jpayne@69: // © 2018 and later: Unicode, Inc. and others.
jpayne@69: // License & terms of use: http://www.unicode.org/copyright.html
jpayne@69: 
jpayne@69: // ucpmap.h
jpayne@69: // created: 2018sep03 Markus W. Scherer
jpayne@69: 
jpayne@69: #ifndef __UCPMAP_H__
jpayne@69: #define __UCPMAP_H__
jpayne@69: 
jpayne@69: #include "unicode/utypes.h"
jpayne@69: 
jpayne@69: U_CDECL_BEGIN
jpayne@69: 
jpayne@69: /**
jpayne@69:  * \file
jpayne@69:  *
jpayne@69:  * This file defines an abstract map from Unicode code points to integer values.
jpayne@69:  *
jpayne@69:  * @see UCPMap
jpayne@69:  * @see UCPTrie
jpayne@69:  * @see UMutableCPTrie
jpayne@69:  */
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Abstract map from Unicode code points (U+0000..U+10FFFF) to integer values.
jpayne@69:  *
jpayne@69:  * @see UCPTrie
jpayne@69:  * @see UMutableCPTrie
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: typedef struct UCPMap UCPMap;
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Selectors for how ucpmap_getRange() etc. should report value ranges overlapping with surrogates.
jpayne@69:  * Most users should use UCPMAP_RANGE_NORMAL.
jpayne@69:  *
jpayne@69:  * @see ucpmap_getRange
jpayne@69:  * @see ucptrie_getRange
jpayne@69:  * @see umutablecptrie_getRange
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: enum UCPMapRangeOption {
jpayne@69:     /**
jpayne@69:      * ucpmap_getRange() enumerates all same-value ranges as stored in the map.
jpayne@69:      * Most users should use this option.
jpayne@69:      * @stable ICU 63
jpayne@69:      */
jpayne@69:     UCPMAP_RANGE_NORMAL,
jpayne@69:     /**
jpayne@69:      * ucpmap_getRange() enumerates all same-value ranges as stored in the map,
jpayne@69:      * except that lead surrogates (U+D800..U+DBFF) are treated as having the
jpayne@69:      * surrogateValue, which is passed to getRange() as a separate parameter.
jpayne@69:      * The surrogateValue is not transformed via filter().
jpayne@69:      * See U_IS_LEAD(c).
jpayne@69:      *
jpayne@69:      * Most users should use UCPMAP_RANGE_NORMAL instead.
jpayne@69:      *
jpayne@69:      * This option is useful for maps that map surrogate code *units* to
jpayne@69:      * special values optimized for UTF-16 string processing
jpayne@69:      * or for special error behavior for unpaired surrogates,
jpayne@69:      * but those values are not to be associated with the lead surrogate code *points*.
jpayne@69:      * @stable ICU 63
jpayne@69:      */
jpayne@69:     UCPMAP_RANGE_FIXED_LEAD_SURROGATES,
jpayne@69:     /**
jpayne@69:      * ucpmap_getRange() enumerates all same-value ranges as stored in the map,
jpayne@69:      * except that all surrogates (U+D800..U+DFFF) are treated as having the
jpayne@69:      * surrogateValue, which is passed to getRange() as a separate parameter.
jpayne@69:      * The surrogateValue is not transformed via filter().
jpayne@69:      * See U_IS_SURROGATE(c).
jpayne@69:      *
jpayne@69:      * Most users should use UCPMAP_RANGE_NORMAL instead.
jpayne@69:      *
jpayne@69:      * This option is useful for maps that map surrogate code *units* to
jpayne@69:      * special values optimized for UTF-16 string processing
jpayne@69:      * or for special error behavior for unpaired surrogates,
jpayne@69:      * but those values are not to be associated with the lead surrogate code *points*.
jpayne@69:      * @stable ICU 63
jpayne@69:      */
jpayne@69:     UCPMAP_RANGE_FIXED_ALL_SURROGATES
jpayne@69: };
jpayne@69: #ifndef U_IN_DOXYGEN
jpayne@69: typedef enum UCPMapRangeOption UCPMapRangeOption;
jpayne@69: #endif
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Returns the value for a code point as stored in the map, with range checking.
jpayne@69:  * Returns an implementation-defined error value if c is not in the range 0..U+10FFFF.
jpayne@69:  *
jpayne@69:  * @param map the map
jpayne@69:  * @param c the code point
jpayne@69:  * @return the map value,
jpayne@69:  *         or an implementation-defined error value if the code point is not in the range 0..U+10FFFF
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI uint32_t U_EXPORT2
jpayne@69: ucpmap_get(const UCPMap *map, UChar32 c);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Callback function type: Modifies a map value.
jpayne@69:  * Optionally called by ucpmap_getRange()/ucptrie_getRange()/umutablecptrie_getRange().
jpayne@69:  * The modified value will be returned by the getRange function.
jpayne@69:  *
jpayne@69:  * Can be used to ignore some of the value bits,
jpayne@69:  * make a filter for one of several values,
jpayne@69:  * return a value index computed from the map value, etc.
jpayne@69:  *
jpayne@69:  * @param context an opaque pointer, as passed into the getRange function
jpayne@69:  * @param value a value from the map
jpayne@69:  * @return the modified value
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: typedef uint32_t U_CALLCONV
jpayne@69: UCPMapValueFilter(const void *context, uint32_t value);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Returns the last code point such that all those from start to there have the same value.
jpayne@69:  * Can be used to efficiently iterate over all same-value ranges in a map.
jpayne@69:  * (This is normally faster than iterating over code points and get()ting each value,
jpayne@69:  * but much slower than a data structure that stores ranges directly.)
jpayne@69:  *
jpayne@69:  * If the UCPMapValueFilter function pointer is not NULL, then
jpayne@69:  * the value to be delivered is passed through that function, and the return value is the end
jpayne@69:  * of the range where all values are modified to the same actual value.
jpayne@69:  * The value is unchanged if that function pointer is NULL.
jpayne@69:  *
jpayne@69:  * Example:
jpayne@69:  * \code
jpayne@69:  * UChar32 start = 0, end;
jpayne@69:  * uint32_t value;
jpayne@69:  * while ((end = ucpmap_getRange(map, start, UCPMAP_RANGE_NORMAL, 0,
jpayne@69:  *                               NULL, NULL, &value)) >= 0) {
jpayne@69:  *     // Work with the range start..end and its value.
jpayne@69:  *     start = end + 1;
jpayne@69:  * }
jpayne@69:  * \endcode
jpayne@69:  *
jpayne@69:  * @param map the map
jpayne@69:  * @param start range start
jpayne@69:  * @param option defines whether surrogates are treated normally,
jpayne@69:  *               or as having the surrogateValue; usually UCPMAP_RANGE_NORMAL
jpayne@69:  * @param surrogateValue value for surrogates; ignored if option==UCPMAP_RANGE_NORMAL
jpayne@69:  * @param filter a pointer to a function that may modify the map data value,
jpayne@69:  *     or NULL if the values from the map are to be used unmodified
jpayne@69:  * @param context an opaque pointer that is passed on to the filter function
jpayne@69:  * @param pValue if not NULL, receives the value that every code point start..end has;
jpayne@69:  *     may have been modified by filter(context, map value)
jpayne@69:  *     if that function pointer is not NULL
jpayne@69:  * @return the range end code point, or -1 if start is not a valid code point
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI UChar32 U_EXPORT2
jpayne@69: ucpmap_getRange(const UCPMap *map, UChar32 start,
jpayne@69:                 UCPMapRangeOption option, uint32_t surrogateValue,
jpayne@69:                 UCPMapValueFilter *filter, const void *context, uint32_t *pValue);
jpayne@69: 
jpayne@69: U_CDECL_END
jpayne@69: 
jpayne@69: #endif