michael@0: /*
michael@0: *******************************************************************************
michael@0: *   Copyright (C) 2001-2011, International Business Machines
michael@0: *   Corporation and others.  All Rights Reserved.
michael@0: *******************************************************************************
michael@0: *
michael@0: * File ucoleitr.cpp
michael@0: *
michael@0: * Modification History:
michael@0: *
michael@0: * Date        Name        Description
michael@0: * 02/15/2001  synwee      Modified all methods to process its own function 
michael@0: *                         instead of calling the equivalent c++ api (coleitr.h)
michael@0: *******************************************************************************/
michael@0: 
michael@0: #ifndef UCOLEITR_H
michael@0: #define UCOLEITR_H
michael@0: 
michael@0: #include "unicode/utypes.h"
michael@0: 
michael@0: #if !UCONFIG_NO_COLLATION
michael@0: 
michael@0: /**  
michael@0:  * This indicates an error has occured during processing or if no more CEs is 
michael@0:  * to be returned.
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: #define UCOL_NULLORDER        ((int32_t)0xFFFFFFFF)
michael@0: 
michael@0: #ifndef U_HIDE_INTERNAL_API
michael@0: /**
michael@0:  * This indicates an error has occured during processing or there are no more CEs 
michael@0:  * to be returned.
michael@0:  *
michael@0:  * @internal
michael@0:  */
michael@0: #define UCOL_PROCESSED_NULLORDER        ((int64_t)U_INT64_MAX)
michael@0: #endif  /* U_HIDE_INTERNAL_API */
michael@0: 
michael@0: #include "unicode/ucol.h"
michael@0: 
michael@0: /** 
michael@0:  * The UCollationElements struct.
michael@0:  * For usage in C programs.
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: typedef struct UCollationElements UCollationElements;
michael@0: 
michael@0: /**
michael@0:  * \file
michael@0:  * \brief C API: UCollationElements
michael@0:  *
michael@0:  * The UCollationElements API is used as an iterator to walk through each 
michael@0:  * character of an international string. Use the iterator to return the
michael@0:  * ordering priority of the positioned character. The ordering priority of a 
michael@0:  * character, which we refer to as a key, defines how a character is collated 
michael@0:  * in the given collation object.
michael@0:  * For example, consider the following in Spanish:
michael@0:  * <pre>
michael@0:  * .       "ca" -> the first key is key('c') and second key is key('a').
michael@0:  * .       "cha" -> the first key is key('ch') and second key is key('a').
michael@0:  * </pre>
michael@0:  * And in German,
michael@0:  * <pre>
michael@0:  * .       "<ae ligature>b"-> the first key is key('a'), the second key is key('e'), and
michael@0:  * .       the third key is key('b').
michael@0:  * </pre>
michael@0:  * <p>Example of the iterator usage: (without error checking)
michael@0:  * <pre>
michael@0:  * .  void CollationElementIterator_Example()
michael@0:  * .  {
michael@0:  * .      UChar *s;
michael@0:  * .      t_int32 order, primaryOrder;
michael@0:  * .      UCollationElements *c;
michael@0:  * .      UCollatorOld *coll;
michael@0:  * .      UErrorCode success = U_ZERO_ERROR;
michael@0:  * .      s=(UChar*)malloc(sizeof(UChar) * (strlen("This is a test")+1) );
michael@0:  * .      u_uastrcpy(s, "This is a test");
michael@0:  * .      coll = ucol_open(NULL, &success);
michael@0:  * .      c = ucol_openElements(coll, str, u_strlen(str), &status);
michael@0:  * .      order = ucol_next(c, &success);
michael@0:  * .      ucol_reset(c);
michael@0:  * .      order = ucol_prev(c, &success);
michael@0:  * .      free(s);
michael@0:  * .      ucol_close(coll);
michael@0:  * .      ucol_closeElements(c);
michael@0:  * .  }
michael@0:  * </pre>
michael@0:  * <p>
michael@0:  * ucol_next() returns the collation order of the next.
michael@0:  * ucol_prev() returns the collation order of the previous character.
michael@0:  * The Collation Element Iterator moves only in one direction between calls to
michael@0:  * ucol_reset. That is, ucol_next() and ucol_prev can not be inter-used. 
michael@0:  * Whenever ucol_prev is to be called after ucol_next() or vice versa, 
michael@0:  * ucol_reset has to be called first to reset the status, shifting pointers to 
michael@0:  * either the end or the start of the string. Hence at the next call of 
michael@0:  * ucol_prev or ucol_next, the first or last collation order will be returned. 
michael@0:  * If a change of direction is done without a ucol_reset, the result is 
michael@0:  * undefined.
michael@0:  * The result of a forward iterate (ucol_next) and reversed result of the  
michael@0:  * backward iterate (ucol_prev) on the same string are equivalent, if 
michael@0:  * collation orders with the value UCOL_IGNORABLE are ignored.
michael@0:  * Character based on the comparison level of the collator.  A collation order 
michael@0:  * consists of primary order, secondary order and tertiary order.  The data 
michael@0:  * type of the collation order is <strong>t_int32</strong>. 
michael@0:  *
michael@0:  * @see UCollator
michael@0:  */
michael@0: 
michael@0: /**
michael@0:  * Open the collation elements for a string.
michael@0:  *
michael@0:  * @param coll The collator containing the desired collation rules.
michael@0:  * @param text The text to iterate over.
michael@0:  * @param textLength The number of characters in text, or -1 if null-terminated
michael@0:  * @param status A pointer to an UErrorCode to receive any errors.
michael@0:  * @return a struct containing collation element information
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE UCollationElements* U_EXPORT2 
michael@0: ucol_openElements(const UCollator  *coll,
michael@0:                   const UChar      *text,
michael@0:                         int32_t    textLength,
michael@0:                         UErrorCode *status);
michael@0: 
michael@0: 
michael@0: /**
michael@0:  * get a hash code for a key... Not very useful!
michael@0:  * @param key    the given key.
michael@0:  * @param length the size of the key array.
michael@0:  * @return       the hash code.
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE int32_t U_EXPORT2 
michael@0: ucol_keyHashCode(const uint8_t* key, int32_t length);
michael@0: 
michael@0: /**
michael@0:  * Close a UCollationElements.
michael@0:  * Once closed, a UCollationElements may no longer be used.
michael@0:  * @param elems The UCollationElements to close.
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE void U_EXPORT2 
michael@0: ucol_closeElements(UCollationElements *elems);
michael@0: 
michael@0: /**
michael@0:  * Reset the collation elements to their initial state.
michael@0:  * This will move the 'cursor' to the beginning of the text.
michael@0:  * Property settings for collation will be reset to the current status.
michael@0:  * @param elems The UCollationElements to reset.
michael@0:  * @see ucol_next
michael@0:  * @see ucol_previous
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE void U_EXPORT2 
michael@0: ucol_reset(UCollationElements *elems);
michael@0: 
michael@0: #ifndef U_HIDE_INTERNAL_API
michael@0: /**
michael@0:  * Set the collation elements to use implicit ordering for Han
michael@0:  * even if they've been tailored. This will also force Hangul
michael@0:  * syllables to be ordered by decomposing them to their component
michael@0:  * Jamo.
michael@0:  *
michael@0:  * @param elems The UCollationElements containing the text.
michael@0:  * @param status A pointer to a UErrorCode to reveive any errors.
michael@0:  *
michael@0:  * @internal
michael@0:  */
michael@0: U_INTERNAL void U_EXPORT2
michael@0: ucol_forceHanImplicit(UCollationElements *elems, UErrorCode *status);
michael@0: #endif  /* U_HIDE_INTERNAL_API */
michael@0: 
michael@0: /**
michael@0:  * Get the ordering priority of the next collation element in the text.
michael@0:  * A single character may contain more than one collation element.
michael@0:  * @param elems The UCollationElements containing the text.
michael@0:  * @param status A pointer to an UErrorCode to receive any errors.
michael@0:  * @return The next collation elements ordering, otherwise returns NULLORDER 
michael@0:  *         if an error has occured or if the end of string has been reached
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE int32_t U_EXPORT2 
michael@0: ucol_next(UCollationElements *elems, UErrorCode *status);
michael@0: 
michael@0: /**
michael@0:  * Get the ordering priority of the previous collation element in the text.
michael@0:  * A single character may contain more than one collation element.
michael@0:  * Note that internally a stack is used to store buffered collation elements. 
michael@0:  * It is very rare that the stack will overflow, however if such a case is 
michael@0:  * encountered, the problem can be solved by increasing the size 
michael@0:  * UCOL_EXPAND_CE_BUFFER_SIZE in ucol_imp.h.
michael@0:  * @param elems The UCollationElements containing the text.
michael@0:  * @param status A pointer to an UErrorCode to receive any errors. Noteably 
michael@0:  *               a U_BUFFER_OVERFLOW_ERROR is returned if the internal stack
michael@0:  *               buffer has been exhausted.
michael@0:  * @return The previous collation elements ordering, otherwise returns 
michael@0:  *         NULLORDER if an error has occured or if the start of string has 
michael@0:  *         been reached.
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE int32_t U_EXPORT2 
michael@0: ucol_previous(UCollationElements *elems, UErrorCode *status);
michael@0: 
michael@0: #ifndef U_HIDE_INTERNAL_API
michael@0: /**
michael@0:  * Get the processed ordering priority of the next collation element in the text.
michael@0:  * A single character may contain more than one collation element.
michael@0:  *
michael@0:  * @param elems The UCollationElements containing the text.
michael@0:  * @param ixLow a pointer to an int32_t to receive the iterator index before fetching the CE.
michael@0:  * @param ixHigh a pointer to an int32_t to receive the iterator index after fetching the CE.
michael@0:  * @param status A pointer to an UErrorCode to receive any errors.
michael@0:  * @return The next collation elements ordering, otherwise returns UCOL_PROCESSED_NULLORDER 
michael@0:  *         if an error has occured or if the end of string has been reached
michael@0:  *
michael@0:  * @internal
michael@0:  */
michael@0: U_INTERNAL int64_t U_EXPORT2
michael@0: ucol_nextProcessed(UCollationElements *elems, int32_t *ixLow, int32_t *ixHigh, UErrorCode *status);
michael@0: 
michael@0: /**
michael@0:  * Get the processed ordering priority of the previous collation element in the text.
michael@0:  * A single character may contain more than one collation element.
michael@0:  * Note that internally a stack is used to store buffered collation elements. 
michael@0:  * It is very rare that the stack will overflow, however if such a case is 
michael@0:  * encountered, the problem can be solved by increasing the size 
michael@0:  * UCOL_EXPAND_CE_BUFFER_SIZE in ucol_imp.h.
michael@0:  *
michael@0:  * @param elems The UCollationElements containing the text.
michael@0:  * @param ixLow A pointer to an int32_t to receive the iterator index after fetching the CE
michael@0:  * @param ixHigh A pointer to an int32_t to receiver the iterator index before fetching the CE
michael@0:  * @param status A pointer to an UErrorCode to receive any errors. Noteably 
michael@0:  *               a U_BUFFER_OVERFLOW_ERROR is returned if the internal stack
michael@0:  *               buffer has been exhausted.
michael@0:  * @return The previous collation elements ordering, otherwise returns 
michael@0:  *         UCOL_PROCESSED_NULLORDER if an error has occured or if the start of
michael@0:  *         string has been reached.
michael@0:  *
michael@0:  * @internal
michael@0:  */
michael@0: U_INTERNAL int64_t U_EXPORT2
michael@0: ucol_previousProcessed(UCollationElements *elems, int32_t *ixLow, int32_t *ixHigh, UErrorCode *status);
michael@0: #endif  /* U_HIDE_INTERNAL_API */
michael@0: 
michael@0: /**
michael@0:  * Get the maximum length of any expansion sequences that end with the 
michael@0:  * specified comparison order.
michael@0:  * This is useful for .... ?
michael@0:  * @param elems The UCollationElements containing the text.
michael@0:  * @param order A collation order returned by previous or next.
michael@0:  * @return maximum size of the expansion sequences ending with the collation 
michael@0:  *         element or 1 if collation element does not occur at the end of any 
michael@0:  *         expansion sequence
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE int32_t U_EXPORT2 
michael@0: ucol_getMaxExpansion(const UCollationElements *elems, int32_t order);
michael@0: 
michael@0: /**
michael@0:  * Set the text containing the collation elements.
michael@0:  * Property settings for collation will remain the same.
michael@0:  * In order to reset the iterator to the current collation property settings,
michael@0:  * the API reset() has to be called.
michael@0:  * @param elems The UCollationElements to set.
michael@0:  * @param text The source text containing the collation elements.
michael@0:  * @param textLength The length of text, or -1 if null-terminated.
michael@0:  * @param status A pointer to an UErrorCode to receive any errors.
michael@0:  * @see ucol_getText
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE void U_EXPORT2 
michael@0: ucol_setText(      UCollationElements *elems, 
michael@0:              const UChar              *text,
michael@0:                    int32_t            textLength,
michael@0:                    UErrorCode         *status);
michael@0: 
michael@0: /**
michael@0:  * Get the offset of the current source character.
michael@0:  * This is an offset into the text of the character containing the current
michael@0:  * collation elements.
michael@0:  * @param elems The UCollationElements to query.
michael@0:  * @return The offset of the current source character.
michael@0:  * @see ucol_setOffset
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE int32_t U_EXPORT2 
michael@0: ucol_getOffset(const UCollationElements *elems);
michael@0: 
michael@0: /**
michael@0:  * Set the offset of the current source character.
michael@0:  * This is an offset into the text of the character to be processed.
michael@0:  * Property settings for collation will remain the same.
michael@0:  * In order to reset the iterator to the current collation property settings,
michael@0:  * the API reset() has to be called.
michael@0:  * @param elems The UCollationElements to set.
michael@0:  * @param offset The desired character offset.
michael@0:  * @param status A pointer to an UErrorCode to receive any errors.
michael@0:  * @see ucol_getOffset
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: U_STABLE void U_EXPORT2 
michael@0: ucol_setOffset(UCollationElements *elems,
michael@0:                int32_t        offset,
michael@0:                UErrorCode         *status);
michael@0: 
michael@0: /**
michael@0: * Get the primary order of a collation order.
michael@0: * @param order the collation order
michael@0: * @return the primary order of a collation order.
michael@0: * @stable ICU 2.6
michael@0: */
michael@0: U_STABLE int32_t U_EXPORT2
michael@0: ucol_primaryOrder (int32_t order); 
michael@0: 
michael@0: /**
michael@0: * Get the secondary order of a collation order.
michael@0: * @param order the collation order
michael@0: * @return the secondary order of a collation order.
michael@0: * @stable ICU 2.6
michael@0: */
michael@0: U_STABLE int32_t U_EXPORT2
michael@0: ucol_secondaryOrder (int32_t order); 
michael@0: 
michael@0: /**
michael@0: * Get the tertiary order of a collation order.
michael@0: * @param order the collation order
michael@0: * @return the tertiary order of a collation order.
michael@0: * @stable ICU 2.6
michael@0: */
michael@0: U_STABLE int32_t U_EXPORT2
michael@0: ucol_tertiaryOrder (int32_t order); 
michael@0: 
michael@0: #endif /* #if !UCONFIG_NO_COLLATION */
michael@0: 
michael@0: #endif