contactrecord.java

这是一个matlab的java实现。里面有许多内容。请大家慢慢捉摸。

/* Copyright (C) 2002 Univ. of Massachusetts Amherst, Computer Science Dept.
   This file is part of "MALLET" (MAchine Learning for LanguagE Toolkit).
   http://www.cs.umass.edu/~mccallum/mallet
   This software is provided under the terms of the Common Public License,
   version 1.0, as published by http://www.opensource.org.  For further
   information, see the file `LICENSE' included with this distribution. */


/** 
   @author Aron Culotta <a href="mailto:[e-mail]">[e-mail]</a>
 */

package edu.umass.cs.mallet.projects.dex.types;

import java.util.Vector;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.Iterator;
import java.util.regex.Pattern;
import java.util.regex.Matcher;
import java.io.*;

/**
	 Stores contact information about a person
 */

public class ContactRecord implements Serializable {
	/** arbitrary list of fields and their values*/
	private HashMap fields = null;
	private ArrayList names;
	
	// regex for types of fields to perform filtering
	private static final Pattern PHONE = Pattern.compile (".*[Nn][Uu][Mm][Bb][Ee][Rr].*");
	private static final Pattern EMAIL = Pattern.compile ("[Ee][Mm][Aa][Ii][Ll]");
	private static final Pattern URL = Pattern.compile (".*[Uu][Rr][Ll]");
	private static final Pattern LASTNAME = Pattern.compile ("[Ll][Aa][Ss][Tt][Nn][Aa][Mm][Ee]");
	private static final Pattern F_PHONE = Pattern.compile ("[x\0-9\\(\\)\\[\\]\\-\\/\\.]+");
	private static final Pattern F_EMAIL = Pattern.compile (".*@.*");
	private static final Pattern F_URL = Pattern.compile (".*www\\.|http.*");
	private static final Pattern F_LASTNAME = Pattern.compile (".*\\p{Alpha}\\p{Alpha}.*");


	public ContactRecord () {
		this.fields = new HashMap();
		this.names = new ArrayList();
	}

	public void freeNames () {this.names = null; this.names = new ArrayList();}
	public void setNames (ArrayList peeps) {this.names = peeps;}
	public ArrayList getNames () {return this.names; }
	public void addName (CountedString n) {names.add (n);}
	public void addNames (ArrayList a) {names.addAll (a);}
	public void addAll (String key, ArrayList values) {
		Vector v = (Vector) fields.get (key);
		if (v == null)
			v = new Vector();
		for (int i=0; i < values.size(); i++) {
			String w = (String)values.get(i);
			w = removeEndline (w);
			if (filter (key, w)) {
				v.add (w);
			}
		}
		if (v.size() > 0)
			this.fields.put (key, v);		
	}
	
	public void setFieldValue (String key, String value)	{
		Vector v = (Vector) fields.get (key);
		if (v == null)
			v = new Vector();		
		value = removeEndline (value);
		if (filter (key, value)) {
			v.add (value);
			this.fields.put (key, v);
		}
	}
	
	public int size () { return numberFields(); }
	
	public boolean hasField (String key) {
		return (((Vector)this.fields.get(key)) != null);
	}
	
	public String getFirstValue (String key) {
		if (hasField (key)) {
			Vector v = getFieldValues (key);
			return (String)v.get (0);
		}
		else return "";
	}
	
	public int numberFields () { return this.fields.keySet().size(); }
	
	public Vector getFieldValues (String key) {
		return (Vector) this.fields.get (key);
	}

	/** Prints record in csv format, where each column corresponds to a
	 * field in <code>fields</code>. NOTE - only prints the top listing
	 * for each field.*/
	public String toCSV (ArrayList fields) {
		String ret = "";
		for (int i=0; i < fields.size(); i++) {
			String s = getFirstValue ((String)fields.get (i));			
			if (s == null)
				s = " ";
			s.replaceAll (",", " ");
			ret += s;
			if (i != fields.size()-1)
				ret += ", ";
		}
		return ret;
	}
	
	public String toString () {
		if (this.fields == null)
			return "";
		String ret = "";		
		Iterator iter = this.fields.keySet().iterator();
		while (iter.hasNext()) {
			String key = (String) iter.next();
			Vector v = (Vector) this.fields.get (key);
			if (v == null)
				throw new IllegalStateException (key + " not in hash.");
			ret += key + ": ";
			for (int i=0; i < v.size(); i++) 
				ret += (String)v.get (i) + " | ";
			ret += "\n";
		}
		return ret;
	}

	private String removeEndline (String s) {
		return s.replaceAll ("ENDLINE", " ").replaceAll (",", " ");
	}
	
	/** Perform some heuristic filters to catch some obvious errors*/
	private boolean filter (String key, String value) {
		// hack - ENDLINE used as a token in CRF, but shouldn't be in output
		Matcher mPhone = PHONE.matcher (key);
		Matcher mEmail = EMAIL.matcher (key);
		Matcher mURL = URL.matcher (key);
		Matcher mLastName = LASTNAME.matcher (key);
		if (mPhone.matches ()) {
			Matcher fPhone = F_PHONE.matcher (value);
			if (fPhone.matches()) {
				//System.err.println (key + ": " + value + " MATCHES");
				return true;
			}
			else {
				System.err.println (key + ": " + value + " DOESN'T MATCH");
				return false;
			}
		}
		else if (mEmail.matches ()) {
			Matcher fEmail = F_EMAIL.matcher (value);
			if (fEmail.matches()) {
				//System.err.println (key + ": " + value + " MATCHES");
				return true;
			}
			else {
				System.err.println (key + ": " + value + " DOESN'T MATCH");
				return false;
			}
		}
		else if (mURL.matches ()) {
			Matcher fURL = F_URL.matcher (value);
			if (fURL.matches()) {
				//System.err.println (key + ": " + value + " MATCHES");
				return true;
			}
			else {
				System.err.println (key + ": " + value + " DOESN'T MATCH");
				return false;
			}
		}
		else if (mLastName.matches ()) {
			Matcher fLastName = F_LASTNAME.matcher (value);
			if (fLastName.matches()) {
				//System.err.println (key + ": " + value + " MATCHES");
				return true;
			}
			else {
				System.err.println (key + ": " + value + " DOESN'T MATCH");
				return false;
			}
		}
		else
			return true;
	}

	private static final long serialVersionUID = 1;
	private static final int CURRENT_SERIAL_VERSION = 0;
	
	private void writeObject (ObjectOutputStream out) throws IOException {
		out.writeInt (CURRENT_SERIAL_VERSION);
		out.writeObject (fields);
		out.writeInt (names.size());
		for (int i=0; i < names.size(); i++) 
			out.writeObject ((CountedString)names.get (i));
	}
	private void readObject (ObjectInputStream in) throws IOException, ClassNotFoundException {
		int version = in.readInt ();
		fields = (HashMap) in.readObject();
		int size = in.readInt();
		names = new ArrayList();
		for (int i=0; i < size; i++) 
			names.add ((CountedString) in.readObject());
	}

}