michael@0: /*
michael@0: ********************************************************************************
michael@0: *   Copyright (C) 1996-2013, International Business Machines
michael@0: *   Corporation and others.  All Rights Reserved.
michael@0: ********************************************************************************
michael@0: */
michael@0: 
michael@0: #include "unicode/utypes.h"
michael@0: 
michael@0: #if !UCONFIG_NO_BREAK_ITERATION
michael@0: 
michael@0: #include "unicode/ubrk.h"
michael@0: 
michael@0: #include "unicode/brkiter.h"
michael@0: #include "unicode/uloc.h"
michael@0: #include "unicode/ustring.h"
michael@0: #include "unicode/uchriter.h"
michael@0: #include "unicode/rbbi.h"
michael@0: #include "rbbirb.h"
michael@0: #include "uassert.h"
michael@0: 
michael@0: U_NAMESPACE_USE
michael@0: 
michael@0: //------------------------------------------------------------------------------
michael@0: //
michael@0: //    ubrk_open      Create a canned type of break iterator based on type (word, line, etc.)
michael@0: //                   and locale.
michael@0: //
michael@0: //------------------------------------------------------------------------------
michael@0: U_CAPI UBreakIterator* U_EXPORT2
michael@0: ubrk_open(UBreakIteratorType type,
michael@0:       const char *locale,
michael@0:       const UChar *text,
michael@0:       int32_t textLength,
michael@0:       UErrorCode *status)
michael@0: {
michael@0: 
michael@0:   if(U_FAILURE(*status)) return 0;
michael@0: 
michael@0:   BreakIterator *result = 0;
michael@0: 
michael@0:   switch(type) {
michael@0: 
michael@0:   case UBRK_CHARACTER:
michael@0:     result = BreakIterator::createCharacterInstance(Locale(locale), *status);
michael@0:     break;
michael@0: 
michael@0:   case UBRK_WORD:
michael@0:     result = BreakIterator::createWordInstance(Locale(locale), *status);
michael@0:     break;
michael@0: 
michael@0:   case UBRK_LINE:
michael@0:     result = BreakIterator::createLineInstance(Locale(locale), *status);
michael@0:     break;
michael@0: 
michael@0:   case UBRK_SENTENCE:
michael@0:     result = BreakIterator::createSentenceInstance(Locale(locale), *status);
michael@0:     break;
michael@0: 
michael@0:   case UBRK_TITLE:
michael@0:     result = BreakIterator::createTitleInstance(Locale(locale), *status);
michael@0:     break;
michael@0: 
michael@0:   default:
michael@0:     *status = U_ILLEGAL_ARGUMENT_ERROR;
michael@0:   }
michael@0: 
michael@0:   // check for allocation error
michael@0:   if (U_FAILURE(*status)) {
michael@0:      return 0;
michael@0:   }
michael@0:   if(result == 0) {
michael@0:     *status = U_MEMORY_ALLOCATION_ERROR;
michael@0:     return 0;
michael@0:   }
michael@0: 
michael@0: 
michael@0:   UBreakIterator *uBI = (UBreakIterator *)result;
michael@0:   if (text != NULL) {
michael@0:       ubrk_setText(uBI, text, textLength, status);
michael@0:   }
michael@0:   return uBI;
michael@0: }
michael@0: 
michael@0: 
michael@0: 
michael@0: //------------------------------------------------------------------------------
michael@0: //
michael@0: //   ubrk_openRules      open a break iterator from a set of break rules.
michael@0: //                       Invokes the rule builder.
michael@0: //
michael@0: //------------------------------------------------------------------------------
michael@0: U_CAPI UBreakIterator* U_EXPORT2
michael@0: ubrk_openRules(  const UChar        *rules,
michael@0:                        int32_t       rulesLength,
michael@0:                  const UChar        *text,
michael@0:                        int32_t       textLength,
michael@0:                        UParseError  *parseErr,
michael@0:                        UErrorCode   *status)  {
michael@0: 
michael@0:     if (status == NULL || U_FAILURE(*status)){
michael@0:         return 0;
michael@0:     }
michael@0: 
michael@0:     BreakIterator *result = 0;
michael@0:     UnicodeString ruleString(rules, rulesLength);
michael@0:     result = RBBIRuleBuilder::createRuleBasedBreakIterator(ruleString, parseErr, *status);
michael@0:     if(U_FAILURE(*status)) {
michael@0:         return 0;
michael@0:     }
michael@0: 
michael@0:     UBreakIterator *uBI = (UBreakIterator *)result;
michael@0:     if (text != NULL) {
michael@0:         ubrk_setText(uBI, text, textLength, status);
michael@0:     }
michael@0:     return uBI;
michael@0: }
michael@0: 
michael@0: 
michael@0: 
michael@0: 
michael@0: 
michael@0: U_CAPI UBreakIterator * U_EXPORT2
michael@0: ubrk_safeClone(
michael@0:           const UBreakIterator *bi,
michael@0:           void * /*stackBuffer*/,
michael@0:           int32_t *pBufferSize,
michael@0:           UErrorCode *status)
michael@0: {
michael@0:     if (status == NULL || U_FAILURE(*status)){
michael@0:         return NULL;
michael@0:     }
michael@0:     if (bi == NULL) {
michael@0:        *status = U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         return NULL;
michael@0:     }
michael@0:     if (pBufferSize != NULL) {
michael@0:         int32_t inputSize = *pBufferSize;
michael@0:         *pBufferSize = 1;
michael@0:         if (inputSize == 0) {
michael@0:             return NULL;  // preflighting for deprecated functionality
michael@0:         }
michael@0:     }
michael@0:     BreakIterator *newBI = ((BreakIterator *)bi)->clone();
michael@0:     if (newBI == NULL) {
michael@0:         *status = U_MEMORY_ALLOCATION_ERROR;
michael@0:     } else {
michael@0:         *status = U_SAFECLONE_ALLOCATED_WARNING;
michael@0:     }
michael@0:     return (UBreakIterator *)newBI;
michael@0: }
michael@0: 
michael@0: 
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: ubrk_close(UBreakIterator *bi)
michael@0: {
michael@0:     delete (BreakIterator *)bi;
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: ubrk_setText(UBreakIterator* bi,
michael@0:              const UChar*    text,
michael@0:              int32_t         textLength,
michael@0:              UErrorCode*     status)
michael@0: {
michael@0:     BreakIterator *brit = (BreakIterator *)bi;
michael@0:     UText  ut = UTEXT_INITIALIZER;
michael@0:     utext_openUChars(&ut, text, textLength, status);
michael@0:     brit->setText(&ut, *status);
michael@0:     // A stack allocated UText wrapping a UChar * string
michael@0:     //   can be dumped without explicitly closing it.
michael@0: }
michael@0: 
michael@0: 
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: ubrk_setUText(UBreakIterator *bi,
michael@0:              UText          *text,
michael@0:              UErrorCode     *status)
michael@0: {
michael@0:     RuleBasedBreakIterator *brit = (RuleBasedBreakIterator *)bi;
michael@0:     brit->RuleBasedBreakIterator::setText(text, *status);
michael@0: }
michael@0: 
michael@0: 
michael@0: 
michael@0: 
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: ubrk_current(const UBreakIterator *bi)
michael@0: {
michael@0: 
michael@0:   return ((RuleBasedBreakIterator*)bi)->RuleBasedBreakIterator::current();
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: ubrk_next(UBreakIterator *bi)
michael@0: {
michael@0: 
michael@0:   return ((RuleBasedBreakIterator*)bi)->RuleBasedBreakIterator::next();
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: ubrk_previous(UBreakIterator *bi)
michael@0: {
michael@0: 
michael@0:   return ((RuleBasedBreakIterator*)bi)->RuleBasedBreakIterator::previous();
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: ubrk_first(UBreakIterator *bi)
michael@0: {
michael@0: 
michael@0:   return ((RuleBasedBreakIterator*)bi)->RuleBasedBreakIterator::first();
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: ubrk_last(UBreakIterator *bi)
michael@0: {
michael@0: 
michael@0:   return ((RuleBasedBreakIterator*)bi)->RuleBasedBreakIterator::last();
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: ubrk_preceding(UBreakIterator *bi,
michael@0:            int32_t offset)
michael@0: {
michael@0: 
michael@0:   return ((RuleBasedBreakIterator*)bi)->RuleBasedBreakIterator::preceding(offset);
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: ubrk_following(UBreakIterator *bi,
michael@0:            int32_t offset)
michael@0: {
michael@0: 
michael@0:   return ((RuleBasedBreakIterator*)bi)->RuleBasedBreakIterator::following(offset);
michael@0: }
michael@0: 
michael@0: U_CAPI const char* U_EXPORT2
michael@0: ubrk_getAvailable(int32_t index)
michael@0: {
michael@0: 
michael@0:   return uloc_getAvailable(index);
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: ubrk_countAvailable()
michael@0: {
michael@0: 
michael@0:   return uloc_countAvailable();
michael@0: }
michael@0: 
michael@0: 
michael@0: U_CAPI  UBool U_EXPORT2
michael@0: ubrk_isBoundary(UBreakIterator *bi, int32_t offset)
michael@0: {
michael@0:     return ((RuleBasedBreakIterator *)bi)->RuleBasedBreakIterator::isBoundary(offset);
michael@0: }
michael@0: 
michael@0: 
michael@0: U_CAPI  int32_t U_EXPORT2
michael@0: ubrk_getRuleStatus(UBreakIterator *bi)
michael@0: {
michael@0:     return ((RuleBasedBreakIterator *)bi)->RuleBasedBreakIterator::getRuleStatus();
michael@0: }
michael@0: 
michael@0: U_CAPI  int32_t U_EXPORT2
michael@0: ubrk_getRuleStatusVec(UBreakIterator *bi, int32_t *fillInVec, int32_t capacity, UErrorCode *status)
michael@0: {
michael@0:     return ((RuleBasedBreakIterator *)bi)->RuleBasedBreakIterator::getRuleStatusVec(fillInVec, capacity, *status);
michael@0: }
michael@0: 
michael@0: 
michael@0: U_CAPI const char* U_EXPORT2
michael@0: ubrk_getLocaleByType(const UBreakIterator *bi,
michael@0:                      ULocDataLocaleType type,
michael@0:                      UErrorCode* status)
michael@0: {
michael@0:     if (bi == NULL) {
michael@0:         if (U_SUCCESS(*status)) {
michael@0:             *status = U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         }
michael@0:         return NULL;
michael@0:     }
michael@0:     return ((BreakIterator*)bi)->getLocaleID(type, *status);
michael@0: }
michael@0: 
michael@0: 
michael@0: void ubrk_refreshUText(UBreakIterator *bi,
michael@0:                        UText          *text,
michael@0:                        UErrorCode     *status)
michael@0: {
michael@0:     BreakIterator *bii = reinterpret_cast<BreakIterator *>(bi);
michael@0:     bii->refreshInputText(text, *status);
michael@0: }
michael@0: 
michael@0: 
michael@0: 
michael@0: #endif /* #if !UCONFIG_NO_BREAK_ITERATION */