+		cchInput -= n;
+		lpInput += n;
+		lpOutput++;
+		nRemain--;
+	}
+
+	return (UINT)(cchOutput - nRemain);
+}

codecnv/ucs2utf8.c

@@ -3,10 +3,13 @@
  * @brief	Implementation of converting UCS2 to UTF-8
  */
 
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
 #include "compiler.h"
+#endif
 #include "codecnv.h"
 
-static UINT ucs2len(const UINT16 *lpString);
 static UINT ucs2toutf8(char *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput);
 
 /**
@@ -17,36 +20,28 @@ static UINT ucs2toutf8(char *lpOutput, UINT cchOutput, const UINT16 *lpInput, UI
  * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
  * @return The number of characters written to the buffer indicated by lpOutput
  */
-UINT codecnv_ucs2toutf8(char *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput)
-{
+UINT codecnv_ucs2toutf8(char *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput) {
+	UINT n = 0;
 	UINT nLength;
 
-	if (lpInput == NULL)
-	{
-		return 0;
-	}
-
-	if (cchOutput == 0)
-	{
+	if(lpOutput != NULL && lpInput != NULL) {
+		if(cchOutput == 0) {
 			lpOutput = NULL;
 			cchOutput = (UINT)-1;
 		}
 
-	if (cchInput != (UINT)-1)
-	{
+		if(cchInput != (UINT)-1) {
 			// Binary mode
-		return ucs2toutf8(lpOutput, cchOutput, lpInput, cchInput);
-	}
-	else
-	{
+			n = ucs2toutf8(lpOutput, cchOutput, lpInput, cchInput);
+		} else {
 			// String mode
-		nLength = ucs2toutf8(lpOutput, cchOutput - 1, lpInput, ucs2len(lpInput));
-		if (lpOutput)
-		{
+			nLength = ucs2toutf8(lpOutput, cchOutput - 1, lpInput, codecnv_ucs2len(lpInput));
 			lpOutput[nLength] = '\0';
+			n = nLength + 1;
 		}
-		return nLength + 1;
 	}
+
+	return n;
 }
 
 /**
@@ -54,13 +49,13 @@ UINT codecnv_ucs2toutf8(char *lpOutput, UINT cchOutput, const UINT16 *lpInput, U
  * @param[in] lpString Null-terminated string
  * @return the number of characters in lpString
  */
-static UINT ucs2len(const UINT16 *lpString)
-{
+UINT codecnv_ucs2len(const UINT16 *lpString) {
 	const UINT16 *p = lpString;
-	while (*p != 0)
-	{
+
+	while (*p != 0) {
 		p++;
 	}
+
 	return (UINT)(p - lpString);
 }
 
@@ -72,52 +67,23 @@ static UINT ucs2len(const UINT16 *lpString)
  * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
  * @return The number of characters written to the buffer indicated by lpOutput
  */
-static UINT ucs2toutf8(char *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput)
-{
+static UINT ucs2toutf8(char *lpOutput, UINT cchOutput, const UINT16 *lpInput, UINT cchInput) {
+	UINT n, m;
 	UINT nRemain;
-	UINT c;
+	UINT32 c[2];
 
+	n = m = 1;
 	nRemain = cchOutput;
-	while ((cchInput > 0) && (nRemain > 0))
-	{
-		c = *lpInput++;
-		cchInput--;
+	while ((cchInput > 0) && (nRemain > 0) && (m > 0) && (n > 0)) {
+		n = codecnv_ucs2toucs4(c, 1, lpInput, 2);
+		if(n) {
+			m = codecnv_ucs4toutf8(lpOutput, nRemain, c, 1);
+			lpInput += n;
+			cchInput -= n;
+			lpOutput += m;
+			nRemain -= m;
+		}
+	}
 
-		if (c < 0x80)
-		{
-			nRemain--;
-			if (lpOutput)
-			{
-				*lpOutput++ = (char)c;
-			}
-		}
-		else if (c < 0x800)
-		{
-			if (nRemain < 2)
-			{
-				break;
-			}
-			nRemain -= 2;
-			if (lpOutput)
-			{
-				*lpOutput++ = (char)(0xc0 + ((c >> 6) & 0x1f));
-				*lpOutput++ = (char)(0x80 + ((c >> 0) & 0x3f));
-			}
-		}
-		else
-		{
-			if (nRemain < 3)
-			{
-				break;
-			}
-			nRemain -= 3;
-			if (lpOutput)
-			{
-				*lpOutput++ = (char)(0xe0 + ((c >> 12) & 0x0f));
-				*lpOutput++ = (char)(0x80 + ((c >> 6) & 0x3f));
-				*lpOutput++ = (char)(0x80 + ((c >> 0) & 0x3f));
-			}
-		}
-	}
 	return (UINT)(cchOutput - nRemain);
 }

codecnv/ucs4ucs2.c

@@ -0,0 +1,111 @@
+/**
+ * @file	ucs4ucs2.c
+ * @brief	Implementation of converting UCS4 to UCS2
+ */
+
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
+#include "compiler.h"
+#endif
+#include "codecnv.h"
+
+static UINT ucs4toucs2(UINT16 *lpOutput, UINT cchOutput, const UINT32 *lpInput, UINT cchInput);
+
+/**
+ * Maps a UTF-32 string to a UTF-16 string
+ * @param[out] lpOutput Pointer to a buffer that receives the converted string
+ * @param[in] cchOutput Size, in characters, of the buffer indicated by lpOutput
+ * @param[in] lpInput Pointer to the character string to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters written to the buffer indicated by lpOutput
+ */
+UINT codecnv_ucs4toucs2(UINT16 *lpOutput, UINT cchOutput, const UINT32 *lpInput, UINT cchInput) {
+	UINT n = 0;
+	UINT nLength;
+
+	if(lpOutput != NULL && lpInput != NULL) {
+		if(cchOutput == 0) {
+			lpOutput = NULL;
+			cchOutput = (UINT)-1;
+		}
+
+		if(cchInput != (UINT)-1) {
+			// Binary mode
+			n = ucs4toucs2(lpOutput, cchOutput, lpInput, cchInput);
+		} else {
+			// String mode
+			nLength = ucs4toucs2(lpOutput, cchOutput - 1, lpInput, codecnv_ucs4len(lpInput));
+			lpOutput[nLength] = '\0';
+			n = nLength + 1;
+		}
+	}
+
+	return n;
+}
+
+/**
+ * Get the length of input string
+ * @param[in] lpString Null-terminated string
+ * @return the number of characters in lpString
+ */
+UINT codecnv_ucs4len(const UINT32 *lpString) {
+	const UINT32 *p = lpString;
+
+	while(*p != 0) {
+		p++;
+	}
+
+	return (UINT)(p - lpString);
+}
+
+/**
+ * Maps a UTF-32 char to a UTF-16 character
+ * @param[out] lpOutput Pointer to a buffer that receives the converted character
+ * @param[in] cchOutput Size, in characters, of the buffer indicated by lpOutput
+ * @param[in] lpInput Pointer to the character to convert
+ * @return The number of characters written to the buffer indicated by lpOutput
+ */
+UINT codecnv_ucs4toucs2_1(UINT16 *lpOutput, UINT cchOutput, const UINT32 *lpInput) {
+	UINT n = 0;
+
+	if (lpInput[0] > 0x10FFFF) {
+		return n;
+	}
+
+	if(lpInput[0] < 0x10000 && cchOutput > 1) {
+		lpOutput[0] = (UINT16)lpInput[0];
+		lpOutput[1] = 0;
+		n = 1;
+	} else if(cchOutput > 2) {
+		lpOutput[0] = (UINT16)((lpInput[0] - 0x10000) / 0x400 + 0xD800);
+		lpOutput[1] = (UINT16)((lpInput[0] - 0x10000) % 0x400 + 0xDC00);
+		n = 2;
+	}
+
+	return n;
+}
+
+/**
+ * Maps a UTF-32 string to a UTF-16 string (inner)
+ * @param[out] lpOutput Pointer to a buffer that receives the converted string
+ * @param[in] cchOutput Size, in characters, of the buffer indicated by lpOutput
+ * @param[in] lpInput Pointer to the character string to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters written to the buffer indicated by lpOutput
+ */
+static UINT ucs4toucs2(UINT16 *lpOutput, UINT cchOutput, const UINT32 *lpInput, UINT cchInput) {
+	UINT nRemain;
+	UINT n = 1;
+
+	nRemain = cchOutput;
+	while ((cchInput > 0) && (nRemain > 0) && (n > 0)) {
+		n = codecnv_ucs4toucs2_1(lpOutput, nRemain, lpInput);
+		lpOutput += n;
+		nRemain -= n;
+		lpInput++;
+		cchInput--;
+	}
+
+	return (UINT)(cchOutput - nRemain);
+}

codecnv/ucs4utf8.c

@@ -0,0 +1,104 @@
+/**
+ * @file	ucs4utf8.c
+ * @brief	Implementation of converting UCS4 to UTF-8
+ */
+
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
+#include "compiler.h"
+#endif
+#include "codecnv.h"
+
+static UINT ucs4toutf8(char *lpOutput, UINT cchOutput, const UINT32 *lpInput, UINT cchInput);
+
+/**
+ * Maps a UTF-32 string to a UTF-8 string
+ * @param[out] lpOutput Pointer to a buffer that receives the converted string
+ * @param[in] cchOutput Size, in characters, of the buffer indicated by lpOutput
+ * @param[in] lpInput Pointer to the character string to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters written to the buffer indicated by lpOutput
+ */
+UINT codecnv_ucs4toutf8(char *lpOutput, UINT cchOutput, const UINT32 *lpInput, UINT cchInput) {
+	UINT n = 0;
+	UINT nLength;
+
+	if(lpOutput != NULL && lpInput != NULL) {
+		if(cchOutput == 0) {
+			lpOutput = NULL;
+			cchOutput = (UINT)-1;
+		}
+
+		if(cchInput != (UINT)-1) {
+			// Binary mode
+			n = ucs4toutf8(lpOutput, cchOutput, lpInput, cchInput);
+		} else {
+			// String mode
+			nLength = ucs4toutf8(lpOutput, cchOutput - 1, lpInput, codecnv_ucs4len(lpInput));
+			lpOutput[nLength] = '\0';
+			n = nLength + 1;
+		}
+	}
+
+	return n;
+}
+
+/**
+ * Maps a UTF-32 string to a UTF-8 string (inner)
+ * @param[out] lpOutput Pointer to a buffer that receives the converted string
+ * @param[in] cchOutput Size, in characters, of the buffer indicated by lpOutput
+ * @param[in] lpInput Pointer to the character string to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters written to the buffer indicated by lpOutput
+ */
+static UINT ucs4toutf8(char *lpOutput, UINT cchOutput, const UINT32 *lpInput, UINT cchInput)
+{
+	UINT nRemain;
+	UINT c;
+
+	nRemain = cchOutput;
+	while((cchInput > 0) && (nRemain > 0)) {
+		c = *lpInput++;
+		cchInput--;
+
+		if(c < 0x80) {
+			nRemain--;
+			if(lpOutput) {
+				*lpOutput++ = (char)c;
+			}
+		} else if(c < 0x800) {
+			if(nRemain < 2) {
+				break;
+			}
+			nRemain -= 2;
+			if(lpOutput) {
+				*lpOutput++ = (char)(0xc0 + ((c >> 6) & 0x1f));
+				*lpOutput++ = (char)(0x80 + ((c >> 0) & 0x3f));
+			}
+		} else if(c < 0x10000) {
+			if (nRemain < 3) {
+				break;
+			}
+			nRemain -= 3;
+			if(lpOutput) {
+				*lpOutput++ = (char)(0xe0 + ((c >> 12) & 0x0f));
+				*lpOutput++ = (char)(0x80 + ((c >>  6) & 0x3f));
+				*lpOutput++ = (char)(0x80 + ((c >>  0) & 0x3f));
+			}
+		} else {
+			if(nRemain < 4) {
+				break;
+			}
+			nRemain -= 4;
+			if(lpOutput) {
+				*lpOutput++ = (char)(0xe0 + ((c >> 18) & 0x07));
+				*lpOutput++ = (char)(0x80 + ((c >> 12) & 0x3f));
+				*lpOutput++ = (char)(0x80 + ((c >>  6) & 0x3f));
+				*lpOutput++ = (char)(0x80 + ((c >>  0) & 0x3f));
+			}
+		}
+	}
+
+	return (UINT)(cchOutput - nRemain);
+}

codecnv/utf8ucs2.c

@@ -3,7 +3,11 @@
  * @brief	Implementation of converting UTF-8 to UCS2
  */
 
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
 #include "compiler.h"
+#endif
 #include "codecnv.h"
 
 static UINT utf8toucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput);
@@ -16,36 +20,28 @@ static UINT utf8toucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UI
  * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
  * @return The number of characters written to the buffer indicated by lpOutput
  */
-UINT codecnv_utf8toucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput)
-{
+UINT codecnv_utf8toucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput)	{
+	UINT n = 0;
 	UINT nLength;
 
-	if (lpInput == NULL)
-	{
-		return 0;
-	}
-
-	if (cchOutput == 0)
-	{
+	if(lpOutput != NULL && lpInput != NULL) {
+		if(cchOutput == 0) {
 			lpOutput = NULL;
 			cchOutput = (UINT)-1;
 		}
 
-	if (cchInput != (UINT)-1)
-	{
+		if(cchInput != (UINT)-1) {
 			// Binary mode
-		return utf8toucs2(lpOutput, cchOutput, lpInput, cchInput);
-	}
-	else
-	{
+			n = utf8toucs2(lpOutput, cchOutput, lpInput, cchInput);
+		} else {
 			// String mode
 			nLength = utf8toucs2(lpOutput, cchOutput - 1, lpInput, (UINT)strlen(lpInput));
-		if (lpOutput)
-		{
 			lpOutput[nLength] = '\0';
+			n = nLength + 1;
 		}
-		return nLength + 1;
 	}
+
+	return n;
 }
 
 /**
@@ -56,42 +52,24 @@ UINT codecnv_utf8toucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, U
  * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
  * @return The number of characters written to the buffer indicated by lpOutput
  */
-static UINT utf8toucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput)
-{
+static UINT utf8toucs2(UINT16 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput) {
+	UINT n, m;
 	UINT nRemain;
-	UINT c;
-	int nBits;
+	UINT32 c[2];
 
+	n = m = 1;
 	nRemain = cchOutput;
-	while ((cchInput > 0) && (nRemain > 0))
-	{
-		c = *lpInput++;
-		cchInput--;
-
-		if (c & 0x80)
-		{
-			nBits = 0;
-			while ((nBits < 6) && (c & (0x80 >> nBits)))
-			{
-				nBits++;
-			}
-
-			c &= (0x7f >> nBits);
-			nBits--;
-
-			while ((nBits > 0) && (cchInput > 0) && (((*lpInput) & 0xc0) == 0x80))
-			{
-				c = (c << 6) | ((*lpInput++) & 0x3f);
-				cchInput--;
-				nBits--;
+	while ((cchInput > 0) && (nRemain > 0) && (m > 0) && (n > 0)) {
+		n = codecnv_utf8toucs4_1(c, lpInput, cchInput);
+		if(n) {
+			c[1] = '\0';
+			m = codecnv_ucs4toucs2(lpOutput, 2, c, 1);
+			lpInput += n;
+			cchInput -= n;
+			lpOutput += m;
+			nRemain -= m;
 		}
 	}
 
-		nRemain--;
-		if (lpOutput)
-		{
-			*lpOutput++ = (UINT16)c;
-		}
-	}
 	return (UINT)(cchOutput - nRemain);
 }

codecnv/utf8ucs4.c

@@ -0,0 +1,121 @@
+/**
+ * @file	utf8ucs4.c
+ * @brief	Implementation of converting UTF-8 to UCS4
+ */
+
+#ifdef CODECNV_TEST
+#include "compiler_base.h"
+#else
+#include "compiler.h"
+#endif
+#include "codecnv.h"
+
+static UINT utf8toucs4(UINT32 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput);
+
+/**
+ * Maps a UTF-8 string to a UTF-32 string
+ * @param[out] lpOutput Pointer to a buffer that receives the converted string
+ * @param[in] cchOutput Size, in characters, of the buffer indicated by lpOutput
+ * @param[in] lpInput Pointer to the character string to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters written to the buffer indicated by lpOutput
+ */
+UINT codecnv_utf8toucs4(UINT32 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput) {
+	UINT n = 0;
+	UINT nLength;
+
+	if(lpOutput != NULL && lpInput != NULL) {
+		if(cchOutput == 0) {
+			cchOutput = (UINT)-1;
+		}
+
+		if(cchInput != (UINT)-1) {
+			// Binary mode
+			n = utf8toucs4(lpOutput, cchOutput, lpInput, cchInput);
+		} else {
+			// String mode
+			nLength = utf8toucs4(lpOutput, cchOutput - 1, lpInput, (UINT)strlen(lpInput));
+			lpOutput[nLength] = '\0';
+			n = nLength + 1;
+		}
+	}
+
+	return n;
+}
+
+/**
+ * Maps a UTF-8 char to a UTF-32 character
+ * @param[out] lpOutput Pointer to a buffer that receives the converted character
+ * @param[in] lpInput Pointer to the character to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters read to the buffer indicated by lpInput
+ */
+UINT codecnv_utf8toucs4_1(UINT32 *lpOutput, const char *lpInput, UINT cchInput) {
+	UINT c;
+	UINT n;
+
+	c = (UINT8)lpInput[0];
+	if(c < 0x80) {
+		n = 1;
+	} else if(0xC2 <= c && c < 0xE0) {
+		n = 2;
+	} else if(0xE0 <= c && c < 0xF0) {
+		n = 3;
+	} else if(0xF0 <= c && c < 0xF8) {
+		n = 4;
+	} else {
+		n = 0;
+	}
+
+	if(cchInput < n) {
+		n = 0;
+	}
+
+	switch(n) {
+	case 1:
+		lpOutput[0] = (UINT8)lpInput[0];
+		break;
+	case 2:
+		lpOutput[0]  = ((UINT8)lpInput[0] & 0x1F) << 6;
+		lpOutput[0] |=  (UINT8)lpInput[1] & 0x3F;
+		break;
+	case 3:
+		lpOutput[0]  = ((UINT8)lpInput[0] & 0x0F) << 12;
+		lpOutput[0] |= ((UINT8)lpInput[1] & 0x3F) <<  6;
+		lpOutput[0] |=  (UINT8)lpInput[2] & 0x3F;
+		break;
+	case 4:
+		lpOutput[0]  = ((UINT8)lpInput[0] & 0x07) << 18;
+		lpOutput[0] |= ((UINT8)lpInput[1] & 0x3F) << 12;
+		lpOutput[0] |= ((UINT8)lpInput[2] & 0x3F) <<  6;
+		lpOutput[0] |=  (UINT8)lpInput[3] & 0x3F;
+		break;
+	}
+
+	return n;
+}
+
+/**
+ * Maps a UTF-8 string to a UTF-32 string (inner)
+ * @param[out] lpOutput Pointer to a buffer that receives the converted string
+ * @param[in] cchOutput Size, in characters, of the buffer indicated by lpOutput
+ * @param[in] lpInput Pointer to the character string to convert
+ * @param[in] cchInput Size, in characters, of the buffer indicated by lpInput
+ * @return The number of characters written to the buffer indicated by lpOutput
+ */
+static UINT utf8toucs4(UINT32 *lpOutput, UINT cchOutput, const char *lpInput, UINT cchInput)
+{
+	UINT nRemain;
+	UINT n = 1;
+
+	nRemain = cchOutput;
+	while ((cchInput > 0) && (nRemain > 0) && (n > 0)) {
+		n = codecnv_utf8toucs4_1(lpOutput, lpInput, nRemain);
+		cchInput -= n;
+		lpInput += n;
+		lpOutput++;
+		nRemain--;
+	}
+
+	return (UINT)(cchOutput - nRemain);
+}

compiler_base.h

@@ -440,8 +440,8 @@ typedef int32_t FILELEN;
 #endif
 
 #include "common.h"
-#include "_memory.h"
-#include "rect.h"
-#include "lstarray.h"
+#include "common/_memory.h"
+#include "common/rect.h"
+#include "common/lstarray.h"
 
 #endif  // _COMPILER_BASE_H_

i386c/ia32/cpu.c

@@ -289,7 +289,6 @@ exec_allstep(void)
 	static int remclock_mul = 1000;
 	int remclockb = 0;
 	int remclkcnt = 0x100;
-	int repflag = 0;
 	static int latecount = 0;
 	static int latecount2 = 0;
 	static int hltflag = 0;
@@ -392,9 +391,6 @@ exec_allstep(void)
 		}
 
 		/* rep */
-#if defined(SUPPORT_ASYNC_CPU)
-		repflag = CPU_ECX;
-#endif
 		CPU_WORKCLOCK(5);
 	#if defined(DEBUG)
 		if (!cpu_debug_rep_cont) {

misc/test_codecnv.c

@@ -0,0 +1,102 @@
+/* === codecnv test === (c) 2020 AZO */
+// Windows(MSVC):
+//   cl test_codecnv.c -DCODECNV_TEST ../codecnv/sjisucs2.c ../codecnv/ucs2sjis.c ../codecnv/ucs2ucs4.c ../codecnv/ucs2utf8.c ../codecnv/ucs4ucs2.c ../codecnv/ucs4utf8.c ../codecnv/utf8ucs2.c ../codecnv/utf8ucs4.c -I../ -I../common -utf-8 -Wall
+// Windows(MSYS/MinGW):
+//   gcc test_codecnv.c -DCODECNV_TEST ../codecnv/sjisucs2.c ../codecnv/ucs2sjis.c ../codecnv/ucs2ucs4.c ../codecnv/ucs2utf8.c ../codecnv/ucs4ucs2.c ../codecnv/ucs4utf8.c ../codecnv/utf8ucs2.c ../codecnv/utf8ucs4.c -I../ -I../codecnv -Wall -Wextra -o test_codecnv
+// Linux:
+//   gcc test_codecnv.c -DCODECNV_TEST ../codecnv/sjisucs2.c ../codecnv/ucs2sjis.c ../codecnv/ucs2ucs4.c ../codecnv/ucs2utf8.c ../codecnv/ucs4ucs2.c ../codecnv/ucs4utf8.c ../codecnv/utf8ucs2.c ../codecnv/utf8ucs4.c -I../ -I../codecnv -Wall -Wextra -o test_codecnv
+
+#include "compiler_base.h"
+#include "codecnv.h"
+
+#define COUNT_BUFFER 256
+
+char strU8String[COUNT_BUFFER];
+char strSJString[COUNT_BUFFER];
+UINT16 su16String[COUNT_BUFFER];
+UINT32 su32String[COUNT_BUFFER];
+
+int main(int iArgc, char* strArgv[]) {
+  int i;
+  int iLen;
+  UINT uiCount;
+
+  // test000
+  printf("=== test000\n");
+  strcpy(strU8String, "0123ABCｺﾝﾆﾁﾊこんにちは今日は");
+  iLen = strlen(strU8String);
+  printf("UTF-8 length: %d\n", iLen);
+  printf("UTF-8 string: %s\n", strU8String);
+  printf("UTF-8 code: ");
+  for(i = 0; i < iLen; i++) {
+    printf(" %02X", ((UINT8*)strU8String)[i]);
+  }
+  printf("\n");
+
+  // test001 : UTF-8 to UTF-32
+  printf("=== test001 : UTF-8 to UTF-32\n");
+  uiCount = codecnv_utf8toucs4(su32String, COUNT_BUFFER, strU8String, -1);
+  iLen = codecnv_ucs4len(su32String);
+  printf("UTF-32 count: %d\n", uiCount);
+  printf("UTF-32 length: %d\n", iLen);
+  printf("UTF-32 code: ");
+  for(i = 0; i < iLen; i++) {
+    printf(" %08X", su32String[i]);
+  }
+  printf("\n");
+
+  // test002 : UTF-32 to UTF-16
+  printf("=== test002 : UTF-32 to UTF-16\n");
+  uiCount = codecnv_ucs4toucs2(su16String, COUNT_BUFFER, su32String, -1);
+  iLen = codecnv_ucs2len(su16String);
+  printf("UTF-16 count: %d\n", uiCount);
+  printf("UTF-16 length: %d\n", iLen);
+  printf("UTF-16 code: ");
+  for(i = 0; i < iLen; i++) {
+    printf(" %04X", su16String[i]);
+  }
+  printf("\n");
+
+  // test003 : UTF-32 to UTF-8
+  printf("=== test003 : UTF-32 to UTF-8\n");
+  uiCount = codecnv_ucs4toutf8(strU8String, COUNT_BUFFER, su32String, -1);
+  iLen = strlen(strU8String);
+  printf("UTF-8 count: %d\n", uiCount);
+  printf("UTF-8 length: %d\n", iLen);
+  printf("UTF-8 string: %s\n", strU8String);
+  printf("UTF-8 code: ");
+  for(i = 0; i < iLen; i++) {
+    printf(" %02X", ((UINT8*)strU8String)[i]);
+  }
+  printf("\n");
+
+  // test004 : UTF-8 to UTF-16
+  printf("=== test004 : UTF-8 to UTF-16\n");
+  uiCount = codecnv_utf8toucs2(su16String, COUNT_BUFFER, strU8String, -1);
+  iLen = codecnv_ucs2len(su16String);
+  printf("UTF-16 count: %d\n", uiCount);
+  printf("UTF-16 length: %d\n", iLen);
+  printf("UTF-16 code: ");
+  for(i = 0; i < iLen; i++) {
+    printf(" %04X", su16String[i]);
+  }
+  printf("\n");
+
+  // test005 : UTF-16 to UTF-8
+  printf("=== test005 : UTF-16 to UTF-8\n");
+  uiCount = codecnv_ucs2toutf8(strU8String, COUNT_BUFFER, su16String, -1);
+  iLen = strlen(strU8String);
+  printf("UTF-8 count: %d\n", uiCount);
+  printf("UTF-8 length: %d\n", iLen);
+  printf("UTF-8 string: %s\n", strU8String);
+  printf("UTF-8 code: ");
+  for(i = 0; i < iLen; i++) {
+    printf(" %02X", ((UINT8*)strU8String)[i]);
+  }
+  printf("\n");
+
+
+
+  return 0;
+}
+

misc/test_milstr.c

@@ -56,10 +56,11 @@ ANK=Alphabet Numeric Katakana=1Byte。
 */
 
 #include "milstr.h"
-#include <stdio.h>
 
-char strString1[1024];
-char strString2[1024];
+#define COUNT_BUFFER 256
+
+char strString1[COUNT_BUFFER];
+char strString2[COUNT_BUFFER];
 
 int main(int iArgc, char* strArgv[]) {
   // test000 : OEMSTRCPY, OEMPRINTFSTR

sdl2/gcw0/compiler.h

@@ -31,8 +31,8 @@
 
 #define	USE_SDL_JOYSTICK
 
-#include "milstr.h"
-#include "trace.h"
+#include "common/milstr.h"
+#include "sdl2/trace.h"
 
 #endif  // COMPILER_H
 

sdl2/libretro/compiler.h

@@ -25,9 +25,9 @@
 
 #define GETTICK() (cpu_features_get_time_usec() / 1000)  // millisecond timer
 
-#include "retro_inline.h"
-#include "milstr.h"
-#include "trace.h"
+//#include "retro_inline.h"
+#include "common/milstr.h"
+#include "sdl2/trace.h"
 
 #endif  // COMPILER_H
 

sdl2/unix/compiler.h

@@ -31,7 +31,7 @@
 
 #define	USE_SDL_JOYSTICK
 
-#include "milstr.h"
-#include "trace.h"
+#include "common/milstr.h"
+#include "sdl2/trace.h"
 
 #endif  // COMPILER_H

sdl2/win32/compiler.h

@@ -31,8 +31,8 @@
 
 #define	USE_SDL_JOYSTICK
 
-#include "milstr.h"
-#include "trace.h"
+#include "common/milstr.h"
+#include "sdl2/trace.h"
 
 #endif  // COMPILER_H
 

win9x/compiler.h

@@ -74,8 +74,8 @@ WINBASEAPI BOOL WINAPI SetFilePointerEx(HANDLE, LARGE_INTEGER, PLARGE_INTEGER, D
 #endif
 #endif	/* (_MSC_VER >= 1400) */
 
-#include "milstr.h"
-#include "trace.h"
+#include "common/milstr.h"
+#include "win9x/misc/trace.h"
 
 #endif  // COMPILER_H
 

win9x/dialog/winfiledlg.c

@@ -0,0 +1,115 @@
+/* === NP2 file dialog for Windows === (c) 2020 AZO */
+
+#include "winfiledlg.h"
+#include "codecnv.h"
+
+#define WINFILEDLG_GET_FLAG (OFN_FILEMUSTEXIST | OFN_HIDEREADONLY)
+#define WINFILEDLG_SET1_FLAG (OFN_FILEMUSTEXIST | OFN_HIDEREADONLY | OFN_SHAREAWARE)
+#define WINFILEDLG_SET2_FLAG (OFN_OVERWRITEPROMPT | OFN_HIDEREADONLY)
+
+BOOL WinFileDialogW_MM(
+  HWND hwnd,
+  const WinFileDialogW_Mode_t mode,
+  char* path,
+  char* name,
+  const char* title,
+  const char* filter
+) {
+  BOOL res;
+  wchar_t wpath[MAX_PATH];
+  wchar_t wname[MAX_PATH];
+
+  res = WinFileDialogW_MW(hwnd, mode, wpath, wname, title, filter);
+
+  codecnv_ucs2toutf8(path, MAX_PATH, wpath, MAX_PATH);
+  codecnv_ucs2toutf8(name, MAX_PATH, wname, MAX_PATH);
+
+  return res;
+}
+
+BOOL WinFileDialogW_MW(
+  HWND hwnd,
+  const WinFileDialogW_Mode_t mode,
+  wchar_t* path,
+  wchar_t* name,
+  const char* title,
+  const char* filter
+) {
+  wchar_t wtitle[256];
+  wchar_t wfilter[MAX_PATH];
+
+  codecnv_utf8toucs2(wtitle, 256, title, 256);
+  codecnv_utf8toucs2(wfilter, MAX_PATH, filter, MAX_PATH);
+
+  return WinFileDialogW_WW(hwnd, mode, wpath, wname, title, filter);
+}
+
+BOOL WinFileDialogW_WM(
+  HWND hwnd,
+  const WinFileDialogW_Mode_t mode,
+  char* path,
+  char* name,
+  const wchar_t* title,
+  const wchar_t* filter
+) {
+  BOOL res;
+  wchar_t wpath[MAX_PATH];
+  wchar_t wname[MAX_PATH];
+
+  res = WinFileDialogW_WW(hwnd, mode, wpath, wname, title, filter);
+
+  codecnv_ucs2toutf8(path, MAX_PATH, wpath, MAX_PATH);
+  codecnv_ucs2toutf8(name, MAX_PATH, wname, MAX_PATH);
+
+  return res;
+}
+
+BOOL WinFileDialogW_WW(
+  HWND hwnd,
+  const WinFileDialogW_Mode_t mode,
+  wchar_t* path,
+  wchar_t* name,
+  const wchar_t* title,
+  const wchar_t* filter
+) {
+  BOOL res = TRUE;
+  OPENFILENAMEW ofnw;
+
+  path[0] = L'\0';
+  name[0] = L'\0';
+
+  memset(&ofnw, 0, sizeof(OPENFILENAMEW));
+
+  ofnw.lStructSize = sizeof(OPENFILENAMEW);
+  ofnw.hwndOwner = hwnd;
+  ofnw.lpstrFile = path;
+  ofnw.nMaxFile = MAX_PATH;
+  ofnw.lpstrFileTitle = name;
+  ofnw.nMaxFileTitle = MAX_PATH;
+  ofnw.lpstrFilter = filter;
+  ofnw.lpstrTitle = title;
+
+  switch(mode) {
+  case WINFILEDIALOGW_MODE_GET:
+    ofnw.Flags = WINFILEDLG_GET_FLAG;
+    if(GetOpenFileNameW(&ofnw) == FALSE) 
+      res = FALSE;
+    break;
+  case WINFILEDIALOGW_MODE_SET1:
+    ofnw.Flags = WINFILEDLG_SET1_FLAG;
+    if(GetSaveFileNameW(&ofnw) == FALSE) 
+      res = FALSE;
+    break;
+  case WINFILEDIALOGW_MODE_SET2:
+    ofnw.Flags = WINFILEDLG_SET2_FLAG;
+    if(GetSaveFileNameW(&ofnw) == FALSE) 
+      res = FALSE;
+    break;
+  default:
+    res = FALSE;
+    break;
+  }
+ 
+  return res;
+}
+

win9x/dialog/winfiledlg.h

@@ -0,0 +1,64 @@
+/* === NP2 file dialog for Windows === (c) 2020 AZO */
+
+#ifndef _WINFILEDLG_H_
+#define _WINFILEDLG_H_
+
+#include "compiler.h"
+
+#define WinFileDialogW(h, m, p, n, t, f) WinFileDialogW_MM(h, m, p, n, t, f)
+
+typedef enum {
+  WINFILEDIALOGW_MODE_GET = 0,
+  WINFILEDIALOGW_MODE_SET1,
+  WINFILEDIALOGW_MODE_SET2,
+} WinFileDialogW_Mode_t;
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+// [I] mode : Get(Exist file) or Set1(Overwrite file) or New(New or Overwrite)
+// [O] path : Full path of file
+// [O] name : Filename
+// [I] title : Title of dialog
+// [I] name : filename
+
+BOOL WinFileDialogW_MM(
+  HWND hwnd,
+  const WinFileDialogW_Mode_t mode,
+  char* path,
+  char* name,
+  const char* title,
+  const char* filter
+);
+BOOL WinFileDialogW_MW(
+  HWND hwnd,
+  const WinFileDialogW_Mode_t mode,
+  wchar_t* path,
+  wchar_t* name,
+  const char* title,
+  const char* filter
+);
+BOOL WinFileDialogW_WM(
+  HWND hwnd,
+  const WinFileDialogW_Mode_t mode,
+  char* path,
+  char* name,
+  const wchar_t* title,
+  const wchar_t* filter
+);
+BOOL WinFileDialogW_WW(
+  HWND hwnd,
+  const WinFileDialogW_Mode_t mode,
+  wchar_t* path,
+  wchar_t* name,
+  const wchar_t* title,
+  const wchar_t* filter
+);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif  // _WINFILEDLG_H_
+

x11/compiler.h

@@ -112,8 +112,8 @@ G_END_DECLS
 extern char timidity_cfgfile_path[MAX_PATH];
 #define	TIMIDITY_CFGFILE	timidity_cfgfile_path
 
-#include "milstr.h"
-#include "trace.h"
-#include "toolkit.h"
+#include "common/milstr.h"
+#include "x11/trace.h"
+#include "x11/toolkit.h"
 
 #endif	/* NP2_X11_COMPILER_H__ */