View Javadoc

1   /*
2    * Copyright 2015 Data Archiving and Networked Services (an institute of
3    * Koninklijke Nederlandse Akademie van Wetenschappen), King's College London,
4    * Georg-August-Universitaet Goettingen Stiftung Oeffentlichen Rechts
5    *
6    * Licensed under the EUPL, Version 1.1 or – as soon they will be approved by
7    * the European Commission - subsequent versions of the EUPL (the "Licence");
8    * You may not use this work except in compliance with the Licence.
9    * You may obtain a copy of the Licence at:
10   *
11   * https://joinup.ec.europa.eu/software/page/eupl
12   *
13   * Unless required by applicable law or agreed to in writing, software
14   * distributed under the Licence is distributed on an "AS IS" basis,
15   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16   * See the Licence for the specific language governing
17   * permissions and limitations under the Licence.
18   */
19  
20  package eu.ehri.project.utils;
21  
22  /**
23   * Class for handling slugification of strings, for use
24   * in global identifiers.
25   */
26  public class Slugify {
27  
28      private static final String DEFAULT_REPLACE = "-";
29  
30      /**
31       * Slugify the input string using the default replacement string.
32       * @param input the String to slugify
33       * @return the slugified copy of the input string
34       */
35      public static String slugify(String input) {
36          return slugify(input, DEFAULT_REPLACE);
37      }
38  
39      /**
40       * Slugify the input string using the given replacement string.
41       *
42       * @param input the String to slugify
43       * @param replacement the replacement string
44       * @return the slugified copy of the input string
45       */
46      public static String slugify(String input, String replacement) {
47          return input
48                  .replaceAll("[\\p{P}=+<>~\\|]", replacement)
49                  .replaceAll("\\s+", replacement)            // whitespace
50                  .replaceAll(replacement + "+", replacement) // replacements
51                  .replaceAll("^" + replacement + "|" + replacement + "$", "") // leading/trailing replacements
52                  .toLowerCase();
53      }
54  }
55