2 * Copyright (c) 2004 Kungliga Tekniska Högskolan
3 * (Royal Institute of Technology, Stockholm, Sweden).
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
17 * 3. Neither the name of the Institute nor the names of its contributors
18 * may be used to endorse or promote products derived from this software
19 * without specific prior written permission.
21 * THIS SOFTWARE IS PROVIDED BY THE INSTITUTE AND CONTRIBUTORS ``AS IS'' AND
22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
24 * ARE DISCLAIMED. IN NO EVENT SHALL THE INSTITUTE OR CONTRIBUTORS BE LIABLE
25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
43 #include "normalize_table.h"
48 translation_cmp(const void *key, const void *data)
50 const struct translation *t1 = (const struct translation *)key;
51 const struct translation *t2 = (const struct translation *)data;
53 return t1->key - t2->key;
56 enum { s_base = 0xAC00};
57 enum { s_count = 11172};
58 enum { l_base = 0x1100};
60 enum { v_base = 0x1161};
62 enum { t_base = 0x11A7};
64 enum { n_count = v_count * t_count};
67 hangul_decomp(const uint32_t *in, size_t in_len,
68 uint32_t *out, size_t *out_len)
75 if (u < s_base || u >= s_base + s_count)
78 l = l_base + s_index / n_count;
79 v = v_base + (s_index % n_count) / t_count;
80 t = t_base + s_index % t_count;
85 return WIND_ERR_OVERRUN;
95 hangul_composition(const uint32_t *in, size_t in_len)
99 if (in[0] >= l_base && in[0] < l_base + l_count) {
100 unsigned l_index = in[0] - l_base;
103 if (in[1] < v_base || in[1] >= v_base + v_count)
105 v_index = in[1] - v_base;
106 return (l_index * v_count + v_index) * t_count + s_base;
107 } else if (in[0] >= s_base && in[0] < s_base + s_count) {
108 unsigned s_index = in[0] - s_base;
111 if (s_index % t_count != 0)
113 if (in[1] < t_base || in[1] >= t_base + t_count)
115 t_index = in[1] - t_base;
116 return in[0] + t_index;
122 compat_decomp(const uint32_t *in, size_t in_len,
123 uint32_t *out, size_t *out_len)
128 for (i = 0; i < in_len; ++i) {
129 struct translation ts = {in[i]};
130 size_t sub_len = *out_len - o;
133 ret = hangul_decomp(in + i, in_len - i,
136 if (ret == WIND_ERR_OVERRUN)
140 void *s = bsearch(&ts,
141 _wind_normalize_table,
142 _wind_normalize_table_size,
143 sizeof(_wind_normalize_table[0]),
146 const struct translation *t = (const struct translation *)s;
148 ret = compat_decomp(_wind_normalize_val_table + t->val_offset,
156 return WIND_ERR_OVERRUN;
167 cc_cmp(const void *a, const void *b)
169 const uint32_t *ua = (const uint32_t *)a;
170 const uint32_t *ub = (const uint32_t *)b;
172 return _wind_combining_class(*ua) - _wind_combining_class(*ub);
176 canonical_reorder(uint32_t *tmp, size_t tmp_len)
180 for (i = 0; i < tmp_len; ++i) {
181 int cc = _wind_combining_class(tmp[i]);
185 j < tmp_len && _wind_combining_class(tmp[j]);
188 qsort(&tmp[i], j - i, sizeof(unsigned),
196 find_composition(const uint32_t *in, unsigned in_len)
198 unsigned short canon_index = 0;
202 cur = hangul_composition(in, in_len);
207 const struct canon_node *c = &_wind_canon_table[canon_index];
217 if (i < c->next_start || i >= c->next_end)
221 _wind_canon_next_table[c->next_offset + i - c->next_start];
222 if (canon_index != 0) {
223 cur = (cur << 4) & 0xFFFFF;
226 } while (canon_index != 0);
231 combine(const uint32_t *in, size_t in_len,
232 uint32_t *out, size_t *out_len)
240 for (i = 0; i < in_len;) {
241 while (i < in_len && (cc = _wind_combining_class(in[i])) != 0) {
246 return WIND_ERR_OVERRUN;
255 v[0] = out[ostarter];
258 cc = _wind_combining_class(in[i]);
259 if (old_cc != cc && (comb = find_composition(v, 2))) {
260 out[ostarter] = comb;
261 } else if (cc == 0) {
265 return WIND_ERR_OVERRUN;
278 _wind_stringprep_normalize(const uint32_t *in, size_t in_len,
279 uint32_t *out, size_t *out_len)
285 tmp_len = in_len * 4;
286 if (tmp_len < MAX_LENGTH_CANON)
287 tmp_len = MAX_LENGTH_CANON;
288 tmp = malloc(tmp_len * sizeof(uint32_t));
292 ret = compat_decomp(in, in_len, tmp, &tmp_len);
297 canonical_reorder(tmp, tmp_len);
298 ret = combine(tmp, tmp_len, out, out_len);