String UCS4toUTF8(UInt32 ucs4char)
{
StringBuffer sb(5); // max 4 chars + null
UCS4toUTF8(ucs4char, sb);
return sb.releaseString();
}
GP<GStringRep>
GStringRep::Unicode::create(
void const * const xbuf,
unsigned int bufsize,
EncodeType t)
{
GP<GStringRep> gretval;
GStringRep *retval=0;
void const * const buf=checkmarks(xbuf,bufsize,t);
if(buf && bufsize)
{
unsigned char const *eptr=(unsigned char *)buf;
unsigned int maxutf8size=0;
void const* const xeptr=(void const *)((size_t)eptr+bufsize);
switch(t)
{
case XUCS4:
case XUCS4BE:
case XUCS4LE:
case XUCS4_2143:
case XUCS4_3412:
{
for(unsigned long w;
(eptr<xeptr)&&(w=*(unsigned long const *)eptr);
eptr+=sizeof(unsigned long))
{
maxutf8size+=(w>0x7f)?6:1;
}
break;
}
case XUTF16:
case XUTF16BE:
case XUTF16LE:
{
for(unsigned short w;
(eptr<xeptr)&&(w=*(unsigned short const *)eptr);
eptr+=sizeof(unsigned short))
{
maxutf8size+=3;
}
break;
}
case XUTF8:
for(;(eptr<xeptr)&&*eptr;maxutf8size++,eptr++)
EMPTY_LOOP;
break;
case XEBCDIC:
for(;(eptr<xeptr)&&*eptr;eptr++)
{
maxutf8size+=(*eptr>0x7f)?2:1;
}
break;
default:
break;
}
unsigned char *utf8buf=0;
GPBuffer<unsigned char> gutf8buf(utf8buf,maxutf8size+1);
utf8buf[0]=0;
if (maxutf8size)
{
unsigned char *optr=utf8buf;
int len=0;
unsigned char const *iptr=(unsigned char *)buf;
unsigned short const *sptr=(unsigned short *)buf;
unsigned long w;
switch(t)
{
case XUCS4:
for(;
(iptr<eptr)&&(w=*(unsigned long const *)iptr);
len++,iptr+=sizeof(unsigned long const))
{
optr=UCS4toUTF8(w,optr);
}
break;
case XUCS4BE:
for(;(w=UCS4BEtoUCS4(iptr,eptr));len++)
{
optr=UCS4toUTF8(w,optr);
}
break;
case XUCS4LE:
for(;(w=UCS4LEtoUCS4(iptr,eptr));len++)
{
optr=UCS4toUTF8(w,optr);
}
break;
case XUCS4_2143:
for(;(w=UCS4_2143toUCS4(iptr,eptr));len++)
{
optr=UCS4toUTF8(w,optr);
}
break;
case XUCS4_3412:
for(;(w=UCS4_3412toUCS4(iptr,eptr));len++)
{
optr=UCS4toUTF8(w,optr);
}
break;
case XUTF16:
for(;(w=xUTF16toUCS4(sptr,eptr));len++)
{
optr=UCS4toUTF8(w,optr);
}
break;
case XUTF16BE:
for(;(w=UTF16BEtoUCS4(iptr,eptr));len++)
{
optr=UCS4toUTF8(w,optr);
}
break;
case XUTF16LE:
for(;(w=UTF16LEtoUCS4(iptr,eptr));len++)
{
optr=UCS4toUTF8(w,optr);
}
break;
case XUTF8:
for(;(w=UTF8toUCS4(iptr,eptr));len++)
{
optr=UCS4toUTF8(w,optr);
}
break;
case XEBCDIC:
for(;(iptr<eptr)&&(w=*iptr++);len++)
{
optr=UCS4toUTF8(w,optr);
}
break;
default:
break;
}
const unsigned int size=(size_t)optr-(size_t)utf8buf;
if(size)
{
retval=(gretval=GStringRep::Unicode::create(size));
memcpy(retval->data,utf8buf,size);
}else
{
retval=(gretval=GStringRep::Unicode::create(1));
retval->size=size;
}
retval->data[size]=0;
gutf8buf.resize(0);
const size_t s=(size_t)eptr-(size_t)iptr;
retval->set_remainder(iptr,s,t);
}
}
if(!retval)
{
retval=(gretval=GStringRep::Unicode::create(1));
retval->data[0]=0;
retval->size=0;
retval->set_remainder(0,0,t);
}
return gretval;
}
String UCS2toUTF8(UInt16 ucs2char)
{
// UCS2 and UCS4 are the same, only different sizes.
return UCS4toUTF8(ucs2char);
}
