michael@0: /*
michael@0:  *******************************************************************************
michael@0:  *   Copyright (C) 1997-2009, International Business Machines
michael@0:  *   Corporation and others.  All Rights Reserved.
michael@0:  *******************************************************************************
michael@0:  *   Date        Name        Description
michael@0:  *   06/21/00    aliu        Creation.
michael@0:  *******************************************************************************
michael@0:  */
michael@0: 
michael@0: #include "unicode/utypes.h"
michael@0: 
michael@0: #if !UCONFIG_NO_TRANSLITERATION
michael@0: 
michael@0: #include "unicode/utrans.h"
michael@0: #include "unicode/putil.h"
michael@0: #include "unicode/rep.h"
michael@0: #include "unicode/translit.h"
michael@0: #include "unicode/unifilt.h"
michael@0: #include "unicode/uniset.h"
michael@0: #include "unicode/ustring.h"
michael@0: #include "unicode/uenum.h"
michael@0: #include "uenumimp.h"
michael@0: #include "cpputils.h"
michael@0: #include "rbt.h"
michael@0: 
michael@0: // Following macro is to be followed by <return value>';' or just ';'
michael@0: #define utrans_ENTRY(s) if ((s)==NULL || U_FAILURE(*(s))) return
michael@0: 
michael@0: /********************************************************************
michael@0:  * Replaceable-UReplaceableCallbacks glue
michael@0:  ********************************************************************/
michael@0: 
michael@0: /**
michael@0:  * Make a UReplaceable + UReplaceableCallbacks into a Replaceable object.
michael@0:  */
michael@0: U_NAMESPACE_BEGIN
michael@0: class ReplaceableGlue : public Replaceable {
michael@0: 
michael@0:     UReplaceable *rep;
michael@0:     UReplaceableCallbacks *func;
michael@0: 
michael@0: public:
michael@0: 
michael@0:     ReplaceableGlue(UReplaceable *replaceable,
michael@0:                     UReplaceableCallbacks *funcCallback);
michael@0: 
michael@0:     virtual ~ReplaceableGlue();
michael@0: 
michael@0:     virtual void handleReplaceBetween(int32_t start,
michael@0:                                       int32_t limit,
michael@0:                                       const UnicodeString& text);
michael@0: 
michael@0:     virtual void extractBetween(int32_t start,
michael@0:                                 int32_t limit,
michael@0:                                 UnicodeString& target) const;
michael@0: 
michael@0:     virtual void copy(int32_t start, int32_t limit, int32_t dest);
michael@0: 
michael@0:     // virtual Replaceable *clone() const { return NULL; } same as default
michael@0: 
michael@0:     /**
michael@0:      * ICU "poor man's RTTI", returns a UClassID for the actual class.
michael@0:      *
michael@0:      * @draft ICU 2.2
michael@0:      */
michael@0:     virtual UClassID getDynamicClassID() const;
michael@0: 
michael@0:     /**
michael@0:      * ICU "poor man's RTTI", returns a UClassID for this class.
michael@0:      *
michael@0:      * @draft ICU 2.2
michael@0:      */
michael@0:     static UClassID U_EXPORT2 getStaticClassID();
michael@0: 
michael@0: protected:
michael@0: 
michael@0:     virtual int32_t getLength() const;
michael@0: 
michael@0:     virtual UChar getCharAt(int32_t offset) const;
michael@0: 
michael@0:     virtual UChar32 getChar32At(int32_t offset) const;
michael@0: };
michael@0: 
michael@0: UOBJECT_DEFINE_RTTI_IMPLEMENTATION(ReplaceableGlue)
michael@0: 
michael@0: ReplaceableGlue::ReplaceableGlue(UReplaceable *replaceable,
michael@0:                                  UReplaceableCallbacks *funcCallback)
michael@0:   : Replaceable()
michael@0: {
michael@0:     this->rep = replaceable;
michael@0:     this->func = funcCallback;
michael@0: }
michael@0: 
michael@0: ReplaceableGlue::~ReplaceableGlue() {}
michael@0: 
michael@0: int32_t ReplaceableGlue::getLength() const {
michael@0:     return (*func->length)(rep);
michael@0: }
michael@0: 
michael@0: UChar ReplaceableGlue::getCharAt(int32_t offset) const {
michael@0:     return (*func->charAt)(rep, offset);
michael@0: }
michael@0: 
michael@0: UChar32 ReplaceableGlue::getChar32At(int32_t offset) const {
michael@0:     return (*func->char32At)(rep, offset);
michael@0: }
michael@0: 
michael@0: void ReplaceableGlue::handleReplaceBetween(int32_t start,
michael@0:                           int32_t limit,
michael@0:                           const UnicodeString& text) {
michael@0:     (*func->replace)(rep, start, limit, text.getBuffer(), text.length());
michael@0: }
michael@0: 
michael@0: void ReplaceableGlue::extractBetween(int32_t start,
michael@0:                                      int32_t limit,
michael@0:                                      UnicodeString& target) const {
michael@0:     (*func->extract)(rep, start, limit, target.getBuffer(limit-start));
michael@0:     target.releaseBuffer(limit-start);
michael@0: }
michael@0: 
michael@0: void ReplaceableGlue::copy(int32_t start, int32_t limit, int32_t dest) {
michael@0:     (*func->copy)(rep, start, limit, dest);
michael@0: }
michael@0: U_NAMESPACE_END
michael@0: /********************************************************************
michael@0:  * General API
michael@0:  ********************************************************************/
michael@0: U_NAMESPACE_USE
michael@0: 
michael@0: U_CAPI UTransliterator* U_EXPORT2
michael@0: utrans_openU(const UChar *id,
michael@0:              int32_t idLength,
michael@0:              UTransDirection dir,
michael@0:              const UChar *rules,
michael@0:              int32_t rulesLength,
michael@0:              UParseError *parseError,
michael@0:              UErrorCode *status) {
michael@0:     if(status==NULL || U_FAILURE(*status)) {
michael@0:         return NULL;
michael@0:     }
michael@0:     if (id == NULL) {
michael@0:         *status = U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         return NULL;
michael@0:     }
michael@0:     UParseError temp;
michael@0:     
michael@0:     if(parseError == NULL){
michael@0:         parseError = &temp;
michael@0:     }
michael@0:     
michael@0:     UnicodeString ID(idLength<0, id, idLength); // r-o alias
michael@0: 
michael@0:     if(rules==NULL){
michael@0: 
michael@0:         Transliterator *trans = NULL;
michael@0: 
michael@0:         trans = Transliterator::createInstance(ID, dir, *parseError, *status);
michael@0:         
michael@0:         if(U_FAILURE(*status)){
michael@0:             return NULL;
michael@0:         }
michael@0:         return (UTransliterator*) trans;
michael@0:     }else{
michael@0:         UnicodeString ruleStr(rulesLength < 0,
michael@0:                               rules,
michael@0:                               rulesLength); // r-o alias
michael@0: 
michael@0:         Transliterator *trans = NULL;
michael@0:         trans = Transliterator::createFromRules(ID, ruleStr, dir, *parseError, *status); 
michael@0:         if(U_FAILURE(*status)) { 
michael@0:             return NULL;
michael@0:         }
michael@0: 
michael@0:         return (UTransliterator*) trans;
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CAPI UTransliterator* U_EXPORT2
michael@0: utrans_open(const char* id,
michael@0:             UTransDirection dir,
michael@0:             const UChar* rules,         /* may be Null */
michael@0:             int32_t rulesLength,        /* -1 if null-terminated */ 
michael@0:             UParseError* parseError,    /* may be Null */
michael@0:             UErrorCode* status) {
michael@0:     UnicodeString ID(id, -1, US_INV); // use invariant converter
michael@0:     return utrans_openU(ID.getBuffer(), ID.length(), dir,
michael@0:                         rules, rulesLength,
michael@0:                         parseError, status);
michael@0: }
michael@0: 
michael@0: U_CAPI UTransliterator* U_EXPORT2
michael@0: utrans_openInverse(const UTransliterator* trans,
michael@0:                    UErrorCode* status) {
michael@0: 
michael@0:     utrans_ENTRY(status) NULL;
michael@0: 
michael@0:     UTransliterator* result =
michael@0:         (UTransliterator*) ((Transliterator*) trans)->createInverse(*status);
michael@0: 
michael@0:     return result;
michael@0: }
michael@0: 
michael@0: U_CAPI UTransliterator* U_EXPORT2
michael@0: utrans_clone(const UTransliterator* trans,
michael@0:              UErrorCode* status) {
michael@0: 
michael@0:     utrans_ENTRY(status) NULL;
michael@0: 
michael@0:     if (trans == NULL) {
michael@0:         *status = U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         return NULL;
michael@0:     }
michael@0: 
michael@0:     Transliterator *t = ((Transliterator*) trans)->clone();
michael@0:     if (t == NULL) {
michael@0:         *status = U_MEMORY_ALLOCATION_ERROR;
michael@0:     }
michael@0:     return (UTransliterator*) t;
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: utrans_close(UTransliterator* trans) {
michael@0:     delete (Transliterator*) trans;
michael@0: }
michael@0: 
michael@0: U_CAPI const UChar * U_EXPORT2
michael@0: utrans_getUnicodeID(const UTransliterator *trans,
michael@0:                     int32_t *resultLength) {
michael@0:     // Transliterator keeps its ID NUL-terminated
michael@0:     const UnicodeString &ID=((Transliterator*) trans)->getID();
michael@0:     if(resultLength!=NULL) {
michael@0:         *resultLength=ID.length();
michael@0:     }
michael@0:     return ID.getBuffer();
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: utrans_getID(const UTransliterator* trans,
michael@0:              char* buf,
michael@0:              int32_t bufCapacity) {
michael@0:     return ((Transliterator*) trans)->getID().extract(0, 0x7fffffff, buf, bufCapacity, US_INV);
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: utrans_register(UTransliterator* adoptedTrans,
michael@0:                 UErrorCode* status) {
michael@0:     utrans_ENTRY(status);
michael@0:     // status currently ignored; may remove later
michael@0:     Transliterator::registerInstance((Transliterator*) adoptedTrans);
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: utrans_unregisterID(const UChar* id, int32_t idLength) {
michael@0:     UnicodeString ID(idLength<0, id, idLength); // r-o alias
michael@0:     Transliterator::unregister(ID);
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: utrans_unregister(const char* id) {
michael@0:     UnicodeString ID(id, -1, US_INV); // use invariant converter
michael@0:     Transliterator::unregister(ID);
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: utrans_setFilter(UTransliterator* trans,
michael@0:                  const UChar* filterPattern,
michael@0:                  int32_t filterPatternLen,
michael@0:                  UErrorCode* status) {
michael@0: 
michael@0:     utrans_ENTRY(status);
michael@0:     UnicodeFilter* filter = NULL;
michael@0:     if (filterPattern != NULL && *filterPattern != 0) {
michael@0:         // Create read only alias of filterPattern:
michael@0:         UnicodeString pat(filterPatternLen < 0, filterPattern, filterPatternLen);
michael@0:         filter = new UnicodeSet(pat, *status);
michael@0:         /* test for NULL */
michael@0:         if (filter == NULL) {
michael@0:             *status = U_MEMORY_ALLOCATION_ERROR;
michael@0:             return;
michael@0:         }
michael@0:         if (U_FAILURE(*status)) {
michael@0:             delete filter;
michael@0:             filter = NULL;
michael@0:         }
michael@0:     }
michael@0:     ((Transliterator*) trans)->adoptFilter(filter);
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: utrans_countAvailableIDs(void) {
michael@0:     return Transliterator::countAvailableIDs();
michael@0: }
michael@0: 
michael@0: U_CAPI int32_t U_EXPORT2
michael@0: utrans_getAvailableID(int32_t index,
michael@0:                       char* buf, // may be NULL
michael@0:                       int32_t bufCapacity) {
michael@0:     return Transliterator::getAvailableID(index).extract(0, 0x7fffffff, buf, bufCapacity, US_INV);
michael@0: }
michael@0: 
michael@0: /* Transliterator UEnumeration ---------------------------------------------- */
michael@0: 
michael@0: typedef struct UTransEnumeration {
michael@0:     UEnumeration uenum;
michael@0:     int32_t index, count;
michael@0: } UTransEnumeration;
michael@0: 
michael@0: U_CDECL_BEGIN
michael@0: static int32_t U_CALLCONV
michael@0: utrans_enum_count(UEnumeration *uenum, UErrorCode *pErrorCode) {
michael@0:     if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
michael@0:         return 0;
michael@0:     }
michael@0:     return ((UTransEnumeration *)uenum)->count;
michael@0: }
michael@0: 
michael@0: static const UChar* U_CALLCONV
michael@0: utrans_enum_unext(UEnumeration *uenum,
michael@0:                   int32_t* resultLength,
michael@0:                   UErrorCode *pErrorCode) {
michael@0:     if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
michael@0:         return 0;
michael@0:     }
michael@0: 
michael@0:     UTransEnumeration *ute=(UTransEnumeration *)uenum;
michael@0:     int32_t index=ute->index;
michael@0:     if(index<ute->count) {
michael@0:         const UnicodeString &ID=Transliterator::getAvailableID(index);
michael@0:         ute->index=index+1;
michael@0:         if(resultLength!=NULL) {
michael@0:             *resultLength=ID.length();
michael@0:         }
michael@0:         // Transliterator keeps its ID NUL-terminated
michael@0:         return ID.getBuffer();
michael@0:     }
michael@0: 
michael@0:     if(resultLength!=NULL) {
michael@0:         *resultLength=0;
michael@0:     }
michael@0:     return NULL;
michael@0: }
michael@0: 
michael@0: static void U_CALLCONV
michael@0: utrans_enum_reset(UEnumeration *uenum, UErrorCode *pErrorCode) {
michael@0:     if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
michael@0:         return;
michael@0:     }
michael@0: 
michael@0:     UTransEnumeration *ute=(UTransEnumeration *)uenum;
michael@0:     ute->index=0;
michael@0:     ute->count=Transliterator::countAvailableIDs();
michael@0: }
michael@0: 
michael@0: static void U_CALLCONV
michael@0: utrans_enum_close(UEnumeration *uenum) {
michael@0:     uprv_free(uenum);
michael@0: }
michael@0: U_CDECL_END
michael@0: 
michael@0: static const UEnumeration utransEnumeration={
michael@0:     NULL,
michael@0:     NULL,
michael@0:     utrans_enum_close,
michael@0:     utrans_enum_count,
michael@0:     utrans_enum_unext,
michael@0:     uenum_nextDefault,
michael@0:     utrans_enum_reset
michael@0: };
michael@0: 
michael@0: U_CAPI UEnumeration * U_EXPORT2
michael@0: utrans_openIDs(UErrorCode *pErrorCode) {
michael@0:     UTransEnumeration *ute;
michael@0: 
michael@0:     if(pErrorCode==NULL || U_FAILURE(*pErrorCode)) {
michael@0:         return NULL;
michael@0:     }
michael@0: 
michael@0:     ute=(UTransEnumeration *)uprv_malloc(sizeof(UTransEnumeration));
michael@0:     if(ute==NULL) {
michael@0:         *pErrorCode=U_MEMORY_ALLOCATION_ERROR;
michael@0:         return NULL;
michael@0:     }
michael@0: 
michael@0:     ute->uenum=utransEnumeration;
michael@0:     ute->index=0;
michael@0:     ute->count=Transliterator::countAvailableIDs();
michael@0:     return (UEnumeration *)ute;
michael@0: }
michael@0: 
michael@0: /********************************************************************
michael@0:  * Transliteration API
michael@0:  ********************************************************************/
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: utrans_trans(const UTransliterator* trans,
michael@0:              UReplaceable* rep,
michael@0:              UReplaceableCallbacks* repFunc,
michael@0:              int32_t start,
michael@0:              int32_t* limit,
michael@0:              UErrorCode* status) {
michael@0: 
michael@0:     utrans_ENTRY(status);
michael@0: 
michael@0:     if (trans == 0 || rep == 0 || repFunc == 0 || limit == 0) {
michael@0:         *status = U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         return;
michael@0:     }
michael@0: 
michael@0:     ReplaceableGlue r(rep, repFunc);
michael@0: 
michael@0:     *limit = ((Transliterator*) trans)->transliterate(r, start, *limit);
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: utrans_transIncremental(const UTransliterator* trans,
michael@0:                         UReplaceable* rep,
michael@0:                         UReplaceableCallbacks* repFunc,
michael@0:                         UTransPosition* pos,
michael@0:                         UErrorCode* status) {
michael@0: 
michael@0:     utrans_ENTRY(status);
michael@0: 
michael@0:     if (trans == 0 || rep == 0 || repFunc == 0 || pos == 0) {
michael@0:         *status = U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         return;
michael@0:     }
michael@0: 
michael@0:     ReplaceableGlue r(rep, repFunc);
michael@0: 
michael@0:     ((Transliterator*) trans)->transliterate(r, *pos, *status);
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: utrans_transUChars(const UTransliterator* trans,
michael@0:                    UChar* text,
michael@0:                    int32_t* textLength,
michael@0:                    int32_t textCapacity,
michael@0:                    int32_t start,
michael@0:                    int32_t* limit,
michael@0:                    UErrorCode* status) {
michael@0: 
michael@0:     utrans_ENTRY(status);
michael@0: 
michael@0:     if (trans == 0 || text == 0 || limit == 0) {
michael@0:         *status = U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         return;
michael@0:     }
michael@0:  
michael@0:     int32_t textLen = (textLength == NULL || *textLength < 0)
michael@0:         ? u_strlen(text) : *textLength;
michael@0:     // writeable alias: for this ct, len CANNOT be -1 (why?)
michael@0:     UnicodeString str(text, textLen, textCapacity);
michael@0: 
michael@0:     *limit = ((Transliterator*) trans)->transliterate(str, start, *limit);
michael@0: 
michael@0:     // Copy the string buffer back to text (only if necessary)
michael@0:     // and fill in *neededCapacity (if neededCapacity != NULL).
michael@0:     textLen = str.extract(text, textCapacity, *status);
michael@0:     if(textLength != NULL) {
michael@0:         *textLength = textLen;
michael@0:     }
michael@0: }
michael@0: 
michael@0: U_CAPI void U_EXPORT2
michael@0: utrans_transIncrementalUChars(const UTransliterator* trans,
michael@0:                               UChar* text,
michael@0:                               int32_t* textLength,
michael@0:                               int32_t textCapacity,
michael@0:                               UTransPosition* pos,
michael@0:                               UErrorCode* status) {
michael@0: 
michael@0:     utrans_ENTRY(status);
michael@0: 
michael@0:     if (trans == 0 || text == 0 || pos == 0) {
michael@0:         *status = U_ILLEGAL_ARGUMENT_ERROR;
michael@0:         return;
michael@0:     }
michael@0: 
michael@0:     int32_t textLen = (textLength == NULL || *textLength < 0)
michael@0:         ? u_strlen(text) : *textLength;
michael@0:     // writeable alias: for this ct, len CANNOT be -1 (why?)
michael@0:     UnicodeString str(text, textLen, textCapacity);
michael@0: 
michael@0:     ((Transliterator*) trans)->transliterate(str, *pos, *status);
michael@0: 
michael@0:     // Copy the string buffer back to text (only if necessary)
michael@0:     // and fill in *neededCapacity (if neededCapacity != NULL).
michael@0:     textLen = str.extract(text, textCapacity, *status);
michael@0:     if(textLength != NULL) {
michael@0:         *textLength = textLen;
michael@0:     }
michael@0: }
michael@0: 
michael@0: #endif /* #if !UCONFIG_NO_TRANSLITERATION */