jpayne@69: // © 2017 and later: Unicode, Inc. and others.
jpayne@69: // License & terms of use: http://www.unicode.org/copyright.html
jpayne@69: 
jpayne@69: // umutablecptrie.h (split out of ucptrie.h)
jpayne@69: // created: 2018jan24 Markus W. Scherer
jpayne@69: 
jpayne@69: #ifndef __UMUTABLECPTRIE_H__
jpayne@69: #define __UMUTABLECPTRIE_H__
jpayne@69: 
jpayne@69: #include "unicode/utypes.h"
jpayne@69: 
jpayne@69: #include "unicode/localpointer.h"
jpayne@69: #include "unicode/ucpmap.h"
jpayne@69: #include "unicode/ucptrie.h"
jpayne@69: #include "unicode/utf8.h"
jpayne@69: 
jpayne@69: U_CDECL_BEGIN
jpayne@69: 
jpayne@69: /**
jpayne@69:  * \file
jpayne@69:  *
jpayne@69:  * This file defines a mutable Unicode code point trie.
jpayne@69:  *
jpayne@69:  * @see UCPTrie
jpayne@69:  * @see UMutableCPTrie
jpayne@69:  */
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Mutable Unicode code point trie.
jpayne@69:  * Fast map from Unicode code points (U+0000..U+10FFFF) to 32-bit integer values.
jpayne@69:  * For details see http://site.icu-project.org/design/struct/utrie
jpayne@69:  *
jpayne@69:  * Setting values (especially ranges) and lookup is fast.
jpayne@69:  * The mutable trie is only somewhat space-efficient.
jpayne@69:  * It builds a compacted, immutable UCPTrie.
jpayne@69:  *
jpayne@69:  * This trie can be modified while iterating over its contents.
jpayne@69:  * For example, it is possible to merge its values with those from another
jpayne@69:  * set of ranges (e.g., another mutable or immutable trie):
jpayne@69:  * Iterate over those source ranges; for each of them iterate over this trie;
jpayne@69:  * add the source value into the value of each trie range.
jpayne@69:  *
jpayne@69:  * @see UCPTrie
jpayne@69:  * @see umutablecptrie_buildImmutable
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: typedef struct UMutableCPTrie UMutableCPTrie;
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Creates a mutable trie that initially maps each Unicode code point to the same value.
jpayne@69:  * It uses 32-bit data values until umutablecptrie_buildImmutable() is called.
jpayne@69:  * umutablecptrie_buildImmutable() takes a valueWidth parameter which
jpayne@69:  * determines the number of bits in the data value in the resulting UCPTrie.
jpayne@69:  * You must umutablecptrie_close() the trie once you are done using it.
jpayne@69:  *
jpayne@69:  * @param initialValue the initial value that is set for all code points
jpayne@69:  * @param errorValue the value for out-of-range code points and ill-formed UTF-8/16
jpayne@69:  * @param pErrorCode an in/out ICU UErrorCode
jpayne@69:  * @return the trie
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI UMutableCPTrie * U_EXPORT2
jpayne@69: umutablecptrie_open(uint32_t initialValue, uint32_t errorValue, UErrorCode *pErrorCode);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Clones a mutable trie.
jpayne@69:  * You must umutablecptrie_close() the clone once you are done using it.
jpayne@69:  *
jpayne@69:  * @param other the trie to clone
jpayne@69:  * @param pErrorCode an in/out ICU UErrorCode
jpayne@69:  * @return the trie clone
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI UMutableCPTrie * U_EXPORT2
jpayne@69: umutablecptrie_clone(const UMutableCPTrie *other, UErrorCode *pErrorCode);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Closes a mutable trie and releases associated memory.
jpayne@69:  *
jpayne@69:  * @param trie the trie
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI void U_EXPORT2
jpayne@69: umutablecptrie_close(UMutableCPTrie *trie);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Creates a mutable trie with the same contents as the UCPMap.
jpayne@69:  * You must umutablecptrie_close() the mutable trie once you are done using it.
jpayne@69:  *
jpayne@69:  * @param map the source map
jpayne@69:  * @param pErrorCode an in/out ICU UErrorCode
jpayne@69:  * @return the mutable trie
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI UMutableCPTrie * U_EXPORT2
jpayne@69: umutablecptrie_fromUCPMap(const UCPMap *map, UErrorCode *pErrorCode);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Creates a mutable trie with the same contents as the immutable one.
jpayne@69:  * You must umutablecptrie_close() the mutable trie once you are done using it.
jpayne@69:  *
jpayne@69:  * @param trie the immutable trie
jpayne@69:  * @param pErrorCode an in/out ICU UErrorCode
jpayne@69:  * @return the mutable trie
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI UMutableCPTrie * U_EXPORT2
jpayne@69: umutablecptrie_fromUCPTrie(const UCPTrie *trie, UErrorCode *pErrorCode);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Returns the value for a code point as stored in the trie.
jpayne@69:  *
jpayne@69:  * @param trie the trie
jpayne@69:  * @param c the code point
jpayne@69:  * @return the value
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI uint32_t U_EXPORT2
jpayne@69: umutablecptrie_get(const UMutableCPTrie *trie, UChar32 c);
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 trie.
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:  * The trie can be modified between calls to this function.
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:  * See the same-signature ucptrie_getRange() for a code sample.
jpayne@69:  *
jpayne@69:  * @param trie the trie
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 trie data value,
jpayne@69:  *     or NULL if the values from the trie 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, trie 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: umutablecptrie_getRange(const UMutableCPTrie *trie, UChar32 start,
jpayne@69:                         UCPMapRangeOption option, uint32_t surrogateValue,
jpayne@69:                         UCPMapValueFilter *filter, const void *context, uint32_t *pValue);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Sets a value for a code point.
jpayne@69:  *
jpayne@69:  * @param trie the trie
jpayne@69:  * @param c the code point
jpayne@69:  * @param value the value
jpayne@69:  * @param pErrorCode an in/out ICU UErrorCode
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI void U_EXPORT2
jpayne@69: umutablecptrie_set(UMutableCPTrie *trie, UChar32 c, uint32_t value, UErrorCode *pErrorCode);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Sets a value for each code point [start..end].
jpayne@69:  * Faster and more space-efficient than setting the value for each code point separately.
jpayne@69:  *
jpayne@69:  * @param trie the trie
jpayne@69:  * @param start the first code point to get the value
jpayne@69:  * @param end the last code point to get the value (inclusive)
jpayne@69:  * @param value the value
jpayne@69:  * @param pErrorCode an in/out ICU UErrorCode
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI void U_EXPORT2
jpayne@69: umutablecptrie_setRange(UMutableCPTrie *trie,
jpayne@69:                         UChar32 start, UChar32 end,
jpayne@69:                         uint32_t value, UErrorCode *pErrorCode);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Compacts the data and builds an immutable UCPTrie according to the parameters.
jpayne@69:  * After this, the mutable trie will be empty.
jpayne@69:  *
jpayne@69:  * The mutable trie stores 32-bit values until buildImmutable() is called.
jpayne@69:  * If values shorter than 32 bits are to be stored in the immutable trie,
jpayne@69:  * then the upper bits are discarded.
jpayne@69:  * For example, when the mutable trie contains values 0x81, -0x7f, and 0xa581,
jpayne@69:  * and the value width is 8 bits, then each of these is stored as 0x81
jpayne@69:  * and the immutable trie will return that as an unsigned value.
jpayne@69:  * (Some implementations may want to make productive temporary use of the upper bits
jpayne@69:  * until buildImmutable() discards them.)
jpayne@69:  *
jpayne@69:  * Not every possible set of mappings can be built into a UCPTrie,
jpayne@69:  * because of limitations resulting from speed and space optimizations.
jpayne@69:  * Every Unicode assigned character can be mapped to a unique value.
jpayne@69:  * Typical data yields data structures far smaller than the limitations.
jpayne@69:  *
jpayne@69:  * It is possible to construct extremely unusual mappings that exceed the data structure limits.
jpayne@69:  * In such a case this function will fail with a U_INDEX_OUTOFBOUNDS_ERROR.
jpayne@69:  *
jpayne@69:  * @param trie the trie trie
jpayne@69:  * @param type selects the trie type
jpayne@69:  * @param valueWidth selects the number of bits in a trie data value; if smaller than 32 bits,
jpayne@69:  *                   then the values stored in the trie will be truncated first
jpayne@69:  * @param pErrorCode an in/out ICU UErrorCode
jpayne@69:  *
jpayne@69:  * @see umutablecptrie_fromUCPTrie
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_CAPI UCPTrie * U_EXPORT2
jpayne@69: umutablecptrie_buildImmutable(UMutableCPTrie *trie, UCPTrieType type, UCPTrieValueWidth valueWidth,
jpayne@69:                               UErrorCode *pErrorCode);
jpayne@69: 
jpayne@69: U_CDECL_END
jpayne@69: 
jpayne@69: #if U_SHOW_CPLUSPLUS_API
jpayne@69: 
jpayne@69: U_NAMESPACE_BEGIN
jpayne@69: 
jpayne@69: /**
jpayne@69:  * \class LocalUMutableCPTriePointer
jpayne@69:  * "Smart pointer" class, closes a UMutableCPTrie via umutablecptrie_close().
jpayne@69:  * For most methods see the LocalPointerBase base class.
jpayne@69:  *
jpayne@69:  * @see LocalPointerBase
jpayne@69:  * @see LocalPointer
jpayne@69:  * @stable ICU 63
jpayne@69:  */
jpayne@69: U_DEFINE_LOCAL_OPEN_POINTER(LocalUMutableCPTriePointer, UMutableCPTrie, umutablecptrie_close);
jpayne@69: 
jpayne@69: U_NAMESPACE_END
jpayne@69: 
jpayne@69: #endif
jpayne@69: 
jpayne@69: #endif