idn_result_t
idn_delimitermap_add(idn_delimitermap_t ctx, unsigned long delimiter) {
idn_result_t r;
assert(ctx != NULL && ctx->ndelimiters <= ctx->delimiter_size);
TRACE(("idn_delimitermap_add(delimiter=\\x%04lx)\n", delimiter));
if (delimiter == 0 || delimiter > UNICODE_MAX ||
IS_SURROGATE_HIGH(delimiter) || IS_SURROGATE_LOW(delimiter)) {
r = idn_invalid_codepoint;
goto ret;
}
if (ctx->ndelimiters == ctx->delimiter_size) {
unsigned long *new_delimiters;
new_delimiters = (unsigned long *) realloc(ctx->delimiters,
sizeof(unsigned long) * ctx->delimiter_size * 2);
if (new_delimiters == NULL) {
r = idn_nomemory;
goto ret;
}
ctx->delimiters = new_delimiters;
ctx->delimiter_size *= 2;
}
ctx->delimiters[ctx->ndelimiters] = delimiter;
ctx->ndelimiters++;
r = idn_success;
ret:
TRACE(("idn_delimitermap_add(): %s\n", idn_result_tostring(r)));
return (r);
}
idn_result_t
idn_ucs4_ucs4toutf16(const unsigned long *ucs4, unsigned short *utf16,
size_t tolen) {
unsigned short *utf16p = utf16;
unsigned long v;
idn_result_t r;
TRACE(("idn_ucs4_ucs4toutf16(ucs4=\"%s\", tolen=%d)\n",
idn__debug_ucs4xstring(ucs4, 50), (int)tolen));
while (*ucs4 != '\0') {
v = *ucs4++;
if (IS_SURROGATE_LOW(v) || IS_SURROGATE_HIGH(v)) {
WARNING(("idn_ucs4_ucs4toutf16: UCS4 string contains "
"surrogate pair\n"));
r = idn_invalid_encoding;
goto ret;
} else if (v > 0xffff) {
/* Convert to surrogate pair */
if (v >= 0x110000) {
r = idn_invalid_encoding;
goto ret;
}
if (tolen < 2) {
r = idn_buffer_overflow;
goto ret;
}
*utf16p++ = SURROGATE_HIGH(v);
*utf16p++ = SURROGATE_LOW(v);
tolen -= 2;
} else {
if (tolen < 1) {
r = idn_buffer_overflow;
goto ret;
}
*utf16p++ = v;
tolen--;
}
}
if (tolen < 1) {
r = idn_buffer_overflow;
goto ret;
}
*utf16p = '\0';
r = idn_success;
ret:
if (r == idn_success) {
TRACE(("idn_ucs4_ucs4toutf16(): success (utf16=\"%s\")\n",
idn__debug_utf16xstring(utf16, 50)));
} else {
TRACE(("idn_ucs4_ucs4toutf16(): %s\n",
idn_result_tostring(r)));
}
return (r);
}
idn_result_t
idn_ucs4_utf16toucs4(const unsigned short *utf16, unsigned long *ucs4,
size_t tolen) {
unsigned long *ucs4p = ucs4;
unsigned short v0, v1;
idn_result_t r;
TRACE(("idn_ucs4_utf16toucs4(utf16=\"%s\", tolen=%d)\n",
idn__debug_utf16xstring(utf16, 50), (int)tolen));
while (*utf16 != '\0') {
v0 = *utf16;
if (tolen < 1) {
r = idn_buffer_overflow;
goto ret;
}
if (IS_SURROGATE_HIGH(v0)) {
v1 = *(utf16 + 1);
if (!IS_SURROGATE_LOW(v1)) {
WARNING(("idn_ucs4_utf16toucs4: "
"corrupted surrogate pair\n"));
r = idn_invalid_encoding;
goto ret;
}
*ucs4p++ = COMBINE_SURROGATE(v0, v1);
tolen--;
utf16 += 2;
} else {
*ucs4p++ = v0;
tolen--;
utf16++;
}
}
if (tolen < 1) {
r = idn_buffer_overflow;
goto ret;
}
*ucs4p = '\0';
r = idn_success;
ret:
if (r == idn_success) {
TRACE(("idn_ucs4_utf16toucs4(): success (ucs4=\"%s\")\n",
idn__debug_ucs4xstring(ucs4, 50)));
} else {
TRACE(("idn_ucs4_utf16toucs4(): %s\n",
idn_result_tostring(r)));
}
return (r);
}
idn_result_t
idn_ucs4_ucs4toutf8(const unsigned long *ucs4, char *utf8, size_t tolen) {
unsigned char *utf8p = (unsigned char *)utf8;
unsigned long v;
int width;
int mask;
int offset;
idn_result_t r;
TRACE(("idn_ucs4_ucs4toutf8(ucs4=\"%s\", tolen=%d)\n",
idn__debug_ucs4xstring(ucs4, 50), (int)tolen));
while (*ucs4 != '\0') {
v = *ucs4++;
if (IS_SURROGATE_LOW(v) || IS_SURROGATE_HIGH(v)) {
WARNING(("idn_ucs4_ucs4toutf8: UCS4 string contains "
"surrogate pair\n"));
r = idn_invalid_encoding;
goto ret;
}
if (v < 0x80) {
mask = 0;
width = 1;
} else if (v < 0x800) {
mask = 0xc0;
width = 2;
} else if (v < 0x10000) {
mask = 0xe0;
width = 3;
} else if (v < 0x200000) {
mask = 0xf0;
width = 4;
} else if (v < 0x4000000) {
mask = 0xf8;
width = 5;
} else if (v < 0x80000000) {
mask = 0xfc;
width = 6;
} else {
WARNING(("idn_ucs4_ucs4toutf8: invalid character\n"));
r = idn_invalid_encoding;
goto ret;
}
if (tolen < width) {
r = idn_buffer_overflow;
goto ret;
}
offset = 6 * (width - 1);
*utf8p++ = (v >> offset) | mask;
mask = 0x80;
while (offset > 0) {
offset -= 6;
*utf8p++ = ((v >> offset) & 0x3f) | mask;
}
tolen -= width;
}
if (tolen < 1) {
r = idn_buffer_overflow;
goto ret;
}
*utf8p = '\0';
r = idn_success;
ret:
if (r == idn_success) {
TRACE(("idn_ucs4_ucs4toutf8(): success (utf8=\"%s\")\n",
idn__debug_xstring(utf8, 50)));
} else {
TRACE(("idn_ucs4_ucs4toutf8(): %s\n",
idn_result_tostring(r)));
}
return (r);
}
idn_result_t
idn_ucs4_utf8toucs4(const char *utf8, unsigned long *ucs4, size_t tolen) {
const unsigned char *utf8p = (const unsigned char *)utf8;
unsigned long *ucs4p = ucs4;
unsigned long v, min;
unsigned char c;
int width;
int i;
idn_result_t r;
TRACE(("idn_ucs4_utf8toucs4(utf8=\"%s\", tolen=%d)\n",
idn__debug_xstring(utf8, 50), (int)tolen));
while(*utf8p != '\0') {
c = *utf8p++;
if (c < 0x80) {
v = c;
min = 0;
width = 1;
} else if (c < 0xc0) {
WARNING(("idn_ucs4_utf8toucs4: invalid character\n"));
r = idn_invalid_encoding;
goto ret;
} else if (c < 0xe0) {
v = c & 0x1f;
min = 0x80;
width = 2;
} else if (c < 0xf0) {
v = c & 0x0f;
min = 0x800;
width = 3;
} else if (c < 0xf8) {
v = c & 0x07;
min = 0x10000;
width = 4;
} else if (c < 0xfc) {
v = c & 0x03;
min = 0x200000;
width = 5;
} else if (c < 0xfe) {
v = c & 0x01;
min = 0x4000000;
width = 6;
} else {
WARNING(("idn_ucs4_utf8toucs4: invalid character\n"));
r = idn_invalid_encoding;
goto ret;
}
for (i = width - 1; i > 0; i--) {
c = *utf8p++;
if (c < 0x80 || 0xc0 <= c) {
WARNING(("idn_ucs4_utf8toucs4: "
"invalid character\n"));
r = idn_invalid_encoding;
goto ret;
}
v = (v << 6) | (c & 0x3f);
}
if (v < min) {
WARNING(("idn_ucs4_utf8toucs4: invalid character\n"));
r = idn_invalid_encoding;
goto ret;
}
if (IS_SURROGATE_LOW(v) || IS_SURROGATE_HIGH(v)) {
WARNING(("idn_ucs4_utf8toucs4: UTF-8 string contains "
"surrogate pair\n"));
r = idn_invalid_encoding;
goto ret;
}
if (tolen < 1) {
r = idn_buffer_overflow;
goto ret;
}
tolen--;
*ucs4p++ = v;
}
if (tolen < 1) {
r = idn_buffer_overflow;
goto ret;
}
*ucs4p = '\0';
r = idn_success;
ret:
if (r == idn_success) {
TRACE(("idn_ucs4_utf8toucs4(): success (ucs4=\"%s\")\n",
idn__debug_ucs4xstring(ucs4, 50)));
} else {
TRACE(("idn_ucs4_utf8toucs4(): %s\n",
idn_result_tostring(r)));
}
return (r);
}