lib/libc/citrus/citrus_csmapper.c

   1 /*      $NetBSD: citrus_csmapper.c,v 1.8.12.1 2009/01/04 17:02:18 christos 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 1.8.12.1 2009/01/04 17:02:18 christos 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/queue.h>
  45
  46 #include "citrus_namespace.h"
  47 #include "citrus_types.h"
  48 #include "citrus_bcs.h"
  49 #include "citrus_region.h"
  50 #include "citrus_memstream.h"
  51 #include "citrus_mmap.h"
  52 #include "citrus_module.h"
  53 #include "citrus_hash.h"
  54 #include "citrus_mapper.h"
  55 #include "citrus_csmapper.h"
  56 #include "citrus_pivot_file.h"
  57 #include "citrus_db.h"
  58 #include "citrus_db_hash.h"
  59 #include "citrus_lookup.h"
  60
  61 #ifdef _REENTRANT
  62 static rwlock_t lock = RWLOCK_INITIALIZER;
  63 #endif
  64 static struct _citrus_mapper_area *maparea = NULL;
  65
  66 #define CS_ALIAS        _PATH_CSMAPPER "/charset.alias"
  67 #define CS_PIVOT        _PATH_CSMAPPER "/charset.pivot"
  68
  69
  70 /* ---------------------------------------------------------------------- */
  71
  72 static int
  73 get32(struct _region *r, uint32_t *rval)
  74 {
  75         if (_region_size(r) != 4)
  76                 return EFTYPE;
  77
  78         memcpy(rval, _region_head(r), (size_t)4);
  79         *rval = be32toh(*rval);
  80
  81         return 0;
  82 }
  83
  84 static int
  85 open_subdb(struct _citrus_db **subdb, struct _citrus_db *db, const char *src)
  86 {
  87         int ret;
  88         struct _region r;
  89
  90         ret = _db_lookup_by_s(db, src, &r, NULL);
  91         if (ret)
  92                 return ret;
  93         ret = _db_open(subdb, &r, _CITRUS_PIVOT_SUB_MAGIC, _db_hash_std, NULL);
  94         if (ret)
  95                 return ret;
  96
  97         return 0;
  98 }
  99
 100
 101 #define NO_SUCH_FILE    EOPNOTSUPP
 102 static int
 103 find_best_pivot_pvdb(const char *src, const char *dst, char *pivot,
 104                      size_t pvlen, unsigned long *rnorm)
 105 {
 106         int ret, num, i;
 107         struct _region fr, r1, r2;
 108         struct _citrus_db *db1, *db2, *db3;
 109         char buf[LINE_MAX];
 110         unsigned long norm;
 111         uint32_t val32;
 112
 113         ret = _map_file(&fr, CS_PIVOT ".pvdb");
 114         if (ret) {
 115                 if (ret == ENOENT)
 116                         ret = NO_SUCH_FILE;
 117                 return ret;
 118         }
 119         ret = _db_open(&db1, &fr, _CITRUS_PIVOT_MAGIC, _db_hash_std, NULL);
 120         if (ret)
 121                 goto quit1;
 122         ret = open_subdb(&db2, db1, src);
 123         if (ret)
 124                 goto quit2;
 125
 126         num = _db_get_num_entries(db2);
 127         *rnorm = ULONG_MAX;
 128         for (i = 0; i < num; i++) {
 129                 /* iterate each pivot */
 130                 ret = _db_get_entry(db2, i, &r1, &r2);
 131                 if (ret)
 132                         goto quit3;
 133                 /* r1:pivot name, r2:norm among src and pivot */
 134                 ret = get32(&r2, &val32);
 135                 if (ret)
 136                         goto quit3;
 137                 norm = val32;
 138                 snprintf(buf, sizeof(buf), "%.*s",
 139                          (int)_region_size(&r1), (char *)_region_head(&r1));
 140                 /* buf: pivot name */
 141                 ret = open_subdb(&db3, db1, buf);
 142                 if (ret)
 143                         goto quit3;
 144                 if (_db_lookup_by_s(db3, dst, &r2, NULL) != 0)
 145                         goto quit4;
 146                 /* r2: norm among pivot and dst */
 147                 ret = get32(&r2, &val32);
 148                 if (ret)
 149                         goto quit4;
 150                 norm += val32;
 151                 /* judge minimum norm */
 152                 if (norm < *rnorm) {
 153                         *rnorm = norm;
 154                         strlcpy(pivot, buf, pvlen);
 155                 }
 156 quit4:
 157                 _db_close(db3);
 158                 if (ret)
 159                         goto quit3;
 160         }
 161 quit3:
 162         _db_close(db2);
 163 quit2:
 164         _db_close(db1);
 165 quit1:
 166         _unmap_file(&fr);
 167         if (ret)
 168                 return ret;
 169
 170         if (*rnorm == ULONG_MAX)
 171                 return ENOENT;
 172
 173         return 0;
 174 }
 175
 176 /* ---------------------------------------------------------------------- */
 177
 178 struct zone {
 179         const char *begin, *end;
 180 };
 181
 182 struct parse_arg {
 183         char dst[PATH_MAX];
 184         unsigned long norm;
 185 };
 186
 187 static int
 188 parse_line(struct parse_arg *pa, struct _region *r)
 189 {
 190         char buf[20];
 191         struct zone z1, z2;
 192         size_t len;
 193
 194         len = _region_size(r);
 195         z1.begin = _bcs_skip_ws_len(_region_head(r), &len);
 196         if (len == 0)
 197                 return EFTYPE;
 198         z1.end = _bcs_skip_nonws_len(z1.begin, &len);
 199         if (len == 0)
 200                 return EFTYPE;
 201         z2.begin = _bcs_skip_ws_len(z1.end, &len);
 202         if (len == 0)
 203                 return EFTYPE;
 204         z2.end = _bcs_skip_nonws_len(z2.begin, &len);
 205
 206         /* z1 : dst name, z2 : norm */
 207         snprintf(pa->dst, sizeof(pa->dst),
 208                  "%.*s", (int)(z1.end-z1.begin), z1.begin);
 209         snprintf(buf, sizeof(buf),
 210                  "%.*s", (int)(z2.end-z2.begin), z2.begin);
 211         pa->norm = _bcs_strtoul(buf, NULL, 0);
 212
 213         return 0;
 214 }
 215
 216 static int
 217 find_dst(struct parse_arg *pasrc, const char *dst)
 218 {
 219         int ret;
 220         struct parse_arg padst;
 221         struct _lookup *cl;
 222         struct _region data;
 223
 224         ret = _lookup_seq_open(&cl, CS_PIVOT, _LOOKUP_CASE_IGNORE);
 225         if (ret)
 226                 return ret;
 227
 228         ret = _lookup_seq_lookup(cl, pasrc->dst, &data);
 229         while (ret == 0) {
 230                 ret = parse_line(&padst, &data);
 231                 if (ret)
 232                         break;
 233                 if (strcmp(dst, padst.dst) == 0) {
 234                         pasrc->norm += padst.norm;
 235                         break;
 236                 }
 237                 ret = _lookup_seq_next(cl, NULL, &data);
 238         }
 239         _lookup_seq_close(cl);
 240
 241         return ret;
 242 }
 243
 244 static int
 245 find_best_pivot_lookup(const char *src, const char *dst, char *pivot,
 246                        size_t pvlen, unsigned long *rnorm)
 247 {
 248         int ret;
 249         struct _lookup *cl;
 250         struct _region data;
 251         struct parse_arg pa;
 252         unsigned long norm_min;
 253         char pivot_min[PATH_MAX];
 254
 255         ret = _lookup_seq_open(&cl, CS_PIVOT, _LOOKUP_CASE_IGNORE);
 256         if (ret)
 257                 return ret;
 258
 259         norm_min = ULONG_MAX;
 260
 261         /* find pivot code */
 262         ret = _lookup_seq_lookup(cl, src, &data);
 263         while (ret == 0) {
 264                 ret = parse_line(&pa, &data);
 265                 if (ret)
 266                         break;
 267                 ret = find_dst(&pa, dst);
 268                 if (ret)
 269                         break;
 270                 if (pa.norm < norm_min) {
 271                         norm_min = pa.norm;
 272                         strlcpy(pivot_min, pa.dst, sizeof(pivot_min));
 273                 }
 274                 ret = _lookup_seq_next(cl, NULL, &data);
 275         }
 276         _lookup_seq_close(cl);
 277
 278         if (ret != ENOENT)
 279                 return ret;
 280         if (norm_min == ULONG_MAX)
 281                 return ENOENT;
 282         strlcpy(pivot, pivot_min, pvlen);
 283         if (rnorm)
 284                 *rnorm = norm_min;
 285
 286         return 0;
 287 }
 288
 289 static int
 290 find_best_pivot(const char *src, const char *dst, char *pivot, size_t pvlen,
 291                 unsigned long *rnorm)
 292 {
 293         int ret;
 294
 295         ret = find_best_pivot_pvdb(src, dst, pivot, pvlen, rnorm);
 296         if (ret == NO_SUCH_FILE)
 297                 ret = find_best_pivot_lookup(src, dst, pivot, pvlen, rnorm);
 298
 299         return ret;
 300 }
 301
 302 static __inline int
 303 open_serial_mapper(struct _citrus_mapper_area *__restrict ma,
 304                    struct _citrus_mapper * __restrict * __restrict rcm,
 305                    const char *src, const char *pivot, const char *dst)
 306 {
 307         char buf[PATH_MAX];
 308
 309         snprintf(buf, sizeof(buf), "%s/%s,%s/%s", src, pivot, pivot, dst);
 310
 311         return _mapper_open_direct(ma, rcm, "mapper_serial", buf);
 312 }
 313
 314 static struct _citrus_csmapper *csm_none = NULL;
 315 static int
 316 get_none(struct _citrus_mapper_area *__restrict ma,
 317          struct _citrus_csmapper *__restrict *__restrict rcsm)
 318 {
 319         int ret;
 320
 321         rwlock_wrlock(&lock);
 322         if (csm_none) {
 323                 *rcsm = csm_none;
 324                 ret = 0;
 325                 goto quit;
 326         }
 327
 328         ret = _mapper_open_direct(ma, &csm_none, "mapper_none", "");
 329         if (ret)
 330                 goto quit;
 331         _mapper_set_persistent(csm_none);
 332
 333         *rcsm = csm_none;
 334         ret = 0;
 335 quit:
 336         rwlock_unlock(&lock);
 337         return ret;
 338 }
 339
 340 int
 341 _citrus_csmapper_open(struct _citrus_csmapper * __restrict * __restrict rcsm,
 342                       const char * __restrict src, const char * __restrict dst,
 343                       uint32_t flags, unsigned long *rnorm)
 344 {
 345         int ret;
 346         char buf1[PATH_MAX], buf2[PATH_MAX], key[PATH_MAX], pivot[PATH_MAX];
 347         const char *realsrc, *realdst;
 348         unsigned long norm;
 349
 350         norm = 0;       /* XXX gcc */
 351
 352         ret = _citrus_mapper_create_area(&maparea, _PATH_CSMAPPER);
 353         if (ret)
 354                 return ret;
 355
 356         realsrc = _lookup_alias(CS_ALIAS, src, buf1, sizeof(buf1),
 357                                 _LOOKUP_CASE_IGNORE);
 358         realdst = _lookup_alias(CS_ALIAS, dst, buf2, sizeof(buf2),
 359                                 _LOOKUP_CASE_IGNORE);
 360         if (!strcmp(realsrc, realdst)) {
 361                 ret = get_none(maparea, rcsm);
 362                 if (ret == 0 && rnorm != NULL)
 363                         *rnorm = 0;
 364                 return ret;
 365         }
 366
 367         snprintf(key, sizeof(key), "%s/%s", realsrc, realdst);
 368
 369         ret = _mapper_open(maparea, rcsm, key);
 370         if (ret == 0) {
 371                 if (rnorm != NULL)
 372                         *rnorm = 0;
 373                 return 0;
 374         }
 375         if (ret != ENOENT || (flags & _CSMAPPER_F_PREVENT_PIVOT)!=0)
 376                 return ret;
 377
 378         ret = find_best_pivot(realsrc, realdst, pivot, sizeof(pivot), &norm);
 379         if (ret)
 380                 return ret;
 381
 382         ret = open_serial_mapper(maparea, rcsm, realsrc, pivot, realdst);
 383         if (ret == 0 && rnorm != NULL)
 384                 *rnorm = norm;
 385
 386         return ret;
 387 }