icuSources/common/uresdata.h

   1 /*
   2 ******************************************************************************
   3 *                                                                            *
   4 * Copyright (C) 1999-2003, 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) ((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_TOP
  52 };
  53
  54 /* number of bytes at the beginning of the bundle before the strings start */
  55 enum {
  56     URES_STRINGS_BOTTOM=(1+URES_INDEX_TOP)*4
  57 };
  58
  59 /*
  60  * File format for .res resource bundle files (formatVersion=1.1)
  61  *
  62  * An ICU4C resource bundle file (.res) is a binary, memory-mappable file
  63  * with nested, hierarchical data structures.
  64  * It physically contains the following:
  65  *
  66  *   Resource root; -- 32-bit Resource item, root item for this bundle's tree;
  67  *                     currently, the root item must be a table or table32 resource item
  68  *   int32_t indexes[indexes[0]]; -- array of indexes for friendly
  69  *                                   reading and swapping; see URES_INDEX_* above
  70  *                                   new in formatVersion 1.1
  71  *   char keys[]; -- characters for key strings
  72  *                   (formatVersion 1.0: up to 65k of characters; 1.1: <2G)
  73  *                   (minus the space for root and indexes[]),
  74  *                   which consist of invariant characters (ASCII/EBCDIC) and are NUL-terminated;
  75  *                   padded to multiple of 4 bytes for 4-alignment of the following data
  76  *   data; -- data directly and indirectly indexed by the root item;
  77  *            the structure is determined by walking the tree
  78  *
  79  * Each resource bundle item has a 32-bit Resource handle (see typedef above)
  80  * which contains the item type number in its upper 4 bits (31..28) and either
  81  * an offset or a direct value in its lower 28 bits (27..0).
  82  * The order of items is undefined and only determined by walking the tree.
  83  * Leaves of the tree may be stored first or last or anywhere in between,
  84  * and it is in theory possible to have unreferenced holes in the file.
  85  *
  86  * Direct values:
  87  * - Empty Unicode strings have an offset value of 0 in the Resource handle itself.
  88  * - Integer values are 28-bit values stored in the Resource handle itself;
  89  *   the interpretation of unsigned vs. signed integers is up to the application.
  90  *
  91  * All other types and values use 28-bit offsets to point to the item's data.
  92  * The offset is an index to the first 32-bit word of the value, relative to the
  93  * start of the resource data (i.e., the root item handle is at offset 0).
  94  * To get byte offsets, the offset is multiplied by 4 (or shifted left by 2 bits).
  95  * All resource item values are 4-aligned.
  96  *
  97  * The structures (memory layouts) for the values for each item type are listed
  98  * in the table above.
  99  *
 100  * Nested, hierarchical structures: -------------
 101  *
 102  * Table items contain key-value pairs where the keys are 16-bit offsets to char * key strings.
 103  * Key string offsets are also relative to the start of the resource data (of the root handle),
 104  * i.e., the first string has an offset of 4 (after the 4-byte root handle).
 105  *
 106  * The values of these pairs are Resource handles.
 107  *
 108  * Array items are simple vectors of Resource handles.
 109  *
 110  * An alias item is special (and new in ICU 2.4): --------------
 111  *
 112  * Its memory layout is just like for a UnicodeString, but at runtime it resolves to
 113  * another resource bundle's item according to the path in the string.
 114  * This is used to share items across bundles that are in different lookup/fallback
 115  * chains (e.g., large collation data among zh_TW and zh_HK).
 116  * This saves space (for large items) and maintenance effort (less duplication of data).
 117  *
 118  * --------------------------------------------------------------------------
 119  *
 120  * Resource types:
 121  *
 122  * Most resources have their values stored at four-byte offsets from the start
 123  * of the resource data. These values are at least 4-aligned.
 124  * Some resource values are stored directly in the offset field of the Resource itself.
 125  * See UResType in unicode/ures.h for enumeration constants for Resource types.
 126  *
 127  * Type Name            Memory layout of values
 128  *                      (in parentheses: scalar, non-offset values)
 129  *
 130  * 0  Unicode String:   int32_t length, UChar[length], (UChar)0, (padding)
 131  *                  or  (empty string ("") if offset==0)
 132  * 1  Binary:           int32_t length, uint8_t[length], (padding)
 133  *                      - this value should be 32-aligned -
 134  * 2  Table:            uint16_t count, uint16_t keyStringOffsets[count], (uint16_t padding), Resource[count]
 135  * 3  Alias:            (physically same value layout as string, new in ICU 2.4)
 136  * 4  Table32:          int32_t count, int32_t keyStringOffsets[count], Resource[count]
 137  *                      (new in formatVersion 1.1/ICU 2.8)
 138  *
 139  * 7  Integer:          (28-bit offset is integer value)
 140  * 8  Array:            int32_t count, Resource[count]
 141  *
 142  * 14 Integer Vector:   int32_t length, int32_t[length]
 143  * 15 Reserved:         This value denotes special purpose resources and is for internal use.
 144  *
 145  * Note that there are 3 types with data vector values:
 146  * - Vectors of 8-bit bytes stored as type Binary.
 147  * - Vectors of 16-bit words stored as type Unicode String
 148  *                     (no value restrictions, all values 0..ffff allowed!).
 149  * - Vectors of 32-bit words stored as type Integer Vector.
 150  */
 151
 152 /*
 153  * Structure for a single, memory-mapped ResourceBundle.
 154  */
 155 typedef struct {
 156     UDataMemory *data;
 157     Resource *pRoot;
 158     Resource rootRes;
 159 } ResourceData;
 160
 161 /*
 162  * Load a resource bundle file.
 163  * The ResourceData structure must be allocated externally.
 164  */
 165 U_CFUNC UBool
 166 res_load(ResourceData *pResData,
 167          const char *path, const char *name, UErrorCode *errorCode);
 168
 169 /*
 170  * Release a resource bundle file.
 171  * This does not release the ResourceData structure itself.
 172  */
 173 U_CFUNC void
 174 res_unload(ResourceData *pResData);
 175
 176 /*
 177  * Return a pointer to a zero-terminated, const UChar* string
 178  * and set its length in *pLength.
 179  * Returns NULL if not found.
 180  */
 181 U_CFUNC const UChar *
 182 res_getString(const ResourceData *pResData, const Resource res, int32_t *pLength);
 183
 184 U_CFUNC const UChar *
 185 res_getAlias(const ResourceData *pResData, const Resource res, int32_t *pLength);
 186
 187 U_CFUNC const uint8_t *
 188 res_getBinary(const ResourceData *pResData, const Resource res, int32_t *pLength);
 189
 190 U_CFUNC const int32_t *
 191 res_getIntVector(const ResourceData *pResData, const Resource res, int32_t *pLength);
 192
 193 U_CFUNC Resource
 194 res_getResource(const ResourceData *pResData, const char *key);
 195
 196 U_CFUNC int32_t
 197 res_countArrayItems(const ResourceData *pResData, const Resource res);
 198
 199 U_CFUNC Resource res_getArrayItem(const ResourceData *pResData, Resource array, const int32_t indexS);
 200 U_CFUNC Resource res_getTableItemByIndex(const ResourceData *pResData, Resource table, int32_t indexS, const char ** key);
 201 U_CFUNC Resource res_getTableItemByKey(const ResourceData *pResData, Resource table, int32_t *indexS, const char* * key);
 202
 203 /*
 204  * Modifies the contents of *path (replacing separators with NULs),
 205  * and also moves *path forward while it finds items.
 206  */
 207 U_CFUNC Resource res_findResource(const ResourceData *pResData, Resource r, char** path, const char** key);
 208
 209 /**
 210  * Swap an ICU resource bundle. See udataswp.h.
 211  * @internal
 212  */
 213 U_CAPI int32_t U_EXPORT2
 214 ures_swap(const UDataSwapper *ds,
 215           const void *inData, int32_t length, void *outData,
 216           UErrorCode *pErrorCode);
 217
 218 #endif