CodeCow/Notepad3
1
0
Fork 0
mirror of https://github.com/rizonesoft/Notepad3.git synced 2026-06-11 21:03:05 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
fe977cddec
Notepad3/src/Encoding.c
Pairi Daiza fe977cddec Change Rizonesoft Copyright © to year "2019"
2018-12-30 00:28:28 +01:00

922 lines
30 KiB
C
Raw Blame History

/******************************************************************************
* *
* *
* Notepad3 *
* *
* Encoding.c *
* Handling and Helpers for File Encoding *
* Based on code from Notepad2, (c) Florian Balmer 1996-2011 *
* *
* *
* *
* (c) Rizonesoft 2015-2019 *
* https://rizonesoft.com *
* *
* *
*******************************************************************************/
#if !defined(WINVER)
#define WINVER 0x601 /*_WIN32_WINNT_WIN7*/
#endif
#if !defined(_WIN32_WINNT)
#define _WIN32_WINNT 0x601 /*_WIN32_WINNT_WIN7*/
#endif
#if !defined(NTDDI_VERSION)
#define NTDDI_VERSION 0x06010000 /*NTDDI_WIN7*/
#endif
#define VC_EXTRALEAN 1
#define WIN32_LEAN_AND_MEAN 1
#define NOMINMAX 1
#include <windows.h>
#include <shellapi.h>
#include <commctrl.h>
#include <stdlib.h>
#include <assert.h>
#include "../uthash/utarray.h"
#include "Scintilla.h"
#include "Helpers.h"
#include "resource.h"
#include "Encoding.h"
//=============================================================================
//
// Encoding Helper Functions
//
int g_DOSEncoding = CPI_NONE;
bool g_bForceCompEncDetection = false;
extern bool g_bIsCJKInputCodePage;
// ============================================================================
// Supported Encodings
WCHAR wchANSI[16] = { L'\0' };
WCHAR wchOEM[16] = { L'\0' };
// ============================================================================
int Encoding_Current(int iEncoding)
{
static int CurrentEncoding = CPI_NONE;
if (iEncoding >= 0) {
if (Encoding_IsValid(iEncoding))
CurrentEncoding = iEncoding;
else
CurrentEncoding = CPI_UTF8;
}
return CurrentEncoding;
}
// ============================================================================
int Encoding_SrcCmdLn(int iSrcEncoding)
{
static int SourceEncoding = CPI_NONE;
if (iSrcEncoding >= 0) {
if (Encoding_IsValid(iSrcEncoding))
SourceEncoding = iSrcEncoding;
else
SourceEncoding = CPI_ANSI_DEFAULT;
}
else if (iSrcEncoding == CPI_NONE) {
SourceEncoding = CPI_NONE;
}
return SourceEncoding;
}
// ============================================================================
int Encoding_SrcWeak(int iSrcWeakEnc)
{
static int SourceWeakEncoding = CPI_NONE;
if (iSrcWeakEnc >= 0) {
if (Encoding_IsValid(iSrcWeakEnc))
SourceWeakEncoding = iSrcWeakEnc;
else
SourceWeakEncoding = CPI_ANSI_DEFAULT;
}
else if (iSrcWeakEnc == CPI_NONE) {
SourceWeakEncoding = CPI_NONE;
}
return SourceWeakEncoding;
}
// ============================================================================
bool Encoding_HasChanged(int iOriginalEncoding) {
static int OriginalEncoding = CPI_NONE;
if (iOriginalEncoding >= CPI_NONE) {
OriginalEncoding = iOriginalEncoding;
}
return (bool)(OriginalEncoding != Encoding_Current(CPI_GET));
}
// ============================================================================
void Encoding_InitDefaults()
{
const UINT uCodePageMBCS[20] = {
42, // (Symbol)
50220,50221,50222,50225,50227,50229, // (Chinese, Japanese, Korean)
54936, // (GB18030)
57002,57003,57004,57005,57006,57007,57008,57009,57010,57011, // (ISCII)
65000, // (UTF-7)
65001 // (UTF-8)
};
UINT const ansiInputCP = GetACP();
ChangeEncodingCodePage(CPI_ANSI_DEFAULT, ansiInputCP); // set ANSI system CP ()
assert(g_Encodings[CPI_ANSI_DEFAULT].uCodePage == ansiInputCP);
StringCchPrintf(wchANSI, COUNTOF(wchANSI), L" (CP-%u)", ansiInputCP);
g_bIsCJKInputCodePage = IsDBCSCodePage(Scintilla_InputCodePage());
for (int i = CPI_UTF7 + 1; i < Encoding_CountOf(); ++i) {
if (Encoding_IsValid(i) && (g_Encodings[i].uCodePage == g_Encodings[CPI_ANSI_DEFAULT].uCodePage)) {
g_Encodings[i].uFlags |= NCP_ANSI;
if (g_Encodings[i].uFlags & NCP_EXTERNAL_8BIT)
g_Encodings[CPI_ANSI_DEFAULT].uFlags |= NCP_EXTERNAL_8BIT;
break;
}
}
ChangeEncodingCodePage(CPI_OEM, GetOEMCP()); // set OEM system CP
StringCchPrintf(wchOEM, COUNTOF(wchOEM), L" (CP-%u)", g_Encodings[CPI_OEM].uCodePage);
for (int i = CPI_UTF7 + 1; i < Encoding_CountOf(); ++i) {
if (Encoding_IsValid(i) && (g_Encodings[i].uCodePage == g_Encodings[CPI_OEM].uCodePage)) {
g_Encodings[i].uFlags |= NCP_OEM;
if (g_Encodings[i].uFlags & NCP_EXTERNAL_8BIT)
g_Encodings[CPI_OEM].uFlags |= NCP_EXTERNAL_8BIT;
break;
}
}
// multi byte character sets
for (int i = 0; i < Encoding_CountOf(); ++i) {
for (int k = 0; k < COUNTOF(uCodePageMBCS); k++) {
if (g_Encodings[i].uCodePage == uCodePageMBCS[k]) {
g_Encodings[i].uFlags |= NCP_MBCS;
}
}
}
g_DOSEncoding = CPI_OEM;
// Try to set the DOS encoding to DOS-437 if the default OEMCP is not DOS-437
if (g_Encodings[g_DOSEncoding].uCodePage != 437) {
for (int i = CPI_UTF7 + 1; i < Encoding_CountOf(); ++i) {
if (Encoding_IsValid(i) && (g_Encodings[i].uCodePage == 437)) {
g_DOSEncoding = i;
break;
}
}
}
}
// ============================================================================
int Encoding_MapIniSetting(bool bLoad, int iSetting) {
if (bLoad) {
switch (iSetting) {
case -1: return CPI_NONE;
case 0: return CPI_ANSI_DEFAULT;
case 1: return CPI_UNICODEBOM;
case 2: return CPI_UNICODEBEBOM;
case 3: return CPI_UTF8;
case 4: return CPI_UTF8SIGN;
case 5: return CPI_OEM;
case 6: return CPI_UNICODE;
case 7: return CPI_UNICODEBE;
case 8: return CPI_UTF7;
default: {
for (int i = CPI_UTF7 + 1; i < Encoding_CountOf(); i++) {
if ((g_Encodings[i].uCodePage == (UINT)iSetting) && Encoding_IsValid(i))
return(i);
}
return CPI_ANSI_DEFAULT;
}
}
}
else {
switch (iSetting) {
case CPI_NONE: return -1;
case CPI_ANSI_DEFAULT: return 0;
case CPI_UNICODEBOM: return 1;
case CPI_UNICODEBEBOM: return 2;
case CPI_UTF8: return 3;
case CPI_UTF8SIGN: return 4;
case CPI_OEM: return 5;
case CPI_UNICODE: return 6;
case CPI_UNICODEBE: return 7;
case CPI_UTF7: return 8;
default:
if (Encoding_IsValid(iSetting)) {
return(g_Encodings[iSetting].uCodePage);
}
return CPI_ANSI_DEFAULT;
}
}
}
// ============================================================================
int Encoding_MapUnicode(int iUni)
{
if (iUni == CPI_UNICODEBOM) {
return CPI_UNICODE;
}
if (iUni == CPI_UNICODEBEBOM) {
return CPI_UNICODEBE;
}
if (iUni == CPI_UTF8SIGN) {
return CPI_UTF8;
}
return iUni;
}
// ============================================================================
void Encoding_SetLabel(int iEncoding)
{
if (g_Encodings[iEncoding].wchLabel[0] == L'\0') {
WCHAR wch1[128] = { L'\0' };
WCHAR wch2[128] = { L'\0' };
GetLngString(g_Encodings[iEncoding].idsName, wch1, COUNTOF(wch1));
WCHAR *pwsz = StrChr(wch1, L';');
if (pwsz) {
pwsz = StrChr(CharNext(pwsz), L';');
if (pwsz) {
pwsz = CharNext(pwsz);
}
}
if (!pwsz)
pwsz = wch1;
StringCchCopyN(wch2, COUNTOF(wch2), pwsz, COUNTOF(wch1));
if (Encoding_IsANSI(iEncoding))
StringCchCatN(wch2, COUNTOF(wch2), wchANSI, COUNTOF(wchANSI));
else if (Encoding_IsOEM(iEncoding))
StringCchCatN(wch2, COUNTOF(wch2), wchOEM, COUNTOF(wchOEM));
StringCchCopyN(g_Encodings[iEncoding].wchLabel, COUNTOF(g_Encodings[iEncoding].wchLabel),
wch2, COUNTOF(g_Encodings[iEncoding].wchLabel));
}
}
// ============================================================================
int Encoding_MatchW(LPCWSTR pwszTest)
{
char tchTest[256] = { '\0' };
WideCharToMultiByte(CP_ACP, 0, pwszTest, -1, tchTest, COUNTOF(tchTest), NULL, NULL);
return(Encoding_MatchA(tchTest));
}
// ============================================================================
int Encoding_MatchA(char *pchTest)
{
char chTest[256] = { '\0' };
char *pchSrc = pchTest;
char *pchDst = chTest;
*pchDst++ = ',';
while (*pchSrc) {
if (IsCharAlphaNumericA(*pchSrc)) {
*pchDst++ = *CharLowerA(pchSrc);
}
++pchSrc;
}
*pchDst++ = ',';
*pchDst = 0;
for (int i = 0; i < Encoding_CountOf(); i++) {
if (StrStrIA(g_Encodings[i].pszParseNames, chTest)) {
CPINFO cpi;
if ((g_Encodings[i].uFlags & NCP_INTERNAL) ||
(IsValidCodePage(g_Encodings[i].uCodePage) &&
GetCPInfo(g_Encodings[i].uCodePage, &cpi))) {
return(i);
}
return(-1);
}
}
return(-1);
}
// ============================================================================
int Encoding_GetByCodePage(UINT cp) {
for (int i = 0; i < Encoding_CountOf(); i++) {
if (cp == g_Encodings[i].uCodePage) {
return i;
}
}
return CPI_ANSI_DEFAULT;
}
// ============================================================================
bool Encoding_IsValid(int iTestEncoding) {
CPINFO cpi;
if ((iTestEncoding >= 0) && (iTestEncoding < Encoding_CountOf())) {
if ((g_Encodings[iTestEncoding].uFlags & NCP_INTERNAL) ||
(IsValidCodePage(g_Encodings[iTestEncoding].uCodePage) &&
GetCPInfo(g_Encodings[iTestEncoding].uCodePage, &cpi))) {
return(true);
}
}
return(false);
}
// ============================================================================
typedef struct _ee {
int id;
WCHAR wch[256];
} ENCODINGENTRY, *PENCODINGENTRY;
int CmpEncoding(const void *s1, const void *s2) {
return StrCmp(((PENCODINGENTRY)s1)->wch, ((PENCODINGENTRY)s2)->wch);
}
// ============================================================================
void Encoding_AddToListView(HWND hwnd, int idSel, bool bRecodeOnly) {
int i;
int iSelItem = -1;
LVITEM lvi;
WCHAR wchBuf[256] = { L'\0' };
PENCODINGENTRY pEE = AllocMem(Encoding_CountOf() * sizeof(ENCODINGENTRY), HEAP_ZERO_MEMORY);
if (pEE) {
for (i = 0; i < Encoding_CountOf(); i++) {
pEE[i].id = i;
GetLngString(g_Encodings[i].idsName, pEE[i].wch, COUNTOF(pEE[i].wch));
}
qsort(pEE, Encoding_CountOf(), sizeof(ENCODINGENTRY), CmpEncoding);
ZeroMemory(&lvi, sizeof(LVITEM));
lvi.mask = LVIF_PARAM | LVIF_TEXT | LVIF_IMAGE;
lvi.pszText = wchBuf;
for (i = 0; i < Encoding_CountOf(); i++) {
int id = pEE[i].id;
if (!bRecodeOnly || (g_Encodings[id].uFlags & NCP_RECODE)) {
lvi.iItem = ListView_GetItemCount(hwnd);
WCHAR *pwsz = StrChr(pEE[i].wch, L';');
if (pwsz) {
StringCchCopyN(wchBuf, COUNTOF(wchBuf), CharNext(pwsz), COUNTOF(wchBuf));
pwsz = StrChr(wchBuf, L';');
if (pwsz)
*pwsz = 0;
}
else
StringCchCopyN(wchBuf, COUNTOF(wchBuf), pEE[i].wch, COUNTOF(wchBuf));
if (Encoding_IsANSI(id))
StringCchCatN(wchBuf, COUNTOF(wchBuf), wchANSI, COUNTOF(wchANSI));
else if (Encoding_IsOEM(id))
StringCchCatN(wchBuf, COUNTOF(wchBuf), wchOEM, COUNTOF(wchOEM));
if (Encoding_IsValid(id))
lvi.iImage = 0;
else
lvi.iImage = 1;
lvi.lParam = (LPARAM)id;
ListView_InsertItem(hwnd, &lvi);
if (idSel == id)
iSelItem = lvi.iItem;
}
}
FreeMem(pEE);
}
if (iSelItem != -1) {
ListView_SetItemState(hwnd, iSelItem, LVIS_SELECTED | LVIS_FOCUSED, LVIS_SELECTED | LVIS_FOCUSED);
ListView_EnsureVisible(hwnd, iSelItem, false);
}
else {
ListView_SetItemState(hwnd, 0, LVIS_FOCUSED, LVIS_FOCUSED);
ListView_EnsureVisible(hwnd, 0, false);
}
}
// ============================================================================
bool Encoding_GetFromListView(HWND hwnd, int *pidEncoding) {
LVITEM lvi;
lvi.iItem = ListView_GetNextItem(hwnd, -1, LVNI_ALL | LVNI_SELECTED);
lvi.iSubItem = 0;
lvi.mask = LVIF_PARAM;
if (ListView_GetItem(hwnd, &lvi)) {
if (Encoding_IsValid((int)lvi.lParam))
*pidEncoding = (int)lvi.lParam;
else
*pidEncoding = -1;
return (true);
}
return(false);
}
// ============================================================================
void Encoding_AddToComboboxEx(HWND hwnd, int idSel, bool bRecodeOnly) {
int i;
int iSelItem = -1;
COMBOBOXEXITEM cbei;
WCHAR wchBuf[256] = { L'\0' };
PENCODINGENTRY pEE = AllocMem(Encoding_CountOf() * sizeof(ENCODINGENTRY), HEAP_ZERO_MEMORY);
if (pEE) {
for (i = 0; i < Encoding_CountOf(); i++) {
pEE[i].id = i;
GetLngString(g_Encodings[i].idsName, pEE[i].wch, COUNTOF(pEE[i].wch));
}
qsort(pEE, Encoding_CountOf(), sizeof(ENCODINGENTRY), CmpEncoding);
ZeroMemory(&cbei, sizeof(COMBOBOXEXITEM));
cbei.mask = CBEIF_TEXT | CBEIF_IMAGE | CBEIF_SELECTEDIMAGE | CBEIF_LPARAM;
cbei.pszText = wchBuf;
cbei.cchTextMax = COUNTOF(wchBuf);
cbei.iImage = 0;
cbei.iSelectedImage = 0;
for (i = 0; i < Encoding_CountOf(); i++) {
int id = pEE[i].id;
if (!bRecodeOnly || (g_Encodings[id].uFlags & NCP_RECODE)) {
cbei.iItem = SendMessage(hwnd, CB_GETCOUNT, 0, 0);
WCHAR *pwsz = StrChr(pEE[i].wch, L';');
if (pwsz) {
StringCchCopyN(wchBuf, COUNTOF(wchBuf), CharNext(pwsz), COUNTOF(wchBuf));
pwsz = StrChr(wchBuf, L';');
if (pwsz)
*pwsz = 0;
}
else
StringCchCopyN(wchBuf, COUNTOF(wchBuf), pEE[i].wch, COUNTOF(wchBuf));
if (Encoding_IsANSI(id))
StringCchCatN(wchBuf, COUNTOF(wchBuf), wchANSI, COUNTOF(wchANSI));
else if (id == CPI_OEM)
StringCchCatN(wchBuf, COUNTOF(wchBuf), wchOEM, COUNTOF(wchOEM));
cbei.iImage = (Encoding_IsValid(id) ? 0 : 1);
cbei.lParam = (LPARAM)id;
SendMessage(hwnd, CBEM_INSERTITEM, 0, (LPARAM)&cbei);
if (idSel == id)
iSelItem = (int)cbei.iItem;
}
}
FreeMem(pEE);
}
if (iSelItem != -1) {
SendMessage(hwnd, CB_SETCURSEL, (WPARAM)iSelItem, 0);
}
}
// ============================================================================
bool Encoding_GetFromComboboxEx(HWND hwnd, int *pidEncoding) {
COMBOBOXEXITEM cbei;
cbei.iItem = SendMessage(hwnd, CB_GETCURSEL, 0, 0);
cbei.mask = CBEIF_LPARAM;
if (SendMessage(hwnd, CBEM_GETITEM, 0, (LPARAM)&cbei)) {
if (Encoding_IsValid((int)cbei.lParam))
*pidEncoding = (int)cbei.lParam;
else
*pidEncoding = -1;
return (true);
}
return(false);
}
// ============================================================================
UINT Encoding_GetCodePage(const int iEncoding) {
return (iEncoding >= 0) ? g_Encodings[iEncoding].uCodePage : CP_ACP;
}
// ============================================================================
bool Encoding_IsDefault(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_DEFAULT) : false;
}
// ============================================================================
bool Encoding_IsANSI(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_ANSI) : false;
}
// ============================================================================
bool Encoding_IsOEM(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_OEM) : false;
}
// ============================================================================
bool Encoding_IsUTF8(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_UTF8) : false;
}
// ============================================================================
bool Encoding_IsUTF8_SIGN(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_UTF8_SIGN) : false;
}
// ============================================================================
bool Encoding_IsMBCS(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_MBCS) : false;
}
// ============================================================================
bool Encoding_IsASCII(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_ASCII_7BIT) : false;
}
// ============================================================================
bool Encoding_IsUNICODE(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_UNICODE) : false;
}
// ============================================================================
bool Encoding_IsUNICODE_BOM(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_UNICODE_BOM) : false;
}
// ============================================================================
bool Encoding_IsUNICODE_REVERSE(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_UNICODE_REVERSE) : false;
}
// ============================================================================
bool Encoding_IsINTERNAL(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_INTERNAL) : false;
}
// ============================================================================
bool Encoding_IsEXTERNAL_8BIT(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_EXTERNAL_8BIT) : false;
}
// ============================================================================
bool Encoding_IsRECODE(const int iEncoding) {
return (iEncoding >= 0) ? (g_Encodings[iEncoding].uFlags & NCP_RECODE) : false;
}
// ============================================================================
void Encoding_SetDefaultFlag(const int iEncoding) {
if (iEncoding >= 0)
g_Encodings[iEncoding].uFlags |= NCP_DEFAULT;
}
// ============================================================================
const WCHAR* Encoding_GetLabel(const int iEncoding) {
return (iEncoding >= 0) ? g_Encodings[iEncoding].wchLabel : NULL;
}
// ============================================================================
const char* Encoding_GetParseNames(const int iEncoding) {
return (iEncoding >= 0) ? g_Encodings[iEncoding].pszParseNames : NULL;
}
// ============================================================================
bool Has_UTF16_LE_BOM(const char* pBuf, int cnt)
{
int iTest = IS_TEXT_UNICODE_SIGNATURE;
bool const ok = IsTextUnicode(pBuf, cnt, &iTest);
return (ok && ((iTest & IS_TEXT_UNICODE_SIGNATURE) != 0));
}
// ----------------------------------------------------------------------------
bool Has_UTF16_BE_BOM(const char* pBuf, int cnt)
{
int iTest = IS_TEXT_UNICODE_REVERSE_SIGNATURE;
bool const ok = IsTextUnicode(pBuf, cnt, &iTest);
return (ok && ((iTest & IS_TEXT_UNICODE_REVERSE_SIGNATURE) != 0));
}
// ============================================================================
bool IsValidUnicode(const char* pBuffer, size_t cb, bool* lpbBOM, bool* lpbReverse)
{
if (!pBuffer || cb < 2) { return false; }
// IS_TEXT_UNICODE_UNICODE_MASK -> IS_TEXT_UNICODE_ASCII16, IS_TEXT_UNICODE_STATISTICS, IS_TEXT_UNICODE_CONTROLS, IS_TEXT_UNICODE_SIGNATURE.
// IS_TEXT_UNICODE_REVERSE_MASK -> IS_TEXT_UNICODE_REVERSE_ASCII16, IS_TEXT_UNICODE_REVERSE_STATISTICS, IS_TEXT_UNICODE_REVERSE_CONTROLS, IS_TEXT_UNICODE_REVERSE_SIGNATURE.
// IS_TEXT_UNICODE_NOT_UNICODE_MASK -> IS_TEXT_UNICODE_ILLEGAL_CHARS, IS_TEXT_UNICODE_ODD_LENGTH, and two currently unused bit flags.
// IS_TEXT_UNICODE_NOT_ASCII_MASK -> IS_TEXT_UNICODE_NULL_BYTES and three currently unused bit flags.
//
int const iAllTests = IS_TEXT_UNICODE_UNICODE_MASK | IS_TEXT_UNICODE_REVERSE_MASK | IS_TEXT_UNICODE_NOT_UNICODE_MASK | IS_TEXT_UNICODE_NOT_ASCII_MASK;
int iTest = iAllTests;
(void) IsTextUnicode(pBuffer, (int)cb, &iTest);
if (iTest == iAllTests) {
iTest = 0; // iTest doesn't seem to have been modified ...
}
bool const bHasBOM = (iTest & IS_TEXT_UNICODE_SIGNATURE);
bool const bHasRBOM = (iTest & IS_TEXT_UNICODE_REVERSE_SIGNATURE);
bool const bIsUnicode = (iTest & IS_TEXT_UNICODE_UNICODE_MASK);
bool const bIsReverse = (iTest & IS_TEXT_UNICODE_REVERSE_MASK);
bool const bIsIllegal = (iTest & IS_TEXT_UNICODE_NOT_UNICODE_MASK);
//bool const bHasNullBytes = (iTest & IS_TEXT_UNICODE_NULL_BYTES);
if (bHasBOM || bHasRBOM || ((bIsUnicode || bIsReverse) && !bIsIllegal && !(bIsUnicode && bIsReverse)))
{
if (lpbBOM) {
*lpbBOM = (bHasBOM || bHasRBOM);
}
if (lpbReverse) {
*lpbReverse = (bHasRBOM || bIsReverse);
}
return true;
}
return false;
}
// ============================================================================
bool IsValidUTF7(const char* pTest, size_t nLength)
{
if (!pTest) { return false; }
char const *pt = pTest;
for (size_t i = 0; i < nLength; ++i) {
if ((*pt & 0x80) || !*pt) { return false; }
++pt;
}
return true;
}
// ============================================================================
#undef _OLD_UTF8_VALIDATOR_
//#define _OLD_UTF8_VALIDATOR_ 1
#ifdef _OLD_UTF8_VALIDATOR_
// ============================================================================
/* byte length of UTF-8 sequence based on value of first byte.
for UTF-16 (21-bit space), max. code length is 4, so we only need to look
at 4 upper bits.
*/
static const size_t utf8_lengths[16] =
{
1,1,1,1,1,1,1,1, /* 0000 to 0111 : 1 byte (plain ASCII) */
0,0,0,0, /* 1000 to 1011 : not valid */
2,2, /* 1100, 1101 : 2 bytes */
3, /* 1110 : 3 bytes */
4 /* 1111 : 4 bytes */
};
// ----------------------------------------------------------------------------
/*++
Function :
UTF8_mbslen_bytes [INTERNAL]
Calculates the byte size of a NULL-terminated UTF-8 string.
Parameters :
char *utf8_string : string to examine
Return value :
size (in bytes) of a NULL-terminated UTF-8 string.
-1 if invalid NULL-terminated UTF-8 string
--*/
size_t UTF8_mbslen_bytes(LPCSTR utf8_string)
{
size_t length = 0;
size_t code_size;
BYTE byte;
while (*utf8_string)
{
byte = (BYTE)*utf8_string;
if ((byte <= 0xF7) && (0 != (code_size = utf8_lengths[byte >> 4]))) {
length += code_size;
utf8_string += code_size;
}
else {
/* we got an invalid byte value but need to count it,
it will be later ignored during the string conversion */
//WARN("invalid first byte value 0x%02X in UTF-8 sequence!\n",byte);
length++;
utf8_string++;
}
}
length++; /* include NULL terminator */
return length;
}
// ----------------------------------------------------------------------------
/*++
Function :
UTF8_mbslen [INTERNAL]
Calculates the character size of a NULL-terminated UTF-8 string.
Parameters :
char *utf8_string : string to examine
int byte_length : byte size of string
Return value :
size (in characters) of a UTF-8 string.
-1 if invalid UTF-8 string
--*/
size_t UTF8_mbslen(LPCSTR utf8_string, size_t byte_length)
{
size_t wchar_length = 0;
size_t code_size;
BYTE byte;
while (byte_length > 0) {
byte = (BYTE)*utf8_string;
/* UTF-16 can't encode 5-byte and 6-byte sequences, so maximum value
for first byte is 11110111. Use lookup table to determine sequence
length based on upper 4 bits of first byte */
if ((byte <= 0xF7) && (0 != (code_size = utf8_lengths[byte >> 4]))) {
/* 1 sequence == 1 character */
wchar_length++;
if (code_size == 4)
wchar_length++;
utf8_string += code_size; /* increment pointer */
byte_length -= code_size; /* decrement counter*/
}
else {
/*
unlike UTF8_mbslen_bytes, we ignore the invalid characters.
we only report the number of valid characters we have encountered
to match the Windows behavior.
*/
//WARN("invalid byte 0x%02X in UTF-8 sequence, skipping it!\n", byte);
utf8_string++;
byte_length--;
}
}
return wchar_length;
}
// ----------------------------------------------------------------------------
bool UTF8_ContainsInvalidChars(LPCSTR utf8_string, size_t byte_length)
{
return ((UTF8_mbslen_bytes(UTF8StringStart(utf8_string)) - 1) !=
UTF8_mbslen(UTF8StringStart(utf8_string), IsUTF8Signature(utf8_string) ? (byte_length - 3) : byte_length));
}
// ----------------------------------------------------------------------------
bool IsValidUTF8(const char* pTest, size_t nLength)
{
static int byte_class_table[256] = {
/* 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F */
/* 00 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 10 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 20 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 30 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 40 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 50 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 60 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 70 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 80 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
/* 90 */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
/* A0 */ 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
/* B0 */ 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
/* C0 */ 4, 4, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,
/* D0 */ 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,
/* E0 */ 6, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 8, 7, 7,
/* F0 */ 9,10,10,10,11, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4
/* 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F */ };
/* state table */
typedef enum {
kSTART = 0, kA, kB, kC, kD, kE, kF, kG, kERROR, kNumOfStates
} utf8_state;
static utf8_state state_table[] = {
/* kSTART, kA, kB, kC, kD, kE, kF, kG, kERROR */
/* 0x00-0x7F: 0 */ kSTART, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR,
/* 0x80-0x8F: 1 */ kERROR, kSTART, kA, kERROR, kA, kB, kERROR, kB, kERROR,
/* 0x90-0x9f: 2 */ kERROR, kSTART, kA, kERROR, kA, kB, kB, kERROR, kERROR,
/* 0xa0-0xbf: 3 */ kERROR, kSTART, kA, kA, kERROR, kB, kB, kERROR, kERROR,
/* 0xc0-0xc1, 0xf5-0xff: 4 */ kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR,
/* 0xc2-0xdf: 5 */ kA, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR,
/* 0xe0: 6 */ kC, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR,
/* 0xe1-0xec, 0xee-0xef: 7 */ kB, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR,
/* 0xed: 8 */ kD, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR,
/* 0xf0: 9 */ kF, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR,
/* 0xf1-0xf3: 10 */ kE, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR,
/* 0xf4: 11 */ kG, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR, kERROR };
#define BYTE_CLASS(b) (byte_class_table[(unsigned char)b])
#define NEXT_STATE(b,cur) (state_table[(BYTE_CLASS(b) * kNumOfStates) + (cur)])
utf8_state current = kSTART;
const char* pt = pTest;
size_t len = nLength;
for (size_t i = 0; i < len; i++, pt++) {
current = NEXT_STATE(*pt, current);
if (kERROR == current)
break;
}
return (current == kSTART) && !UTF8_ContainsInvalidChars(pTest, nLength);
}
// ============================================================================
#else // new UTF-8 validator
// ============================================================================
// Copyright (c) 2008-2010 Bjoern Hoehrmann <bjoern@hoehrmann.de>
// See http://bjoern.hoehrmann.de/utf-8/decoder/dfa/ for details.
bool IsValidUTF8(const char* pTest, size_t nLength)
{
enum {
UTF8_ACCEPT = 0,
UTF8_REJECT = 12
};
static const unsigned char utf8_dfa[] = {
// The first part of the table maps bytes to character classes that
// to reduce the size of the transition table and create bitmasks.
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, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,
7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
8,8,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
10,3,3,3,3,3,3,3,3,3,3,3,3,4,3,3, 11,6,6,6,5,8,8,8,8,8,8,8,8,8,8,8,
// The second part is a transition table that maps a combination
// of a state of the automaton and a character class to a state.
0,12,24,36,60,96,84,12,12,12,48,72, 12,12,12,12,12,12,12,12,12,12,12,12,
12, 0,12,12,12,12,12, 0,12, 0,12,12, 12,24,12,12,12,12,12,24,12,24,12,12,
12,12,12,12,12,12,12,24,12,12,12,12, 12,24,12,12,12,12,12,12,12,24,12,12,
12,12,12,12,12,12,12,36,12,36,12,12, 12,36,12,12,12,12,12,36,12,36,12,12,
12,36,12,12,12,12,12,12,12,12,12,12,
};
const unsigned char *pt = (const unsigned char *)pTest;
const unsigned char *end = pt + nLength;
UINT state = UTF8_ACCEPT;
while (pt < end && *pt) {
state = utf8_dfa[256 + state + utf8_dfa[*pt++]];
if (state == UTF8_REJECT) {
return false;
}
}
return (state == UTF8_ACCEPT);
}
// ----------------------------------------------------------------------------
#endif
// ============================================================================
Reference in New Issue View Git Blame Copy Permalink