JavadocTokenizer.java

/*
 *  Author:  Chris Seguin
 *
 *  This software has been developed under the copyleft
 *  rules of the GNU General Public License.  Please
 *  consult the GNU General Public License for more
 *  details about use and distribution of this software.
 */
package org.acm.seguin.pretty;

import org.acm.seguin.parser.Token;

/**
 *  Parses a javadoc comment
 *
 *@author    Chris Seguin
 */
public class JavadocTokenizer
{
	private             String       value;
	private             int          index;
	private             StringBuffer buffer;
	private             int          last;

	/**
	 *  Represents spaces
	 */
	public final static int SPACE = 0;

	/**
	 *  Represents newline
	 */
	public final static int NEWLINE = 1;

	/**
	 *  Represents a word
	 */
	public final static int WORD = 2;


	/**
	 *  Constructor for the JavadocTokenizer object
	 *
	 *@param  init  Description of Parameter
	 */
	public JavadocTokenizer(String init)
	{
		value = init;
		index = 0;
		buffer = new StringBuffer();
		last = value.length();
	}


	/**
	 *  Description of the Method
	 *
	 *@return    Description of the Returned Value
	 */
	public boolean hasNext()
	{
		return index < last;
	}


	/**
	 *  Description of the Method
	 *
	 *@return    Description of the Returned Value
	 */
	public Token next()
	{
		Token result = new Token();
		if (index == last) {
			result.kind = SPACE;
			result.image = " ";
			return result;
		}

		buffer.setLength(0);
		if (((index == 0) && (value.charAt(index) == '*')) ||
				((index == 0) && (value.charAt(index) == '/')) ||
				(value.charAt(index) == '\r') ||
				(value.charAt(index) == '\n'))
		{
			if (value.charAt(index) == '/')
			{
				index++;
			}

			loadNewline();
			result.kind = NEWLINE;
			result.image = buffer.toString();

			//System.out.println("Found a newline:  [" + result.image + "]");
		}
		else if (Character.isWhitespace(value.charAt(index)))
		{
			loadSpace();
			result.kind = SPACE;
			result.image = buffer.toString();
			//System.out.println("Found a space:  [" + result.image + "]");
		}
		else
		{
			loadWord();
			result.kind = WORD;
			result.image = checkEnd(buffer.toString());
			//System.out.println("Found a word:  [" + result.image + "]");
		}

		return result;
	}


	/**
	 *  Description of the Method
	 */
	private void loadNewline()
	{
		while ((index < last) && Character.isWhitespace(value.charAt(index)))
		{
			buffer.append(value.charAt(index));
			index++;
		}

		while ((index < last) && (value.charAt(index) == '*'))
		{
			buffer.append(value.charAt(index));
			index++;
		}

		if ((index < last) && (value.charAt(index) == '/'))
		{
			buffer.append(value.charAt(index));
			index++;
		}
	}


	/**
	 *  Description of the Method
	 */
	private void loadSpace()
	{
		while ((index < last) && Character.isWhitespace(value.charAt(index))
				 && (value.charAt(index) != '\n') && (value.charAt(index) != '\r'))
		{
			buffer.append(value.charAt(index));
			index++;
		}
	}


	/**
	 *  Description of the Method
	 */
	private void loadWord()
	{
		int start = index;
		while ((index < last) && !Character.isWhitespace(value.charAt(index))
				 && ((value.charAt(index) != '<') || (index == start)))
		{
			buffer.append(value.charAt(index));
			index++;

			if (value.charAt(index - 1) == '>')
			{
				return;
			}
		}
	}


	/**
	 *  Checks that the end of the word doesn't end with end of comment
	 *
	 *@param  value  the value to check
	 *@return        the revised value
	 */
	private String checkEnd(String value)
	{
		if (value.endsWith("*/"))
		{
			return value.substring(0, value.length() - 2);
		}
		return value;
	}

	public static boolean hasContent(String value) {
		if (value == null)
			return false;

		int valueLength = value.length();
		if (valueLength == 0)
			return false;

		int start = 0;
		if (value.charAt(0) == '/')
			start++;

		int last = valueLength - 1;
		for (int ndx = start; ndx < last; ndx++) {
			char ch = value.charAt(ndx);
			if (!(Character.isWhitespace(ch) || (ch == '*')))
				return true;
		}

		char ch = value.charAt(last);
		return !(Character.isWhitespace(ch) || (ch == '*') || (ch == '/'));
	}
}