mirrors
/
FreeRDP
mirror of https://github.com/FreeRDP/FreeRDP
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

[winpr,crt] Added unicode conversion backends 

* Support apple with NSString
* Support android via JNI
* Split ICU to own file
This commit is contained in:
akallabeth 2022-11-15 08:27:31 +01:00 committed by akallabeth
parent 00ee213f97
commit 5ae159303f
10 changed files with 634 additions and 1213 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
winpr/CMakeLists.txt
View File

@ -64,7 +64,6 @@ option(WITH_DEBUG_SYMBOLS "Pack debug symbols to installer" OFF)
option(WITH_NATIVE_SSPI "Use native SSPI modules" ${NATIVE_SSPI})
option(WITH_SMARTCARD_INSPECT "Enable SmartCard API Inspector" OFF)
option(WITH_DEBUG_MUTEX "Print mutex debug messages" ${DEFAULT_DEBUG_OPTION})
option(WITH_ICU "Use ICU for unicode conversion" OFF)
option(WITH_GSSAPI "Compile support for kerberos authentication. (EXPERIMENTAL)" OFF)
option(WITH_INTERNAL_MD4 "Use compiled in md4 hash functions instead of OpenSSL/MBedTLS" OFF)
option(WITH_INTERNAL_MD5 "Use compiled in md5 hash functions instead of OpenSSL/MBedTLS" OFF)

1
winpr/include/config/config.h.in
View File

@ -27,7 +27,6 @@
#cmakedefine WITH_EVENTFD_READ_WRITE
#cmakedefine WITH_NATIVE_SSPI
#cmakedefine WITH_ICU
#cmakedefine WITH_SPNEGO
#cmakedefine WITH_INTERNAL_MD4
#cmakedefine WITH_INTERNAL_MD5

17
winpr/libwinpr/crt/CMakeLists.txt
View File

@ -22,17 +22,18 @@ set (CRT_FILES alignment.c
unicode.c
string.c)
if (NOT WITH_ICU)
set (CRT_FILES ${CRT_FILES}
utf.c
utf.h)
endif(NOT WITH_ICU)
if (WITH_ICU)
IF (ANDROID)
list(APPEND CRT_FILES unicode_android.c)
elseif (NOT APPLE AND NOT WIN32)
find_package(ICU REQUIRED i18n uc io)
list(APPEND CRT_FILES unicode_icu.c)
winpr_include_directory_add(${ICU_INCLUDE_DIRS})
winpr_library_add_private(${ICU_LIBRARIES})
endif (WITH_ICU)
elseif(APPLE)
list(APPEND CRT_FILES unicode_apple.m)
find_library(FOUNDATION_FRAMEWORK Foundation REQUIRED)
winpr_library_add_private(${FOUNDATION_FRAMEWORK})
endif()
winpr_module_add(${CRT_FILES})

182
winpr/libwinpr/crt/unicode.c
View File

@ -3,6 +3,8 @@
* Unicode Conversion (CRT)
*
* Copyright 2012 Marc-Andre Moreau <marcandre.moreau@gmail.com>
* Copyright 2022 Armin Novak <anovak@thincast.com>
* Copyright 2022 Thincast Technologies GmbH
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
@ -18,6 +20,7 @@
*/
#include <winpr/config.h>
#include <winpr/assert.h>
#include <errno.h>
#include <wctype.h>
@ -26,14 +29,13 @@
#include <winpr/error.h>
#include <winpr/print.h>
#ifndef MIN
#define MIN(a, b) (a) < (b) ? (a) : (b)
#endif
#ifndef _WIN32
#if defined(WITH_ICU)
#include <unicode/ucnv.h>
#include <unicode/ustring.h>
#else
#include "utf.h"
#endif
#include "unicode.h"
#include "../log.h"
#define TAG WINPR_TAG("unicode")
@ -152,87 +154,8 @@
int MultiByteToWideChar(UINT CodePage, DWORD dwFlags, LPCSTR lpMultiByteStr, int cbMultiByte,
LPWSTR lpWideCharStr, int cchWideChar)
{
LPWSTR targetStart;
#if !defined(WITH_ICU)
const BYTE* sourceStart;
int length;
ConversionResult result;
#endif
/* If cbMultiByte is 0, the function fails */
if ((cbMultiByte == 0) || (cbMultiByte < -1))
return 0;
/* If cbMultiByte is -1, the string is null-terminated */
if (cbMultiByte == -1)
{
size_t len = strnlen((const char*)lpMultiByteStr, INT32_MAX);
if (len >= INT32_MAX)
return 0;
cbMultiByte = (int)len + 1;
}
/*
* if cchWideChar is 0, the function returns the required buffer size
* in characters for lpWideCharStr and makes no use of the output parameter itself.
*/
#if defined(WITH_ICU)
{
UErrorCode error;
int32_t targetLength;
int32_t targetCapacity;
switch (CodePage)
{
case CP_ACP:
case CP_UTF8:
break;
default:
WLog_ERR(TAG, "Unsupported encoding %u", CodePage);
return 0;
}
targetStart = lpWideCharStr;
targetCapacity = cchWideChar;
error = U_ZERO_ERROR;
if (cchWideChar == 0)
{
u_strFromUTF8(NULL, 0, &targetLength, lpMultiByteStr, cbMultiByte, &error);
cchWideChar = targetLength;
}
else
{
u_strFromUTF8(targetStart, targetCapacity, &targetLength, lpMultiByteStr, cbMultiByte,
&error);
cchWideChar = U_SUCCESS(error) ? targetLength : 0;
}
}
#else
if (cchWideChar == 0)
{
sourceStart = (const BYTE*)lpMultiByteStr;
targetStart = (WCHAR*)NULL;
result = ConvertUTF8toUTF16(&sourceStart, &sourceStart[cbMultiByte], &targetStart, NULL,
strictConversion);
length = targetStart - ((WCHAR*)NULL);
}
else
{
sourceStart = (const BYTE*)lpMultiByteStr;
targetStart = lpWideCharStr;
result = ConvertUTF8toUTF16(&sourceStart, &sourceStart[cbMultiByte], &targetStart,
&targetStart[cchWideChar], strictConversion);
length = targetStart - ((WCHAR*)lpWideCharStr);
}
cchWideChar = (result == conversionOK) ? length : 0;
#endif
return cchWideChar;
return int_MultiByteToWideChar(CodePage, dwFlags, lpMultiByteStr, cbMultiByte, lpWideCharStr,
cchWideChar);
}
/*
@ -276,89 +199,8 @@ int WideCharToMultiByte(UINT CodePage, DWORD dwFlags, LPCWSTR lpWideCharStr, int
LPSTR lpMultiByteStr, int cbMultiByte, LPCSTR lpDefaultChar,
LPBOOL lpUsedDefaultChar)
{
#if !defined(WITH_ICU)
int length;
const WCHAR* sourceStart;
ConversionResult result;
BYTE* targetStart;
#else
char* targetStart;
#endif
/* If cchWideChar is 0, the function fails */
if ((cchWideChar == 0) || (cchWideChar < -1))
return 0;
/* If cchWideChar is -1, the string is null-terminated */
if (cchWideChar == -1)
{
size_t len = _wcslen(lpWideCharStr);
if (len >= INT32_MAX)
return 0;
cchWideChar = (int)len + 1;
}
/*
* if cbMultiByte is 0, the function returns the required buffer size
* in bytes for lpMultiByteStr and makes no use of the output parameter itself.
*/
#if defined(WITH_ICU)
{
UErrorCode error;
int32_t targetLength;
int32_t targetCapacity;
switch (CodePage)
{
case CP_ACP:
case CP_UTF8:
break;
default:
WLog_ERR(TAG, "Unsupported encoding %u", CodePage);
return 0;
}
targetStart = lpMultiByteStr;
targetCapacity = cbMultiByte;
error = U_ZERO_ERROR;
if (cbMultiByte == 0)
{
u_strToUTF8(NULL, 0, &targetLength, lpWideCharStr, cchWideChar, &error);
cbMultiByte = targetLength;
}
else
{
u_strToUTF8(targetStart, targetCapacity, &targetLength, lpWideCharStr, cchWideChar,
&error);
cbMultiByte = U_SUCCESS(error) ? targetLength : 0;
}
}
#else
if (cbMultiByte == 0)
{
sourceStart = (const WCHAR*)lpWideCharStr;
targetStart = (BYTE*)NULL;
result = ConvertUTF16toUTF8(&sourceStart, &sourceStart[cchWideChar], &targetStart, NULL,
strictConversion);
length = targetStart - ((BYTE*)NULL);
}
else
{
sourceStart = (const WCHAR*)lpWideCharStr;
targetStart = (BYTE*)lpMultiByteStr;
result = ConvertUTF16toUTF8(&sourceStart, &sourceStart[cchWideChar], &targetStart,
&targetStart[cbMultiByte], strictConversion);
length = targetStart - ((BYTE*)lpMultiByteStr);
}
cbMultiByte = (result == conversionOK) ? length : 0;
#endif
return cbMultiByte;
return int_WideCharToMultiByte(CodePage, dwFlags, lpWideCharStr, cchWideChar, lpMultiByteStr,
cbMultiByte, lpDefaultChar, lpUsedDefaultChar);
}
#endif

32
winpr/libwinpr/crt/unicode.h Normal file
View File

@ -0,0 +1,32 @@
/**
* WinPR: Windows Portable Runtime
* Unicode Conversion (CRT)
*
* Copyright 2022 Armin Novak <anovak@thincast.com>
* Copyright 2022 Thincast Technologies GmbH
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#ifndef WINPR_CRT_UNICODE_INTERNAL
#define WINPR_CRT_UNICODE_INTERNAL
#include <winpr/wtypes.h>
int int_MultiByteToWideChar(UINT CodePage, DWORD dwFlags, LPCSTR lpMultiByteStr, int cbMultiByte,
LPWSTR lpWideCharStr, int cchWideChar);
int int_WideCharToMultiByte(UINT CodePage, DWORD dwFlags, LPCWSTR lpWideCharStr, int cchWideChar,
LPSTR lpMultiByteStr, int cbMultiByte, LPCSTR lpDefaultChar,
LPBOOL lpUsedDefaultChar);
#endif

188
winpr/libwinpr/crt/unicode_android.c Normal file
View File

@ -0,0 +1,188 @@
/**
* WinPR: Windows Portable Runtime
* Unicode Conversion (CRT)
*
* Copyright 2022 Armin Novak <anovak@thincast.com>
* Copyright 2022 Thincast Technologies GmbH
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include <winpr/config.h>
#include <winpr/assert.h>
#include <winpr/string.h>
#include "../utils/android.h"
#ifndef MIN
#define MIN(a, b) (a) < (b) ? (a) : (b)
#endif
#include "../log.h"
#define TAG WINPR_TAG("unicode")
static int convert_int(JNIEnv* env, const void* data, size_t size, void* buffer, size_t buffersize,
BOOL toUTF16)
{
WINPR_ASSERT(env);
WINPR_ASSERT(data || (size == 0));
WINPR_ASSERT(buffer || (buffersize == 0));
jstring utf8 = (*env)->NewStringUTF(env, "UTF-8");
jstring utf16 = (*env)->NewStringUTF(env, "UTF-16LE");
jclass stringClass = (*env)->FindClass(env, "java/lang/String");
if (!utf8 || !utf16 || !stringClass)
{
WLog_ERR(TAG, "[%s] utf8-%p, utf16=%p, stringClass=%p", __func__, utf8, utf16, stringClass);
return -1;
}
jmethodID constructorID =
(*env)->GetMethodID(env, stringClass, "<init>", "([BLjava/lang/String;)V");
jmethodID getBytesID =
(*env)->GetMethodID(env, stringClass, "getBytes", "(Ljava/lang/String;)[B");
if (!constructorID || !getBytesID)
{
WLog_ERR(TAG, "[%s] constructorID=%p, getBytesID=%p", __func__, constructorID, getBytesID);
return -2;
}
jbyteArray ret = (*env)->NewByteArray(env, size);
if (!ret)
{
WLog_ERR(TAG, "[%s] NewByteArray(%" PRIuz ") failed", __func__, size);
return -3;
}
(*env)->SetByteArrayRegion(env, ret, 0, size, data);
jobject obj = (*env)->NewObject(env, stringClass, constructorID, ret, toUTF16 ? utf8 : utf16);
if (!obj)
{
WLog_ERR(TAG, "[%s] NewObject(String, byteArray, UTF-%d) failed", __func__,
toUTF16 ? 16 : 8);
return -4;
}
jbyteArray res = (*env)->CallObjectMethod(env, obj, getBytesID, toUTF16 ? utf16 : utf8);
if (!res)
{
WLog_ERR(TAG, "[%s] CallObjectMethod(String, getBytes, UTF-%d) failed", __func__,
toUTF16 ? 16 : 8);
return -4;
}
jsize rlen = (*env)->GetArrayLength(env, res);
if (buffersize > 0)
{
rlen = MIN(rlen, buffersize);
(*env)->GetByteArrayRegion(env, res, 0, rlen, buffer);
}
if (toUTF16)
rlen /= sizeof(WCHAR);
return rlen;
}
static int convert(const void* data, size_t size, void* buffer, size_t buffersize, BOOL toUTF16)
{
int rc;
JNIEnv* env = NULL;
jboolean attached = winpr_jni_attach_thread(&env);
rc = convert_int(env, data, size, buffer, buffersize, toUTF16);
if (attached)
winpr_jni_detach_thread();
return rc;
}
int int_MultiByteToWideChar(UINT CodePage, DWORD dwFlags, LPCSTR lpMultiByteStr, int cbMultiByte,
LPWSTR lpWideCharStr, int cchWideChar)
{
size_t cbCharLen = (size_t)cbMultiByte;
WINPR_UNUSED(dwFlags);
/* If cbMultiByte is 0, the function fails */
if ((cbMultiByte == 0) || (cbMultiByte < -1))
return 0;
if (cchWideChar < 0)
return -1;
if (cbMultiByte < 0)
{
const size_t len = strlen(lpMultiByteStr);
if (len >= INT32_MAX)
return 0;
cbCharLen = (int)len + 1;
}
else
{
cbCharLen = strnlen(lpMultiByteStr, cbMultiByte);
if (cbCharLen < cbMultiByte)
cbCharLen++;
}
WINPR_ASSERT(lpMultiByteStr);
switch (CodePage)
{
case CP_ACP:
case CP_UTF8:
break;
default:
WLog_ERR(TAG, "Unsupported encoding %u", CodePage);
return 0;
}
return convert(lpMultiByteStr, cbCharLen, lpWideCharStr, cchWideChar * sizeof(WCHAR), TRUE);
}
int int_WideCharToMultiByte(UINT CodePage, DWORD dwFlags, LPCWSTR lpWideCharStr, int cchWideChar,
LPSTR lpMultiByteStr, int cbMultiByte, LPCSTR lpDefaultChar,
LPBOOL lpUsedDefaultChar)
{
size_t cbCharLen = (size_t)cchWideChar;
WINPR_UNUSED(dwFlags);
/* If cchWideChar is 0, the function fails */
if ((cchWideChar == 0) || (cchWideChar < -1))
return 0;
if (cbMultiByte < 0)
return -1;
WINPR_ASSERT(lpWideCharStr);
/* If cchWideChar is -1, the string is null-terminated */
if (cchWideChar == -1)
{
const size_t len = _wcslen(lpWideCharStr);
if (len >= INT32_MAX)
return 0;
cbCharLen = (int)len + 1;
}
else
{
cbCharLen = (int)_wcsnlen(lpWideCharStr, cchWideChar);
if (cbCharLen < cchWideChar)
cbCharLen++;
}
/*
* if cbMultiByte is 0, the function returns the required buffer size
* in bytes for lpMultiByteStr and makes no use of the output parameter itself.
*/
return convert(lpWideCharStr, cbCharLen * sizeof(WCHAR), lpMultiByteStr, cbMultiByte, FALSE);
}

169
winpr/libwinpr/crt/unicode_apple.m Normal file
View File

@ -0,0 +1,169 @@
/**
* WinPR: Windows Portable Runtime
* Unicode Conversion (CRT)
*
* Copyright 2022 Armin Novak <anovak@thincast.com>
* Copyright 2022 Thincast Technologies GmbH
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#import <Foundation/Foundation.h>
#include <winpr/config.h>
#include <winpr/assert.h>
#include <errno.h>
#include <wctype.h>
#include <winpr/crt.h>
#include <winpr/error.h>
#include <winpr/print.h>
#ifndef MIN
#define MIN(a, b) (a) < (b) ? (a) : (b)
#endif
#include "../log.h"
#define TAG WINPR_TAG("unicode")
int int_MultiByteToWideChar(UINT CodePage, DWORD dwFlags, LPCSTR lpMultiByteStr, int cbMultiByte,
LPWSTR lpWideCharStr, int cchWideChar)
{
BOOL addNullTerminator = FALSE;
/* If cbMultiByte is 0, the function fails */
if ((cbMultiByte == 0) || (cbMultiByte < -1))
return 0;
/* If cbMultiByte is -1, the string is null-terminated */
if (cbMultiByte == -1)
{
size_t len = strnlen((const char *)lpMultiByteStr, INT32_MAX);
if (len >= INT32_MAX)
return 0;
cbMultiByte = (int)len;
addNullTerminator = TRUE;
}
else
{
const size_t len = strnlen(lpMultiByteStr, (size_t)cbMultiByte);
addNullTerminator = len < (size_t)cbMultiByte; /* If len == cbMultiByte no '\0' was found */
cbMultiByte = (int)len;
if (addNullTerminator)
cbMultiByte++;
}
NSString *utf = [[NSString alloc] initWithBytes:lpMultiByteStr
length:cbMultiByte
encoding:NSUTF8StringEncoding];
if (!utf)
{
WLog_WARN(TAG, "[NSString alloc] NSUTF8StringEncoding failed [%d] '%s'", cbMultiByte,
lpMultiByteStr);
return -1;
}
const WCHAR *utf16 =
(const WCHAR *)[utf cStringUsingEncoding:NSUTF16LittleEndianStringEncoding];
const size_t utf16ByteLen = [utf lengthOfBytesUsingEncoding:NSUTF16LittleEndianStringEncoding];
const size_t utf16CharLen = utf16ByteLen / sizeof(WCHAR);
if (!utf16)
{
WLog_WARN(TAG, "[utf cStringUsingEncoding:NSUTF16LittleEndianStringEncoding] failed");
return -1;
}
if (cchWideChar == 0)
{
cchWideChar = _wcsnlen(utf16, utf16CharLen);
if (addNullTerminator)
cchWideChar++;
}
else
{
const size_t len = _wcsnlen(utf16, MIN(utf16CharLen, (size_t)cchWideChar));
memcpy(lpWideCharStr, utf16, len * sizeof(WCHAR));
if ((len < (size_t)cchWideChar) && (len > 0) && (lpWideCharStr[len - 1] != '\0'))
lpWideCharStr[len] = '\0';
if (addNullTerminator && (len < cchWideChar))
cchWideChar = (int)len + 1;
else
cchWideChar = (int)len;
}
return cchWideChar;
}
int int_WideCharToMultiByte(UINT CodePage, DWORD dwFlags, LPCWSTR lpWideCharStr, int cchWideChar,
LPSTR lpMultiByteStr, int cbMultiByte, LPCSTR lpDefaultChar,
LPBOOL lpUsedDefaultChar)
{
BOOL addNullTerminator = FALSE;
/* If cchWideChar is 0, the function fails */
if ((cchWideChar == 0) || (cchWideChar < -1))
return 0;
/* If cchWideChar is -1, the string is null-terminated */
if (cchWideChar == -1)
{
size_t len = _wcslen(lpWideCharStr);
if (len >= INT32_MAX)
return 0;
cchWideChar = (int)len;
addNullTerminator = TRUE;
}
else
{
const size_t len = _wcsnlen(lpWideCharStr, (size_t)cchWideChar);
addNullTerminator = len < (size_t)cchWideChar; /* If len == cchWideChar no '\0' was found */
cchWideChar = (int)len;
if (addNullTerminator)
cchWideChar++;
}
NSString *utf = [[NSString alloc] initWithCharacters:lpWideCharStr length:cchWideChar];
if (!utf)
{
WLog_WARN(TAG, "[NSString alloc] initWithCharacters failed [%d] 'XXX'", cchWideChar);
return -1;
}
const char *utf8 = [utf cStringUsingEncoding:NSUTF8StringEncoding];
const size_t utf8Len = [utf lengthOfBytesUsingEncoding:NSUTF8StringEncoding];
if (!utf8)
{
WLog_WARN(TAG, "[utf cStringUsingEncoding:NSUTF8StringEncoding] failed");
return -1;
}
if (cbMultiByte == 0)
{
cbMultiByte = strnlen(utf8, utf8Len);
if (addNullTerminator)
cbMultiByte++;
}
else
{
const size_t len = strnlen(utf8, MIN((size_t)cbMultiByte, utf8Len));
memcpy(lpMultiByteStr, utf8, len * sizeof(char));
if ((len < (size_t)cbMultiByte) && (len > 0) && (lpMultiByteStr[len - 1] != '\0'))
lpMultiByteStr[len] = '\0';
if (addNullTerminator && (len < cbMultiByte))
cbMultiByte = (int)len + 1;
else
cbMultiByte = (int)len;
}
return cbMultiByte;
}

224
winpr/libwinpr/crt/unicode_icu.c Normal file
View File

@ -0,0 +1,224 @@
/**
* WinPR: Windows Portable Runtime
* Unicode Conversion (CRT)
*
* Copyright 2012 Marc-Andre Moreau <marcandre.moreau@gmail.com>
* Copyright 2022 Armin Novak <anovak@thincast.com>
* Copyright 2022 Thincast Technologies GmbH
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include <winpr/config.h>
#include <winpr/assert.h>
#include <errno.h>
#include <wctype.h>
#include <winpr/crt.h>
#include <winpr/error.h>
#include <winpr/print.h>
#ifndef MIN
#define MIN(a, b) (a) < (b) ? (a) : (b)
#endif
#include <unicode/ucnv.h>
#include <unicode/ustring.h>
#include "unicode.h"
#include "../log.h"
#define TAG WINPR_TAG("unicode")
int int_MultiByteToWideChar(UINT CodePage, DWORD dwFlags, LPCSTR lpMultiByteStr, int cbMultiByte,
LPWSTR lpWideCharStr, int cchWideChar)
{
BOOL isNullTerminated = TRUE;
LPWSTR targetStart;
WINPR_UNUSED(dwFlags);
/* If cbMultiByte is 0, the function fails */
if ((cbMultiByte == 0) || (cbMultiByte < -1))
return 0;
size_t len;
if (cbMultiByte == -1)
len = strlen(lpMultiByteStr) + 1;
else
{
len = strnlen(lpMultiByteStr, cbMultiByte);
isNullTerminated = (len < cbMultiByte);
if (isNullTerminated)
len++;
}
if (len >= INT_MAX)
return -1;
cbMultiByte = (int)len;
/*
* if cchWideChar is 0, the function returns the required buffer size
* in characters for lpWideCharStr and makes no use of the output parameter itself.
*/
{
UErrorCode error;
int32_t targetLength;
int32_t targetCapacity;
switch (CodePage)
{
case CP_ACP:
case CP_UTF8:
break;
default:
WLog_ERR(TAG, "Unsupported encoding %u", CodePage);
return 0;
}
targetStart = lpWideCharStr;
targetCapacity = cchWideChar;
error = U_ZERO_ERROR;
u_strFromUTF8(targetStart, targetCapacity, &targetLength, lpMultiByteStr, cbMultiByte,
&error);
switch (error)
{
case U_BUFFER_OVERFLOW_ERROR:
if (targetCapacity > 0)
{
cchWideChar = 0;
WLog_ERR(TAG, "[%s] insufficient buffer supplied, got %d, required %d",
__FUNCTION__, targetCapacity, targetLength);
SetLastError(ERROR_INSUFFICIENT_BUFFER);
}
else
cchWideChar = targetLength;
break;
case U_STRING_NOT_TERMINATED_WARNING:
cchWideChar = targetLength;
break;
case U_ZERO_ERROR:
cchWideChar = targetLength;
break;
default:
WLog_WARN(TAG, "[%s] unexpected ICU error code %s [0x%08" PRIx32 "]", __func__,
u_errorName(error), error);
if (U_FAILURE(error))
{
WLog_ERR(TAG, "[%s] unexpected ICU error code %s [0x%08" PRIx32 "] is fatal",
__func__, u_errorName(error), error);
cchWideChar = 0;
SetLastError(ERROR_NO_UNICODE_TRANSLATION);
}
else
cchWideChar = targetLength;
break;
}
}
return cchWideChar;
}
int int_WideCharToMultiByte(UINT CodePage, DWORD dwFlags, LPCWSTR lpWideCharStr, int cchWideChar,
LPSTR lpMultiByteStr, int cbMultiByte, LPCSTR lpDefaultChar,
LPBOOL lpUsedDefaultChar)
{
BOOL isNullTerminated = TRUE;
char* targetStart;
/* If cchWideChar is 0, the function fails */
if ((cchWideChar == 0) || (cchWideChar < -1))
return 0;
/* If cchWideChar is -1, the string is null-terminated */
size_t len;
if (cchWideChar == -1)
len = _wcslen(lpWideCharStr) + 1;
else
{
len = _wcsnlen(lpWideCharStr, cchWideChar);
isNullTerminated = len < cchWideChar;
if (isNullTerminated)
len++;
}
if (len >= INT32_MAX)
return 0;
cchWideChar = (int)len;
/*
* if cbMultiByte is 0, the function returns the required buffer size
* in bytes for lpMultiByteStr and makes no use of the output parameter itself.
*/
{
UErrorCode error;
int32_t targetLength;
int32_t targetCapacity;
switch (CodePage)
{
case CP_ACP:
case CP_UTF8:
break;
default:
WLog_ERR(TAG, "Unsupported encoding %u", CodePage);
return 0;
}
targetStart = lpMultiByteStr;
targetCapacity = cbMultiByte;
error = U_ZERO_ERROR;
u_strToUTF8(targetStart, targetCapacity, &targetLength, lpWideCharStr, cchWideChar, &error);
switch (error)
{
case U_BUFFER_OVERFLOW_ERROR:
if (targetCapacity > 0)
{
WLog_ERR(TAG, "[%s] insufficient buffer supplied, got %d, required %d",
__FUNCTION__, targetCapacity, targetLength);
cbMultiByte = 0;
SetLastError(ERROR_INSUFFICIENT_BUFFER);
}
else
cbMultiByte = targetLength;
break;
case U_STRING_NOT_TERMINATED_WARNING:
cbMultiByte = targetLength;
break;
case U_ZERO_ERROR:
cbMultiByte = targetLength;
break;
default:
WLog_WARN(TAG, "[%s] unexpected ICU error code %s [0x%08" PRIx32 "]", __func__,
u_errorName(error), error);
if (U_FAILURE(error))
{
WLog_ERR(TAG, "[%s] unexpected ICU error code %s [0x%08" PRIx32 "] is fatal",
__func__, u_errorName(error), error);
cbMultiByte = 0;
SetLastError(ERROR_NO_UNICODE_TRANSLATION);
}
else
cbMultiByte = targetLength;
break;
}
}
return cbMultiByte;
}

883
winpr/libwinpr/crt/utf.c
View File

@ -1,883 +0,0 @@
/*
* Copyright 2001-2004 Unicode, Inc.
*
* Disclaimer
*
* This source code is provided as is by Unicode, Inc. No claims are
* made as to fitness for any particular purpose. No warranties of any
* kind are expressed or implied. The recipient agrees to determine
* applicability of information provided. If this file has been
* purchased on magnetic or optical media from Unicode, Inc., the
* sole remedy for any claim will be exchange of defective media
* within 90 days of receipt.
*
* Limitations on Rights to Redistribute This Code
*
* Unicode, Inc. hereby grants the right to freely use the information
* supplied in this file in the creation of products supporting the
* Unicode Standard, and to make copies of this file in any form
* for internal or external distribution as long as this notice
* remains attached.
*/
/* ---------------------------------------------------------------------
Conversions between UTF32, UTF-16, and UTF-8. Source code file.
Author: Mark E. Davis, 1994.
Rev History: Rick McGowan, fixes & updates May 2001.
Sept 2001: fixed const & error conditions per
mods suggested by S. Parent & A. Lillich.
June 2002: Tim Dodd added detection and handling of incomplete
source sequences, enhanced error detection, added casts
to eliminate compiler warnings.
July 2003: slight mods to back out aggressive FFFE detection.
Jan 2004: updated switches in from-UTF8 conversions.
Oct 2004: updated to use UNI_MAX_LEGAL_UTF32 in UTF-32 conversions.
See the header file "utf.h" for complete documentation.
------------------------------------------------------------------------ */
#include "utf.h"
#include <winpr/endian.h>
#if __GNUC__
#pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wimplicit-fallthrough"
#endif
static const int halfShift = 10; /* used for shifting by 10 bits */
static const DWORD halfBase = 0x0010000UL;
static const DWORD halfMask = 0x3FFUL;
#define UNI_SUR_HIGH_START (DWORD)0xD800
#define UNI_SUR_HIGH_END (DWORD)0xDBFF
#define UNI_SUR_LOW_START (DWORD)0xDC00
#define UNI_SUR_LOW_END (DWORD)0xDFFF
/* --------------------------------------------------------------------- */
ConversionResult ConvertUTF32toUTF16(const DWORD** sourceStart, const DWORD* sourceEnd,
WCHAR** targetStart, WCHAR* targetEnd, ConversionFlags flags)
{
ConversionResult result = conversionOK;
const DWORD* source = *sourceStart;
WCHAR* target = *targetStart;
while (source < sourceEnd)
{
DWORD ch;
if (target >= targetEnd)
{
result = targetExhausted;
break;
}
ch = *source++;
if (ch <= UNI_MAX_BMP) /* Target is a character <= 0xFFFF */
{
/* UTF-16 surrogate values are illegal in UTF-32; 0xffff or 0xfffe are both reserved
* values */
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END)
{
if (flags == strictConversion)
{
--source; /* return to the illegal value itself */
result = sourceIllegal;
break;
}
else
{
*target++ = UNI_REPLACEMENT_CHAR;
}
}
else
{
*target++ = (WCHAR)ch; /* normal case */
}
}
else if (ch > UNI_MAX_LEGAL_UTF32)
{
if (flags == strictConversion)
{
result = sourceIllegal;
}
else
{
*target++ = UNI_REPLACEMENT_CHAR;
}
}
else
{
/* target is a character in range 0xFFFF - 0x10FFFF. */
if (target + 1 >= targetEnd)
{
--source; /* Back up source pointer! */
result = targetExhausted;
break;
}
ch -= halfBase;
*target++ = (WCHAR)((ch >> halfShift) + UNI_SUR_HIGH_START);
*target++ = (WCHAR)((ch & halfMask) + UNI_SUR_LOW_START);
}
}
*sourceStart = source;
*targetStart = target;
return result;
}
/* --------------------------------------------------------------------- */
ConversionResult ConvertUTF16toUTF32(const WCHAR** sourceStart, const WCHAR* sourceEnd,
DWORD** targetStart, DWORD* targetEnd, ConversionFlags flags)
{
ConversionResult result = conversionOK;
const WCHAR* source = *sourceStart;
DWORD* target = *targetStart;
DWORD ch, ch2;
while (source < sourceEnd)
{
const WCHAR* oldSource =
source; /* In case we have to back up because of target overflow. */
ch = *source++;
/* If we have a surrogate pair, convert to UTF32 first. */
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END)
{
/* If the 16 bits following the high surrogate are in the source buffer... */
if (source < sourceEnd)
{
ch2 = *source;
/* If it's a low surrogate, convert to UTF32. */
if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END)
{
ch = ((ch - UNI_SUR_HIGH_START) << halfShift) + (ch2 - UNI_SUR_LOW_START) +
halfBase;
++source;
}
else if (flags == strictConversion) /* it's an unpaired high surrogate */
{
--source; /* return to the illegal value itself */
result = sourceIllegal;
break;
}
}
else /* We don't have the 16 bits following the high surrogate. */
{
--source; /* return to the high surrogate */
result = sourceExhausted;
break;
}
}
else if (flags == strictConversion)
{
/* UTF-16 surrogate values are illegal in UTF-32 */
if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END)
{
--source; /* return to the illegal value itself */
result = sourceIllegal;
break;
}
}
if (target >= targetEnd)
{
source = oldSource; /* Back up source pointer! */
result = targetExhausted;
break;
}
*target++ = ch;
}
*sourceStart = source;
*targetStart = target;
#ifdef CVTUTF_DEBUG
if (result == sourceIllegal)
{
WLOG_WARN(TAG, "ConvertUTF16toUTF32 illegal seq 0x%04x,%04x", ch, ch2);
}
#endif
return result;
}
/* --------------------------------------------------------------------- */
/*
* Index into the table below with the first byte of a UTF-8 sequence to
* get the number of trailing bytes that are supposed to follow it.
* Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is
* left as-is for anyone who may want to do such conversion, which was
* allowed in earlier algorithms.
*/
static const char trailingBytesForUTF8[256] = {
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5
};
/*
* Magic values subtracted from a buffer value during UTF8 conversion.
* This table contains as many values as there might be trailing bytes
* in a UTF-8 sequence.
*/
static const DWORD offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL,
0x03C82080UL, 0xFA082080UL, 0x82082080UL };
/*
* Once the bits are split out into bytes of UTF-8, this is a mask OR-ed
* into the first byte, depending on how many bytes follow. There are
* as many entries in this table as there are UTF-8 sequence types.
* (I.e., one byte sequence, two byte... etc.). Remember that sequencs
* for *legal* UTF-8 will be 4 or fewer bytes total.
*/
static const BYTE firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC };
/* --------------------------------------------------------------------- */
/* The interface converts a whole buffer to avoid function-call overhead.
* Constants have been gathered. Loops & conditionals have been removed as
* much as possible for efficiency, in favor of drop-through switches.
* (See "Note A" at the bottom of the file for equivalent code.)
* If your compiler supports it, the "isLegalUTF8" call can be turned
* into an inline function.
*/
/* --------------------------------------------------------------------- */
ConversionResult ConvertUTF16toUTF8(const WCHAR** sourceStart, const WCHAR* sourceEnd,
BYTE** targetStart, BYTE* te, ConversionFlags flags)
{
size_t pos = 0;
size_t end = 0;
const WCHAR* source;
const BOOL computeLength = (!te) ? TRUE : FALSE;
ConversionResult result = conversionOK;
if (targetStart && te)
{
const size_t s = (size_t)*targetStart;
const size_t e = (size_t)te;
if (s > e)
return sourceIllegal;
end = e - s;
}
source = *sourceStart;
while (source < sourceEnd)
{
DWORD ch;
unsigned short bytesToWrite = 0;
const DWORD byteMask = 0xBF;
const DWORD byteMark = 0x80;
const WCHAR* oldSource =
source; /* In case we have to back up because of target overflow. */
Data_Read_UINT16(source, ch);
source++;
/* If we have a surrogate pair, convert to UTF32 first. */
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END)
{
/* If the 16 bits following the high surrogate are in the source buffer... */
if (source < sourceEnd)
{
DWORD ch2;
Data_Read_UINT16(source, ch2);
/* If it's a low surrogate, convert to UTF32. */
if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END)
{
ch = ((ch - UNI_SUR_HIGH_START) << halfShift) + (ch2 - UNI_SUR_LOW_START) +
halfBase;
++source;
}
else if (flags == strictConversion)
{
/* it's an unpaired high surrogate */
--source; /* return to the illegal value itself */
result = sourceIllegal;
break;
}
}
else
{
/* We don't have the 16 bits following the high surrogate. */
--source; /* return to the high surrogate */
result = sourceExhausted;
break;
}
}
else if (flags == strictConversion)
{
/* UTF-16 surrogate values are illegal in UTF-32 */
if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END)
{
--source; /* return to the illegal value itself */
result = sourceIllegal;
break;
}
}
/* Figure out how many bytes the result will require */
if (ch < (DWORD)0x80)
{
bytesToWrite = 1;
}
else if (ch < (DWORD)0x800)
{
bytesToWrite = 2;
}
else if (ch < (DWORD)0x10000)
{
bytesToWrite = 3;
}
else if (ch < (DWORD)0x110000)
{
bytesToWrite = 4;
}
else
{
bytesToWrite = 3;
ch = UNI_REPLACEMENT_CHAR;
}
pos += bytesToWrite;
if ((pos > end) && (!computeLength))
{
source = oldSource; /* Back up source pointer! */
pos -= bytesToWrite;
result = targetExhausted;
break;
}
if (!computeLength)
{
switch (bytesToWrite)
{
/* note: everything falls through. */
case 4:
(*targetStart)[--pos] = (BYTE)((ch | byteMark) & byteMask);
ch >>= 6;
case 3:
(*targetStart)[--pos] = (BYTE)((ch | byteMark) & byteMask);
ch >>= 6;
case 2:
(*targetStart)[--pos] = (BYTE)((ch | byteMark) & byteMask);
ch >>= 6;
case 1:
(*targetStart)[--pos] = (BYTE)(ch | firstByteMark[bytesToWrite]);
}
}
else
{
switch (bytesToWrite)
{
/* note: everything falls through. */
case 4:
--pos;
case 3:
--pos;
case 2:
--pos;
case 1:
--pos;
}
}
pos += bytesToWrite;
}
*sourceStart = source;
if (targetStart && *targetStart)
*targetStart = &(*targetStart)[pos];
else if (targetStart)
*targetStart = (BYTE*)pos;
return result;
}
/* --------------------------------------------------------------------- */
/*
* Utility routine to tell whether a sequence of bytes is legal UTF-8.
* This must be called with the length pre-determined by the first byte.
* If not calling this from ConvertUTF8to*, then the length can be set by:
* length = trailingBytesForUTF8[*source]+1;
* and the sequence is illegal right away if there aren't that many bytes
* available.
* If presented with a length > 4, this returns FALSE. The Unicode
* definition of UTF-8 goes up to 4-byte sequences.
*/
static BOOL isLegalUTF8(const BYTE* source, int length)
{
BYTE a;
const BYTE* srcptr = source + length;
switch (length)
{
default:
return FALSE;
/* Everything else falls through when "TRUE"... */
case 4:
if ((a = (*--srcptr)) < 0x80 || a > 0xBF)
return FALSE;
case 3:
if ((a = (*--srcptr)) < 0x80 || a > 0xBF)
return FALSE;
case 2:
if ((a = (*--srcptr)) > 0xBF)
return FALSE;
switch (*source)
{
/* no fall-through in this inner switch */
case 0xE0:
if (a < 0xA0)
return FALSE;
break;
case 0xED:
if (a > 0x9F)
return FALSE;
break;
case 0xF0:
if (a < 0x90)
return FALSE;
break;
case 0xF4:
if (a > 0x8F)
return FALSE;
break;
default:
if (a < 0x80)
return FALSE;
}
case 1:
if (*source >= 0x80 && *source < 0xC2)
return FALSE;
}
if (*source > 0xF4)
return FALSE;
return TRUE;
}
/* --------------------------------------------------------------------- */
/*
* Exported function to return whether a UTF-8 sequence is legal or not.
* This is not used here; it's just exported.
*/
BOOL isLegalUTF8Sequence(const BYTE* source, const BYTE* sourceEnd)
{
int length = trailingBytesForUTF8[*source] + 1;
if (source + length > sourceEnd)
return FALSE;
return isLegalUTF8(source, length);
}
/* --------------------------------------------------------------------- */
ConversionResult ConvertUTF8toUTF16(const BYTE** sourceStart, const BYTE* sourceEnd,
WCHAR** targetStart, WCHAR* targetEnd, ConversionFlags flags)
{
size_t target = 0;
size_t end = 0;
const BYTE* source;
BOOL computeLength;
ConversionResult result;
computeLength = (!targetEnd) ? TRUE : FALSE;
result = conversionOK;
source = *sourceStart;
if (targetStart && targetEnd)
{
const size_t s = (size_t)*targetStart;
const size_t e = (size_t)targetEnd;
if (s > e)
return sourceIllegal;
end = ((size_t)(targetEnd)) - ((size_t)(*targetStart));
}
while (source < sourceEnd)
{
DWORD ch = 0;
unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
if ((source + extraBytesToRead) >= sourceEnd)
{
result = sourceExhausted;
break;
}
/* Do this check whether lenient or strict */
if (!isLegalUTF8(source, extraBytesToRead + 1))
{
result = sourceIllegal;
break;
}
/*
* The cases all fall through. See "Note A" below.
*/
switch (extraBytesToRead)
{
case 5:
ch += *source++;
ch <<= 6; /* remember, illegal UTF-8 */
case 4:
ch += *source++;
ch <<= 6; /* remember, illegal UTF-8 */
case 3:
ch += *source++;
ch <<= 6;
case 2:
ch += *source++;
ch <<= 6;
case 1:
ch += *source++;
ch <<= 6;
case 0:
ch += *source++;
}
ch -= offsetsFromUTF8[extraBytesToRead];
if ((target * sizeof(WCHAR) >= end) && (!computeLength))
{
source -= (extraBytesToRead + 1); /* Back up source pointer! */
result = targetExhausted;
break;
}
if (ch <= UNI_MAX_BMP)
{
/* Target is a character <= 0xFFFF */
/* UTF-16 surrogate values are illegal in UTF-32 */
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END)
{
if (flags == strictConversion)
{
source -= (extraBytesToRead + 1); /* return to the illegal value itself */
result = sourceIllegal;
break;
}
else
{
if (!computeLength)
Data_Write_UINT16(&(*targetStart)[target], UNI_REPLACEMENT_CHAR);
target++;
}
}
else
{
if (!computeLength)
Data_Write_UINT16(&(*targetStart)[target], ch); /* normal case */
target++;
}
}
else if (ch > UNI_MAX_UTF16)
{
if (flags == strictConversion)
{
result = sourceIllegal;
source -= (extraBytesToRead + 1); /* return to the start */
break; /* Bail out; shouldn't continue */
}
else
{
if (!computeLength)
Data_Write_UINT16(&(*targetStart)[target], UNI_REPLACEMENT_CHAR);
target++;
}
}
else
{
/* target is a character in range 0xFFFF - 0x10FFFF. */
if (((target + 1) * sizeof(WCHAR) >= end) && (!computeLength))
{
source -= (extraBytesToRead + 1); /* Back up source pointer! */
result = targetExhausted;
break;
}
ch -= halfBase;
if (!computeLength)
{
WCHAR wchar;
wchar = (WCHAR)((ch >> halfShift) + UNI_SUR_HIGH_START);
Data_Write_UINT16(&(*targetStart)[target++], wchar);
wchar = (WCHAR)((ch & halfMask) + UNI_SUR_LOW_START);
Data_Write_UINT16(&(*targetStart)[target++], wchar);
}
else
{
target++;
target++;
}
}
}
*sourceStart = source;
if (targetStart && (*targetStart))
*targetStart = &(*targetStart)[target];
else if (targetStart)
*targetStart = (WCHAR*)(target * sizeof(WCHAR));
return result;
}
/* --------------------------------------------------------------------- */
ConversionResult ConvertUTF32toUTF8(const DWORD** sourceStart, const DWORD* sourceEnd,
BYTE** targetStart, BYTE* targetEnd, ConversionFlags flags)
{
ConversionResult result = conversionOK;
const DWORD* source = *sourceStart;
BYTE* target = *targetStart;
while (source < sourceEnd)
{
DWORD ch;
unsigned short bytesToWrite = 0;
const DWORD byteMask = 0xBF;
const DWORD byteMark = 0x80;
ch = *source++;
if (flags == strictConversion)
{
/* UTF-16 surrogate values are illegal in UTF-32 */
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END)
{
--source; /* return to the illegal value itself */
result = sourceIllegal;
break;
}
}
/*
* Figure out how many bytes the result will require. Turn any
* illegally large UTF32 things (> Plane 17) into replacement chars.
*/
if (ch < (DWORD)0x80)
{
bytesToWrite = 1;
}
else if (ch < (DWORD)0x800)
{
bytesToWrite = 2;
}
else if (ch < (DWORD)0x10000)
{
bytesToWrite = 3;
}
else if (ch <= UNI_MAX_LEGAL_UTF32)
{
bytesToWrite = 4;
}
else
{
bytesToWrite = 3;
ch = UNI_REPLACEMENT_CHAR;
result = sourceIllegal;
}
target += bytesToWrite;
if (target > targetEnd)
{
--source; /* Back up source pointer! */
target -= bytesToWrite;
result = targetExhausted;
break;
}
switch (bytesToWrite) /* note: everything falls through. */
{
case 4:
*--target = (BYTE)((ch | byteMark) & byteMask);
ch >>= 6;
case 3:
*--target = (BYTE)((ch | byteMark) & byteMask);
ch >>= 6;
case 2:
*--target = (BYTE)((ch | byteMark) & byteMask);
ch >>= 6;
case 1:
*--target = (BYTE)(ch | firstByteMark[bytesToWrite]);
}
target += bytesToWrite;
}
*sourceStart = source;
*targetStart = target;
return result;
}
/* --------------------------------------------------------------------- */
ConversionResult ConvertUTF8toUTF32(const BYTE** sourceStart, const BYTE* sourceEnd,
DWORD** targetStart, DWORD* targetEnd, ConversionFlags flags)
{
ConversionResult result = conversionOK;
const BYTE* source = *sourceStart;
DWORD* target = *targetStart;
while (source < sourceEnd)
{
DWORD ch = 0;
unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
if (source + extraBytesToRead >= sourceEnd)
{
result = sourceExhausted;
break;
}
/* Do this check whether lenient or strict */
if (!isLegalUTF8(source, extraBytesToRead + 1))
{
result = sourceIllegal;
break;
}
/*
* The cases all fall through. See "Note A" below.
*/
switch (extraBytesToRead)
{
case 5:
ch += *source++;
ch <<= 6;
case 4:
ch += *source++;
ch <<= 6;
case 3:
ch += *source++;
ch <<= 6;
case 2:
ch += *source++;
ch <<= 6;
case 1:
ch += *source++;
ch <<= 6;
case 0:
ch += *source++;
}
ch -= offsetsFromUTF8[extraBytesToRead];
if (target >= targetEnd)
{
source -= (extraBytesToRead + 1); /* Back up the source pointer! */
result = targetExhausted;
break;
}
if (ch <= UNI_MAX_LEGAL_UTF32)
{
/*
* UTF-16 surrogate values are illegal in UTF-32, and anything
* over Plane 17 (> 0x10FFFF) is illegal.
*/
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END)
{
if (flags == strictConversion)
{
source -= (extraBytesToRead + 1); /* return to the illegal value itself */
result = sourceIllegal;
break;
}
else
{
*target++ = UNI_REPLACEMENT_CHAR;
}
}
else
{
*target++ = ch;
}
}
else /* i.e., ch > UNI_MAX_LEGAL_UTF32 */
{
result = sourceIllegal;
*target++ = UNI_REPLACEMENT_CHAR;
}
}
*sourceStart = source;
*targetStart = target;
return result;
}
/* ---------------------------------------------------------------------
Note A.
The fall-through switches in UTF-8 reading code save a
temp variable, some decrements & conditionals. The switches
are equivalent to the following loop:
{
int tmpBytesToRead = extraBytesToRead+1;
do {
ch += *source++;
--tmpBytesToRead;
if (tmpBytesToRead) ch <<= 6;
} while (tmpBytesToRead > 0);
}
In UTF-8 writing code, the switches on "bytesToWrite" are
similarly unrolled loops.
--------------------------------------------------------------------- */
#if __GNUC__
#pragma GCC diagnostic pop
#endif

150
winpr/libwinpr/crt/utf.h
View File

@ -1,150 +0,0 @@
/*
* Copyright 2001-2004 Unicode, Inc.
*
* Disclaimer
*
* This source code is provided as is by Unicode, Inc. No claims are
* made as to fitness for any particular purpose. No warranties of any
* kind are expressed or implied. The recipient agrees to determine
* applicability of information provided. If this file has been
* purchased on magnetic or optical media from Unicode, Inc., the
* sole remedy for any claim will be exchange of defective media
* within 90 days of receipt.
*
* Limitations on Rights to Redistribute This Code
*
* Unicode, Inc. hereby grants the right to freely use the information
* supplied in this file in the creation of products supporting the
* Unicode Standard, and to make copies of this file in any form
* for internal or external distribution as long as this notice
* remains attached.
*/
/* ---------------------------------------------------------------------
Conversions between UTF32, UTF-16, and UTF-8. Header file.
Several funtions are included here, forming a complete set of
conversions between the three formats. UTF-7 is not included
here, but is handled in a separate source file.
Each of these routines takes pointers to input buffers and output
buffers. The input buffers are const.
Each routine converts the text between *sourceStart and sourceEnd,
putting the result into the buffer between *targetStart and
targetEnd. Note: the end pointers are *after* the last item: e.g.
*(sourceEnd - 1) is the last item.
The return result indicates whether the conversion was successful,
and if not, whether the problem was in the source or target buffers.
(Only the first encountered problem is indicated.)
After the conversion, *sourceStart and *targetStart are both
updated to point to the end of last text successfully converted in
the respective buffers.
Input parameters:
sourceStart - pointer to a pointer to the source buffer.
The contents of this are modified on return so that
it points at the next thing to be converted.
targetStart - similarly, pointer to pointer to the target buffer.
sourceEnd, targetEnd - respectively pointers to the ends of the
two buffers, for overflow checking only.
These conversion functions take a ConversionFlags argument. When this
flag is set to strict, both irregular sequences and isolated surrogates
will cause an error. When the flag is set to lenient, both irregular
sequences and isolated surrogates are converted.
Whether the flag is strict or lenient, all illegal sequences will cause
an error return. This includes sequences such as: <F4 90 80 80>, <C0 80>,
or <A0> in UTF-8, and values above 0x10FFFF in UTF-32. Conformant code
must check for illegal sequences.
When the flag is set to lenient, characters over 0x10FFFF are converted
to the replacement character; otherwise (when the flag is set to strict)
they constitute an error.
Output parameters:
The value "sourceIllegal" is returned from some routines if the input
sequence is malformed. When "sourceIllegal" is returned, the source
value will point to the illegal value that caused the problem. E.g.,
in UTF-8 when a sequence is malformed, it points to the start of the
malformed sequence.
Author: Mark E. Davis, 1994.
Rev History: Rick McGowan, fixes & updates May 2001.
Fixes & updates, Sept 2001.
------------------------------------------------------------------------ */
#ifndef WINPR_UNICODE_CONVERT_UTF_H
#define WINPR_UNICODE_CONVERT_UTF_H
#include <winpr/wtypes.h>
/*
* Character Types:
*
* UTF8: BYTE 8 bits
* UTF16: WCHAR 16 bits
* UTF32: DWORD 32 bits
*/
/* Some fundamental constants */
#define UNI_REPLACEMENT_CHAR (DWORD)0x0000FFFD
#define UNI_MAX_BMP (DWORD)0x0000FFFF
#define UNI_MAX_UTF16 (DWORD)0x0010FFFF
#define UNI_MAX_UTF32 (DWORD)0x7FFFFFFF
#define UNI_MAX_LEGAL_UTF32 (DWORD)0x0010FFFF
typedef enum
{
conversionOK, /* conversion successful */
sourceExhausted, /* partial character in source, but hit end */
targetExhausted, /* insuff. room in target for conversion */
sourceIllegal /* source sequence is illegal/malformed */
} ConversionResult;
typedef enum
{
strictConversion = 0,
lenientConversion
} ConversionFlags;
/* This is for C++ and does no harm in C */
#ifdef __cplusplus
extern "C"
{
#endif
ConversionResult ConvertUTF8toUTF16(const BYTE** sourceStart, const BYTE* sourceEnd,
WCHAR** targetStart, WCHAR* targetEnd,
ConversionFlags flags);
ConversionResult ConvertUTF16toUTF8(const WCHAR** sourceStart, const WCHAR* sourceEnd,
BYTE** targetStart, BYTE* targetEnd, ConversionFlags flags);
ConversionResult ConvertUTF8toUTF32(const BYTE** sourceStart, const BYTE* sourceEnd,
DWORD** targetStart, DWORD* targetEnd,
ConversionFlags flags);
ConversionResult ConvertUTF32toUTF8(const DWORD** sourceStart, const DWORD* sourceEnd,
BYTE** targetStart, BYTE* targetEnd, ConversionFlags flags);
ConversionResult ConvertUTF16toUTF32(const WCHAR** sourceStart, const WCHAR* sourceEnd,
DWORD** targetStart, DWORD* targetEnd,
ConversionFlags flags);
ConversionResult ConvertUTF32toUTF16(const DWORD** sourceStart, const DWORD* sourceEnd,
WCHAR** targetStart, WCHAR* targetEnd,
ConversionFlags flags);
BOOL isLegalUTF8Sequence(const BYTE* source, const BYTE* sourceEnd);
#ifdef __cplusplus
}
#endif
#endif /* WINPR_UNICODE_CONVERT_UTF_H */