001/*
002 *  Copyright 2013 Anyware Services
003 *
004 *  Licensed under the Apache License, Version 2.0 (the "License");
005 *  you may not use this file except in compliance with the License.
006 *  You may obtain a copy of the License at
007 *
008 *      http://www.apache.org/licenses/LICENSE-2.0
009 *
010 *  Unless required by applicable law or agreed to in writing, software
011 *  distributed under the License is distributed on an "AS IS" BASIS,
012 *  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
013 *  See the License for the specific language governing permissions and
014 *  limitations under the License.
015 */
016package org.ametys.cms;
017
018import java.text.Normalizer;
019import java.util.regex.Matcher;
020import java.util.regex.Pattern;
021
022import org.apache.commons.lang.StringUtils;
023
024/**
025 * Helper for filtering name in order to use it into
026 * a JCR node name or into a segment of an UR.
027 */
028public final class FilterNameHelper
029{
030    private static final Pattern __PAGE_PATTERN = Pattern.compile("^([0-9-_]*)[a-z].*$");
031
032    private FilterNameHelper()
033    {
034        // empty
035    }
036
037    /**
038     * Filter a name for using it into an URI.
039     * @param name the name to filter.
040     * @return the name filtered.
041     */
042    public static String filterName(String name)
043    {
044        // FIXME Page name can not start with a number
045
046        // Use lower case
047        // then remove accents
048        // then replace contiguous spaces with one dash
049        // and finally remove non-alphanumeric characters except -
050        String filteredName = Normalizer.normalize(name.toLowerCase(), Normalizer.Form.NFD).replaceAll("[\\p{InCombiningDiacriticalMarks}]", "").trim(); 
051        filteredName = filteredName.replaceAll("œ", "oe").replaceAll("æ", "ae").replaceAll(" +", "-").replaceAll("[^\\w-]", "-").replaceAll("-+", "-");
052
053        Matcher m = __PAGE_PATTERN.matcher(filteredName);
054        if (!m.matches())
055        {
056            throw new IllegalArgumentException(filteredName + " doesn't match the expected regular expression : " + __PAGE_PATTERN.pattern());
057        }
058
059        filteredName = filteredName.substring(m.end(1));
060
061        // Remove characters '-' and '_' at the start and the end of the string
062        return StringUtils.strip(filteredName, "-_");
063    }
064}