*/ void Debug_Uni(const REBSER *ser)
/*
** Print debug unicode string followed by a newline.
**
***********************************************************************/
{
REBCNT ul;
REBCNT bl;
REBYTE buf[1024];
REBUNI *up = UNI_HEAD(ser);
REBINT size = Length_As_UTF8(up, SERIES_TAIL(ser), TRUE, OS_CRLF);
REBINT disabled = GC_Disabled;
GC_Disabled = 1;
while (size > 0) {
ul = Encode_UTF8(buf, MIN(size, 1020), up, &bl, TRUE, OS_CRLF);
Debug_String(buf, bl, 0, 0);
size -= ul;
up += ul;
}
Debug_Line();
assert(GC_Disabled == 1);
GC_Disabled = disabled;
}
*/ RL_API REBCNT RL_Encode_UTF8(REBYTE *dst, REBINT max, const void *src, REBCNT *len, REBFLG uni, REBFLG ccr)
/*
** Encode the unicode into UTF8 byte string.
**
** Returns:
** Number of dst bytes used.
**
** Arguments:
** dst - destination for encoded UTF8 bytes
** max - maximum size of the result in bytes
** src - source array of bytes or wide characters
** len - input is source length, updated to reflect src chars used
** uni - true if src is in wide character format
** ccr - convert linefeed + carriage-return into just linefeed
**
** Notes:
** Does not add a terminator.
**
** !!! Host code is not supposed to call any Rebol routines except
** for those in the RL_Api. This exposes Rebol's internal UTF8
** length routine, as it was being used by the Linux host code by
** Atronix. Should be reviewed along with the rest of the RL_Api.
**
***********************************************************************/
{
return Encode_UTF8(
dst,
max,
src,
len,
(uni ? OPT_ENC_UNISRC : 0) | (ccr ? OPT_ENC_CRLF : 0)
);
}
*/ REBSER *Encode_UTF8_String(void *src, REBCNT len, REBFLG uni, REBFLG opts)
/*
** Do all the details to encode a string as UTF8.
** No_copy means do not make a copy.
** Result can be a shared buffer!
**
***********************************************************************/
{
REBSER *ser = BUF_FORM; // a shared buffer
REBCNT size;
REBYTE *cp;
REBFLG ccr = GET_FLAG(opts, ENC_OPT_CRLF);
if (uni) {
REBUNI *up = (REBUNI*)src;
size = Length_As_UTF8(up, len, TRUE, (REBOOL)ccr);
cp = Reset_Buffer(ser, size + (GET_FLAG(opts, ENC_OPT_BOM) ? 3 : 0));
UNUSED(cp);
Encode_UTF8(Reset_Buffer(ser, size), size, up, &len, TRUE, ccr);
}
else {
REBYTE *bp = (REBYTE*)src;
if (Is_Not_ASCII(bp, len)) {
size = Length_As_UTF8((REBUNI*)bp, len, FALSE, (REBOOL)ccr);
cp = Reset_Buffer(ser, size + (GET_FLAG(opts, ENC_OPT_BOM) ? 3 : 0));
Encode_UTF8(cp, size, bp, &len, FALSE, ccr);
}
else if (GET_FLAG(opts, ENC_OPT_NO_COPY)) return 0;
else return Copy_Bytes(bp, len);
}
SERIES_TAIL(ser) = len;
STR_TERM(ser);
return Copy_Bytes(BIN_HEAD(ser), len);
}
//
// RL_Encode_UTF8: C
//
// Encode the unicode into UTF8 byte string.
//
// Returns:
// Number of dst bytes used.
//
// Arguments:
// dst - destination for encoded UTF8 bytes
// max - maximum size of the result in bytes
// src - source array of bytes or wide characters
// len - input is source length, updated to reflect src chars used
// unicode - true if src is in wide character format
// crlf_to_lf - convert carriage-return + linefeed into just linefeed
//
// Notes:
// Does not add a terminator.
//
// !!! Host code is not supposed to call any Rebol routines except
// for those in the RL_Api. This exposes Rebol's internal UTF8
// length routine, as it was being used by the Linux host code by
// Atronix. Should be reviewed along with the rest of the RL_Api.
//
RL_API REBCNT RL_Encode_UTF8(
REBYTE *dst,
REBINT max,
const void *src,
REBCNT *len,
REBOOL unicode,
REBOOL crlf_to_lf
) {
return Encode_UTF8(
dst,
max,
src,
len,
(unicode ? OPT_ENC_UNISRC : 0) | (crlf_to_lf ? OPT_ENC_CRLF : 0)
);
}