C++ UnicodeString::extractBetween方法代码示例

void RBBIAPITest::doTest(UnicodeString& testString, int32_t start, int32_t gotoffset, int32_t expectedOffset, const char* expectedString){
    UnicodeString selected;
    UnicodeString expected=CharsToUnicodeString(expectedString);

    if(gotoffset != expectedOffset)
         errln((UnicodeString)"ERROR:****returned #" + gotoffset + (UnicodeString)" instead of #" + expectedOffset);
    if(start <= gotoffset){
        testString.extractBetween(start, gotoffset, selected);
    }
    else{
        testString.extractBetween(gotoffset, start, selected);
    }
    if(selected.compare(expected) != 0)
         errln(prettify((UnicodeString)"ERROR:****selected \"" + selected + "\" instead of \"" + expected + "\""));
    else
        logln(prettify("****selected \"" + selected + "\""));
}

/**
 * Parse an ID into pieces.  Take IDs of the form T, T/V, S-T,
 * S-T/V, or S/V-T.  If the source is missing, return a source of
 * ANY.
 * @param id the id string, in any of several forms
 * @return an array of 4 strings: source, target, variant, and
 * isSourcePresent.  If the source is not present, ANY will be
 * given as the source, and isSourcePresent will be NULL.  Otherwise
 * isSourcePresent will be non-NULL.  The target may be empty if the
 * id is not well-formed.  The variant may be empty.
 */
void TransliteratorIDParser::IDtoSTV(const UnicodeString & id,
                                     UnicodeString & source,
                                     UnicodeString & target,
                                     UnicodeString & variant,
                                     UBool & isSourcePresent)
{
	source = ANY;
	target.truncate(0);
	variant.truncate(0);

	int32_t sep = id.indexOf(TARGET_SEP);
	int32_t var = id.indexOf(VARIANT_SEP);
	if (var < 0)
	{
		var = id.length();
	}
	isSourcePresent = FALSE;

	if (sep < 0)
	{
		// Form: T/V or T (or /V)
		id.extractBetween(0, var, target);
		id.extractBetween(var, id.length(), variant);
	}
	else if (sep < var)
	{
		// Form: S-T/V or S-T (or -T/V or -T)
		if (sep > 0)
		{
			id.extractBetween(0, sep, source);
			isSourcePresent = TRUE;
		}
		id.extractBetween(++sep, var, target);
		id.extractBetween(var, id.length(), variant);
	}
	else
	{
		// Form: (S/V-T or /V-T)
		if (var > 0)
		{
			id.extractBetween(0, var, source);
			isSourcePresent = TRUE;
		}
		id.extractBetween(var, sep++, variant);
		id.extractBetween(sep, id.length(), target);
	}

	if (variant.length() > 0)
	{
		variant.remove(0, 1);
	}
}

StringMatcher::StringMatcher(const UnicodeString& theString,
                             int32_t start,
                             int32_t limit,
                             int32_t segmentNum,
                             const TransliterationRuleData& theData) :
    data(&theData),
    segmentNumber(segmentNum),
    matchStart(-1),
    matchLimit(-1)
{
    theString.extractBetween(start, limit, pattern);
}

void TransliteratorAPITest::displayOutput(const UnicodeString& got, const UnicodeString& expected, UnicodeString& log, UTransPosition& index){
 // Show the start index '{' and the cursor '|'
    UnicodeString a, b, c, d, e;
    got.extractBetween(0, index.contextStart, a);
    got.extractBetween(index.contextStart, index.start, b);
    got.extractBetween(index.start, index.limit, c);
    got.extractBetween(index.limit, index.contextLimit, d);
    got.extractBetween(index.contextLimit, got.length(), e);
    log.append(a).
        append((UChar)0x7b/*{*/).
        append(b).
        append((UChar)0x7c/*|*/).
        append(c).
        append((UChar)0x7c).
        append(d).
        append((UChar)0x7d/*}*/).
        append(e);
    if (got == expected) 
        logln("OK:" + prettify(log));
    else 
        errln("FAIL: " + prettify(log)  + ", expected " + prettify(expected));
}

//
// RBBISymbolTable::parseReference   This function from the abstract symbol table interface
//                                   looks for a $variable name in the source text.
//                                   It does not look it up, only scans for it.
//                                   It is used by the UnicodeSet parser.
//
//                                   This implementation is lifted pretty much verbatim
//                                   from the rules based transliterator implementation.
//                                   I didn't see an obvious way of sharing it.
//
UnicodeString   RBBISymbolTable::parseReference(const UnicodeString& text,
                                                ParsePosition& pos, int32_t limit) const
{
    int32_t start = pos.getIndex();
    int32_t i = start;
    UnicodeString result;
    while (i < limit) {
        UChar c = text.charAt(i);
        if ((i==start && !u_isIDStart(c)) || !u_isIDPart(c)) {
            break;
        }
        ++i;
    }
    if (i == start) { // No valid name chars
        return result; // Indicate failure with empty string
    }
    pos.setIndex(i);
    text.extractBetween(start, i, result);
    return result;
}

 void extractBetween(int32_t start, int32_t limit, UnicodeString& result) const {
     chars.extractBetween(start, limit, result);
 }

/**
 * @bug 4052408
 */
void DateFormatRegressionTest::Test4052408(void) 
{

    DateFormat *fmt = DateFormat::createDateTimeInstance(DateFormat::SHORT,
                                                DateFormat::SHORT, Locale::getUS());
    if (fmt == NULL) {
        dataerrln("Error calling DateFormat::createDateTimeInstance");
        return;
    }

    UDate dt = date(97, UCAL_MAY, 3, 8, 55);
    UnicodeString str;
    str = fmt->format(dt, str);
    logln(str);
    
    if(str != "5/3/97 8:55 AM")
        errln("Fail: Test broken; Want 5/3/97 8:55 AM Got " + str);   
    
    UnicodeString expected[] = {
        (UnicodeString) "", //"ERA_FIELD",
        (UnicodeString) "97", //"YEAR_FIELD",
        (UnicodeString) "5", //"MONTH_FIELD",
        (UnicodeString) "3", //"DATE_FIELD",
        (UnicodeString) "", //"HOUR_OF_DAY1_FIELD",
        (UnicodeString) "", //"HOUR_OF_DAY0_FIELD",
        (UnicodeString) "55", //"MINUTE_FIELD",
        (UnicodeString) "", //"SECOND_FIELD",
        (UnicodeString) "", //"MILLISECOND_FIELD",
        (UnicodeString) "", //"DAY_OF_WEEK_FIELD",
        (UnicodeString) "", //"DAY_OF_YEAR_FIELD",
        (UnicodeString) "", //"DAY_OF_WEEK_IN_MONTH_FIELD",
        (UnicodeString) "", //"WEEK_OF_YEAR_FIELD",
        (UnicodeString) "", //"WEEK_OF_MONTH_FIELD",
        (UnicodeString) "AM", //"AM_PM_FIELD",
        (UnicodeString) "8", //"HOUR1_FIELD",
        (UnicodeString) "", //"HOUR0_FIELD",
        (UnicodeString) "" //"TIMEZONE_FIELD"
    };
    
    //Hashtable expected;// = new Hashtable();
    //expected.put(new LongKey(DateFormat.MONTH_FIELD), "5");
    //expected.put(new LongKey(DateFormat.DATE_FIELD), "3");
    //expected.put(new LongKey(DateFormat.YEAR_FIELD), "97");
    //expected.put(new LongKey(DateFormat.HOUR1_FIELD), "8");
    //expected.put(new LongKey(DateFormat.MINUTE_FIELD), "55");
    //expected.put(new LongKey(DateFormat.AM_PM_FIELD), "AM");
    
    //StringBuffer buf = new StringBuffer();
    UnicodeString fieldNames[] = {
        (UnicodeString) "ERA_FIELD",
        (UnicodeString) "YEAR_FIELD",
        (UnicodeString) "MONTH_FIELD",
        (UnicodeString) "DATE_FIELD",
        (UnicodeString) "HOUR_OF_DAY1_FIELD",
        (UnicodeString) "HOUR_OF_DAY0_FIELD",
        (UnicodeString) "MINUTE_FIELD",
        (UnicodeString) "SECOND_FIELD",
        (UnicodeString) "MILLISECOND_FIELD",
        (UnicodeString) "DAY_OF_WEEK_FIELD",
        (UnicodeString) "DAY_OF_YEAR_FIELD",
        (UnicodeString) "DAY_OF_WEEK_IN_MONTH_FIELD",
        (UnicodeString) "WEEK_OF_YEAR_FIELD",
        (UnicodeString) "WEEK_OF_MONTH_FIELD",
        (UnicodeString) "AM_PM_FIELD",
        (UnicodeString) "HOUR1_FIELD",
        (UnicodeString) "HOUR0_FIELD",
        (UnicodeString) "TIMEZONE_FIELD"
    };

    UBool pass = TRUE;
    for(int i = 0; i <= 17; ++i) {
        FieldPosition pos(i);
        UnicodeString buf;
        fmt->format(dt, buf, pos);
        //char[] dst = new char[pos.getEndIndex() - pos.getBeginIndex()];
        UnicodeString dst;
        buf.extractBetween(pos.getBeginIndex(), pos.getEndIndex(), dst);
        UnicodeString str(dst);
        logln((UnicodeString)"" + i + (UnicodeString)": " + fieldNames[i] +
                (UnicodeString)", \"" + str + (UnicodeString)"\", " +
                pos.getBeginIndex() + (UnicodeString)", " +
                pos.getEndIndex());
        UnicodeString exp = expected[i];
        if((exp.length() == 0 && str.length() == 0) || str == exp)
            logln(" ok");
        else {
            errln(UnicodeString(" expected ") + exp);
            pass = FALSE;
        }
        
    }
    if( ! pass) 
        errln("Fail: FieldPosition not set right by DateFormat");
    
    delete fmt;
}