tokentowords.java

这是java 开发的的免费语音播放插件,很值得学习参考!!!!!!!!!!!!111

     * Converts the given Token into (word) Items in the WordRelation.
     *
     * @param  tokenVal the String value of the token, which may or may not be
     *                  same as the one in called "name" in flite
     *
     */
    private void tokenToWords(String tokenVal) {

	FeatureSet tokenFeatures = tokenItem.getFeatures();
	String itemName = tokenFeatures.getString("name");
	int tokenLength = tokenVal.length();

	if (tokenFeatures.isPresent("phones")) {
	    wordRelation.addWord(tokenVal);

	} else if ((tokenVal.equals("a") || tokenVal.equals("A")) &&
                ((tokenItem.getNext() == null) ||
                 !(tokenVal.equals(itemName)) ||
                 !(((String) tokenItem.findFeature("punc")).equals("")))) {
	    /* if A is a sub part of a token, then its ey not ah */
	    wordRelation.addWord("_a");

	} else if (matches(alphabetPattern, tokenVal)) {

	    if (matches(romanNumbersPattern, tokenVal)) {
		
		/* XVIII */
		romanToWords(tokenVal);
		
	    } else if (matches(illionPattern, tokenVal) &&
		       matches(usMoneyPattern, 
			       (String) tokenItem.findFeature("p.name"))) {
		/* $ X -illion */
		wordRelation.addWord(tokenVal);
		wordRelation.addWord("dollars");	    
		
	    } else if (matches(drStPattern, tokenVal)) {
		
		/* St Andrew's St, Dr King Dr */
		drStToWords(tokenVal);
		
	    } else if (tokenVal.equals("Mr")) {
		
		tokenItem.getFeatures().setString("punc", "");
		wordRelation.addWord("mister");
		
	    } else if (tokenVal.equals("Mrs")) {
		
		tokenItem.getFeatures().setString("punc", "");
		wordRelation.addWord("missus");
		
	    } else if (tokenLength == 1
		       && isUppercaseLetter(tokenVal.charAt(0))
		       && ((String)tokenItem.findFeature("n.whitespace")).equals(" ")
		       && isUppercaseLetter
		       (((String) tokenItem.findFeature("n.name")).charAt(0))) {
		
		tokenFeatures.setString("punc", "");
		String aaa = tokenVal.toLowerCase();
		if (aaa.equals("a")) {
		    wordRelation.addWord("_a");
		} else {
		    wordRelation.addWord(aaa);
		}
	    } else if (isStateName(tokenVal)) {
		/*
		  The name of a US state
		  isStateName() has already added the full name of the
		  state, so we're all set.
		*/
	    } else if (tokenLength > 1 && !isPronounceable(tokenVal)) {
		/* Need common exception list */
		/* unpronouncable list of alphas */
		NumberExpander.expandLetters
		    (tokenVal, wordRelation);
		
	    } else {
		/* just a word */
		wordRelation.addWord(tokenVal.toLowerCase());
	    }
	    
	} else if (matches(dottedAbbrevPattern, tokenVal)) {
	    
	    /* U.S.A. */
	    // remove all dots
	    String aaa = Utilities.deleteChar(tokenVal, '.'); 
	    NumberExpander.expandLetters(aaa, wordRelation);
	    
	} else if (matches(commaIntPattern, tokenVal)) {
	    
	    /* 99,999,999 */
	    String aaa = Utilities.deleteChar(tokenVal, ',');
	    NumberExpander.expandReal(aaa, wordRelation);
	    
	} else if (matches(sevenPhoneNumberPattern, tokenVal)) {
	    
	    /* 234-3434  telephone numbers */
	    int dashIndex = tokenVal.indexOf('-');
	    String aaa = tokenVal.substring(0, dashIndex);
	    String bbb = tokenVal.substring(dashIndex+1);
	    
	    NumberExpander.expandDigits(aaa, wordRelation);
	    wordRelation.addBreak();
	    NumberExpander.expandDigits(bbb, wordRelation);
	    
	} else if (matchesPartPhoneNumber(tokenVal)) {
	    
	    /* part of a telephone number */
	    String punctuation = (String) tokenItem.findFeature("punc");
	    if (punctuation.equals("")) {
		tokenItem.getFeatures().setString("punc", ",");
	    }
	    NumberExpander.expandDigits(tokenVal, wordRelation);
	    wordRelation.addBreak();
		
	} else if (matches(numberTimePattern, tokenVal)) {
	    
	    /* 12:35 */
	    int colonIndex = tokenVal.indexOf(':');
	    String aaa = tokenVal.substring(0, colonIndex);
	    String bbb = tokenVal.substring(colonIndex+1);
	    
	    NumberExpander.expandNumber(aaa, wordRelation);
	    if (!(bbb.equals("00"))) {
		NumberExpander.expandID(bbb, wordRelation);
	    }
	    
	} else if (matches(digits2DashPattern, tokenVal)) {
	    
	    /* 999-999-999 */
	    digitsDashToWords(tokenVal);
	    
	} else if (matches(digitsPattern, tokenVal)) {
	    
	    digitsToWords(tokenVal);
	    
	} else if (tokenLength == 1
		   && isUppercaseLetter(tokenVal.charAt(0))
		   && ((String)tokenItem.findFeature("n.whitespace")).equals
                   (" ")
		   && isUppercaseLetter
		   (((String) tokenItem.findFeature("n.name")).charAt(0))) {
	    
	    tokenFeatures.setString("punc", "");
	    String aaa = tokenVal.toLowerCase();
	    if (aaa.equals("a")) {
		wordRelation.addWord("_a");
	    } else {
		wordRelation.addWord(aaa);
	    }
	} else if (matches(doublePattern, tokenVal)) {

	    NumberExpander.expandReal(tokenVal, wordRelation);

	} else if (matches(ordinalPattern, tokenVal)) {
	    
	    /* explicit ordinals */
	    String aaa = tokenVal.substring(0, tokenLength - 2);
	    NumberExpander.expandOrdinal(aaa, wordRelation);

	} else if (matches(usMoneyPattern, tokenVal)) {

	    /* US money */
	    usMoneyToWords(tokenVal);

	} else if (tokenLength > 0
		   && tokenVal.charAt(tokenLength - 1) == '%') {
	    
	    /* Y% */
	    tokenToWords(tokenVal.substring(0, tokenLength - 1));
	    wordRelation.addWord("per");
	    wordRelation.addWord("cent");

	} else if (matches(numessPattern, tokenVal)) {

	    /* 60s and 7s and 9s */
	    tokenToWords(tokenVal.substring(0, tokenLength - 1));
	    wordRelation.addWord("'s");
	    
	} else if (tokenVal.indexOf('\'') != -1) {
	    
	    postropheToWords(tokenVal);
	    
	} else if (matches(digitsSlashDigitsPattern, tokenVal) &&
		   tokenVal.equals(itemName)) {

	    digitsSlashDigitsToWords(tokenVal);

	} else if (tokenVal.indexOf('-') != -1) {
	    
	    dashToWords(tokenVal);
	    
	} else if (tokenLength > 1 &&
		   !matches(alphabetPattern, tokenVal)) {
	    
	    notJustAlphasToWords(tokenVal);

	} else {
	    /* just a word */
	    wordRelation.addWord(tokenVal.toLowerCase());
	}   
    }

	
    /**
     * Convert the given digit token with dashes (e.g. 999-999-999)
     * into (word) Items in the WordRelation.
     *
     * @param tokenVal  the digit string
     */
    private void digitsDashToWords(String tokenVal) {
	int tokenLength = tokenVal.length();
	int a = 0;
	for (int p = 0; p <= tokenLength; p++) {
	    if (p == tokenLength || tokenVal.charAt(p) == '-') {
		String aaa = tokenVal.substring(a, p);
		NumberExpander.expandDigits(aaa, wordRelation);
		wordRelation.addBreak();
		a = p+1;
	    }
	}
    }

	
    /**
     * Convert the given digit token into (word) Items in the WordRelation.
     *
     * @param tokenVal  the digit string
     */
    private void digitsToWords(String tokenVal) {
	FeatureSet featureSet = tokenItem.getFeatures();
	String nsw = "";
	if (featureSet.isPresent("nsw")) {
	    nsw = featureSet.getString("nsw");
	}

	if (nsw.equals("nide")) {
	    NumberExpander.expandID(tokenVal, wordRelation);
	} else {
	    String rName = featureSet.getString("name");
	    String digitsType = null;
	    
	    if (tokenVal.equals(rName)) {
		digitsType = (String) cart.interpret(tokenItem);
	    } else {
		featureSet.setString("name", tokenVal);
		digitsType = (String) cart.interpret(tokenItem);
		featureSet.setString("name", rName);
	    }
	    
	    if (digitsType.equals("ordinal")) {
		NumberExpander.expandOrdinal(tokenVal, wordRelation);
	    } else if (digitsType.equals("digits")) {
		NumberExpander.expandDigits(tokenVal, wordRelation);
	    } else if (digitsType.equals("year")) {
		NumberExpander.expandID(tokenVal, wordRelation);
	    } else {
		NumberExpander.expandNumber(tokenVal, wordRelation);
	    }
	}
    }
    
    
    /**
     * Converts the given Roman numeral string into (word) Items in the
     * WordRelation.
     *
     * @param romanString the roman numeral string
     */
    private void romanToWords(String romanString) {
	String punctuation = (String) tokenItem.findFeature("p.punc");
	
	if (punctuation.equals("")) {
	    /* no preceeding punctuation */
	    String n = String.valueOf(NumberExpander.expandRoman(romanString));
	    
	    if (kingLike(tokenItem)) {
		wordRelation.addWord("the");
		NumberExpander.expandOrdinal(n, wordRelation);
	    } else if (sectionLike(tokenItem)) {
		NumberExpander.expandNumber(n, wordRelation);
	    } else {
		NumberExpander.expandLetters(romanString, wordRelation);
	    }
	} else {
	    NumberExpander.expandLetters(romanString, wordRelation);
	}
    }
    

    /**
     * Returns true if the given key is in the kingSectionLikeHash
     * Hashtable, and the value is the same as the given value.
     *
     * @param key key to look for in the hashtable
     * @param value the value to match
     *
     * @return true if it matches, or false if it does not or if
     * the key is not mapped to any value in the hashtable.
     */
    private static boolean inKingSectionLikeHash(String key, String value) {
	String hashValue = (String) kingSectionLikeHash.get(key);
	if (hashValue != null) {
	    return (hashValue.equals(value));
	} else {
	    return false;
	}
    }



    /**
     * Returns true if the given token item contains a token that is
     * in a king-like context, e.g., "King" or "Louis".
     *
     * @param tokenItem the token item to check
     *
     * @return true or false
     */
    public static boolean kingLike(Item tokenItem) {
	String kingName = 
	    ((String) tokenItem.findFeature("p.name")).toLowerCase();
	if (inKingSectionLikeHash(kingName, KING_NAMES)) {
	    return true;
	} else {
	    String kingTitle =
		((String) tokenItem.findFeature("p.p.name")).toLowerCase();
	    return inKingSectionLikeHash(kingTitle, KING_TITLES);
	}
    }

    
    /**
     * Returns true if the given token item contains a token that is
     * in a section-like context, e.g., "chapter" or "act".
     *
     * @param tokenItem the token item to check
     *
     * @return true or false
     */
    public static boolean sectionLike(Item tokenItem) {
	String sectionType =
	    ((String) tokenItem.findFeature("p.name")).toLowerCase();
	return inKingSectionLikeHash(sectionType, SECTION_TYPES);
    }


    /**
     * Converts the given string containing "St" and "Dr" to (word) Items
     * in the WordRelation.
     *
     * @param drStString the string with "St" and "Dr"
     */
    private void drStToWords(String drStString) {
	String street = null;
	String saint = null;
	char c0 = drStString.charAt(0);

	if (c0 == 's' || c0 == 'S') {
	    street = "street";
	    saint = "saint";
	} else {
	    street = "drive";
	    saint = "doctor";
	}
	
	FeatureSet featureSet = tokenItem.getFeatures();
	String punctuation = featureSet.getString("punc");

	String featPunctuation = (String) tokenItem.findFeature("punc");

	if (tokenItem.getNext() == null ||
	    punctuation.indexOf(',') != -1) {
	    wordRelation.addWord(street);
	} else if (featPunctuation.equals(",")) {
	    wordRelation.addWord(saint);