C++ UnicodeString函数代码示例

void TCopyParamType::DoGetInfoStr(
  const UnicodeString & Separator, intptr_t Options,
  UnicodeString & Result, bool & SomeAttrIncluded,
  const UnicodeString & Link, UnicodeString & ScriptArgs, bool & NoScriptArgs, /*TAssemblyLanguage Language, UnicodeString & AssemblyCode,*/
  bool & NoCodeProperties) const
{
  TCopyParamType Defaults;

  bool SomeAttrExcluded = false;
  NoScriptArgs = false;
  NoCodeProperties = false;
  SomeAttrIncluded = false;
  #define ADD(STR, EXCEPT) \
    if (FLAGCLEAR(Options, EXCEPT)) \
    { \
      AddToList(Result, (STR), Separator); \
      SomeAttrIncluded = true; \
    } \
    else \
    { \
      SomeAttrExcluded = true; \
    }

  bool AsciiFileMaskDiffers = (GetTransferMode() == tmAutomatic) && !(GetAsciiFileMask() == Defaults.GetAsciiFileMask());
  bool TransferModeDiffers = ((GetTransferMode() != Defaults.GetTransferMode()) || AsciiFileMaskDiffers);

  if (FLAGCLEAR(Options, cpaIncludeMaskOnly | cpaNoTransferMode))
  {
    // Adding Transfer type unconditionally
    bool FormatMask;
    int Ident;
    switch (GetTransferMode())
    {
      case tmBinary:
        FormatMask = false;
        Ident = 2;
        break;
      case tmAscii:
        FormatMask = false;
        Ident = 3;
        break;
      case tmAutomatic:
      default:
        FormatMask = !(GetAsciiFileMask() == Defaults.GetAsciiFileMask());
        Ident = FormatMask ? 4 : 5;
        break;
    }
    UnicodeString S = FORMAT(LoadStrPart(COPY_INFO_TRANSFER_TYPE2, 1).c_str(),
      LoadStrPart(COPY_INFO_TRANSFER_TYPE2, Ident).c_str());
    if (FormatMask)
    {
      S = FORMAT(S.c_str(), GetAsciiFileMask().GetMasks().c_str());
    }
    AddToList(Result, S, Separator);

    if (TransferModeDiffers)
    {
      ADD("", cpaIncludeMaskOnly | cpaNoTransferMode);

      /*ScriptArgs += RtfSwitchValue(TRANSFER_SWITCH, Link, TransferModeNames[TransferMode]);
      const wchar_t * TransferModeMembers[] = { L"Binary", L"Ascii", L"Automatic" };
      AssemblyCode += AssemblyProperty(
        Language, TransferOptionsClassName, L"TransferMode", L"TransferMode", TransferModeMembers[TransferMode], false);
      if (AsciiFileMaskDiffers)
      {
        NoScriptArgs = true;
        NoCodeProperties = true;
      }*/
    }
  }
  else
  {
    if (TransferModeDiffers)
    {
      SomeAttrExcluded = true;
      NoScriptArgs = true;
      NoCodeProperties = true;
    }
  }

  if (GetFileNameCase() != Defaults.GetFileNameCase())
  {
    ADD(FORMAT(LoadStrPart(COPY_INFO_FILENAME, 1).c_str(),
      LoadStrPart(COPY_INFO_FILENAME, GetFileNameCase() + 2).c_str()),
      cpaIncludeMaskOnly);

    NoScriptArgs = true;
    NoCodeProperties = true;
  }

  if ((GetInvalidCharsReplacement() == NoReplacement) !=
        (Defaults.GetInvalidCharsReplacement() == NoReplacement))
  {
    DebugAssert(GetInvalidCharsReplacement() == NoReplacement);
    if (GetInvalidCharsReplacement() == NoReplacement)
    {
      ADD(LoadStr(COPY_INFO_DONT_REPLACE_INV_CHARS).c_str(), cpaIncludeMaskOnly);
    }

    NoScriptArgs = true;
//.........这里部分代码省略.........

UnicodeString&
DefaultLocaleDisplayNames::keyDisplayName(const char* key,
                                          UnicodeString& result) const {
  return result = UnicodeString(key, -1, US_INV);
}

UnicodeString
IndexedString::getPrimaryIndexAtCharacter(int32_t Position) const
{
  auto const Found = lookupPrimaryIndexAtCharacter(Position);
  return Found != Needles.end() ? Found->first : UnicodeString();
}

void CharIterTest::TestConstructionAndEqualityUChariter() {
    U_STRING_DECL(testText, "Now is the time for all good men to come to the aid of their country.", 69);
    U_STRING_DECL(testText2, "Don't bother using this string.", 31);

    U_STRING_INIT(testText, "Now is the time for all good men to come to the aid of their country.", 69);
    U_STRING_INIT(testText2, "Don't bother using this string.", 31);

    UnicodeString result, result4, result5;

    UCharCharacterIterator* test1 = new UCharCharacterIterator(testText, u_strlen(testText));
    UCharCharacterIterator* test2 = new UCharCharacterIterator(testText, u_strlen(testText), 5);
    UCharCharacterIterator* test3 = new UCharCharacterIterator(testText, u_strlen(testText), 2, 20, 5);
    UCharCharacterIterator* test4 = new UCharCharacterIterator(testText2, u_strlen(testText2));
    UCharCharacterIterator* test5 = (UCharCharacterIterator*)test1->clone();
    UCharCharacterIterator* test6 = new UCharCharacterIterator(*test1);

    // j785: length=-1 will use u_strlen()
    UCharCharacterIterator* test7a = new UCharCharacterIterator(testText, -1);
    UCharCharacterIterator* test7b = new UCharCharacterIterator(testText, -1);
    UCharCharacterIterator* test7c = new UCharCharacterIterator(testText, -1, 2, 20, 5);

    // Bad parameters.
    UCharCharacterIterator* test8a = new UCharCharacterIterator(testText, -1, -1, 20, 5);
    UCharCharacterIterator* test8b = new UCharCharacterIterator(testText, -1, 2, 100, 5);
    UCharCharacterIterator* test8c = new UCharCharacterIterator(testText, -1, 2, 20, 100);

    if (test8a->startIndex() < 0)
        errln("Construction failed: startIndex is negative");
    if (test8b->endIndex() != u_strlen(testText))
        errln("Construction failed: endIndex is different from the text length");
    if (test8c->getIndex() < test8c->startIndex() || test8c->endIndex() < test8c->getIndex())
        errln("Construction failed: index is invalid");

    if (*test1 == *test2 || *test1 == *test3 || *test1 == *test4 )
        errln("Construction or operator== failed: Unequal objects compared equal");
    if (*test1 != *test5 )
        errln("clone() or equals() failed: Two clones tested unequal");

    if (*test6 != *test1 )
        errln("copy construction or equals() failed: Two copies tested unequal");

    if (test1->hashCode() == test2->hashCode() || test1->hashCode() == test3->hashCode()
                    || test1->hashCode() == test4->hashCode())
        errln("hashCode() failed:  different objects have same hash code");

    if (test1->hashCode() != test5->hashCode())
        errln("hashCode() failed:  identical objects have different hash codes");
     
    test7a->getText(result);
    test7b->getText(result4);
    test7c->getText(result5);

    if(result != UnicodeString(testText) || result4 != result || result5 != result)
        errln("error in construction");
    
    test1->getText(result);
    test4->getText(result4);
    test5->getText(result5); 
    if(result != result5 || result == result4)
        errln("getText() failed");
    test5->setText(testText2, u_strlen(testText2));
    test5->getText(result5);
    if(result == result5 || result4 != result5)
        errln("setText() or getText() failed");
    test5->setText(testText, u_strlen(testText));
    test5->getText(result5);
    if(result != result5 || result == result4)
        errln("setText() or getText() round-trip failed"); 


    test1->setIndex(5);
    if (*test1 != *test2 || *test1 == *test5)
        errln("setIndex() failed");
    test8b->setIndex32(5);
    if (test8b->getIndex()!=5)
        errln("setIndex32() failed");

    *test1 = *test3;
    if (*test1 != *test3 || *test1 == *test5)
        errln("operator= failed");

    delete test1;
    delete test2;
    delete test3;
    delete test4;
    delete test5;
    delete test6;
    delete test7a;
    delete test7b;
    delete test7c;
    delete test8a;
    delete test8b;
    delete test8c;
}

void CharIterTest::TestUCharIterator() {
    // test string of length 8
    UnicodeString s=UnicodeString("a \\U00010001b\\U0010fffdz", "").unescape();
    const char *const moves=
        "0+++++++++" // 10 moves per line
        "----0-----"
        ">>|>>>>>>>"
        "<<|<<<<<<<"
        "22+>8>-8+2";

    StringCharacterIterator sci(s), compareCI(s);

    UCharIterator sIter, cIter, rIter;

    uiter_setString(&sIter, s.getBuffer(), s.length());
    uiter_setCharacterIterator(&cIter, &sci);
    uiter_setReplaceable(&rIter, &s);

    TestUCharIterator(&sIter, compareCI, moves, "uiter_setString");
    compareCI.setIndex(0);
    TestUCharIterator(&cIter, compareCI, moves, "uiter_setCharacterIterator");
    compareCI.setIndex(0);
    TestUCharIterator(&rIter, compareCI, moves, "uiter_setReplaceable");

    // test move & getIndex some more
    sIter.start=2;
    sIter.index=3;
    sIter.limit=5;
    if( sIter.getIndex(&sIter, UITER_ZERO)!=0 ||
        sIter.getIndex(&sIter, UITER_START)!=2 ||
        sIter.getIndex(&sIter, UITER_CURRENT)!=3 ||
        sIter.getIndex(&sIter, UITER_LIMIT)!=5 ||
        sIter.getIndex(&sIter, UITER_LENGTH)!=s.length()
    ) {
        errln("error: UCharIterator(string).getIndex returns wrong index");
    }

    if( sIter.move(&sIter, 4, UITER_ZERO)!=4 ||
        sIter.move(&sIter, 1, UITER_START)!=3 ||
        sIter.move(&sIter, 3, UITER_CURRENT)!=5 ||
        sIter.move(&sIter, -1, UITER_LIMIT)!=4 ||
        sIter.move(&sIter, -5, UITER_LENGTH)!=3 ||
        sIter.move(&sIter, 0, UITER_CURRENT)!=sIter.getIndex(&sIter, UITER_CURRENT) ||
        sIter.getIndex(&sIter, UITER_CURRENT)!=3
    ) {
        errln("error: UCharIterator(string).move sets/returns wrong index");
    }

    sci=StringCharacterIterator(s, 2, 5, 3);
    uiter_setCharacterIterator(&cIter, &sci);
    if( cIter.getIndex(&cIter, UITER_ZERO)!=0 ||
        cIter.getIndex(&cIter, UITER_START)!=2 ||
        cIter.getIndex(&cIter, UITER_CURRENT)!=3 ||
        cIter.getIndex(&cIter, UITER_LIMIT)!=5 ||
        cIter.getIndex(&cIter, UITER_LENGTH)!=s.length()
    ) {
        errln("error: UCharIterator(character iterator).getIndex returns wrong index");
    }

    if( cIter.move(&cIter, 4, UITER_ZERO)!=4 ||
        cIter.move(&cIter, 1, UITER_START)!=3 ||
        cIter.move(&cIter, 3, UITER_CURRENT)!=5 ||
        cIter.move(&cIter, -1, UITER_LIMIT)!=4 ||
        cIter.move(&cIter, -5, UITER_LENGTH)!=3 ||
        cIter.move(&cIter, 0, UITER_CURRENT)!=cIter.getIndex(&cIter, UITER_CURRENT) ||
        cIter.getIndex(&cIter, UITER_CURRENT)!=3
    ) {
        errln("error: UCharIterator(character iterator).move sets/returns wrong index");
    }


    if(cIter.getIndex(&cIter, (enum UCharIteratorOrigin)-1) != -1)
    {
        errln("error: UCharIterator(char iter).getIndex did not return error value");
    }

    if(cIter.move(&cIter, 0, (enum UCharIteratorOrigin)-1) != -1)
    {
        errln("error: UCharIterator(char iter).move did not return error value");
    }


    if(rIter.getIndex(&rIter, (enum UCharIteratorOrigin)-1) != -1)
    {
        errln("error: UCharIterator(repl iter).getIndex did not return error value");
    }

    if(rIter.move(&rIter, 0, (enum UCharIteratorOrigin)-1) != -1)
    {
        errln("error: UCharIterator(repl iter).move did not return error value");
    }


    if(sIter.getIndex(&sIter, (enum UCharIteratorOrigin)-1) != -1)
    {
        errln("error: UCharIterator(string iter).getIndex did not return error value");
    }

    if(sIter.move(&sIter, 0, (enum UCharIteratorOrigin)-1) != -1)
    {
//.........这里部分代码省略.........

void DateFormatRoundTripTest::test(const Locale& loc)
{
    UnicodeString temp;
#if !INFINITE
    logln("Locale: " + loc.getDisplayName(temp));
#endif

    // Total possibilities = 24
    //  4 date
    //  4 time
    //  16 date-time
    UBool TEST_TABLE [24];//= new boolean[24];
    int32_t i = 0;
    for(i = 0; i < 24; ++i)
        TEST_TABLE[i] = TRUE;

    // If we have some sparseness, implement it here.  Sparseness decreases
    // test time by eliminating some tests, up to 23.
    for(i = 0; i < SPARSENESS; ) {
        int random = (int)(randFraction() * 24);
        if (random >= 0 && random < 24 && TEST_TABLE[i]) {
            TEST_TABLE[i] = FALSE;
            ++i;
        }
    }

    int32_t itable = 0;
    int32_t style = 0;
    for(style = DateFormat::FULL; style <= DateFormat::SHORT; ++style) {
        if(TEST_TABLE[itable++]) {
            logln("Testing style " + UnicodeString(styleName((DateFormat::EStyle)style)));
            DateFormat *df = DateFormat::createDateInstance((DateFormat::EStyle)style, loc);
            if(df == NULL) {
                errln(UnicodeString("Could not DF::createDateInstance ") + UnicodeString(styleName((DateFormat::EStyle)style)) +      " Locale: " + loc.getDisplayName(temp));
            } else {
                test(df, loc);
                delete df;
            }
        }
    }

    for(style = DateFormat::FULL; style <= DateFormat::SHORT; ++style) {
        if (TEST_TABLE[itable++]) {
            logln("Testing style " + UnicodeString(styleName((DateFormat::EStyle)style)));
            DateFormat *df = DateFormat::createTimeInstance((DateFormat::EStyle)style, loc);
            if(df == NULL) {
                errln(UnicodeString("Could not DF::createTimeInstance ") + UnicodeString(styleName((DateFormat::EStyle)style)) + " Locale: " + loc.getDisplayName(temp));
            } else {
                test(df, loc, TRUE);
                delete df;
            }
        }
    }

    for(int32_t dstyle = DateFormat::FULL; dstyle <= DateFormat::SHORT; ++dstyle) {
        for(int32_t tstyle = DateFormat::FULL; tstyle <= DateFormat::SHORT; ++tstyle) {
            if(TEST_TABLE[itable++]) {
                logln("Testing dstyle" + UnicodeString(styleName((DateFormat::EStyle)dstyle)) + ", tstyle" + UnicodeString(styleName((DateFormat::EStyle)tstyle)) );
                DateFormat *df = DateFormat::createDateTimeInstance((DateFormat::EStyle)dstyle, (DateFormat::EStyle)tstyle, loc);
                if(df == NULL) {
                    dataerrln(UnicodeString("Could not DF::createDateTimeInstance ") + UnicodeString(styleName((DateFormat::EStyle)dstyle)) + ", tstyle" + UnicodeString(styleName((DateFormat::EStyle)tstyle))    + "Locale: " + loc.getDisplayName(temp));
                } else {
                    test(df, loc);
                    delete df;
                }
            }
        }
    }
}

void DateFormatRoundTripTest::test(DateFormat *fmt, const Locale &origLocale, UBool timeOnly)
{
    UnicodeString pat;
    if(fmt->getDynamicClassID() != SimpleDateFormat::getStaticClassID()) {
        errln("DateFormat wasn't a SimpleDateFormat");
        return;
    }

    UBool isGregorian = FALSE;
    UErrorCode minStatus = U_ZERO_ERROR;
    if(fmt->getCalendar() == NULL) {
        errln((UnicodeString)"DateFormatRoundTripTest::test, DateFormat getCalendar() returns null for " + origLocale.getName());
        return;
    }
    UDate minDate = CalendarTest::minDateOfCalendar(*fmt->getCalendar(), isGregorian, minStatus);
    if(U_FAILURE(minStatus)) {
        errln((UnicodeString)"Failure getting min date for " + origLocale.getName());
        return;
    }
    //logln(UnicodeString("Min date is ") + fullFormat(minDate)  + " for " + origLocale.getName());

    pat = ((SimpleDateFormat*)fmt)->toPattern(pat);

    // NOTE TO MAINTAINER
    // This indexOf check into the pattern needs to be refined to ignore
    // quoted characters.  Currently, this isn't a problem with the locale
    // patterns we have, but it may be a problem later.

    UBool hasEra = (pat.indexOf(UnicodeString("G")) != -1);
    UBool hasZoneDisplayName = (pat.indexOf(UnicodeString("z")) != -1) || (pat.indexOf(UnicodeString("v")) != -1)
                               || (pat.indexOf(UnicodeString("V")) != -1);

    // Because patterns contain incomplete data representing the Date,
    // we must be careful of how we do the roundtrip.  We start with
    // a randomly generated Date because they're easier to generate.
    // From this we get a string.  The string is our real starting point,
    // because this string should parse the same way all the time.  Note
    // that it will not necessarily parse back to the original date because
    // of incompleteness in patterns.  For example, a time-only pattern won't
    // parse back to the same date.

    //try {
    for(int i = 0; i < TRIALS; ++i) {
        UDate *d                = new UDate    [DEPTH];
        UnicodeString *s    = new UnicodeString[DEPTH];

        if(isGregorian == TRUE) {
            d[0] = generateDate();
        } else {
            d[0] = generateDate(minDate);
        }

        UErrorCode status = U_ZERO_ERROR;

        // We go through this loop until we achieve a match or until
        // the maximum loop count is reached.  We record the points at
        // which the date and the string starts to match.  Once matching
        // starts, it should continue.
        int loop;
        int dmatch = 0; // d[dmatch].getTime() == d[dmatch-1].getTime()
        int smatch = 0; // s[smatch].equals(s[smatch-1])
        for(loop = 0; loop < DEPTH; ++loop) {
            if (loop > 0)  {
                d[loop] = fmt->parse(s[loop-1], status);
                failure(status, "fmt->parse", s[loop-1]+" in locale: " + origLocale.getName() + " with pattern: " + pat);
                status = U_ZERO_ERROR; /* any error would have been reported */
            }

            s[loop] = fmt->format(d[loop], s[loop]);

            // For displaying which date is being tested
            //logln(s[loop] + " = " + fullFormat(d[loop]));

            if(s[loop].length() == 0) {
                errln("FAIL: fmt->format gave 0-length string in " + pat + " with number " + d[loop] + " in locale " + origLocale.getName());
            }

            if(loop > 0) {
                if(smatch == 0) {
                    UBool match = s[loop] == s[loop-1];
                    if(smatch == 0) {
                        if(match)
                            smatch = loop;
                    }
                    else if( ! match)
                        errln("FAIL: String mismatch after match");
                }

                if(dmatch == 0) {
                    // {sfb} watch out here, this might not work
                    UBool match = d[loop]/*.getTime()*/ == d[loop-1]/*.getTime()*/;
                    if(dmatch == 0) {
                        if(match)
                            dmatch = loop;
                    }
                    else if( ! match)
                        errln("FAIL: Date mismatch after match");
                }

                if(smatch != 0 && dmatch != 0)
//.........这里部分代码省略.........

/*write indentation for formatting*/
static void write_tabs(FileStream* os){
    int i=0;
    for(;i<=tabCount;i++){
        write_utf8_file(os,UnicodeString("    "));
    }
}

void
ICUServiceTest::testRBF()
{
    // resource bundle factory.
    UErrorCode status = U_ZERO_ERROR;
    TestStringService service;
    service.registerFactory(new ICUResourceBundleFactory(), status);

    // list all of the resources 
    {
        UErrorCode status = U_ZERO_ERROR;
        UVector ids(uprv_deleteUObject, uhash_compareUnicodeString, 0, status);
        service.getVisibleIDs(ids, status);
        logln("all visible ids:");
        for (int i = 0; i < ids.size(); ++i) {
            const UnicodeString* id = (const UnicodeString*)ids[i];
            logln(*id);
        }
    }

    // get all the display names of these resources
    // this should be fast since the display names were cached.
    {
        UErrorCode status = U_ZERO_ERROR;
        UVector names(status);
        service.getDisplayNames(names, Locale::getGermany(), status);
        logln("service display names for de_DE");
        for (int i = 0; i < names.size(); ++i) {
            const StringPair* pair = (const StringPair*)names[i];
            logln("  " + pair->displayName + " --> " + pair->id);
        }
    }

    service.registerFactory(new CalifornioLanguageFactory(), status);

    // get all the display names of these resources
    {
        logln("californio language factory:");
        const char* idNames[] = {
            CalifornioLanguageFactory::californio, 
            CalifornioLanguageFactory::valley, 
            CalifornioLanguageFactory::surfer, 
            CalifornioLanguageFactory::geek,
        };
        int32_t count = sizeof(idNames)/sizeof(idNames[0]);

        for (int i = 0; i < count; ++i) {
            logln(UnicodeString("\n  --- ") + idNames[i] + " ---");
            {
                UErrorCode status = U_ZERO_ERROR;
                UVector names(status);
                service.getDisplayNames(names, idNames[i], status);
                for (int i = 0; i < names.size(); ++i) {
                    const StringPair* pair = (const StringPair*)names[i];
                    logln("  " + pair->displayName + " --> " + pair->id);
                }
            }
        }
    }
    CalifornioLanguageFactory::cleanup();
}

/*
 * @test @(#)bug4117335.java    1.1 3/5/98
 *
 * @bug 4117335
 */
void
DateFormatMiscTests::test4117335()
{
    //UnicodeString bc = "\u7d00\u5143\u524d";
    UChar bcC [] = {
        0x7D00,
        0x5143,
        0x524D
    };
    UnicodeString bc(bcC, 3, 3);

    //UnicodeString ad = "\u897f\u66a6";
    UChar adC [] = {
        0x897F,
        0x66A6
    };
    UnicodeString ad(adC, 2, 2);
    
    //UnicodeString jstLong = "\u65e5\u672c\u6a19\u6e96\u6642";
    UChar jstLongC [] = {
        0x65e5,
        0x672c,
        0x6a19,
        0x6e96,
        0x6642
    };
    UnicodeString jstLong(jstLongC, 5, 5);

    UnicodeString jstShort = "JST";

    
    UErrorCode status = U_ZERO_ERROR;
    DateFormatSymbols *symbols = new DateFormatSymbols(Locale::getJapan(), status);
    if(U_FAILURE(status)) {
      errln("Failure creating DateFormatSymbols, %s", u_errorName(status));
      delete symbols;
      return;
    }
    failure(status, "new DateFormatSymbols");
    int32_t eraCount = 0;
    const UnicodeString *eras = symbols->getEras(eraCount);
    
    logln(UnicodeString("BC = ") + eras[0]);
    if (eras[0] != bc) {
        errln("*** Should have been " + bc);
        //throw new Exception("Error in BC");
    }

    logln(UnicodeString("AD = ") + eras[1]);
    if (eras[1] != ad) {
        errln("*** Should have been " + ad);
        //throw new Exception("Error in AD");
    }

    int32_t rowCount, colCount;
    const UnicodeString **zones = symbols->getZoneStrings(rowCount, colCount);
    logln(UnicodeString("Long zone name = ") + zones[24][1]);
    if (zones[24][1] != jstLong) {
        errln("*** Should have been " + jstLong);
        //throw new Exception("Error in long TZ name");
    }
    logln(UnicodeString("Short zone name = ") + zones[24][2]);
    if (zones[24][2] != jstShort) {
        errln("*** Should have been " + jstShort);
        //throw new Exception("Error in short TZ name");
    }
    logln(UnicodeString("Long zone name = ") + zones[24][3]);
    if (zones[24][3] != jstLong) {
        errln("*** Should have been " + jstLong);
        //throw new Exception("Error in long TZ name");
    }
    logln(UnicodeString("SHORT zone name = ") + zones[24][4]);
    if (zones[24][4] != jstShort) {
        errln("*** Should have been " + jstShort);
        //throw new Exception("Error in short TZ name");
    }
    delete symbols;

}

/**
 * See if the decomposition of cp2 is at segment starting at segmentPos 
 * (with canonical rearrangment!)
 * If so, take the remainder, and return the equivalents 
 */
Hashtable *CanonicalIterator::extract(Hashtable *fillinResult, UChar32 comp, const UChar *segment, int32_t segLen, int32_t segmentPos, UErrorCode &status) {
//Hashtable *CanonicalIterator::extract(UChar32 comp, const UnicodeString &segment, int32_t segLen, int32_t segmentPos, UErrorCode &status) {
    //if (PROGRESS) printf(" extract: %s, ", UToS(Tr(UnicodeString(comp))));
    //if (PROGRESS) printf("%s, %i\n", UToS(Tr(segment)), segmentPos);

    if (U_FAILURE(status)) {
        return NULL;
    }

    UnicodeString temp(comp);
    int32_t inputLen=temp.length();
    UnicodeString decompString;
    nfd.normalize(temp, decompString, status);
    if (U_FAILURE(status)) {
        return NULL;
    }
    if (decompString.isBogus()) {
        status = U_MEMORY_ALLOCATION_ERROR;
        return NULL;
    }
    const UChar *decomp=decompString.getBuffer();
    int32_t decompLen=decompString.length();

    // See if it matches the start of segment (at segmentPos)
    UBool ok = FALSE;
    UChar32 cp;
    int32_t decompPos = 0;
    UChar32 decompCp;
    U16_NEXT(decomp, decompPos, decompLen, decompCp);

    int32_t i = segmentPos;
    while(i < segLen) {
        U16_NEXT(segment, i, segLen, cp);

        if (cp == decompCp) { // if equal, eat another cp from decomp

            //if (PROGRESS) printf("  matches: %s\n", UToS(Tr(UnicodeString(cp))));

            if (decompPos == decompLen) { // done, have all decomp characters!
                temp.append(segment+i, segLen-i);
                ok = TRUE;
                break;
            }
            U16_NEXT(decomp, decompPos, decompLen, decompCp);
        } else {
            //if (PROGRESS) printf("  buffer: %s\n", UToS(Tr(UnicodeString(cp))));

            // brute force approach
            temp.append(cp);

            /* TODO: optimize
            // since we know that the classes are monotonically increasing, after zero
            // e.g. 0 5 7 9 0 3
            // we can do an optimization
            // there are only a few cases that work: zero, less, same, greater
            // if both classes are the same, we fail
            // if the decomp class < the segment class, we fail

            segClass = getClass(cp);
            if (decompClass <= segClass) return null;
            */
        }
    }
    if (!ok)
        return NULL; // we failed, characters left over

    //if (PROGRESS) printf("Matches\n");

    if (inputLen == temp.length()) {
        fillinResult->put(UnicodeString(), new UnicodeString(), status);
        return fillinResult; // succeed, but no remainder
    }

    // brute force approach
    // check to make sure result is canonically equivalent
    UnicodeString trial;
    nfd.normalize(temp, trial, status);
    if(U_FAILURE(status) || trial.compare(segment+segmentPos, segLen - segmentPos) != 0) {
        return NULL;
    }

    return getEquivalents2(fillinResult, temp.getBuffer()+inputLen, temp.length()-inputLen, status);
}

/*
 * @bug 4097450
 */
void
DateFormatMiscTests::test4097450()
{
    //
    // Date parse requiring 4 digit year.
    //
    UnicodeString  dstring [] = {
        UnicodeString("97"),
        UnicodeString("1997"),  
        UnicodeString("97"),
        UnicodeString("1997"),
        UnicodeString("01"),
        UnicodeString("2001"),  
        UnicodeString("01"),
        UnicodeString("2001"),  
        UnicodeString("1"),
        UnicodeString("1"),
        UnicodeString("11"),  
        UnicodeString("11"),
        UnicodeString("111"), 
        UnicodeString("111")
    };
    
    UnicodeString dformat [] = {
        UnicodeString("yy"),  
        UnicodeString("yy"),
        UnicodeString("yyyy"),
        UnicodeString("yyyy"),
        UnicodeString("yy"),  
        UnicodeString("yy"),
        UnicodeString("yyyy"),
        UnicodeString("yyyy"),
        UnicodeString("yy"),
        UnicodeString("yyyy"),
        UnicodeString("yy"),
        UnicodeString("yyyy"), 
        UnicodeString("yy"),
        UnicodeString("yyyy")
    };
    
/*    UBool dresult [] = {
        TRUE, 
        FALSE, 
        FALSE,  
        TRUE,
        TRUE, 
        FALSE, 
        FALSE,  
        TRUE,
        FALSE,
        FALSE,
        TRUE, 
        FALSE,
        FALSE, 
        FALSE
    };*/

    UErrorCode status = U_ZERO_ERROR;
    SimpleDateFormat *formatter;
    SimpleDateFormat *resultFormatter = new SimpleDateFormat((UnicodeString)"yyyy", status);
    failure(status, "new SimpleDateFormat");

    logln("Format\tSource\tResult");
    logln("-------\t-------\t-------");
    for (int i = 0; i < 14/*dstring.length*/; i++)
    {
        log(dformat[i] + "\t" + dstring[i] + "\t");
        formatter = new SimpleDateFormat(dformat[i], status);
        failure(status, "new SimpleDateFormat");
        //try {
        UnicodeString str;
        FieldPosition pos(FieldPosition::DONT_CARE);
        logln(resultFormatter->format(formatter->parse(dstring[i], status), str, pos));
        failure(status, "resultFormatter->format");
            //if ( !dresult[i] ) System.out.print("   <-- error!");
        /*}
        catch (ParseException exception) {
            //if ( dresult[i] ) System.out.print("   <-- error!");
            System.out.print("exception --> " + exception);
        }*/
        delete formatter;
        logln();
    }

    delete resultFormatter;
}

void DataDrivenCalendarTest::testConvert(int32_t n,
        const CalendarFieldsSet &fromSet, Calendar *fromCalendar,
        const CalendarFieldsSet &toSet, Calendar *toCalendar, UBool forward) {
    UErrorCode status = U_ZERO_ERROR;
    UnicodeString thisString = (UnicodeString)"#"+n+" "+(forward ? "forward"
            : "reverse")+" "+fromCalendar->getType()+"->"+toCalendar->getType();

    fromCalendar->clear();

    fromSet.setOnCalendar(fromCalendar, status);
    if (U_FAILURE(status)) {
        errln("FAIL: Failed to set on Source calendar: %s", u_errorName(status));
        return;
    }

    CalendarFieldsSet diffSet;

    diffSet.clear();
    // Is the calendar sane at the first?
    if (!fromSet.matches(fromCalendar, diffSet, status)) {
        UnicodeString diffs = diffSet.diffFrom(fromSet, status);
        errln((UnicodeString)"FAIL: "+thisString
                +", SOURCE calendar was not set: Differences: "+ diffs
                +"', status: "+ u_errorName(status));
    } else if (U_FAILURE(status)) {
        errln("FAIL: "+thisString+" SOURCE calendar Failed to match: "
                +u_errorName(status));
    } else {
        logln("PASS: "+thisString+" SOURCE calendar match.");
    }

    //logln("Set Source calendar: " + from);

    UDate fromTime = fromCalendar->getTime(status);
    if (U_FAILURE(status)) {
        errln("FAIL: Failed to get Source time: %s", u_errorName(status));
        return;
    }

    diffSet.clear();
    // Is the calendar sane after being set?
    if (!fromSet.matches(fromCalendar, diffSet, status)) {
        UnicodeString diffs = diffSet.diffFrom(fromSet, status);
        errln((UnicodeString)"FAIL: "+thisString
                +", SET SOURCE calendar was not set: Differences: "+ diffs
                +"', status: "+ u_errorName(status));
    } else if (U_FAILURE(status)) {
        errln("FAIL: "+thisString+" SET SOURCE calendar Failed to match: "
                +u_errorName(status));
    } else {
        logln("PASS: "+thisString+" SET SOURCE calendar match.");
    }

    toCalendar->clear();
    toCalendar->setTime(fromTime, status);
    if (U_FAILURE(status)) {
        errln("FAIL: Failed to set Target time: %s", u_errorName(status));
        return;
    }

    diffSet.clear();
    if (!toSet.matches(toCalendar, diffSet, status)) {
        UnicodeString diffs = diffSet.diffFrom(toSet, status);
        errln((UnicodeString)"FAIL: "+thisString+", Differences: "+ diffs
                +"', status: "+ u_errorName(status));
        SimpleDateFormat fmt(UnicodeString("EEE MMM dd yyyy G"), status);
        UnicodeString fromString;
        fmt.format(fromTime, fromString);
        logln("Source Time: "+fromString+", Source Calendar: "
                +fromCalendar->getType());
    } else if (U_FAILURE(status)) {
        errln("FAIL: "+thisString+" Failed to match: "+u_errorName(status));
    } else {
        logln("PASS: "+thisString+" match.");
    }
}

seec::Maybe<IndexedString>
IndexedString::from(UnicodeString const &String)
{
  if (String.isBogus())
    return seec::Maybe<IndexedString>();
  
  UnicodeString const NeedleStart("@[");
  UnicodeString const NeedleEscape("@[[");
  UnicodeString const NeedleEnd("]");
  
  UnicodeString CleanedString; // String with index indicators removed.
  std::multimap<UnicodeString, Needle> Needles;
  
  std::vector<std::pair<UnicodeString, int32_t>> IndexStack;
  
  int32_t SearchFrom = 0; // Current offset in String.
  int32_t FoundStart; // Position of matched index indicator.
  
  while ((FoundStart = String.indexOf(NeedleStart, SearchFrom)) != -1) {
    // Copy all the literal string data.
    CleanedString.append(String, SearchFrom, FoundStart - SearchFrom);
    
    // Check if this is an escape sequence.
    if (String.compare(FoundStart, NeedleEscape.length(), NeedleEscape) == 0) {
      CleanedString.append(NeedleStart);
      SearchFrom = FoundStart + NeedleEscape.length();
      continue;
    }
    
    // Find the end of this sequence.
    int32_t FoundEnd = String.indexOf(NeedleEnd, SearchFrom);
    if (FoundEnd == -1)
      return seec::Maybe<IndexedString>();
    
    if (FoundEnd == FoundStart + NeedleStart.length()) {
      // This is a closing sequence.
      if (IndexStack.size() == 0)
        return seec::Maybe<IndexedString>();
      
      // Pop the starting details of the last-opened sequence.
      auto const Start = IndexStack.back();
      IndexStack.pop_back();
      
      // Store the needle for this sequence.
      Needles.insert(std::make_pair(Start.first,
                                    Needle(Start.second,
                                           CleanedString.countChar32())));
    }
    else {
      // This is an opening sequence.
      int32_t const NameStart = FoundStart + NeedleStart.length();
      int32_t const NameLength = FoundEnd - NameStart;
      
      IndexStack.emplace_back(UnicodeString(String, NameStart, NameLength),
                              CleanedString.countChar32());
    }
    
    SearchFrom = FoundEnd + NeedleEnd.length();
  }
  
  // Copy all remaining literal data.
  CleanedString.append(String, SearchFrom, String.length() - SearchFrom);
  
  return IndexedString(std::move(CleanedString), std::move(Needles));
}

void DataDrivenCalendarTest::testConvert(TestData *testData,
        const DataMap *settings, UBool forward) {
    UErrorCode status = U_ZERO_ERROR;
    Calendar *toCalendar= NULL;
    const DataMap *currentCase= NULL;
    char toCalLoc[256] = "";
    char fromCalLoc[256] = "";
    // build to calendar
    UnicodeString testSetting = settings->getString("ToCalendar", status);
    if (U_SUCCESS(status)) {
        testSetting.extract(0, testSetting.length(), toCalLoc, (const char*)0);
        toCalendar = Calendar::createInstance(toCalLoc, status);
        if (U_FAILURE(status)) {
            dataerrln(UnicodeString("Unable to instantiate ToCalendar for ")+testSetting);
            return;
        }
    }

    CalendarFieldsSet fromSet, toSet, diffSet;
    SimpleDateFormat fmt(UnicodeString("EEE MMM dd yyyy / YYYY'-W'ww-ee"),
            status);
    if (U_FAILURE(status)) {
        errcheckln(status, "FAIL: Couldn't create SimpleDateFormat: %s",
                u_errorName(status));
        return;
    }
    // Start the processing
    int n = 0;
    while (testData->nextCase(currentCase, status)) {
        ++n;
        Calendar *fromCalendar= NULL;
        UnicodeString locale = currentCase->getString("locale", status);
        if (U_SUCCESS(status)) {
            locale.extract(0, locale.length(), fromCalLoc, (const char*)0); // default codepage.  Invariant codepage doesn't have '@'!
            fromCalendar = Calendar::createInstance(fromCalLoc, status);
            if (U_FAILURE(status)) {
                errln("Unable to instantiate fromCalendar for "+locale);
                return;
            }
        } else {
            errln("No 'locale' line.");
            continue;
        }

        fromSet.clear();
        toSet.clear();

        UnicodeString from = currentCase->getString("from", status);
        if (U_FAILURE(status)) {
            errln("No 'from' line.");
            continue;
        }
        fromSet.parseFrom(from, status);
        if (U_FAILURE(status)) {
            errln("Failed to parse 'from' parameter: "+from);
            continue;
        }
        UnicodeString to = currentCase->getString("to", status);
        if (U_FAILURE(status)) {
            errln("No 'to' line.");
            continue;
        }
        toSet.parseFrom(to, &fromSet, status);
        if (U_FAILURE(status)) {
            errln("Failed to parse 'to' parameter: "+to);
            continue;
        }

        // now, do it.
        if (forward) {
            logln((UnicodeString)"#"+n+" "+locale+"/"+from+" >>> "+toCalLoc+"/"
                    +to);
            testConvert(n, fromSet, fromCalendar, toSet, toCalendar, forward);
        } else {
            logln((UnicodeString)"#"+n+" "+locale+"/"+from+" <<< "+toCalLoc+"/"
                    +to);
            testConvert(n, toSet, toCalendar, fromSet, fromCalendar, forward);
        }

        delete fromCalendar;
    }
    delete toCalendar;
}