зеркало из https://github.com/mozilla/gecko-dev.git
Bug 73292: Add const nsACString& constructor to NS_Convert(ASCII|UTF8)toUCS2. r=dbaron, sr=scc.
This commit is contained in:
jaggernaut%netscape.com
Родитель
43fc51fb89
Коммит
79e0a6c9a2
|
|
@ -1500,144 +1500,226 @@ nsAutoString::nsAutoString(const CBufDescriptor& aBuffer) : nsString() {
|
|||
AddNullTerminator(*this);
|
||||
}
|
||||
|
||||
NS_ConvertASCIItoUCS2::NS_ConvertASCIItoUCS2( const char* aCString, PRUint32 aLength )
|
||||
void
|
||||
NS_ConvertASCIItoUCS2::Init( const char* aCString, PRUint32 aLength )
|
||||
{
|
||||
Initialize(*this,mBuffer,(sizeof(mBuffer)>>eTwoByte)-1,0,eTwoByte,PR_FALSE);
|
||||
AddNullTerminator(*this);
|
||||
AppendWithConversion(aCString,aLength);
|
||||
}
|
||||
|
||||
NS_ConvertASCIItoUCS2::NS_ConvertASCIItoUCS2( const char* aCString )
|
||||
NS_ConvertASCIItoUCS2::NS_ConvertASCIItoUCS2( const nsACString& aCString )
|
||||
{
|
||||
Initialize(*this,mBuffer,(sizeof(mBuffer)>>eTwoByte)-1,0,eTwoByte,PR_FALSE);
|
||||
AddNullTerminator(*this);
|
||||
AppendWithConversion(aCString);
|
||||
SetCapacity(aCString.Length());
|
||||
|
||||
nsACString::const_iterator start; aCString.BeginReading(start);
|
||||
nsACString::const_iterator end; aCString.EndReading(end);
|
||||
|
||||
while (start != end)
|
||||
{
|
||||
const nsReadableFragment<char>& frag = start.fragment();
|
||||
AppendWithConversion(frag.mStart, frag.mEnd - frag.mStart);
|
||||
start.advance(start.size_forward());
|
||||
}
|
||||
}
|
||||
|
||||
class UTF8traits
|
||||
{
|
||||
public:
|
||||
static PRBool isASCII(char c) { return (c & 0x80) == 0x00; }
|
||||
static PRBool isInSeq(char c) { return (c & 0xC0) == 0x80; }
|
||||
static PRBool is2byte(char c) { return (c & 0xE0) == 0xC0; }
|
||||
static PRBool is3byte(char c) { return (c & 0xF0) == 0xE0; }
|
||||
static PRBool is4byte(char c) { return (c & 0xF8) == 0xF0; }
|
||||
static PRBool is5byte(char c) { return (c & 0xFC) == 0xF8; }
|
||||
static PRBool is6byte(char c) { return (c & 0xFE) == 0xFC; }
|
||||
};
|
||||
|
||||
class CalculateUTF8Length
|
||||
{
|
||||
public:
|
||||
typedef nsACString::char_type value_type;
|
||||
|
||||
CalculateUTF8Length() : mLength(0) { }
|
||||
|
||||
size_t Length() const { return mLength; }
|
||||
|
||||
PRUint32 write( const value_type* start, PRUint32 N )
|
||||
{
|
||||
// algorithm assumes utf8 units won't
|
||||
// be spread across fragments
|
||||
const value_type* p = start;
|
||||
const value_type* end = start + N;
|
||||
for ( ; p != end /* && *p */; ++mLength )
|
||||
{
|
||||
if ( UTF8traits::isASCII(*p) )
|
||||
p += 1;
|
||||
else if ( UTF8traits::is2byte(*p) )
|
||||
p += 2;
|
||||
else if ( UTF8traits::is3byte(*p) )
|
||||
p += 3;
|
||||
else if ( UTF8traits::is4byte(*p) )
|
||||
p += 4;
|
||||
else if ( UTF8traits::is5byte(*p) )
|
||||
p += 5;
|
||||
else if ( UTF8traits::is6byte(*p) )
|
||||
p += 6;
|
||||
else
|
||||
{
|
||||
NS_ERROR("not a UTF-8 string");
|
||||
break;
|
||||
}
|
||||
}
|
||||
return p - start;
|
||||
}
|
||||
|
||||
private:
|
||||
size_t mLength;
|
||||
};
|
||||
|
||||
class ConvertUTF8toUCS2
|
||||
{
|
||||
public:
|
||||
typedef nsACString::char_type value_type;
|
||||
typedef nsAString::char_type buffer_type;
|
||||
|
||||
ConvertUTF8toUCS2( buffer_type* aBuffer ) : mStart(aBuffer), mBuffer(aBuffer) {}
|
||||
|
||||
size_t Length() const { return mBuffer - mStart; }
|
||||
|
||||
PRUint32 write( const value_type* start, PRUint32 N )
|
||||
{
|
||||
// algorithm assumes utf8 units won't
|
||||
// be spread across fragments
|
||||
const value_type* p = start;
|
||||
const value_type* end = start + N;
|
||||
for ( ; p != end /* && *p */; )
|
||||
{
|
||||
char c = *p++;
|
||||
|
||||
if ( UTF8traits::isASCII(c) )
|
||||
{
|
||||
*mBuffer++ = buffer_type(c);
|
||||
continue;
|
||||
}
|
||||
|
||||
PRUint32 ucs4;
|
||||
PRUint32 minUcs4;
|
||||
PRInt32 state = 0;
|
||||
|
||||
if ( UTF8traits::is2byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 6) & 0x000007C0L;
|
||||
state = 1;
|
||||
minUcs4 = 0x00000080;
|
||||
}
|
||||
else if ( UTF8traits::is3byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 12) & 0x0000F000L;
|
||||
state = 2;
|
||||
minUcs4 = 0x00000800;
|
||||
}
|
||||
else if ( UTF8traits::is4byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 18) & 0x001F0000L;
|
||||
state = 3;
|
||||
minUcs4 = 0x00010000;
|
||||
}
|
||||
else if ( UTF8traits::is5byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 24) & 0x03000000L;
|
||||
state = 4;
|
||||
minUcs4 = 0x00200000;
|
||||
}
|
||||
else if ( UTF8traits::is6byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 30) & 0x40000000L;
|
||||
state = 5;
|
||||
minUcs4 = 0x04000000;
|
||||
}
|
||||
else
|
||||
{
|
||||
NS_ERROR("not a UTF8 string");
|
||||
break;
|
||||
}
|
||||
|
||||
while ( state-- )
|
||||
{
|
||||
c = *p++;
|
||||
|
||||
if ( UTF8traits::isInSeq(c) )
|
||||
{
|
||||
PRInt32 shift = state * 6;
|
||||
ucs4 |= (PRUint32(c) & 0x3F) << shift;
|
||||
}
|
||||
else
|
||||
{
|
||||
NS_ERROR("not a UTF8 string");
|
||||
return p - start;
|
||||
}
|
||||
}
|
||||
|
||||
if ( ucs4 < minUcs4 )
|
||||
{
|
||||
// Overlong sequence
|
||||
*mBuffer++ = 0xFFFD;
|
||||
}
|
||||
else if ( ucs4 <= 0xD7FF )
|
||||
{
|
||||
*mBuffer++ = ucs4;
|
||||
}
|
||||
else if ( /* ucs4 >= 0xD800 && */ ucs4 <= 0xDFFF )
|
||||
{
|
||||
// Surrogates
|
||||
*mBuffer++ = 0xFFFD;
|
||||
}
|
||||
else if ( ucs4 == 0xFFFE || ucs4 == 0xFFFF )
|
||||
{
|
||||
// Prohibited characters
|
||||
*mBuffer++ = 0xFFFD;
|
||||
}
|
||||
else if ( ucs4 >= 0x00010000 )
|
||||
{
|
||||
*mBuffer++ = 0xFFFD;
|
||||
}
|
||||
else
|
||||
{
|
||||
if ( ucs4 != 0xFEFF ) // ignore BOM
|
||||
*mBuffer++ = ucs4;
|
||||
}
|
||||
}
|
||||
return p - start;
|
||||
}
|
||||
|
||||
private:
|
||||
buffer_type* mStart;
|
||||
buffer_type* mBuffer;
|
||||
};
|
||||
|
||||
void
|
||||
NS_ConvertUTF8toUCS2::Init( const char* aCString, PRUint32 aLength )
|
||||
NS_ConvertUTF8toUCS2::Init( const nsACString& aCString )
|
||||
{
|
||||
// Handle null string by just leaving us as a brand-new
|
||||
// uninitialized nsAutoString.
|
||||
if (! aCString)
|
||||
return;
|
||||
|
||||
// Compute space required: do this once so we don't incur multiple
|
||||
// allocations. This "optimization" is probably of dubious value...
|
||||
const char* p;
|
||||
PRUint32 count;
|
||||
for (p = aCString, count = 0; *p && count < aLength; ++count) {
|
||||
if ( 0 == (*p & 0x80) )
|
||||
p += 1; // ASCII
|
||||
else if ( 0xC0 == (*p & 0xE0) )
|
||||
p += 2; // 2 byte UTF8
|
||||
else if ( 0xE0 == (*p & 0xF0) )
|
||||
p += 3; // 3 byte UTF8
|
||||
else if ( 0xF0 == (*p & 0xF8) )
|
||||
p += 4; // 4 byte UTF8
|
||||
else if ( 0xF8 == (*p & 0xFC) )
|
||||
p += 5; // 5 byte UTF8
|
||||
else if ( 0xFC == (*p & 0xFE) )
|
||||
p += 6;
|
||||
else {
|
||||
NS_ERROR("not a UTF-8 string");
|
||||
return;
|
||||
}
|
||||
}
|
||||
|
||||
nsACString::const_iterator start, end;
|
||||
CalculateUTF8Length calculator;
|
||||
copy_string(aCString.BeginReading(start), aCString.EndReading(end), calculator);
|
||||
|
||||
PRUint32 count = calculator.Length();
|
||||
|
||||
// Grow the buffer if we need to.
|
||||
if ((count * sizeof(PRUnichar)) >= sizeof(mBuffer))
|
||||
SetCapacity(count + 1);
|
||||
SetCapacity(count);
|
||||
// |SetCapacity| normally doesn't guarantee the use we are putting it to here (see its interface comment in nsAWritableString.h),
|
||||
// we can only use it since our local implementation, |nsString::SetCapacity|, is known to do what we want
|
||||
|
||||
// We'll write directly into the new string's buffer
|
||||
PRUnichar* out = mUStr;
|
||||
// All ready? Time to convert
|
||||
|
||||
// Convert the characters.
|
||||
for (p = aCString, count = 0; *p && count < aLength; ++count) {
|
||||
char c = *p++;
|
||||
ConvertUTF8toUCS2 converter(mUStr);
|
||||
copy_string(aCString.BeginReading(start), aCString.EndReading(end), converter);
|
||||
mLength = converter.Length();
|
||||
if (mCapacity)
|
||||
mUStr[mLength] = '\0'; // null terminate
|
||||
|
||||
if( 0 == (0x80 & c)) { // ASCII
|
||||
*out++ = PRUnichar(c);
|
||||
continue;
|
||||
}
|
||||
|
||||
PRUint32 ucs4;
|
||||
PRUint32 minUcs4;
|
||||
PRInt32 state = 0;
|
||||
|
||||
if ( 0xC0 == (0xE0 & c) ) { // 2 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 6) & 0x000007C0L;
|
||||
state = 1;
|
||||
minUcs4 = 0x00000080;
|
||||
}
|
||||
else if ( 0xE0 == (0xF0 & c) ) { // 3 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 12) & 0x0000F000L;
|
||||
state = 2;
|
||||
minUcs4 = 0x00000800;
|
||||
}
|
||||
else if ( 0xF0 == (0xF8 & c) ) { // 4 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 18) & 0x001F0000L;
|
||||
state = 3;
|
||||
minUcs4 = 0x00010000;
|
||||
}
|
||||
else if ( 0xF8 == (0xFC & c) ) { // 5 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 24) & 0x03000000L;
|
||||
state = 4;
|
||||
minUcs4 = 0x00200000;
|
||||
}
|
||||
else if ( 0xFC == (0xFE & c) ) { // 6 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 30) & 0x40000000L;
|
||||
state = 5;
|
||||
minUcs4 = 0x04000000;
|
||||
}
|
||||
else {
|
||||
NS_ERROR("not a UTF8 string");
|
||||
break;
|
||||
}
|
||||
|
||||
while (state--) {
|
||||
c = *p++;
|
||||
|
||||
if ( 0x80 == (0xC0 & c) ) {
|
||||
PRInt32 shift = state * 6;
|
||||
ucs4 |= (PRUint32(c) & 0x3F) << shift;
|
||||
}
|
||||
else {
|
||||
NS_ERROR("not a UTF8 string");
|
||||
goto done; // so we minimally clean up
|
||||
}
|
||||
}
|
||||
|
||||
if (ucs4 < minUcs4) {
|
||||
// Overlong sequence
|
||||
*out++ = 0xFFFD;
|
||||
} else if (ucs4 >= 0xD800 && ucs4 <= 0xDFFF) {
|
||||
// Surrogates
|
||||
*out++ = 0xFFFD;
|
||||
} else if (ucs4 == 0xFFFE || ucs4 == 0xFFFF) {
|
||||
// Prohibited characters
|
||||
*out++ = 0xFFFD;
|
||||
} else if (ucs4 >= 0x00010000) {
|
||||
if (ucs4 >= 0x001F0000) {
|
||||
*out++ = 0xFFFD;
|
||||
}
|
||||
else {
|
||||
ucs4 -= 0x00010000;
|
||||
*out++ = 0xD800 | (0x000003FF & (ucs4 >> 10));
|
||||
*out++ = 0xDC00 | (0x000003FF & ucs4);
|
||||
}
|
||||
}
|
||||
else {
|
||||
if (0xfeff != ucs4) // ignore BOM
|
||||
*out++ = ucs4;
|
||||
}
|
||||
}
|
||||
|
||||
done:
|
||||
*out = '\0'; // null terminate
|
||||
mLength = count;
|
||||
NS_ASSERTION(count == mLength, "calculator calculated incorrect length");
|
||||
}
|
||||
|
||||
#if 0
|
||||
|
|
|
|||
|
|
@ -547,8 +547,25 @@ class NS_COM NS_ConvertASCIItoUCS2
|
|||
*/
|
||||
{
|
||||
public:
|
||||
explicit NS_ConvertASCIItoUCS2( const char* );
|
||||
NS_ConvertASCIItoUCS2( const char*, PRUint32 );
|
||||
explicit
|
||||
NS_ConvertASCIItoUCS2( const nsACString& aCString );
|
||||
|
||||
explicit
|
||||
NS_ConvertASCIItoUCS2( const nsAFlatCString& aCString )
|
||||
{
|
||||
Init( aCString.get(), aCString.Length() );
|
||||
}
|
||||
|
||||
explicit
|
||||
NS_ConvertASCIItoUCS2( const char* aCString )
|
||||
{
|
||||
Init( aCString, ~PRUint32(0) /* MAXINT */ );
|
||||
}
|
||||
|
||||
NS_ConvertASCIItoUCS2( const char* aCString, PRUint32 aLength )
|
||||
{
|
||||
Init( aCString, aLength );
|
||||
}
|
||||
|
||||
#if 0
|
||||
operator const nsDependentString() const
|
||||
|
|
@ -557,6 +574,9 @@ class NS_COM NS_ConvertASCIItoUCS2
|
|||
}
|
||||
#endif
|
||||
|
||||
protected:
|
||||
void Init( const char* aCString, PRUint32 aLength );
|
||||
|
||||
private:
|
||||
// NOT TO BE IMPLEMENTED
|
||||
NS_ConvertASCIItoUCS2( PRUnichar );
|
||||
|
|
@ -566,19 +586,25 @@ class NS_COM NS_ConvertUTF8toUCS2
|
|||
: public nsAutoString
|
||||
{
|
||||
public:
|
||||
explicit
|
||||
NS_ConvertUTF8toUCS2( const nsACString& aCString )
|
||||
{
|
||||
Init( aCString );
|
||||
}
|
||||
|
||||
explicit
|
||||
NS_ConvertUTF8toUCS2( const char* aCString )
|
||||
{
|
||||
Init( aCString, ~PRUint32(0) /* MAXINT */ );
|
||||
Init( nsDependentCString( aCString ) );
|
||||
}
|
||||
|
||||
NS_ConvertUTF8toUCS2( const char* aCString, PRUint32 aLength )
|
||||
{
|
||||
Init( aCString, aLength );
|
||||
Init( nsDependentCString( aCString, aLength ) );
|
||||
}
|
||||
|
||||
protected:
|
||||
void Init( const char* aCString, PRUint32 aLength );
|
||||
void Init( const nsACString& aCString );
|
||||
|
||||
private:
|
||||
NS_ConvertUTF8toUCS2( PRUnichar );
|
||||
|
|
|
|||
|
|
@ -1500,144 +1500,226 @@ nsAutoString::nsAutoString(const CBufDescriptor& aBuffer) : nsString() {
|
|||
AddNullTerminator(*this);
|
||||
}
|
||||
|
||||
NS_ConvertASCIItoUCS2::NS_ConvertASCIItoUCS2( const char* aCString, PRUint32 aLength )
|
||||
void
|
||||
NS_ConvertASCIItoUCS2::Init( const char* aCString, PRUint32 aLength )
|
||||
{
|
||||
Initialize(*this,mBuffer,(sizeof(mBuffer)>>eTwoByte)-1,0,eTwoByte,PR_FALSE);
|
||||
AddNullTerminator(*this);
|
||||
AppendWithConversion(aCString,aLength);
|
||||
}
|
||||
|
||||
NS_ConvertASCIItoUCS2::NS_ConvertASCIItoUCS2( const char* aCString )
|
||||
NS_ConvertASCIItoUCS2::NS_ConvertASCIItoUCS2( const nsACString& aCString )
|
||||
{
|
||||
Initialize(*this,mBuffer,(sizeof(mBuffer)>>eTwoByte)-1,0,eTwoByte,PR_FALSE);
|
||||
AddNullTerminator(*this);
|
||||
AppendWithConversion(aCString);
|
||||
SetCapacity(aCString.Length());
|
||||
|
||||
nsACString::const_iterator start; aCString.BeginReading(start);
|
||||
nsACString::const_iterator end; aCString.EndReading(end);
|
||||
|
||||
while (start != end)
|
||||
{
|
||||
const nsReadableFragment<char>& frag = start.fragment();
|
||||
AppendWithConversion(frag.mStart, frag.mEnd - frag.mStart);
|
||||
start.advance(start.size_forward());
|
||||
}
|
||||
}
|
||||
|
||||
class UTF8traits
|
||||
{
|
||||
public:
|
||||
static PRBool isASCII(char c) { return (c & 0x80) == 0x00; }
|
||||
static PRBool isInSeq(char c) { return (c & 0xC0) == 0x80; }
|
||||
static PRBool is2byte(char c) { return (c & 0xE0) == 0xC0; }
|
||||
static PRBool is3byte(char c) { return (c & 0xF0) == 0xE0; }
|
||||
static PRBool is4byte(char c) { return (c & 0xF8) == 0xF0; }
|
||||
static PRBool is5byte(char c) { return (c & 0xFC) == 0xF8; }
|
||||
static PRBool is6byte(char c) { return (c & 0xFE) == 0xFC; }
|
||||
};
|
||||
|
||||
class CalculateUTF8Length
|
||||
{
|
||||
public:
|
||||
typedef nsACString::char_type value_type;
|
||||
|
||||
CalculateUTF8Length() : mLength(0) { }
|
||||
|
||||
size_t Length() const { return mLength; }
|
||||
|
||||
PRUint32 write( const value_type* start, PRUint32 N )
|
||||
{
|
||||
// algorithm assumes utf8 units won't
|
||||
// be spread across fragments
|
||||
const value_type* p = start;
|
||||
const value_type* end = start + N;
|
||||
for ( ; p != end /* && *p */; ++mLength )
|
||||
{
|
||||
if ( UTF8traits::isASCII(*p) )
|
||||
p += 1;
|
||||
else if ( UTF8traits::is2byte(*p) )
|
||||
p += 2;
|
||||
else if ( UTF8traits::is3byte(*p) )
|
||||
p += 3;
|
||||
else if ( UTF8traits::is4byte(*p) )
|
||||
p += 4;
|
||||
else if ( UTF8traits::is5byte(*p) )
|
||||
p += 5;
|
||||
else if ( UTF8traits::is6byte(*p) )
|
||||
p += 6;
|
||||
else
|
||||
{
|
||||
NS_ERROR("not a UTF-8 string");
|
||||
break;
|
||||
}
|
||||
}
|
||||
return p - start;
|
||||
}
|
||||
|
||||
private:
|
||||
size_t mLength;
|
||||
};
|
||||
|
||||
class ConvertUTF8toUCS2
|
||||
{
|
||||
public:
|
||||
typedef nsACString::char_type value_type;
|
||||
typedef nsAString::char_type buffer_type;
|
||||
|
||||
ConvertUTF8toUCS2( buffer_type* aBuffer ) : mStart(aBuffer), mBuffer(aBuffer) {}
|
||||
|
||||
size_t Length() const { return mBuffer - mStart; }
|
||||
|
||||
PRUint32 write( const value_type* start, PRUint32 N )
|
||||
{
|
||||
// algorithm assumes utf8 units won't
|
||||
// be spread across fragments
|
||||
const value_type* p = start;
|
||||
const value_type* end = start + N;
|
||||
for ( ; p != end /* && *p */; )
|
||||
{
|
||||
char c = *p++;
|
||||
|
||||
if ( UTF8traits::isASCII(c) )
|
||||
{
|
||||
*mBuffer++ = buffer_type(c);
|
||||
continue;
|
||||
}
|
||||
|
||||
PRUint32 ucs4;
|
||||
PRUint32 minUcs4;
|
||||
PRInt32 state = 0;
|
||||
|
||||
if ( UTF8traits::is2byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 6) & 0x000007C0L;
|
||||
state = 1;
|
||||
minUcs4 = 0x00000080;
|
||||
}
|
||||
else if ( UTF8traits::is3byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 12) & 0x0000F000L;
|
||||
state = 2;
|
||||
minUcs4 = 0x00000800;
|
||||
}
|
||||
else if ( UTF8traits::is4byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 18) & 0x001F0000L;
|
||||
state = 3;
|
||||
minUcs4 = 0x00010000;
|
||||
}
|
||||
else if ( UTF8traits::is5byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 24) & 0x03000000L;
|
||||
state = 4;
|
||||
minUcs4 = 0x00200000;
|
||||
}
|
||||
else if ( UTF8traits::is6byte(c) )
|
||||
{
|
||||
ucs4 = (PRUint32(c) << 30) & 0x40000000L;
|
||||
state = 5;
|
||||
minUcs4 = 0x04000000;
|
||||
}
|
||||
else
|
||||
{
|
||||
NS_ERROR("not a UTF8 string");
|
||||
break;
|
||||
}
|
||||
|
||||
while ( state-- )
|
||||
{
|
||||
c = *p++;
|
||||
|
||||
if ( UTF8traits::isInSeq(c) )
|
||||
{
|
||||
PRInt32 shift = state * 6;
|
||||
ucs4 |= (PRUint32(c) & 0x3F) << shift;
|
||||
}
|
||||
else
|
||||
{
|
||||
NS_ERROR("not a UTF8 string");
|
||||
return p - start;
|
||||
}
|
||||
}
|
||||
|
||||
if ( ucs4 < minUcs4 )
|
||||
{
|
||||
// Overlong sequence
|
||||
*mBuffer++ = 0xFFFD;
|
||||
}
|
||||
else if ( ucs4 <= 0xD7FF )
|
||||
{
|
||||
*mBuffer++ = ucs4;
|
||||
}
|
||||
else if ( /* ucs4 >= 0xD800 && */ ucs4 <= 0xDFFF )
|
||||
{
|
||||
// Surrogates
|
||||
*mBuffer++ = 0xFFFD;
|
||||
}
|
||||
else if ( ucs4 == 0xFFFE || ucs4 == 0xFFFF )
|
||||
{
|
||||
// Prohibited characters
|
||||
*mBuffer++ = 0xFFFD;
|
||||
}
|
||||
else if ( ucs4 >= 0x00010000 )
|
||||
{
|
||||
*mBuffer++ = 0xFFFD;
|
||||
}
|
||||
else
|
||||
{
|
||||
if ( ucs4 != 0xFEFF ) // ignore BOM
|
||||
*mBuffer++ = ucs4;
|
||||
}
|
||||
}
|
||||
return p - start;
|
||||
}
|
||||
|
||||
private:
|
||||
buffer_type* mStart;
|
||||
buffer_type* mBuffer;
|
||||
};
|
||||
|
||||
void
|
||||
NS_ConvertUTF8toUCS2::Init( const char* aCString, PRUint32 aLength )
|
||||
NS_ConvertUTF8toUCS2::Init( const nsACString& aCString )
|
||||
{
|
||||
// Handle null string by just leaving us as a brand-new
|
||||
// uninitialized nsAutoString.
|
||||
if (! aCString)
|
||||
return;
|
||||
|
||||
// Compute space required: do this once so we don't incur multiple
|
||||
// allocations. This "optimization" is probably of dubious value...
|
||||
const char* p;
|
||||
PRUint32 count;
|
||||
for (p = aCString, count = 0; *p && count < aLength; ++count) {
|
||||
if ( 0 == (*p & 0x80) )
|
||||
p += 1; // ASCII
|
||||
else if ( 0xC0 == (*p & 0xE0) )
|
||||
p += 2; // 2 byte UTF8
|
||||
else if ( 0xE0 == (*p & 0xF0) )
|
||||
p += 3; // 3 byte UTF8
|
||||
else if ( 0xF0 == (*p & 0xF8) )
|
||||
p += 4; // 4 byte UTF8
|
||||
else if ( 0xF8 == (*p & 0xFC) )
|
||||
p += 5; // 5 byte UTF8
|
||||
else if ( 0xFC == (*p & 0xFE) )
|
||||
p += 6;
|
||||
else {
|
||||
NS_ERROR("not a UTF-8 string");
|
||||
return;
|
||||
}
|
||||
}
|
||||
|
||||
nsACString::const_iterator start, end;
|
||||
CalculateUTF8Length calculator;
|
||||
copy_string(aCString.BeginReading(start), aCString.EndReading(end), calculator);
|
||||
|
||||
PRUint32 count = calculator.Length();
|
||||
|
||||
// Grow the buffer if we need to.
|
||||
if ((count * sizeof(PRUnichar)) >= sizeof(mBuffer))
|
||||
SetCapacity(count + 1);
|
||||
SetCapacity(count);
|
||||
// |SetCapacity| normally doesn't guarantee the use we are putting it to here (see its interface comment in nsAWritableString.h),
|
||||
// we can only use it since our local implementation, |nsString::SetCapacity|, is known to do what we want
|
||||
|
||||
// We'll write directly into the new string's buffer
|
||||
PRUnichar* out = mUStr;
|
||||
// All ready? Time to convert
|
||||
|
||||
// Convert the characters.
|
||||
for (p = aCString, count = 0; *p && count < aLength; ++count) {
|
||||
char c = *p++;
|
||||
ConvertUTF8toUCS2 converter(mUStr);
|
||||
copy_string(aCString.BeginReading(start), aCString.EndReading(end), converter);
|
||||
mLength = converter.Length();
|
||||
if (mCapacity)
|
||||
mUStr[mLength] = '\0'; // null terminate
|
||||
|
||||
if( 0 == (0x80 & c)) { // ASCII
|
||||
*out++ = PRUnichar(c);
|
||||
continue;
|
||||
}
|
||||
|
||||
PRUint32 ucs4;
|
||||
PRUint32 minUcs4;
|
||||
PRInt32 state = 0;
|
||||
|
||||
if ( 0xC0 == (0xE0 & c) ) { // 2 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 6) & 0x000007C0L;
|
||||
state = 1;
|
||||
minUcs4 = 0x00000080;
|
||||
}
|
||||
else if ( 0xE0 == (0xF0 & c) ) { // 3 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 12) & 0x0000F000L;
|
||||
state = 2;
|
||||
minUcs4 = 0x00000800;
|
||||
}
|
||||
else if ( 0xF0 == (0xF8 & c) ) { // 4 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 18) & 0x001F0000L;
|
||||
state = 3;
|
||||
minUcs4 = 0x00010000;
|
||||
}
|
||||
else if ( 0xF8 == (0xFC & c) ) { // 5 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 24) & 0x03000000L;
|
||||
state = 4;
|
||||
minUcs4 = 0x00200000;
|
||||
}
|
||||
else if ( 0xFC == (0xFE & c) ) { // 6 bytes UTF8
|
||||
ucs4 = (PRUint32(c) << 30) & 0x40000000L;
|
||||
state = 5;
|
||||
minUcs4 = 0x04000000;
|
||||
}
|
||||
else {
|
||||
NS_ERROR("not a UTF8 string");
|
||||
break;
|
||||
}
|
||||
|
||||
while (state--) {
|
||||
c = *p++;
|
||||
|
||||
if ( 0x80 == (0xC0 & c) ) {
|
||||
PRInt32 shift = state * 6;
|
||||
ucs4 |= (PRUint32(c) & 0x3F) << shift;
|
||||
}
|
||||
else {
|
||||
NS_ERROR("not a UTF8 string");
|
||||
goto done; // so we minimally clean up
|
||||
}
|
||||
}
|
||||
|
||||
if (ucs4 < minUcs4) {
|
||||
// Overlong sequence
|
||||
*out++ = 0xFFFD;
|
||||
} else if (ucs4 >= 0xD800 && ucs4 <= 0xDFFF) {
|
||||
// Surrogates
|
||||
*out++ = 0xFFFD;
|
||||
} else if (ucs4 == 0xFFFE || ucs4 == 0xFFFF) {
|
||||
// Prohibited characters
|
||||
*out++ = 0xFFFD;
|
||||
} else if (ucs4 >= 0x00010000) {
|
||||
if (ucs4 >= 0x001F0000) {
|
||||
*out++ = 0xFFFD;
|
||||
}
|
||||
else {
|
||||
ucs4 -= 0x00010000;
|
||||
*out++ = 0xD800 | (0x000003FF & (ucs4 >> 10));
|
||||
*out++ = 0xDC00 | (0x000003FF & ucs4);
|
||||
}
|
||||
}
|
||||
else {
|
||||
if (0xfeff != ucs4) // ignore BOM
|
||||
*out++ = ucs4;
|
||||
}
|
||||
}
|
||||
|
||||
done:
|
||||
*out = '\0'; // null terminate
|
||||
mLength = count;
|
||||
NS_ASSERTION(count == mLength, "calculator calculated incorrect length");
|
||||
}
|
||||
|
||||
#if 0
|
||||
|
|
|
|||
|
|
@ -547,8 +547,25 @@ class NS_COM NS_ConvertASCIItoUCS2
|
|||
*/
|
||||
{
|
||||
public:
|
||||
explicit NS_ConvertASCIItoUCS2( const char* );
|
||||
NS_ConvertASCIItoUCS2( const char*, PRUint32 );
|
||||
explicit
|
||||
NS_ConvertASCIItoUCS2( const nsACString& aCString );
|
||||
|
||||
explicit
|
||||
NS_ConvertASCIItoUCS2( const nsAFlatCString& aCString )
|
||||
{
|
||||
Init( aCString.get(), aCString.Length() );
|
||||
}
|
||||
|
||||
explicit
|
||||
NS_ConvertASCIItoUCS2( const char* aCString )
|
||||
{
|
||||
Init( aCString, ~PRUint32(0) /* MAXINT */ );
|
||||
}
|
||||
|
||||
NS_ConvertASCIItoUCS2( const char* aCString, PRUint32 aLength )
|
||||
{
|
||||
Init( aCString, aLength );
|
||||
}
|
||||
|
||||
#if 0
|
||||
operator const nsDependentString() const
|
||||
|
|
@ -557,6 +574,9 @@ class NS_COM NS_ConvertASCIItoUCS2
|
|||
}
|
||||
#endif
|
||||
|
||||
protected:
|
||||
void Init( const char* aCString, PRUint32 aLength );
|
||||
|
||||
private:
|
||||
// NOT TO BE IMPLEMENTED
|
||||
NS_ConvertASCIItoUCS2( PRUnichar );
|
||||
|
|
@ -566,19 +586,25 @@ class NS_COM NS_ConvertUTF8toUCS2
|
|||
: public nsAutoString
|
||||
{
|
||||
public:
|
||||
explicit
|
||||
NS_ConvertUTF8toUCS2( const nsACString& aCString )
|
||||
{
|
||||
Init( aCString );
|
||||
}
|
||||
|
||||
explicit
|
||||
NS_ConvertUTF8toUCS2( const char* aCString )
|
||||
{
|
||||
Init( aCString, ~PRUint32(0) /* MAXINT */ );
|
||||
Init( nsDependentCString( aCString ) );
|
||||
}
|
||||
|
||||
NS_ConvertUTF8toUCS2( const char* aCString, PRUint32 aLength )
|
||||
{
|
||||
Init( aCString, aLength );
|
||||
Init( nsDependentCString( aCString, aLength ) );
|
||||
}
|
||||
|
||||
protected:
|
||||
void Init( const char* aCString, PRUint32 aLength );
|
||||
void Init( const nsACString& aCString );
|
||||
|
||||
private:
|
||||
NS_ConvertUTF8toUCS2( PRUnichar );
|
||||
|
|
|
|||
Загрузка…
Ссылка в новой задаче