001/*
002 * Copyright (C) 2009 The Guava Authors
003 *
004 * Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except
005 * in compliance with the License. You may obtain a copy of the License at
006 *
007 * http://www.apache.org/licenses/LICENSE-2.0
008 *
009 * Unless required by applicable law or agreed to in writing, software distributed under the License
010 * is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express
011 * or implied. See the License for the specific language governing permissions and limitations under
012 * the License.
013 */
014
015package com.google.common.net;
016
017import com.google.common.annotations.GwtCompatible;
018import com.google.common.escape.Escaper;
019
020/**
021 * {@code Escaper} instances suitable for strings to be included in particular sections of URLs.
022 *
023 * <p>If the resulting URLs are inserted into an HTML or XML document, they will require additional
024 * escaping with {@link com.google.common.html.HtmlEscapers} or {@link
025 * com.google.common.xml.XmlEscapers}.
026 *
027 * @author David Beaumont
028 * @author Chris Povirk
029 * @since 15.0
030 */
031@GwtCompatible
032public final class UrlEscapers {
033  private UrlEscapers() {}
034
035  // For each xxxEscaper() method, please add links to external reference pages
036  // that are considered authoritative for the behavior of that escaper.
037
038  static final String URL_FORM_PARAMETER_OTHER_SAFE_CHARS = "-_.*";
039
040  static final String URL_PATH_OTHER_SAFE_CHARS_LACKING_PLUS =
041      "-._~" // Unreserved characters.
042          + "!$'()*,;&=" // The subdelim characters (excluding '+').
043          + "@:"; // The gendelim characters permitted in paths.
044
045  /**
046   * Returns an {@link Escaper} instance that escapes strings so they can be safely included in <a
047   * href="https://goo.gl/MplK6I">URL form parameter names and values</a>. Escaping is performed
048   * with the UTF-8 character encoding. The caller is responsible for <a
049   * href="https://goo.gl/9EfkM1">replacing any unpaired carriage return or line feed characters
050   * with a CR+LF pair</a> on any non-file inputs before escaping them with this escaper.
051   *
052   * <p>When escaping a String, the following rules apply:
053   *
054   * <ul>
055   *   <li>The alphanumeric characters "a" through "z", "A" through "Z" and "0" through "9" remain
056   *       the same.
057   *   <li>The special characters ".", "-", "*", and "_" remain the same.
058   *   <li>The space character " " is converted into a plus sign "+".
059   *   <li>All other characters are converted into one or more bytes using UTF-8 encoding and each
060   *       byte is then represented by the 3-character string "%XY", where "XY" is the two-digit,
061   *       uppercase, hexadecimal representation of the byte value.
062   * </ul>
063   *
064   * <p>This escaper is suitable for escaping parameter names and values even when <a
065   * href="https://goo.gl/utn6M">using the non-standard semicolon</a>, rather than the ampersand, as
066   * a parameter delimiter. Nevertheless, we recommend using the ampersand unless you must
067   * interoperate with systems that require semicolons.
068   *
069   * <p><b>Note:</b> Unlike other escapers, URL escapers produce <a
070   * href="https://url.spec.whatwg.org/#percent-encode">uppercase</a> hexadecimal sequences.
071   *
072   */
073  public static Escaper urlFormParameterEscaper() {
074    return URL_FORM_PARAMETER_ESCAPER;
075  }
076
077  private static final Escaper URL_FORM_PARAMETER_ESCAPER =
078      new PercentEscaper(URL_FORM_PARAMETER_OTHER_SAFE_CHARS, true);
079
080  /**
081   * Returns an {@link Escaper} instance that escapes strings so they can be safely included in <a
082   * href="https://goo.gl/m2MIf0">URL path segments</a>. The returned escaper escapes all non-ASCII
083   * characters, even though <a href="https://goo.gl/e7E0In">many of these are accepted in modern
084   * URLs</a>. (<a href="https://goo.gl/jfVxXW">If the escaper were to leave these characters
085   * unescaped, they would be escaped by the consumer at parse time, anyway.</a>) Additionally, the
086   * escaper escapes the slash character ("/"). While slashes are acceptable in URL paths, they are
087   * considered by the specification to be separators between "path segments." This implies that, if
088   * you wish for your path to contain slashes, you must escape each segment separately and then
089   * join them.
090   *
091   * <p>When escaping a String, the following rules apply:
092   *
093   * <ul>
094   *   <li>The alphanumeric characters "a" through "z", "A" through "Z" and "0" through "9" remain
095   *       the same.
096   *   <li>The unreserved characters ".", "-", "~", and "_" remain the same.
097   *   <li>The general delimiters "@" and ":" remain the same.
098   *   <li>The subdelimiters "!", "$", "&amp;", "'", "(", ")", "*", "+", ",", ";", and "=" remain
099   *       the same.
100   *   <li>The space character " " is converted into %20.
101   *   <li>All other characters are converted into one or more bytes using UTF-8 encoding and each
102   *       byte is then represented by the 3-character string "%XY", where "XY" is the two-digit,
103   *       uppercase, hexadecimal representation of the byte value.
104   * </ul>
105   *
106   * <p><b>Note:</b> Unlike other escapers, URL escapers produce <a
107   * href="https://url.spec.whatwg.org/#percent-encode">uppercase</a> hexadecimal sequences.
108   */
109  public static Escaper urlPathSegmentEscaper() {
110    return URL_PATH_SEGMENT_ESCAPER;
111  }
112
113  private static final Escaper URL_PATH_SEGMENT_ESCAPER =
114      new PercentEscaper(URL_PATH_OTHER_SAFE_CHARS_LACKING_PLUS + "+", false);
115
116  /**
117   * Returns an {@link Escaper} instance that escapes strings so they can be safely included in a <a
118   * href="https://goo.gl/xXEq4p">URL fragment</a>. The returned escaper escapes all non-ASCII
119   * characters, even though <a href="https://goo.gl/e7E0In">many of these are accepted in modern
120   * URLs</a>.
121   *
122   * <p>When escaping a String, the following rules apply:
123   *
124   * <ul>
125   *   <li>The alphanumeric characters "a" through "z", "A" through "Z" and "0" through "9" remain
126   *       the same.
127   *   <li>The unreserved characters ".", "-", "~", and "_" remain the same.
128   *   <li>The general delimiters "@" and ":" remain the same.
129   *   <li>The subdelimiters "!", "$", "&amp;", "'", "(", ")", "*", "+", ",", ";", and "=" remain
130   *       the same.
131   *   <li>The space character " " is converted into %20.
132   *   <li>Fragments allow unescaped "/" and "?", so they remain the same.
133   *   <li>All other characters are converted into one or more bytes using UTF-8 encoding and each
134   *       byte is then represented by the 3-character string "%XY", where "XY" is the two-digit,
135   *       uppercase, hexadecimal representation of the byte value.
136   * </ul>
137   *
138   * <p><b>Note:</b> Unlike other escapers, URL escapers produce <a
139   * href="https://url.spec.whatwg.org/#percent-encode">uppercase</a> hexadecimal sequences.
140   */
141  public static Escaper urlFragmentEscaper() {
142    return URL_FRAGMENT_ESCAPER;
143  }
144
145  private static final Escaper URL_FRAGMENT_ESCAPER =
146      new PercentEscaper(URL_PATH_OTHER_SAFE_CHARS_LACKING_PLUS + "+/?", false);
147}