2 /* $NetBSD: citrus_csmapper.c,v 1.11 2011/11/20 07:43:52 tnozaki Exp $ */
5 * Copyright (c)2003 Citrus Project,
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
11 * 1. Redistributions of source code must retain the above copyright
12 * 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 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
30 #include <sys/cdefs.h>
31 #include <sys/endian.h>
32 #include <sys/types.h>
33 #include <sys/queue.h>
43 #include "citrus_namespace.h"
44 #include "citrus_types.h"
45 #include "citrus_bcs.h"
46 #include "citrus_region.h"
47 #include "citrus_lock.h"
48 #include "citrus_memstream.h"
49 #include "citrus_mmap.h"
50 #include "citrus_module.h"
51 #include "citrus_hash.h"
52 #include "citrus_mapper.h"
53 #include "citrus_csmapper.h"
54 #include "citrus_pivot_file.h"
55 #include "citrus_db.h"
56 #include "citrus_db_hash.h"
57 #include "citrus_lookup.h"
59 static struct _citrus_mapper_area *maparea = NULL;
61 static pthread_rwlock_t ma_lock = PTHREAD_RWLOCK_INITIALIZER;
63 #define CS_ALIAS _PATH_CSMAPPER "/charset.alias"
64 #define CS_PIVOT _PATH_CSMAPPER "/charset.pivot"
67 /* ---------------------------------------------------------------------- */
70 get32(struct _region *r, uint32_t *rval)
73 if (_region_size(r) != 4)
76 memcpy(rval, _region_head(r), (size_t)4);
77 *rval = be32toh(*rval);
83 open_subdb(struct _citrus_db **subdb, struct _citrus_db *db, const char *src)
88 ret = _db_lookup_by_s(db, src, &r, NULL);
91 ret = _db_open(subdb, &r, _CITRUS_PIVOT_SUB_MAGIC, _db_hash_std, NULL);
99 #define NO_SUCH_FILE EOPNOTSUPP
101 find_best_pivot_pvdb(const char *src, const char *dst, char *pivot,
102 size_t pvlen, unsigned long *rnorm)
104 struct _citrus_db *db1, *db2, *db3;
105 struct _region fr, r1, r2;
111 ret = _map_file(&fr, CS_PIVOT ".pvdb");
117 ret = _db_open(&db1, &fr, _CITRUS_PIVOT_MAGIC, _db_hash_std, NULL);
120 ret = open_subdb(&db2, db1, src);
124 num = _db_get_num_entries(db2);
126 for (i = 0; i < num; i++) {
127 /* iterate each pivot */
128 ret = _db_get_entry(db2, i, &r1, &r2);
131 /* r1:pivot name, r2:norm among src and pivot */
132 ret = get32(&r2, &val32);
136 snprintf(buf, sizeof(buf), "%.*s",
137 (int)_region_size(&r1), (char *)_region_head(&r1));
138 /* buf: pivot name */
139 ret = open_subdb(&db3, db1, buf);
142 if (_db_lookup_by_s(db3, dst, &r2, NULL) != 0)
143 /* don't break the loop, test all src/dst pairs. */
145 /* r2: norm among pivot and dst */
146 ret = get32(&r2, &val32);
150 /* judge minimum norm */
153 strlcpy(pivot, buf, pvlen);
169 if (*rnorm == ULONG_MAX)
175 /* ---------------------------------------------------------------------- */
178 const char *begin, *end;
187 parse_line(struct parse_arg *pa, struct _region *r)
193 len = _region_size(r);
194 z1.begin = _bcs_skip_ws_len(_region_head(r), &len);
197 z1.end = _bcs_skip_nonws_len(z1.begin, &len);
200 z2.begin = _bcs_skip_ws_len(z1.end, &len);
203 z2.end = _bcs_skip_nonws_len(z2.begin, &len);
205 /* z1 : dst name, z2 : norm */
206 snprintf(pa->dst, sizeof(pa->dst),
207 "%.*s", (int)(z1.end-z1.begin), z1.begin);
208 snprintf(buf, sizeof(buf),
209 "%.*s", (int)(z2.end-z2.begin), z2.begin);
210 pa->norm = _bcs_strtoul(buf, NULL, 0);
216 find_dst(struct parse_arg *pasrc, const char *dst)
219 struct parse_arg padst;
223 ret = _lookup_seq_open(&cl, CS_PIVOT, _LOOKUP_CASE_IGNORE);
227 ret = _lookup_seq_lookup(cl, pasrc->dst, &data);
229 ret = parse_line(&padst, &data);
232 if (strcmp(dst, padst.dst) == 0) {
233 pasrc->norm += padst.norm;
236 ret = _lookup_seq_next(cl, NULL, &data);
238 _lookup_seq_close(cl);
244 find_best_pivot_lookup(const char *src, const char *dst, char *pivot,
245 size_t pvlen, unsigned long *rnorm)
250 char pivot_min[PATH_MAX];
251 unsigned long norm_min;
254 ret = _lookup_seq_open(&cl, CS_PIVOT, _LOOKUP_CASE_IGNORE);
258 norm_min = ULONG_MAX;
260 /* find pivot code */
261 ret = _lookup_seq_lookup(cl, src, &data);
263 ret = parse_line(&pa, &data);
266 ret = find_dst(&pa, dst);
269 if (pa.norm < norm_min) {
271 strlcpy(pivot_min, pa.dst, sizeof(pivot_min));
273 ret = _lookup_seq_next(cl, NULL, &data);
275 _lookup_seq_close(cl);
279 if (norm_min == ULONG_MAX)
281 strlcpy(pivot, pivot_min, pvlen);
289 find_best_pivot(const char *src, const char *dst, char *pivot, size_t pvlen,
290 unsigned long *rnorm)
294 ret = find_best_pivot_pvdb(src, dst, pivot, pvlen, rnorm);
295 if (ret == NO_SUCH_FILE)
296 ret = find_best_pivot_lookup(src, dst, pivot, pvlen, rnorm);
302 open_serial_mapper(struct _citrus_mapper_area *__restrict ma,
303 struct _citrus_mapper * __restrict * __restrict rcm,
304 const char *src, const char *pivot, const char *dst)
308 snprintf(buf, sizeof(buf), "%s/%s,%s/%s", src, pivot, pivot, dst);
310 return (_mapper_open_direct(ma, rcm, "mapper_serial", buf));
313 static struct _citrus_csmapper *csm_none = NULL;
315 get_none(struct _citrus_mapper_area *__restrict ma,
316 struct _citrus_csmapper *__restrict *__restrict rcsm)
327 ret = _mapper_open_direct(ma, &csm_none, "mapper_none", "");
330 _mapper_set_persistent(csm_none);
340 _citrus_csmapper_open(struct _citrus_csmapper * __restrict * __restrict rcsm,
341 const char * __restrict src, const char * __restrict dst, uint32_t flags,
342 unsigned long *rnorm)
344 const char *realsrc, *realdst;
345 char buf1[PATH_MAX], buf2[PATH_MAX], key[PATH_MAX], pivot[PATH_MAX];
351 ret = _citrus_mapper_create_area(&maparea, _PATH_CSMAPPER);
355 realsrc = _lookup_alias(CS_ALIAS, src, buf1, sizeof(buf1),
356 _LOOKUP_CASE_IGNORE);
357 realdst = _lookup_alias(CS_ALIAS, dst, buf2, sizeof(buf2),
358 _LOOKUP_CASE_IGNORE);
359 if (!strcmp(realsrc, realdst)) {
360 ret = get_none(maparea, rcsm);
361 if (ret == 0 && rnorm != NULL)
366 snprintf(key, sizeof(key), "%s/%s", realsrc, realdst);
368 ret = _mapper_open(maparea, rcsm, key);
374 if (ret != ENOENT || (flags & _CSMAPPER_F_PREVENT_PIVOT)!=0)
377 ret = find_best_pivot(realsrc, realdst, pivot, sizeof(pivot), &norm);
381 ret = open_serial_mapper(maparea, rcsm, realsrc, pivot, realdst);
382 if (ret == 0 && rnorm != NULL)