001/* 002 * (C) Copyright 2012 Nuxeo SA (http://nuxeo.com/) and contributors. 003 * 004 * All rights reserved. This program and the accompanying materials 005 * are made available under the terms of the GNU Lesser General Public License 006 * (LGPL) version 2.1 which accompanies this distribution, and is available at 007 * http://www.gnu.org/licenses/lgpl.html 008 * 009 * This library is distributed in the hope that it will be useful, 010 * but WITHOUT ANY WARRANTY; without even the implied warranty of 011 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 012 * Lesser General Public License for more details. 013 * 014 * Contributors: 015 * Antoine Taillefer 016 */ 017package org.nuxeo.ecm.diff.content; 018 019import java.util.regex.Matcher; 020import java.util.regex.Pattern; 021 022/** 023 * Handles HTML detection in a string. 024 * 025 * @author <a href="mailto:ataillefer@nuxeo.com">Antoine Taillefer</a> 026 * @since 5.6 027 */ 028public final class HtmlGuesser { 029 030 public static final String HTML_TAG_REGEXP = "<.*?>.*?</.*?>"; 031 032 /** 033 * Checks if the specified text can be considered as HTML code. 034 * 035 * @param text the text 036 * @return true, if the specified text contains at least one HTML tag. 037 */ 038 public static boolean isHtml(String text) { 039 Pattern pattern = Pattern.compile(HTML_TAG_REGEXP); 040 Matcher matcher = pattern.matcher(text); 041 return matcher.find(); 042 } 043}