EDIT

/**

* Copyright (C) 2007 EDIT

* European Distributed Institute of Taxonomy

* http://www.e-taxonomy.eu

*

* The contents of this file are subject to the Mozilla Public License Version 1.1

* See LICENSE.TXT at the top of this package for the full license terms.

*/

package eu.etaxonomy.cdm.common;

import java.util.regex.Matcher;

import java.util.regex.Pattern;

import org.apache.commons.lang3.StringUtils;

/**

 * A class for handling DOIs (http://www.doi.org).

 * It offers parsing and formatting functionality as well as validation.

 * A {@link DOI} object can only be created by syntactic valid input.

 * It internally stores 2 strings, the first one being the registrant number

 * (including sub numbers), the second being the suffix.

 *

 *

 * @author a.mueller

 * @since 2013-09-04

 */

public final class DOI implements java.io.Serializable{

	/**

     * Explicit serialVersionUID for interoperability.

     */

	private static final long serialVersionUID = -3871039785359980553L;

	public static final int MAX_LENGTH = 1000;

    public static final String DOI_ORG = "doi.org/";

    public static final String OLD_DOI_ORG = "dx." + DOI_ORG;

	/**

	 * The default public DOI proxy server

	 */

	public static final String HTTP_DOI_ORG = "http://" + DOI_ORG;

	/**

	 * The former default public DOI proxy server, still supported but no longer preferred.

	 * @see #HTTP_DOI_ORG

	 */

	public static final String HTTP_OLD_DOI_ORG = "http://" + OLD_DOI_ORG;

    private volatile transient int hashCode = -1;	// Zero ==> undefined

	//http://www.doi.org/doi_handbook/2_Numbering.html#2.2.1

//	prefix + suffix, no defined length, case-insensitive, any printable characters

//********************************* VARIABLES *************************************/

	/**

	 * The directory indicator for DOIs as registered at

	 */

	public static final String DIRECTORY_INDICATOR = "10";

	private String prefix_registrantCode;

	private String suffix;

// ***************************** FACTORY METHODS ***************************************/

	public static DOI fromString(String doi) throws IllegalArgumentException{

		return new DOI(doi);

	}

	public static DOI fromRegistrantCodeAndSuffix(String registrantCode, String suffix) throws IllegalArgumentException{

		return new DOI(registrantCode, suffix);

	}

// ******************************* CONSTRUCTOR ************************************/

	private DOI(){} //empty constructor required for JAXB

    /**

     * Creates a doi by its registrantCode and its suffix

     * @param registrantCode the registrant code, the is the part following the directoryIndicator "10."

     * 	and preceding the first forward slash (followed by the suffix)

     * @param suffix the suffix is the part of the DOI following the first forward slash. It is provided

     * by the registrant

     */

    private DOI(String registrantCode, String suffix) {

    	//preliminary until prefix_registrantCode and suffix validation is implemented

		this("10." + registrantCode + "/" + suffix);

		//use only after validation of both parts

//		this.prefix_registrantCode = registrantCode;

//		this.suffix = suffix;

	}

    private DOI(String doiString) {

		super();

		parseDoiString(doiString);

	}

//************************************ GETTER ***********************************/

	public String getPrefix() {

		return makePrefix();

	}

	public String getPrefix_registrantCode() {

		return prefix_registrantCode;

	}

	public String getSuffix() {

		return suffix;

	}

	private static Pattern doiPattern = Pattern.compile("^doi:\\s*", Pattern.CASE_INSENSITIVE);

// ********************************************* PARSER *******************************/

	private void parseDoiString(String doi){

		boolean isUrn = false;

		if (StringUtils.isBlank(doi)){

			throw new IllegalArgumentException("Doi string must not be null or blank");

		}

		doi = doi.trim();

		if (doi.startsWith("https") ){

			doi = doi.replaceFirst("https", "http").trim();

		}

		Matcher matcher = doiPattern.matcher(doi);

		if (matcher.find()){

			doi = matcher.replaceFirst("").trim();

		}

		//replace URI prefix

		if (doi.startsWith(HTTP_DOI_ORG)){

			doi = doi.replaceFirst(HTTP_DOI_ORG, "");

		}else if (doi.startsWith(HTTP_OLD_DOI_ORG)){

			doi = doi.replaceFirst(HTTP_OLD_DOI_ORG, "");

		}else if (doi.startsWith(DOI_ORG)){

            doi = doi.replaceFirst(DOI_ORG, "");

        }else if (doi.startsWith(OLD_DOI_ORG)){

            doi = doi.replaceFirst(OLD_DOI_ORG, "");

        }

		//handle URN prefix

		if (doi.startsWith("urn:doi:")){

			doi = doi.replaceFirst("urn:doi:","");

			isUrn = true;

		}

		//now we should have the pure doi

		if (doi.length() > MAX_LENGTH){

			//for persistence reason we currently restrict the length of DOIs to 1000

			throw new IllegalArgumentException("DOIs may have a maximum length of 1000 in the CDM.");

		}

		if (! doi.startsWith("10.")){

			throw new IllegalArgumentException("DOI not parsable. DOI must start with 10. or an URI or URN prefix. But was " + doi);

		}

		doi = doi.substring(3);

		String sep = isUrn? ":" : "/";

//		registrant

		String registrant = doi.split(sep)[0];

		if (!registrant.matches("[0-9]{2,}(?:[.][0-9]+)*")){   //per definition the number of digits may also be 1, however the lowest known number is 3 so we may be on the safe side here

			String message = "Invalid prefix '10.%s'";

			throw new IllegalArgumentException(String.format(message, registrant));

		}

		//suffix

		String suffix = doi.replaceFirst(registrant + sep,"");

		if (suffix.equals("")){

            String message = "Suffix must not be empty";

            throw new IllegalArgumentException(message);

        }

		if (! suffix.matches("\\p{Print}+")){

			String message = "Suffix should only include printable characters";

			throw new IllegalArgumentException(message + ": " + suffix);

		}

		if (isUrn){

			//TODO do some other replacements according to http://www.doi.org/doi_handbook/2_Numbering.html#2.6.3

			//e.g. slash becomes : in URN

			//TODO do we need this also for other URIs? According to http://www.doi.org/doi_handbook/2_Numbering.html#2.6 it is only required for URNs

			suffix = UrlUtf8Coder.unescape(suffix);

		}

		//success

		this.prefix_registrantCode = registrant;

		this.suffix = suffix;

	}

	private String makePrefix(){

		return DIRECTORY_INDICATOR + "." + this.prefix_registrantCode;

	}

	private String makeDoi(){

		return makePrefix() + "/" + this.suffix;

	}

	public String asURI(){

		return HTTP_DOI_ORG + makePrefix() + "/" + uriEncodedSuffix();

	}

	private String uriEncodedSuffix() {

		String result = UrlUtf8Coder.encode(this.suffix);

		return result;

	}

//************************************************* toString/equals /hashCode *********************/

	@Override

	public int hashCode() {

		if (hashCode == -1) {

            hashCode = 31 * prefix_registrantCode.toUpperCase().hashCode() + suffix.toUpperCase().hashCode();

        }

        return hashCode;

	}

	@Override

	public boolean equals(Object obj) {

		if (obj instanceof DOI){

			DOI doi = (DOI)obj;

			if (this.prefix_registrantCode.toUpperCase().equals(doi.prefix_registrantCode.toUpperCase()) &&

					this.suffix.toUpperCase().equals(doi.suffix.toUpperCase())){

				return true;

			}

		}

		return false;

	}

	@Override

	public String toString(){

		return makeDoi();

	}

}