icuSources/common/uresdata.h

   1 /*
   2 ******************************************************************************
   3 *                                                                            *
   4 * Copyright (C) 1999-2006, International Business Machines                   *
   5 *                Corporation and others. All Rights Reserved.                *
   6 *                                                                            *
   7 ******************************************************************************
   8 *   file name:  uresdata.h
   9 *   encoding:   US-ASCII
  10 *   tab size:   8 (not used)
  11 *   indentation:4
  12 *
  13 *   created on: 1999dec08
  14 *   created by: Markus W. Scherer
  15 *   06/24/02    weiv        Added support for resource sharing
  16 */
  17
  18 #ifndef __RESDATA_H__
  19 #define __RESDATA_H__
  20
  21 #include "unicode/utypes.h"
  22 #include "unicode/udata.h"
  23 #include "udataswp.h"
  24
  25 /*
  26  * A Resource is a 32-bit value that has 2 bit fields:
  27  * 31..28   4-bit type, see enum below
  28  * 27..0    28-bit four-byte-offset or value according to the type
  29  */
  30 typedef uint32_t Resource;
  31
  32 #define RES_BOGUS 0xffffffff
  33
  34 #define RES_GET_TYPE(res) ((UResType)((res)>>28UL))
  35 #define RES_GET_OFFSET(res) ((res)&0x0fffffff)
  36 #define RES_GET_POINTER(pRoot, res) ((pRoot)+RES_GET_OFFSET(res))
  37
  38 /* get signed and unsigned integer values directly from the Resource handle */
  39 #define RES_GET_INT(res) (((int32_t)((res)<<4L))>>4L)
  40 #define RES_GET_UINT(res) ((res)&0x0fffffff)
  41
  42 /* indexes[] value names; indexes are generally 32-bit (Resource) indexes */
  43 enum {
  44     URES_INDEX_LENGTH,          /* [0] contains URES_INDEX_TOP==the length of indexes[] */
  45     URES_INDEX_STRINGS_TOP,     /* [1] contains the top of the strings, */
  46                                 /*     same as the bottom of resources, rounded up */
  47     URES_INDEX_RESOURCES_TOP,   /* [2] contains the top of all resources */
  48     URES_INDEX_BUNDLE_TOP,      /* [3] contains the top of the bundle, */
  49                                 /*     in case it were ever different from [2] */
  50     URES_INDEX_MAX_TABLE_LENGTH,/* [4] max. length of any table */
  51     URES_INDEX_ATTRIBUTES,      /* [5] attributes bit set, see URES_ATT_* (new in formatVersion 1.2) */
  52     URES_INDEX_TOP
  53 };
  54
  55 /* number of bytes at the beginning of the bundle before the strings start */
  56 enum {
  57     URES_STRINGS_BOTTOM=(1+URES_INDEX_TOP)*4
  58 };
  59
  60 /*
  61  * Nofallback attribute, attribute bit 0 in indexes[URES_INDEX_ATTRIBUTES].
  62  * New in formatVersion 1.2 (ICU 3.6).
  63  *
  64  * If set, then this resource bundle is a standalone bundle.
  65  * If not set, then the bundle participates in locale fallback, eventually
  66  * all the way to the root bundle.
  67  * If indexes[] is missing or too short, then the attribute cannot be determined
  68  * reliably. Dependency checking should ignore such bundles, and loading should
  69  * use fallbacks.
  70  */
  71 #define URES_ATT_NO_FALLBACK 1
  72
  73 /*
  74  * File format for .res resource bundle files (formatVersion=1.2)
  75  *
  76  * An ICU4C resource bundle file (.res) is a binary, memory-mappable file
  77  * with nested, hierarchical data structures.
  78  * It physically contains the following:
  79  *
  80  *   Resource root; -- 32-bit Resource item, root item for this bundle's tree;
  81  *                     currently, the root item must be a table or table32 resource item
  82  *   int32_t indexes[indexes[0]]; -- array of indexes for friendly
  83  *                                   reading and swapping; see URES_INDEX_* above
  84  *                                   new in formatVersion 1.1 (ICU 2.8)
  85  *   char keys[]; -- characters for key strings
  86  *                   (formatVersion 1.0: up to 65k of characters; 1.1: <2G)
  87  *                   (minus the space for root and indexes[]),
  88  *                   which consist of invariant characters (ASCII/EBCDIC) and are NUL-terminated;
  89  *                   padded to multiple of 4 bytes for 4-alignment of the following data
  90  *   data; -- data directly and indirectly indexed by the root item;
  91  *            the structure is determined by walking the tree
  92  *
  93  * Each resource bundle item has a 32-bit Resource handle (see typedef above)
  94  * which contains the item type number in its upper 4 bits (31..28) and either
  95  * an offset or a direct value in its lower 28 bits (27..0).
  96  * The order of items is undefined and only determined by walking the tree.
  97  * Leaves of the tree may be stored first or last or anywhere in between,
  98  * and it is in theory possible to have unreferenced holes in the file.
  99  *
 100  * Direct values:
 101  * - Empty Unicode strings have an offset value of 0 in the Resource handle itself.
 102  * - Integer values are 28-bit values stored in the Resource handle itself;
 103  *   the interpretation of unsigned vs. signed integers is up to the application.
 104  *
 105  * All other types and values use 28-bit offsets to point to the item's data.
 106  * The offset is an index to the first 32-bit word of the value, relative to the
 107  * start of the resource data (i.e., the root item handle is at offset 0).
 108  * To get byte offsets, the offset is multiplied by 4 (or shifted left by 2 bits).
 109  * All resource item values are 4-aligned.
 110  *
 111  * The structures (memory layouts) for the values for each item type are listed
 112  * in the table above.
 113  *
 114  * Nested, hierarchical structures: -------------
 115  *
 116  * Table items contain key-value pairs where the keys are 16-bit offsets to char * key strings.
 117  * Key string offsets are also relative to the start of the resource data (of the root handle),
 118  * i.e., the first string has an offset of 4 (after the 4-byte root handle).
 119  *
 120  * The values of these pairs are Resource handles.
 121  *
 122  * Array items are simple vectors of Resource handles.
 123  *
 124  * An alias item is special (and new in ICU 2.4): --------------
 125  *
 126  * Its memory layout is just like for a UnicodeString, but at runtime it resolves to
 127  * another resource bundle's item according to the path in the string.
 128  * This is used to share items across bundles that are in different lookup/fallback
 129  * chains (e.g., large collation data among zh_TW and zh_HK).
 130  * This saves space (for large items) and maintenance effort (less duplication of data).
 131  *
 132  * --------------------------------------------------------------------------
 133  *
 134  * Resource types:
 135  *
 136  * Most resources have their values stored at four-byte offsets from the start
 137  * of the resource data. These values are at least 4-aligned.
 138  * Some resource values are stored directly in the offset field of the Resource itself.
 139  * See UResType in unicode/ures.h for enumeration constants for Resource types.
 140  *
 141  * Type Name            Memory layout of values
 142  *                      (in parentheses: scalar, non-offset values)
 143  *
 144  * 0  Unicode String:   int32_t length, UChar[length], (UChar)0, (padding)
 145  *                  or  (empty string ("") if offset==0)
 146  * 1  Binary:           int32_t length, uint8_t[length], (padding)
 147  *                      - this value should be 32-aligned -
 148  * 2  Table:            uint16_t count, uint16_t keyStringOffsets[count], (uint16_t padding), Resource[count]
 149  * 3  Alias:            (physically same value layout as string, new in ICU 2.4)
 150  * 4  Table32:          int32_t count, int32_t keyStringOffsets[count], Resource[count]
 151  *                      (new in formatVersion 1.1/ICU 2.8)
 152  *
 153  * 7  Integer:          (28-bit offset is integer value)
 154  * 8  Array:            int32_t count, Resource[count]
 155  *
 156  * 14 Integer Vector:   int32_t length, int32_t[length]
 157  * 15 Reserved:         This value denotes special purpose resources and is for internal use.
 158  *
 159  * Note that there are 3 types with data vector values:
 160  * - Vectors of 8-bit bytes stored as type Binary.
 161  * - Vectors of 16-bit words stored as type Unicode String
 162  *                     (no value restrictions, all values 0..ffff allowed!).
 163  * - Vectors of 32-bit words stored as type Integer Vector.
 164  */
 165
 166 /*
 167  * Structure for a single, memory-mapped ResourceBundle.
 168  */
 169 typedef struct {
 170     UDataMemory *data;
 171     Resource *pRoot;
 172     Resource rootRes;
 173     UBool noFallback; /* see URES_ATT_NO_FALLBACK */
 174 } ResourceData;
 175
 176 /*
 177  * Load a resource bundle file.
 178  * The ResourceData structure must be allocated externally.
 179  */
 180 U_CFUNC UBool
 181 res_load(ResourceData *pResData,
 182          const char *path, const char *name, UErrorCode *errorCode);
 183
 184 /*
 185  * Release a resource bundle file.
 186  * This does not release the ResourceData structure itself.
 187  */
 188 U_CFUNC void
 189 res_unload(ResourceData *pResData);
 190
 191 /*
 192  * Return a pointer to a zero-terminated, const UChar* string
 193  * and set its length in *pLength.
 194  * Returns NULL if not found.
 195  */
 196 U_CFUNC const UChar *
 197 res_getString(const ResourceData *pResData, const Resource res, int32_t *pLength);
 198
 199 U_CFUNC const UChar *
 200 res_getAlias(const ResourceData *pResData, const Resource res, int32_t *pLength);
 201
 202 U_CFUNC const uint8_t *
 203 res_getBinary(const ResourceData *pResData, const Resource res, int32_t *pLength);
 204
 205 U_CFUNC const int32_t *
 206 res_getIntVector(const ResourceData *pResData, const Resource res, int32_t *pLength);
 207
 208 U_CFUNC Resource
 209 res_getResource(const ResourceData *pResData, const char *key);
 210
 211 U_CFUNC int32_t
 212 res_countArrayItems(const ResourceData *pResData, const Resource res);
 213
 214 U_CFUNC Resource res_getArrayItem(const ResourceData *pResData, Resource array, const int32_t indexS);
 215 U_CFUNC Resource res_getTableItemByIndex(const ResourceData *pResData, Resource table, int32_t indexS, const char ** key);
 216 U_CFUNC Resource res_getTableItemByKey(const ResourceData *pResData, Resource table, int32_t *indexS, const char* * key);
 217
 218 /*
 219  * Modifies the contents of *path (replacing separators with NULs),
 220  * and also moves *path forward while it finds items.
 221  */
 222 U_CFUNC Resource res_findResource(const ResourceData *pResData, Resource r, char** path, const char** key);
 223
 224 /**
 225  * Swap an ICU resource bundle. See udataswp.h.
 226  * @internal
 227  */
 228 U_CAPI int32_t U_EXPORT2
 229 ures_swap(const UDataSwapper *ds,
 230           const void *inData, int32_t length, void *outData,
 231           UErrorCode *pErrorCode);
 232
 233 #endif