All Downloads are FREE. Search and download functionalities are using the official Maven repository.

com.google.gwt.safehtml.shared.SafeHtmlUtils Maven / Gradle / Ivy

There is a newer version: 2.7.0.vaadin7
Show newest version
/*
 * Copyright 2010 Google Inc.
 *
 * Licensed under the Apache License, Version 2.0 (the "License"); you may not
 * use this file except in compliance with the License. You may obtain a copy of
 * the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
 * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
 * License for the specific language governing permissions and limitations under
 * the License.
 */
package com.google.gwt.safehtml.shared;

import com.google.gwt.regexp.shared.RegExp;

/**
 * Utility class containing static methods for escaping and sanitizing strings.
 */
public final class SafeHtmlUtils {

  private static final String HTML_ENTITY_REGEX = "[a-z]+|#[0-9]+|#x[0-9a-fA-F]+";

  /**
   * An empty String.
   */
  public static final SafeHtml EMPTY_SAFE_HTML = new SafeHtmlString("");

  private static final RegExp AMP_RE = RegExp.compile("&", "g");
  private static final RegExp GT_RE = RegExp.compile(">", "g");
  private static final RegExp LT_RE = RegExp.compile("<", "g");
  private static final RegExp SQUOT_RE = RegExp.compile("\'", "g");
  private static final RegExp QUOT_RE = RegExp.compile("\"", "g");

  /**
   * Returns a {@link SafeHtml} constructed from a safe string, i.e., without escaping
   * the string.
   *
   * 

* Important: For this method to be able to honor the {@link SafeHtml} * contract, all uses of this method must satisfy the following constraints: * *

    * *
  1. The argument expression must be fully determined at compile time. * *
  2. The value of the argument must end in "inner HTML" context and not * contain incomplete HTML tags. I.e., the following is not a correct use of * this method, because the {@code } tag is incomplete: * *
       * {@code shb.appendConstantHtml("':
            return ">";
          case '"':
            return """;
          case '\'':
            return "'";
          default:
            return "" + c;
        }
      }
    
      /**
       * HTML-escapes a string.
       *
       * Note: The following variants of this function were profiled on FF36,
       * Chrome6, IE8:
       * 
      *
    1. For each case, check indexOf, then use s.replace(regex, string)
    2. *
    3. For each case, check indexOf, then use s.replaceAll()
    4. *
    5. Check if any metachar is present using a regex, then use #1
    6. *
    7. For each case, use s.replace(regex, string)
    8. *
    * * #1 was found to be the fastest, and is used below. * * @param s the string to be escaped * @return the input string, with all occurrences of HTML meta-characters * replaced with their corresponding HTML Entity References */ public static String htmlEscape(String s) { if (s.indexOf("&") != -1) { s = AMP_RE.replace(s, "&"); } if (s.indexOf("<") != -1) { s = LT_RE.replace(s, "<"); } if (s.indexOf(">") != -1) { s = GT_RE.replace(s, ">"); } if (s.indexOf("\"") != -1) { s = QUOT_RE.replace(s, """); } if (s.indexOf("'") != -1) { s = SQUOT_RE.replace(s, "'"); } return s; } /** * HTML-escapes a string, but does not double-escape HTML-entities already * present in the string. * * @param text the string to be escaped * @return the input string, with all occurrences of HTML meta-characters * replaced with their corresponding HTML Entity References, with the * exception that ampersand characters are not double-escaped if they * form the start of an HTML Entity Reference */ public static String htmlEscapeAllowEntities(String text) { StringBuilder escaped = new StringBuilder(); boolean firstSegment = true; for (String segment : text.split("&", -1)) { if (firstSegment) { /* * The first segment is never part of an entity reference, so we always * escape it. * Note that if the input starts with an ampersand, we will get an empty * segment before that. */ firstSegment = false; escaped.append(htmlEscape(segment)); continue; } int entityEnd = segment.indexOf(';'); if (entityEnd > 0 && segment.substring(0, entityEnd).matches(HTML_ENTITY_REGEX)) { // Append the entity without escaping. escaped.append("&").append(segment.substring(0, entityEnd + 1)); // Append the rest of the segment, escaped. escaped.append(htmlEscape(segment.substring(entityEnd + 1))); } else { // The segment did not start with an entity reference, so escape the // whole segment. escaped.append("&").append(htmlEscape(segment)); } } return escaped.toString(); } // prevent instantiation private SafeHtmlUtils() { } }




© 2015 - 2024 Weber Informatics LLC | Privacy Policy