1 /* $NetBSD: citrus_csmapper.c,v 1.13 2025/12/16 12:39:01 nia Exp $ */ 2 3 /*- 4 * Copyright (c)2003 Citrus Project, 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 19 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 20 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 21 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 22 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 23 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 24 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 25 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 26 * SUCH DAMAGE. 27 */ 28 29 #include <sys/cdefs.h> 30 #if defined(LIBC_SCCS) && !defined(lint) 31 __RCSID("$NetBSD: citrus_csmapper.c,v 1.13 2025/12/16 12:39:01 nia Exp $"); 32 #endif /* LIBC_SCCS and not lint */ 33 34 #include "namespace.h" 35 #include "reentrant.h" 36 #include <assert.h> 37 #include <stdio.h> 38 #include <stdlib.h> 39 #include <string.h> 40 #include <errno.h> 41 #include <limits.h> 42 #include <paths.h> 43 #include <sys/types.h> 44 #include <sys/endian.h> 45 #include <sys/queue.h> 46 47 #include "citrus_namespace.h" 48 #include "citrus_types.h" 49 #include "citrus_bcs.h" 50 #include "citrus_region.h" 51 #include "citrus_memstream.h" 52 #include "citrus_mmap.h" 53 #include "citrus_module.h" 54 #include "citrus_hash.h" 55 #include "citrus_mapper.h" 56 #include "citrus_csmapper.h" 57 #include "citrus_pivot_file.h" 58 #include "citrus_db.h" 59 #include "citrus_db_hash.h" 60 #include "citrus_lookup.h" 61 62 #ifdef _REENTRANT 63 static rwlock_t lock = RWLOCK_INITIALIZER; 64 #endif 65 static struct _citrus_mapper_area *maparea = NULL; 66 67 #define CS_ALIAS _PATH_CSMAPPER "/charset.alias" 68 #define CS_PIVOT _PATH_CSMAPPER "/charset.pivot" 69 70 71 /* ---------------------------------------------------------------------- */ 72 73 static int 74 get32(struct _region *r, uint32_t *rval) 75 { 76 if (_region_size(r) != 4) 77 return EFTYPE; 78 79 memcpy(rval, _region_head(r), (size_t)4); 80 *rval = be32toh(*rval); 81 82 return 0; 83 } 84 85 static int 86 open_subdb(struct _citrus_db **subdb, struct _citrus_db *db, const char *src) 87 { 88 int ret; 89 struct _region r; 90 91 ret = _db_lookup_by_s(db, src, &r, NULL); 92 if (ret) 93 return ret; 94 ret = _db_open(subdb, &r, _CITRUS_PIVOT_SUB_MAGIC, _db_hash_std, NULL); 95 if (ret) 96 return ret; 97 98 return 0; 99 } 100 101 102 #define NO_SUCH_FILE EOPNOTSUPP 103 static int 104 find_best_pivot_pvdb(const char *src, const char *dst, char *pivot, 105 size_t pvlen, unsigned long *rnorm) 106 { 107 int ret, num, i; 108 struct _region fr, r1, r2; 109 struct _citrus_db *db1, *db2, *db3; 110 char buf[LINE_MAX]; 111 unsigned long norm; 112 uint32_t val32; 113 114 ret = _map_file(&fr, CS_PIVOT ".pvdb"); 115 if (ret) { 116 if (ret == ENOENT) 117 ret = NO_SUCH_FILE; 118 return ret; 119 } 120 ret = _db_open(&db1, &fr, _CITRUS_PIVOT_MAGIC, _db_hash_std, NULL); 121 if (ret) 122 goto quit1; 123 ret = open_subdb(&db2, db1, src); 124 if (ret) 125 goto quit2; 126 127 num = _db_get_num_entries(db2); 128 *rnorm = ULONG_MAX; 129 for (i = 0; i < num; i++) { 130 /* iterate each pivot */ 131 ret = _db_get_entry(db2, i, &r1, &r2); 132 if (ret) 133 goto quit3; 134 /* r1:pivot name, r2:norm among src and pivot */ 135 ret = get32(&r2, &val32); 136 if (ret) 137 goto quit3; 138 norm = val32; 139 snprintf(buf, sizeof(buf), "%.*s", 140 (int)_region_size(&r1), (char *)_region_head(&r1)); 141 /* buf: pivot name */ 142 ret = open_subdb(&db3, db1, buf); 143 if (ret) 144 goto quit3; 145 if (_db_lookup_by_s(db3, dst, &r2, NULL) != 0) 146 /* don't break the loop, test all src/dst pairs. */ 147 goto quit4; 148 /* r2: norm among pivot and dst */ 149 ret = get32(&r2, &val32); 150 if (ret) 151 goto quit4; 152 norm += val32; 153 /* judge minimum norm */ 154 if (norm < *rnorm) { 155 *rnorm = norm; 156 strlcpy(pivot, buf, pvlen); 157 } 158 quit4: 159 _db_close(db3); 160 if (ret) 161 goto quit3; 162 } 163 quit3: 164 _db_close(db2); 165 quit2: 166 _db_close(db1); 167 quit1: 168 _unmap_file(&fr); 169 if (ret) 170 return ret; 171 172 if (*rnorm == ULONG_MAX) 173 return ENOENT; 174 175 return 0; 176 } 177 178 /* ---------------------------------------------------------------------- */ 179 180 struct zone { 181 const char *begin, *end; 182 }; 183 184 struct parse_arg { 185 char dst[PATH_MAX]; 186 unsigned long norm; 187 }; 188 189 static int 190 parse_line(struct parse_arg *pa, struct _region *r) 191 { 192 char buf[20]; 193 struct zone z1, z2; 194 size_t len; 195 196 len = _region_size(r); 197 z1.begin = _bcs_skip_ws_len(_region_head(r), &len); 198 if (len == 0) 199 return EFTYPE; 200 z1.end = _bcs_skip_nonws_len(z1.begin, &len); 201 if (len == 0) 202 return EFTYPE; 203 z2.begin = _bcs_skip_ws_len(z1.end, &len); 204 if (len == 0) 205 return EFTYPE; 206 z2.end = _bcs_skip_nonws_len(z2.begin, &len); 207 208 /* z1 : dst name, z2 : norm */ 209 snprintf(pa->dst, sizeof(pa->dst), 210 "%.*s", (int)(z1.end-z1.begin), z1.begin); 211 snprintf(buf, sizeof(buf), 212 "%.*s", (int)(z2.end-z2.begin), z2.begin); 213 pa->norm = _bcs_strtoul(buf, NULL, 0); 214 215 return 0; 216 } 217 218 static int 219 find_dst(struct parse_arg *pasrc, const char *dst) 220 { 221 int ret; 222 struct parse_arg padst; 223 struct _lookup *cl; 224 struct _region data; 225 226 ret = _lookup_seq_open(&cl, CS_PIVOT, _LOOKUP_CASE_IGNORE); 227 if (ret) 228 return ret; 229 230 ret = _lookup_seq_lookup(cl, pasrc->dst, &data); 231 while (ret == 0) { 232 ret = parse_line(&padst, &data); 233 if (ret) 234 break; 235 if (strcmp(dst, padst.dst) == 0) { 236 pasrc->norm += padst.norm; 237 break; 238 } 239 ret = _lookup_seq_next(cl, NULL, &data); 240 } 241 _lookup_seq_close(cl); 242 243 return ret; 244 } 245 246 static int 247 find_best_pivot_lookup(const char *src, const char *dst, char *pivot, 248 size_t pvlen, unsigned long *rnorm) 249 { 250 int ret; 251 struct _lookup *cl; 252 struct _region data; 253 struct parse_arg pa; 254 unsigned long norm_min; 255 char pivot_min[PATH_MAX]; 256 257 ret = _lookup_seq_open(&cl, CS_PIVOT, _LOOKUP_CASE_IGNORE); 258 if (ret) 259 return ret; 260 261 norm_min = ULONG_MAX; 262 263 /* find pivot code */ 264 ret = _lookup_seq_lookup(cl, src, &data); 265 while (ret == 0) { 266 ret = parse_line(&pa, &data); 267 if (ret) 268 break; 269 ret = find_dst(&pa, dst); 270 if (ret) 271 break; 272 if (pa.norm < norm_min) { 273 norm_min = pa.norm; 274 strlcpy(pivot_min, pa.dst, sizeof(pivot_min)); 275 } 276 ret = _lookup_seq_next(cl, NULL, &data); 277 } 278 _lookup_seq_close(cl); 279 280 if (ret != ENOENT) 281 return ret; 282 if (norm_min == ULONG_MAX) 283 return ENOENT; 284 strlcpy(pivot, pivot_min, pvlen); 285 if (rnorm) 286 *rnorm = norm_min; 287 288 return 0; 289 } 290 291 static int 292 find_best_pivot(const char *src, const char *dst, char *pivot, size_t pvlen, 293 unsigned long *rnorm) 294 { 295 int ret; 296 297 ret = find_best_pivot_pvdb(src, dst, pivot, pvlen, rnorm); 298 if (ret == NO_SUCH_FILE) 299 ret = find_best_pivot_lookup(src, dst, pivot, pvlen, rnorm); 300 301 return ret; 302 } 303 304 static __inline int 305 open_serial_mapper(struct _citrus_mapper_area *__restrict ma, 306 struct _citrus_mapper * __restrict * __restrict rcm, 307 const char *src, const char *pivot, const char *dst) 308 { 309 char buf[4 * PATH_MAX]; 310 311 snprintf(buf, sizeof(buf), "%s/%s,%s/%s", src, pivot, pivot, dst); 312 313 return _mapper_open_direct(ma, rcm, "mapper_serial", buf); 314 } 315 316 static struct _citrus_csmapper *csm_none = NULL; 317 static int 318 get_none(struct _citrus_mapper_area *__restrict ma, 319 struct _citrus_csmapper *__restrict *__restrict rcsm) 320 { 321 int ret; 322 323 rwlock_wrlock(&lock); 324 if (csm_none) { 325 *rcsm = csm_none; 326 ret = 0; 327 goto quit; 328 } 329 330 ret = _mapper_open_direct(ma, &csm_none, "mapper_none", ""); 331 if (ret) 332 goto quit; 333 _mapper_set_persistent(csm_none); 334 335 *rcsm = csm_none; 336 ret = 0; 337 quit: 338 rwlock_unlock(&lock); 339 return ret; 340 } 341 342 int 343 _citrus_csmapper_open(struct _citrus_csmapper * __restrict * __restrict rcsm, 344 const char * __restrict src, const char * __restrict dst, 345 uint32_t flags, unsigned long *rnorm) 346 { 347 int ret; 348 char buf1[PATH_MAX], buf2[PATH_MAX], key[PATH_MAX], pivot[PATH_MAX]; 349 const char *realsrc, *realdst; 350 unsigned long norm; 351 352 norm = 0; /* XXX gcc */ 353 354 ret = _citrus_mapper_create_area(&maparea, _PATH_CSMAPPER); 355 if (ret) 356 return ret; 357 358 realsrc = _lookup_alias(CS_ALIAS, src, buf1, sizeof(buf1), 359 _LOOKUP_CASE_IGNORE); 360 realdst = _lookup_alias(CS_ALIAS, dst, buf2, sizeof(buf2), 361 _LOOKUP_CASE_IGNORE); 362 if (!strcmp(realsrc, realdst)) { 363 ret = get_none(maparea, rcsm); 364 if (ret == 0 && rnorm != NULL) 365 *rnorm = 0; 366 return ret; 367 } 368 369 snprintf(key, sizeof(key), "%s/%s", realsrc, realdst); 370 371 ret = _mapper_open(maparea, rcsm, key); 372 if (ret == 0) { 373 if (rnorm != NULL) 374 *rnorm = 0; 375 return 0; 376 } 377 if (ret != ENOENT || (flags & _CSMAPPER_F_PREVENT_PIVOT)!=0) 378 return ret; 379 380 ret = find_best_pivot(realsrc, realdst, pivot, sizeof(pivot), &norm); 381 if (ret) 382 return ret; 383 384 ret = open_serial_mapper(maparea, rcsm, realsrc, pivot, realdst); 385 if (ret == 0 && rnorm != NULL) 386 *rnorm = norm; 387 388 return ret; 389 } 390
Indexes created Sat Feb 21 16:20:20 UTC 2026