
AIX qsort() is unstable and might change the order of the elements if they are equal, libwind require them to be in the same order as they were on the input, pull in FreeBSD qsort and use that instead of AIX version.
307 lines
7.0 KiB
C
307 lines
7.0 KiB
C
/*
|
|
* Copyright (c) 2004 Kungliga Tekniska Högskolan
|
|
* (Royal Institute of Technology, Stockholm, Sweden).
|
|
* All rights reserved.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
*
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
*
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
*
|
|
* 3. Neither the name of the Institute nor the names of its contributors
|
|
* may be used to endorse or promote products derived from this software
|
|
* without specific prior written permission.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE INSTITUTE AND CONTRIBUTORS ``AS IS'' AND
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE INSTITUTE OR CONTRIBUTORS BE LIABLE
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
* SUCH DAMAGE.
|
|
*/
|
|
|
|
#ifdef HAVE_CONFIG_H
|
|
#include <config.h>
|
|
#endif
|
|
#include "windlocl.h"
|
|
|
|
#include <assert.h>
|
|
#include <stdlib.h>
|
|
#include <errno.h>
|
|
#include <stdio.h>
|
|
|
|
#include "roken.h"
|
|
|
|
#include "normalize_table.h"
|
|
|
|
static int
|
|
translation_cmp(const void *key, const void *data)
|
|
{
|
|
const struct translation *t1 = (const struct translation *)key;
|
|
const struct translation *t2 = (const struct translation *)data;
|
|
|
|
return t1->key - t2->key;
|
|
}
|
|
|
|
enum { s_base = 0xAC00};
|
|
enum { s_count = 11172};
|
|
enum { l_base = 0x1100};
|
|
enum { l_count = 19};
|
|
enum { v_base = 0x1161};
|
|
enum { v_count = 21};
|
|
enum { t_base = 0x11A7};
|
|
enum { t_count = 28};
|
|
enum { n_count = v_count * t_count};
|
|
|
|
static int
|
|
hangul_decomp(const uint32_t *in, size_t in_len,
|
|
uint32_t *out, size_t *out_len)
|
|
{
|
|
uint32_t u = *in;
|
|
unsigned s_index;
|
|
unsigned l, v, t;
|
|
unsigned o;
|
|
|
|
if (u < s_base || u >= s_base + s_count)
|
|
return 0;
|
|
s_index = u - s_base;
|
|
l = l_base + s_index / n_count;
|
|
v = v_base + (s_index % n_count) / t_count;
|
|
t = t_base + s_index % t_count;
|
|
o = 2;
|
|
if (t != t_base)
|
|
++o;
|
|
if (*out_len < o)
|
|
return WIND_ERR_OVERRUN;
|
|
out[0] = l;
|
|
out[1] = v;
|
|
if (t != t_base)
|
|
out[2] = t;
|
|
*out_len = o;
|
|
return 1;
|
|
}
|
|
|
|
static uint32_t
|
|
hangul_composition(const uint32_t *in, size_t in_len)
|
|
{
|
|
if (in_len < 2)
|
|
return 0;
|
|
if (in[0] >= l_base && in[0] < l_base + l_count) {
|
|
unsigned l_index = in[0] - l_base;
|
|
unsigned v_index;
|
|
|
|
if (in[1] < v_base || in[1] >= v_base + v_count)
|
|
return 0;
|
|
v_index = in[1] - v_base;
|
|
return (l_index * v_count + v_index) * t_count + s_base;
|
|
} else if (in[0] >= s_base && in[0] < s_base + s_count) {
|
|
unsigned s_index = in[0] - s_base;
|
|
unsigned t_index;
|
|
|
|
if (s_index % t_count != 0)
|
|
return 0;
|
|
if (in[1] < t_base || in[1] >= t_base + t_count)
|
|
return 0;
|
|
t_index = in[1] - t_base;
|
|
return in[0] + t_index;
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static int
|
|
compat_decomp(const uint32_t *in, size_t in_len,
|
|
uint32_t *out, size_t *out_len)
|
|
{
|
|
unsigned i;
|
|
unsigned o = 0;
|
|
|
|
for (i = 0; i < in_len; ++i) {
|
|
struct translation ts = {in[i]};
|
|
size_t sub_len = *out_len - o;
|
|
int ret;
|
|
|
|
ret = hangul_decomp(in + i, in_len - i,
|
|
out + o, &sub_len);
|
|
if (ret) {
|
|
if (ret == WIND_ERR_OVERRUN)
|
|
return ret;
|
|
o += sub_len;
|
|
} else {
|
|
void *s = bsearch(&ts,
|
|
_wind_normalize_table,
|
|
_wind_normalize_table_size,
|
|
sizeof(_wind_normalize_table[0]),
|
|
translation_cmp);
|
|
if (s != NULL) {
|
|
const struct translation *t = (const struct translation *)s;
|
|
|
|
ret = compat_decomp(_wind_normalize_val_table + t->val_offset,
|
|
t->val_len,
|
|
out + o, &sub_len);
|
|
if (ret)
|
|
return ret;
|
|
o += sub_len;
|
|
} else {
|
|
if (o >= *out_len)
|
|
return WIND_ERR_OVERRUN;
|
|
out[o++] = in[i];
|
|
|
|
}
|
|
}
|
|
}
|
|
*out_len = o;
|
|
return 0;
|
|
}
|
|
|
|
static int
|
|
cc_cmp(const void *a, const void *b)
|
|
{
|
|
const uint32_t *ua = (const uint32_t *)a;
|
|
const uint32_t *ub = (const uint32_t *)b;
|
|
|
|
return _wind_combining_class(*ua) - _wind_combining_class(*ub);
|
|
}
|
|
|
|
static void
|
|
canonical_reorder(uint32_t *tmp, size_t tmp_len)
|
|
{
|
|
size_t i;
|
|
|
|
for (i = 0; i < tmp_len; ++i) {
|
|
int cc = _wind_combining_class(tmp[i]);
|
|
if (cc) {
|
|
size_t j;
|
|
for (j = i + 1;
|
|
j < tmp_len && _wind_combining_class(tmp[j]);
|
|
++j)
|
|
;
|
|
qsort(&tmp[i], j - i, sizeof(tmp[0]), cc_cmp);
|
|
i = j;
|
|
}
|
|
}
|
|
}
|
|
|
|
static uint32_t
|
|
find_composition(const uint32_t *in, unsigned in_len)
|
|
{
|
|
unsigned short canon_index = 0;
|
|
uint32_t cur;
|
|
unsigned n = 0;
|
|
|
|
cur = hangul_composition(in, in_len);
|
|
if (cur)
|
|
return cur;
|
|
|
|
do {
|
|
const struct canon_node *c = &_wind_canon_table[canon_index];
|
|
unsigned i;
|
|
|
|
if (n % 5 == 0) {
|
|
cur = *in++;
|
|
if (in_len-- == 0)
|
|
return c->val;
|
|
}
|
|
|
|
i = cur >> 16;
|
|
if (i < c->next_start || i >= c->next_end)
|
|
canon_index = 0;
|
|
else
|
|
canon_index =
|
|
_wind_canon_next_table[c->next_offset + i - c->next_start];
|
|
if (canon_index != 0) {
|
|
cur = (cur << 4) & 0xFFFFF;
|
|
++n;
|
|
}
|
|
} while (canon_index != 0);
|
|
return 0;
|
|
}
|
|
|
|
static int
|
|
combine(const uint32_t *in, size_t in_len,
|
|
uint32_t *out, size_t *out_len)
|
|
{
|
|
unsigned i;
|
|
int ostarter;
|
|
unsigned o = 0;
|
|
int old_cc;
|
|
|
|
for (i = 0; i < in_len;) {
|
|
while (i < in_len && _wind_combining_class(in[i]) != 0) {
|
|
out[o++] = in[i++];
|
|
}
|
|
if (i < in_len) {
|
|
if (o >= *out_len)
|
|
return WIND_ERR_OVERRUN;
|
|
ostarter = o;
|
|
out[o++] = in[i++];
|
|
old_cc = -1;
|
|
|
|
while (i < in_len) {
|
|
uint32_t comb;
|
|
uint32_t v[2];
|
|
int cc;
|
|
|
|
v[0] = out[ostarter];
|
|
v[1] = in[i];
|
|
|
|
cc = _wind_combining_class(in[i]);
|
|
if (old_cc != cc && (comb = find_composition(v, 2))) {
|
|
out[ostarter] = comb;
|
|
} else if (cc == 0) {
|
|
break;
|
|
} else {
|
|
if (o >= *out_len)
|
|
return WIND_ERR_OVERRUN;
|
|
out[o++] = in[i];
|
|
old_cc = cc;
|
|
}
|
|
++i;
|
|
}
|
|
}
|
|
}
|
|
*out_len = o;
|
|
return 0;
|
|
}
|
|
|
|
int
|
|
_wind_stringprep_normalize(const uint32_t *in, size_t in_len,
|
|
uint32_t *out, size_t *out_len)
|
|
{
|
|
size_t tmp_len;
|
|
uint32_t *tmp;
|
|
int ret;
|
|
|
|
if (in_len == 0) {
|
|
*out_len = 0;
|
|
return 0;
|
|
}
|
|
|
|
tmp_len = in_len * 4;
|
|
if (tmp_len < MAX_LENGTH_CANON)
|
|
tmp_len = MAX_LENGTH_CANON;
|
|
tmp = malloc(tmp_len * sizeof(uint32_t));
|
|
if (tmp == NULL)
|
|
return ENOMEM;
|
|
|
|
ret = compat_decomp(in, in_len, tmp, &tmp_len);
|
|
if (ret) {
|
|
free(tmp);
|
|
return ret;
|
|
}
|
|
canonical_reorder(tmp, tmp_len);
|
|
ret = combine(tmp, tmp_len, out, out_len);
|
|
free(tmp);
|
|
return ret;
|
|
}
|