idn_result_t
idn_decodename2(idn_action_t actions, const char *from, char *to, size_t tolen,
const char *auxencoding) {
idn_result_t r;
assert(from != NULL && to != NULL);
TRACE(("idn_decodename2(actions=%s, from=\"%s\", tolen=%d)\n",
idn__res_actionstostring(actions),
idn__debug_xstring(from, 50), (int)tolen));
if (!initialized && ((r = idn_nameinit(0)) != idn_success))
goto ret;
r = idn_res_decodename2(default_conf, actions, from, to, tolen,
auxencoding);
ret:
if (r == idn_success) {
TRACE(("idn_decodename2(): success (to=\"%s\")\n",
idn__debug_xstring(to, 50)));
} else {
TRACE(("idn_decodename2(): %s\n", idn_result_tostring(r)));
}
return (r);
}
/*
* Add an alias to 'ctx'.
*/
idn_result_t
idn__langalias_add(idn__langalias_t ctx, const char *alias_name,
const char *real_name) {
idn_result_t r = idn_success;
idn__langalias_entry_t new_entry = NULL;
assert(ctx != NULL && alias_name != NULL && real_name != NULL);
TRACE(("idn__langalias_add(alias=\"%s\", real=\"%s\")\n",
idn__debug_xstring(alias_name), idn__debug_xstring(real_name)));
new_entry = (idn__langalias_entry_t)malloc(sizeof(*new_entry));
if (new_entry == NULL) {
r = idn_nomemory;
goto ret;
}
new_entry->alias_name = NULL;
new_entry->real_name = NULL;
new_entry->next = NULL;
new_entry->alias_name = idn__util_strdup(alias_name);
if (new_entry->alias_name == NULL) {
r = idn_nomemory;
goto ret;
}
idn__util_asclower(new_entry->alias_name);
new_entry->real_name = idn__util_strdup(real_name);
if (new_entry->real_name == NULL) {
r = idn_nomemory;
goto ret;
}
idn__util_asclower(new_entry->real_name);
r = idn__strhash8_put(ctx->hash, new_entry->alias_name,
new_entry->real_name);
if (r != idn_success)
goto ret;
if (ctx->entries_head == NULL) {
ctx->entries_head = new_entry;
ctx->entries_tail = new_entry;
} else {
ctx->entries_tail->next = new_entry;
ctx->entries_tail = new_entry;
}
ret:
TRACE(("idn__langalias_add(): %s\n", idn_result_tostring(r)));
if (r != idn_success) {
if (new_entry != NULL) {
free(new_entry->alias_name);
free(new_entry->real_name);
free(new_entry);
}
}
return (r);
}
idn_result_t
idn_converter_convtoucs4(idn_converter_t ctx, const char *from,
unsigned long *to, size_t tolen) {
idn_result_t r;
assert(ctx != NULL && from != NULL && to != NULL);
TRACE(("idn_converter_convtoucs4(ctx=%s, from=\"%s\", tolen=%d)\n",
ctx->local_encoding_name, idn__debug_xstring(from, 50),
(int)tolen));
if (!ctx->opened_convtoucs4) {
r = (*ctx->ops->opentoucs4)(ctx, &(ctx->private_data));
if (r != idn_success)
goto ret;
ctx->opened_convtoucs4 = 1;
}
r = (*ctx->ops->convtoucs4)(ctx, ctx->private_data, from, to, tolen);
ret:
if (r == idn_success) {
TRACE(("idn_converter_convtoucs4(): success (to=\"%s\")\n",
idn__debug_ucs4xstring(to, 50)));
} else {
TRACE(("idn_converter_convtoucs4(): %s\n",
idn_result_tostring(r)));
}
return (r);
}
/*
* Find a real name for 'alias_name'.
*/
const char *
idn__langalias_find(idn__langalias_t ctx, const char *alias_name) {
const char *real_name;
const char *result;
assert(ctx != NULL && alias_name != NULL);
TRACE(("idn__langalias_find(alias=\"%s\")\n",
idn__debug_xstring(alias_name)));
real_name = idn__strhash8_get(ctx->hash, alias_name);
result = (real_name != NULL) ? real_name : alias_name;
TRACE(("idn__langalias_find(): success (real=\"%s\")\n",
idn__debug_xstring(result)));
return (result);
}
idn_result_t
mdn_decodename(int actions, const char *from, char *to, size_t tolen) {
idn_result_t r;
assert(from != NULL && to != NULL);
TRACE(("idn_decodename(actions=%s, from=\"%s\", tolen=%d)\n",
idn__res_actionstostring(actions),
idn__debug_xstring(from, 50), (int)tolen));
if (!initialized && ((r = idn_nameinit(1)) != idn_success))
return (r);
return (idn_res_decodename(default_conf, actions, from, to, tolen));
}
idn_result_t
idn_selectiveencode_findregion(const char *s,
char **startp, char **endp)
{
char *non_ascii;
char *start, *end;
assert(s != NULL && startp != NULL && endp != NULL);
TRACE(("idn_selectiveencode_findregion(s=\"%s\")\n",
idn__debug_xstring(s, 20)));
/*
* Scan the specified string looking for non-ascii character.
*/
if ((non_ascii = find_nonascii(s)) == NULL)
return (idn_notfound);
/*
* Non-ascii character found.
* Determine the region to encode.
*/
/*
* First, we scan backwards to find the beginning of the region
* that should be converted.
*/
start = non_ascii;
while (start > s) {
char *prev = idn_utf8_findfirstbyte(start - 1, s);
if (is_domain_delimiter(*prev))
break; /* Found */
start = prev;
}
*startp = start;
/*
* Next we scan forwards looking for the end of the region.
*/
end = non_ascii + idn_utf8_mblen(non_ascii);
while (!is_domain_delimiter(*end))
end += idn_utf8_mblen(end);
*endp = end;
return (idn_success);
}
/*
* Set the local encoding to 'name'.
*/
void
idn__localencoding_setname(idn__localencoding_t ctx, const char *name) {
idn_result_t r = idn_success;
assert(ctx != NULL);
TRACE(("idn__localencoding_setname(name=\"%s\")\n",
idn__debug_xstring(name)));
if (name == NULL)
ctx->is_static = 0;
else {
idn__util_strcpy(ctx->name, sizeof(ctx->name),
idn__encodingalias_find(aliases, name));
ctx->is_static = 1;
}
TRACE(("idn__localencoding_setname(): %s\n", idn_result_tostring(r)));
}
const char *
idn__localencoding_getname(idn__localencoding_t ctx) {
const char *key = NULL;
#ifdef WIN32
char cp_str[40]; /* enough */
#endif
assert(ctx != NULL);
TRACE(("idn__localencoding_getname()\n"));
if (!ctx->is_static) {
key = getenv(IDN_LOCALCS_ENV);
#ifdef WIN32
if (key == NULL) {
sprintf(cp_str, "CP%u", GetACP());
key = cp_str;
}
#endif
#if defined(HAVE_NL_LANGINFO) && defined(CODESET)
if (key == NULL)
key = nl_langinfo(CODESET);
#endif
#ifdef HAVE_SETLOCALE
if (key == NULL)
key = setlocale(LC_CTYPE, NULL);
#endif
if (key == NULL)
key = getenv("LC_ALL");
if (key == NULL)
key = getenv("LC_CTYPE");
if (key == NULL)
key = getenv("LANG");
if (key == NULL)
key = getenv("LANG");
if (key == NULL)
key = IDN__UTF8_ENCODINGNAME;
idn__util_strcpy(ctx->name, sizeof(ctx->name),
idn__encodingalias_find(aliases, key));
}
TRACE(("idn__localencoding_getname(): success "
"(name=\"%s\")\n", idn__debug_xstring(ctx->name)));
return (ctx->name);
}
/*
* Perform UTF32 to UTF-8 to conversion.
*/
idn_result_t
idn__res_utf32toutf8(idn_resconf_t ctx, const unsigned long *from, char **to) {
idn_result_t r = idn_success;
size_t tolen = 0;
assert(ctx != NULL && from != NULL && to != NULL);
TRACE(("idn__res_utf32toutf8(label=\"%s\")\n",
idn__debug_utf32xstring(from)));
*to = NULL;
tolen = 256; /* may be large enough. */
for (;;) {
void *new_to;
new_to = realloc(*to, sizeof(**to) * tolen);
if (new_to == NULL) {
r = idn_nomemory;
goto ret;
}
*to = (char *)new_to;
r = idn__utf32_toutf8(from, *to, tolen);
if (r == idn_success)
break;
else if (r != idn_buffer_overflow)
goto ret;
tolen *= 2;
}
ret:
if (r == idn_success) {
TRACE(("idn__res_utf32toutf8(): success (to=\"%s\")\n",
idn__debug_xstring(*to)));
} else {
TRACE(("idn__res_utf32toutf8(): %s\n",
idn_result_tostring(r)));
free(*to);
*to = NULL;
}
return (r);
}
/*
* Read an alias definition file and add aliases in it to 'ctx'.
*/
idn_result_t
idn__langalias_addfromfile(idn__langalias_t ctx, const char *file) {
idn_result_t r = idn_success;
FILE *fp = NULL;
int line_no;
char line[200], alias[200], real[200];
assert(ctx != NULL && file != NULL);
TRACE(("idn__langalias_addfromfile(file=\"%s\")\n",
idn__debug_xstring(file)));
fp = fopen(file, "r");
if (fp == NULL) {
r = idn_nofile;
goto ret;
}
for (line_no = 1; fgets(line, sizeof(line), fp) != NULL; line_no++) {
unsigned char *p = (unsigned char *)line;
while (isascii(*p) && isspace(*p))
p++;
if (*p == '#' || *p == '\n' || *p == '\0')
continue;
if (sscanf((char *)p, "%s %s", alias, real) == 2) {
r = idn__langalias_add(ctx, alias, real);
if (r != idn_success)
break;
} else {
INFO(("idn__langalias_addfromfile: syntax error, "
"line %d\n",line_no));
r = idn_invalid_syntax;
break;
}
}
ret:
TRACE(("idn__langalias_addfromfile(): %s\n",
idn_result_tostring(r)));
return (r);
}
idn_result_t
idn_converter_convfromucs4(idn_converter_t ctx, const unsigned long *from,
char *to, size_t tolen) {
idn_result_t r;
assert(ctx != NULL && from != NULL && to != NULL);
TRACE(("idn_converter_convfromucs4(ctx=%s, from=\"%s\", tolen=%d)\n",
ctx->local_encoding_name, idn__debug_ucs4xstring(from, 50),
(int)tolen));
if (!ctx->opened_convfromucs4) {
r = (*ctx->ops->openfromucs4)(ctx, &(ctx->private_data));
if (r != idn_success)
goto ret;
ctx->opened_convfromucs4 = 1;
}
r = (*ctx->ops->convfromucs4)(ctx, ctx->private_data, from, to, tolen);
if (r != idn_success)
goto ret;
if ((ctx->flags & IDN_CONVERTER_RTCHECK) != 0) {
r = roundtrip_check(ctx, from, to);
if (r != idn_success)
goto ret;
}
r = idn_success;
ret:
if (r == idn_success) {
TRACE(("idn_converter_convfromucs4(): success (to=\"%s\")\n",
idn__debug_xstring(to, 50)));
} else {
TRACE(("idn_converter_convfromucs4(): %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);
}
idn_result_t
idn__punycode_decode(idn_converter_t ctx, void *privdata,
const char *from, unsigned long *to, size_t tolen) {
unsigned long *to_org = to;
unsigned long c, idx;
size_t prefixlen = strlen(IDN_PUNYCODE_PREFIX);
size_t fromlen;
size_t uidx, fidx, ucslen;
int first, bias;
idn_result_t r;
assert(ctx != NULL);
TRACE(("idn__punycode_decode(from=\"%s\", tolen=%d)\n",
idn__debug_xstring(from, 50), (int)tolen));
if (!idn__util_asciihaveaceprefix(from, IDN_PUNYCODE_PREFIX)) {
if (*from == '\0') {
r = idn_ucs4_utf8toucs4(from, to, tolen);
goto ret;
}
r = idn_invalid_encoding;
goto ret;
}
from += prefixlen;
fromlen = strlen(from);
/*
* Find the last delimiter, and copy the characters
* before it verbatim.
*/
ucslen = 0;
for (fidx = fromlen; fidx > 0; fidx--) {
if (from[fidx - 1] == '-') {
if (tolen < fidx) {
r = idn_buffer_overflow;
goto ret;
}
for (uidx = 0; uidx < fidx - 1; uidx++) {
to[uidx] = from[uidx];
}
ucslen = uidx;
break;
}
}
first = 1;
bias = PUNYCODE_INITIAL_BIAS;
c = PUNYCODE_INITIAL_N;
idx = 0;
while (fidx < fromlen) {
int len;
unsigned long delta;
int i;
len = punycode_getwc(from + fidx, fromlen - fidx, bias, &delta);
if (len == 0) {
r = idn_invalid_encoding;
goto ret;
}
fidx += len;
bias = punycode_update_bias(delta, ucslen + 1, first);
first = 0;
idx += delta;
c += idx / (ucslen + 1);
uidx = idx % (ucslen + 1);
/* Insert 'c' at uidx. */
if (tolen-- <= 0) {
r = idn_buffer_overflow;
goto ret;
}
for (i = ucslen; i > uidx; i--)
to[i] = to[i - 1];
to[uidx] = c;
ucslen++;
idx = uidx + 1;
}
/* Terminate with NUL. */
if (tolen <= 0) {
r = idn_buffer_overflow;
goto ret;
}
to[ucslen] = '\0';
r = idn_success;
ret:
if (r == idn_success) {
TRACE(("idn__punycode_decode(): succcess (to=\"%s\")\n",
idn__debug_ucs4xstring(to_org, 50)));
} else {
TRACE(("idn__punycode_decode(): %s\n", idn_result_tostring(r)));
}
return (r);
}
idn_result_t
idn_checker_add(idn_checker_t ctx, const char *scheme_name) {
idn_result_t r;
check_scheme_t *scheme;
const char *scheme_prefix;
const char *scheme_parameter;
void *scheme_context = NULL;
char *buffer = NULL;
assert(scheme_hash != NULL);
assert(ctx != NULL);
TRACE(("idn_checker_add(scheme_name=%s)\n",
idn__debug_xstring(scheme_name, 50)));
/*
* Split `scheme_name' into `scheme_prefix' and `scheme_parameter'.
*/
scheme_parameter = strchr(scheme_name, ':');
if (scheme_parameter == NULL) {
scheme_prefix = scheme_name;
scheme_parameter = NULL;
} else {
ptrdiff_t scheme_prefixlen;
scheme_prefixlen = scheme_parameter - scheme_name;
buffer = (char *) malloc(scheme_prefixlen + 1);
if (buffer == NULL) {
r = idn_nomemory;
goto ret;
}
memcpy(buffer, scheme_name, scheme_prefixlen);
*(buffer + scheme_prefixlen) = '\0';
scheme_prefix = buffer;
scheme_parameter++;
}
/*
* Find a scheme.
*/
if (idn__strhash_get(scheme_hash, scheme_prefix, (void **)&scheme)
!= idn_success) {
ERROR(("idn_checker_add(): invalid scheme \"%-.30s\"\n",
scheme_name));
r = idn_invalid_name;
goto ret;
}
if (scheme_parameter == NULL && scheme->parameter != NULL)
scheme_parameter = scheme->parameter;
/*
* Add the scheme.
*/
assert(ctx->nschemes <= ctx->scheme_size);
if (ctx->nschemes == ctx->scheme_size) {
check_scheme_t *new_schemes;
new_schemes = (check_scheme_t *) realloc(ctx->schemes,
sizeof(check_scheme_t) * ctx->scheme_size * 2);
if (new_schemes == NULL) {
r = idn_nomemory;
goto ret;
}
ctx->schemes = new_schemes;
ctx->scheme_size *= 2;
}
r = scheme->create(scheme_parameter, &scheme_context);
if (r != idn_success)
goto ret;
memcpy(ctx->schemes + ctx->nschemes, scheme, sizeof(check_scheme_t));
ctx->schemes[ctx->nschemes].context = scheme_context;
ctx->nschemes++;
r = idn_success;
ret:
free(buffer);
if (r != idn_success)
free(scheme_context);
TRACE(("idn_checker_add(): %s\n", idn_result_tostring(r)));
return (r);
}
idn_result_t
idn__punycode_encode(idn_converter_t ctx, void *privdata,
const unsigned long *from, char *to, size_t tolen) {
char *to_org = to;
unsigned long cur_code, next_code, delta;
size_t prefixlen = strlen(IDN_PUNYCODE_PREFIX);
size_t fromlen;
size_t ucsdone;
size_t toidx;
int uidx, bias, first;
idn_result_t r;
assert(ctx != NULL);
TRACE(("idn__punycode_encode(from=\"%s\", tolen=%d)\n",
idn__debug_ucs4xstring(from, 50), (int)tolen));
if (*from == '\0') {
r = idn_ucs4_ucs4toutf8(from, to, tolen);
goto ret;
} else if (idn__util_ucs4haveaceprefix(from, IDN_PUNYCODE_PREFIX)) {
r = idn_prohibited;
goto ret;
}
if (tolen < prefixlen) {
r = idn_buffer_overflow;
goto ret;
}
memcpy(to, IDN_PUNYCODE_PREFIX, prefixlen);
to += prefixlen;
tolen -= prefixlen;
fromlen = idn_ucs4_strlen(from);
/*
* If the input string is too long (actually too long to be sane),
* return failure in order to prevent possible overflow.
*/
if (fromlen > PUNYCODE_MAXINPUT) {
ERROR(("idn__punycode_encode(): "
"the input string is too long to convert Punycode\n",
idn__debug_ucs4xstring(from, 50)));
r = idn_failure;
goto ret;
}
ucsdone = 0; /* number of characters processed */
toidx = 0;
/*
* First, pick up basic code points and copy them to 'to'.
*/
for (uidx = 0; uidx < fromlen; uidx++) {
if (from[uidx] < 0x80) {
if (toidx >= tolen) {
r = idn_buffer_overflow;
goto ret;
}
to[toidx++] = from[uidx];
ucsdone++;
}
}
/*
* If there are any basic code points, output a delimiter
* (hyphen-minus).
*/
if (toidx > 0) {
if (toidx >= tolen) {
r = idn_buffer_overflow;
goto ret;
}
to[toidx++] = '-';
to += toidx;
tolen -= toidx;
}
/*
* Then encode non-basic characters.
*/
first = 1;
cur_code = PUNYCODE_INITIAL_N;
bias = PUNYCODE_INITIAL_BIAS;
delta = 0;
while (ucsdone < fromlen) {
int limit = -1, rest;
/*
* Find the smallest code point equal to or greater
* than 'cur_code'. Also remember the index of the
* last occurence of the code point.
*/
for (next_code = MAX_UCS, uidx = fromlen - 1;
uidx >= 0; uidx--) {
if (from[uidx] >= cur_code && from[uidx] < next_code) {
next_code = from[uidx];
limit = uidx;
}
}
/* There must be such code point. */
assert(limit >= 0);
delta += (next_code - cur_code) * (ucsdone + 1);
cur_code = next_code;
/*
* Scan the input string again, and encode characters
* whose code point is 'cur_code'. Use 'limit' to avoid
* unnecessary scan.
*/
for (uidx = 0, rest = ucsdone; uidx <= limit; uidx++) {
if (from[uidx] < cur_code) {
delta++;
rest--;
} else if (from[uidx] == cur_code) {
int sz = punycode_putwc(to, tolen, delta, bias);
if (sz == 0) {
r = idn_buffer_overflow;
goto ret;
}
to += sz;
tolen -= sz;
ucsdone++;
bias = punycode_update_bias(delta, ucsdone,
first);
delta = 0;
first = 0;
}
}
delta += rest + 1;
cur_code++;
}
/*
* Terminate with NUL.
*/
if (tolen <= 0) {
r = idn_buffer_overflow;
goto ret;
}
*to = '\0';
r = idn_success;
ret:
if (r == idn_success) {
TRACE(("idn__punycode_encode(): succcess (to=\"%s\")\n",
idn__debug_xstring(to_org, 50)));
} else {
TRACE(("idn__punycode_encode(): %s\n", idn_result_tostring(r)));
}
return (r);
}
idn_result_t
idn_res_decodename(idn_resconf_t ctx, idn_action_t actions, const char *from,
char *to, size_t tolen) {
idn_converter_t local_converter = NULL;
idn_converter_t idn_converter = NULL;
idn_delimitermap_t delimiter_mapper;
idn_result_t r;
labellist_t labels = NULL, l;
unsigned long *buffer = NULL;
unsigned long *saved_name = NULL;
size_t buffer_length;
int idn_is_ace;
assert(ctx != NULL && from != NULL && to != NULL);
TRACE(("idn_res_decodename(actions=%s, from=\"%s\", tolen=%d)\n",
idn__res_actionstostring(actions),
idn__debug_xstring(from, 50), (int)tolen));
if (actions & ~DECODE_MASK) {
WARNING(("idn_res_decodename: invalid actions 0x%x\n",
actions));
r = idn_invalid_action;
goto ret;
}
if (!initialized)
idn_res_initialize();
if (!enabled || actions == 0) {
r = copy_verbatim(from, to, tolen);
goto ret;
} else if (tolen <= 0) {
r = idn_buffer_overflow;
goto ret;
}
if (actions & IDN_DECODE_QUERY) {
#ifndef WITHOUT_ICONV
actions |= (IDN_DELIMMAP | IDN_MAP | IDN_NORMALIZE | \
IDN_PROHCHECK | IDN_BIDICHECK | IDN_IDNCONV | \
IDN_RTCHECK | IDN_LOCALCONV);
#else
actions |= (IDN_DELIMMAP | IDN_MAP | IDN_NORMALIZE | \
IDN_PROHCHECK | IDN_BIDICHECK | IDN_IDNCONV | \
IDN_RTCHECK);
#endif
}
/*
* Convert `from' to UCS4.
*/
local_converter = idn_resconf_getlocalconverter(ctx);
#ifndef WITHOUT_ICONV
if (local_converter == NULL) {
r = idn_invalid_name;
goto ret;
}
#endif
idn_converter = idn_resconf_getidnconverter(ctx);
if (idn_converter != NULL &&
idn_converter_isasciicompatible(idn_converter))
idn_is_ace = 1;
else
idn_is_ace = 0;
buffer_length = tolen * 2;
TRACE(("res idndecode(name=\"%s\")\n", idn__debug_xstring(from, 50)));
for (;;) {
void *new_buffer;
new_buffer = realloc(buffer, sizeof(*buffer) * buffer_length);
if (new_buffer == NULL) {
r = idn_nomemory;
goto ret;
}
buffer = (unsigned long *)new_buffer;
if ((actions & IDN_IDNCONV) &&
idn_converter != NULL && !idn_is_ace) {
r = idn_converter_convtoucs4(idn_converter, from,
buffer, buffer_length);
} else {
r = idn_ucs4_utf8toucs4(from, buffer, buffer_length);
}
if (r == idn_success)
break;
else if (r != idn_buffer_overflow)
goto ret;
buffer_length *= 2;
}
if (*buffer == '\0') {
if (tolen <= 0) {
r = idn_buffer_overflow;
goto ret;
}
*to = '\0';
r = idn_success;
goto ret;
}
/*
* Delimiter map.
*/
if (actions & IDN_DELIMMAP) {
TRACE(("res delimitermap(name=\"%s\")\n",
idn__debug_ucs4xstring(buffer, 50)));
delimiter_mapper = idn_resconf_getdelimitermap(ctx);
if (delimiter_mapper != NULL) {
r = idn_delimitermap_map(delimiter_mapper, buffer,
buffer, buffer_length);
idn_delimitermap_destroy(delimiter_mapper);
if (r != idn_success)
goto ret;
}
TRACE(("res delimitermap(): success (name=\"%s\")\n",
idn__debug_ucs4xstring(buffer, 50)));
}
/*
* Split the name into a list of labels.
*/
r = labellist_create(buffer, &labels);
if (r != idn_success)
goto ret;
/*
* Perform conversions and tests.
*/
for (l = labellist_tail(labels); l != NULL;
l = labellist_previous(l)) {
free(saved_name);
saved_name = NULL;
if (!idn__util_ucs4isasciirange(labellist_getname(l))) {
if (actions & IDN_MAP) {
r = label_map(ctx, l);
if (r != idn_success)
goto ret;
}
if (actions & IDN_NORMALIZE) {
r = label_normalize(ctx, l);
if (r != idn_success)
goto ret;
}
if (actions & IDN_PROHCHECK) {
r = label_prohcheck(ctx, l);
if (r == idn_prohibited) {
labellist_undo(l);
continue;
} else if (r != idn_success) {
goto ret;
}
}
if (actions & IDN_UNASCHECK) {
r = label_unascheck(ctx, l);
if (r == idn_prohibited) {
labellist_undo(l);
continue;
} else if (r != idn_success) {
goto ret;
}
}
if (actions & IDN_BIDICHECK) {
r = label_bidicheck(ctx, l);
if (r == idn_prohibited) {
labellist_undo(l);
continue;
} else if (r != idn_success) {
goto ret;
}
}
}
if ((actions & IDN_IDNCONV) && idn_is_ace) {
saved_name = idn_ucs4_strdup(labellist_getname(l));
if (saved_name == NULL) {
r = idn_nomemory;
goto ret;
}
r = label_idndecode(ctx, l);
if (r == idn_invalid_encoding) {
labellist_undo(l);
continue;
} else if (r != idn_success) {
goto ret;
}
}
if ((actions & IDN_RTCHECK) && saved_name != NULL) {
r = label_rtcheck(ctx, actions, l, saved_name);
if (r == idn_invalid_encoding) {
labellist_undo(l);
continue;
} else if (r != idn_success) {
goto ret;
}
}
#ifndef WITHOUT_ICONV
if (actions & IDN_LOCALCONV) {
r = label_localdecodecheck(ctx, l);
if (r != idn_success)
goto ret;
}
#endif
}
/*
* Concat a list of labels to a name.
*/
for (;;) {
void *new_buffer;
new_buffer = realloc(buffer, sizeof(*buffer) * buffer_length);
if (new_buffer == NULL) {
r = idn_nomemory;
goto ret;
}
buffer = (unsigned long *)new_buffer;
r = labellist_getnamelist(labels, buffer, buffer_length);
if (r == idn_success)
break;
else if (r != idn_buffer_overflow)
goto ret;
buffer_length *= 2;
}
if (actions & IDN_LOCALCONV) {
r = idn_converter_convfromucs4(local_converter, buffer, to,
tolen);
} else {
r = idn_ucs4_ucs4toutf8(buffer, to, tolen);
}
ret:
if (r == idn_success) {
TRACE(("idn_res_decodename(): success (to=\"%s\")\n",
idn__debug_xstring(to, 50)));
} else {
TRACE(("idn_res_decodename(): %s\n", idn_result_tostring(r)));
}
free(saved_name);
free(buffer);
if (local_converter != NULL)
idn_converter_destroy(local_converter);
if (idn_converter != NULL)
idn_converter_destroy(idn_converter);
if (labels != NULL)
labellist_destroy(labels);
return (r);
}
/*
* Read a mapping definition file and add mappings in it to 'ctx'.
*/
idn_result_t
idn__tldlocalmap_addfromfile(idn__tldlocalmap_t ctx, const char *tld,
const char *file) {
static const unsigned long default_map_tld[] = {
'<', 'd', 'e', 'f', 'a', 'u', 'l', 't', '>', '\0'
};
idn_result_t r = idn_success;
unsigned long tld_utf32[IDN__TLDLOCALMAP_MAXTLDLENGTH + 1];
idn__tldlocalmap_entry_t new_entry = NULL;
idn__foreignmap_t new_mapctx = NULL;
assert(ctx != NULL && file != NULL);
TRACE(("idn__tldlocalmap_addfromfile(tld=\"%s\", file=\"%s\")\n",
idn__debug_xstring(tld), idn__debug_xstring(file)));
if (tld != NULL && *tld == '.')
tld++;
if (tld != NULL && *tld == '\0') {
r = idn_invalid_syntax;
goto ret;
}
if (tld != NULL && strcmp(tld, "*") == 0)
tld = NULL;
new_entry = (idn__tldlocalmap_entry_t)malloc(sizeof(*new_entry));
if (new_entry == NULL) {
r = idn_nomemory;
goto ret;
}
new_entry->tld = NULL;
new_entry->map = NULL;
new_entry->next = NULL;
r = idn__foreignmap_create(&new_mapctx);
if (r != idn_success)
goto ret;
r = idn__foreignmap_addfromfile(new_mapctx, file);
if (r != idn_success)
goto ret;
new_entry->map = new_mapctx;
if (tld != NULL) {
r = idn__utf32_fromutf8(tld, tld_utf32,
IDN__TLDLOCALMAP_MAXTLDLENGTH + 1);
if (r != idn_success) {
r = idn_invalid_name;
goto ret;
}
idn__utf32_asclower(tld_utf32);
new_entry->tld = idn__utf32_strdup(tld_utf32);
if (new_entry->tld == NULL) {
r = idn_nomemory;
goto ret;
}
r = idn__strhash32_put(ctx->hash, tld_utf32, new_entry);
if (r != idn_success)
goto ret;
} else {
new_entry->tld = idn__utf32_strdup(default_map_tld);
if (new_entry->tld == NULL) {
r = idn_nomemory;
goto ret;
}
ctx->default_map = new_entry;
}
if (ctx->entries_head == NULL) {
ctx->entries_head = new_entry;
ctx->entries_tail = new_entry;
} else {
ctx->entries_tail->next = new_entry;
ctx->entries_tail = new_entry;
}
ret:
if (r == idn_success) {
TRACE(("idn__tldlocalmap_addfromfile(): success "
"(tld=\"%s\")\n", idn__debug_xstring(tld)));
} else {
TRACE(("idn__tldlocalmap_addfromfile(): %s\n",
idn_result_tostring(r)));
if (new_mapctx != NULL)
idn__foreignmap_destroy(new_mapctx);
if (new_entry != NULL) {
free(new_entry->tld);
free(new_entry);
}
}
return (r);
}
idn_result_t
idn_res_decodename2(idn_resconf_t ctx, idn_action_t actions, const char *from,
char *to, size_t tolen, const char *auxencoding) {
#ifdef WITHOUT_ICONV
return idn_failure;
#else /* WITHOUT_ICONV */
idn_result_t r;
idn_converter_t aux_converter = NULL;
unsigned long *buffer_ucs4 = NULL;
char *buffer_utf8 = NULL;
size_t buffer_length;
assert(ctx != NULL && from != NULL && to != NULL);
TRACE(("idn_res_decodename2(actions=%s, from=\"%s\", tolen=%d, "
"auxencoding=\"%s\")\n",
idn__res_actionstostring(actions),
idn__debug_xstring(from, 50), (int)tolen,
(auxencoding != NULL) ? auxencoding : "<null>"));
if (!initialized)
idn_res_initialize();
if (!enabled || actions == 0) {
r = copy_verbatim(from, to, tolen);
goto ret;
} else if (tolen <= 0) {
r = idn_buffer_overflow;
goto ret;
}
if (auxencoding == NULL ||
strcmp(auxencoding, IDN_UTF8_ENCODING_NAME) == 0 ||
strcmp(auxencoding, "UTF-8") == 0) {
return idn_res_decodename(ctx, actions, from, to, tolen);
}
/*
* Convert `from' to UCS4.
*/
r = idn_resconf_setauxidnconvertername(ctx, auxencoding,
IDN_CONVERTER_DELAYEDOPEN);
if (r != idn_success) {
goto ret;
}
aux_converter = idn_resconf_getauxidnconverter(ctx);
if (aux_converter == NULL) {
r = idn_failure;
goto ret;
}
buffer_length = tolen * 2;
for (;;) {
void *new_buffer;
new_buffer = realloc(buffer_ucs4,
sizeof(*buffer_ucs4) * buffer_length);
if (new_buffer == NULL) {
r = idn_nomemory;
goto ret;
}
buffer_ucs4 = (unsigned long *)new_buffer;
r = idn_converter_convtoucs4(aux_converter, from,
buffer_ucs4,
buffer_length);
if (r == idn_success)
break;
else if (r != idn_buffer_overflow)
goto ret;
buffer_length *= 2;
}
if (*buffer_ucs4 == '\0') {
if (tolen <= 0) {
r = idn_buffer_overflow;
goto ret;
}
*to = '\0';
r = idn_success;
goto ret;
}
/*
* Convert `buffer_ucs4' to UTF-8.
*/
buffer_length = tolen * 2;
for (;;) {
void *new_buffer;
new_buffer = realloc(buffer_utf8,
sizeof(*buffer_utf8) * buffer_length);
if (new_buffer == NULL) {
r = idn_nomemory;
goto ret;
}
buffer_utf8 = (char *)new_buffer;
r = idn_ucs4_ucs4toutf8(buffer_ucs4, buffer_utf8,
buffer_length);
if (r == idn_success)
break;
else if (r != idn_buffer_overflow)
goto ret;
buffer_length *= 2;
}
if (*buffer_utf8 == '\0') {
if (tolen <= 0) {
r = idn_buffer_overflow;
goto ret;
}
*to = '\0';
r = idn_success;
goto ret;
}
r = idn_res_decodename(ctx, actions, buffer_utf8, to, tolen);
ret:
if (r == idn_success) {
TRACE(("idn_res_decodename2(): success (to=\"%s\")\n",
idn__debug_xstring(to, 50)));
} else {
TRACE(("idn_res_decodename2(): %s\n", idn_result_tostring(r)));
}
free(buffer_ucs4);
free(buffer_utf8);
if (aux_converter != NULL)
idn_converter_destroy(aux_converter);
return (r);
#endif /* WITHOUT_ICONV */
}
idn_result_t
idn__race_decode(idn_converter_t ctx, void *privdata,
const char *from, unsigned long *to, size_t tolen) {
unsigned short *buf = NULL;
size_t prefixlen = strlen(IDN_RACE_PREFIX);
size_t fromlen;
size_t buflen;
idn_result_t r;
assert(ctx != NULL);
TRACE(("idn__race_decode(from=\"%s\", tolen=%d)\n",
idn__debug_xstring(from, 50), (int)tolen));
if (!idn__util_asciihaveaceprefix(from, IDN_RACE_PREFIX)) {
if (*from == '\0') {
r = idn_ucs4_utf8toucs4(from, to, tolen);
goto ret;
}
r = idn_invalid_encoding;
goto ret;
}
from += prefixlen;
fromlen = strlen(from);
/*
* Allocate sufficient buffer.
*/
buflen = fromlen + 1;
buf = malloc(sizeof(*buf) * buflen);
if (buf == NULL) {
r = idn_nomemory;
goto ret;
}
/*
* Decode base32 and decompress.
*/
r = race_decode_decompress(from, buf, buflen);
if (r != idn_success)
goto ret;
/*
* Now 'buf' points the decompressed string, which must contain
* UTF-16 characters.
*/
/*
* Convert to UCS4.
*/
r = idn_ucs4_utf16toucs4(buf, to, tolen);
if (r != idn_success)
goto ret;
ret:
free(buf);
if (r == idn_success) {
TRACE(("idn__race_decode(): succcess (to=\"%s\")\n",
idn__debug_ucs4xstring(to, 50)));
} else {
TRACE(("idn__race_decode(): %s\n", idn_result_tostring(r)));
}
return (r);
}