static int
pg_sjis_verifier(const unsigned char *s, int len)
{
int l, mbl;
unsigned char c1, c2;
l = mbl = pg_sjis_mblen(s);
if (len < l)
return -1;
if (l == 1) /* pg_sjis_mblen already verified it */
return mbl;
c1 = *s++;
c2 = *s;
if (!ISSJISHEAD(c1) || !ISSJISTAIL(c2))
return -1;
return mbl;
}
/* ----------
* conv_proc(
* INTEGER, -- source encoding id
* INTEGER, -- destination encoding id
* CSTRING, -- source string (null terminated C string)
* CSTRING, -- destination string (null terminated C string)
* INTEGER -- source string length
* ) returns VOID;
* ----------
*/
Datum
sjis_eudc_to_utf8(PG_FUNCTION_ARGS)
{
unsigned char *src = (unsigned char *) PG_GETARG_CSTRING(2);
unsigned char *dest = (unsigned char *) PG_GETARG_CSTRING(3);
unsigned char *p;
unsigned char *fallback_character = NULL;
int len = PG_GETARG_INT32(4);
int sjis_len;
int clen;
CHECK_ENCODING_CONVERSION_ARGS(PG_SJIS, PG_UTF8);
if (sjis_to_utf8 == NULL)
sjis_to_utf8 = load_external_function(
"utf8_and_sjis", "sjis_to_utf8", true, NULL);
*dest = '\0';
p = src;
sjis_len = 0;
for (; len > 0; len -= clen)
{
const unsigned char *c = p + sjis_len;
if (c[0] == '\0')
report_invalid_encoding(PG_SJIS, (const char *) p + sjis_len, len);
if (c[0] >= 0xf0 && c[0] <= 0xf9 && len >= 2 && ISSJISTAIL(c[1]))
{
int ucs;
int m;
int n;
clen = 2;
/* SJIS to UTF8 */
if (sjis_len > 0)
{
DirectFunctionCall5(sjis_to_utf8, PG_SJIS, PG_UTF8,
CStringGetDatum(p), CStringGetDatum(dest),
sjis_len);
dest = dest + strlen((char *) dest);
p += sjis_len;
sjis_len = 0;
}
p += clen;
elog(eudc_log_level,
"eudc character found: %02x%02x in SJIS to UTF8 conversion",
c[0], c[1]);
/* SJIS EUDC to UTF8 */
if (eudc_fallback_character && eudc_fallback_character[0])
{
/* map to fallback character */
int i;
if (fallback_character == NULL)
{
fallback_character = pg_do_encoding_conversion(
(unsigned char *) eudc_fallback_character,
strlen(eudc_fallback_character),
GetDatabaseEncoding(),
PG_UTF8);
}
for (i = 0; fallback_character[i]; i++)
*dest++ = fallback_character[i];
}
else
{
/* linear mapping */
n = c[0] - 0xf0;
m = c[1] - 0x40;
if (m >= 0x40)
m--;
ucs = 0xe000 + n * 188 + m;
*dest++ = (ucs >> 12) | 0xe0;
*dest++ = (ucs & 0x0fc0) >> 6 | 0x80;
*dest++ = (ucs & 0x003f) | 0x80;
}
*dest = '\0';
}
else
{
