001/*
002 * $Id: PdfTextExtractor.java 4784 2011-03-15 08:33:00Z blowagie $
003 *
004 * This file is part of the iText (R) project.
005 * Copyright (c) 1998-2011 1T3XT BVBA
006 * Authors: Kevin Day, Bruno Lowagie, Paulo Soares, et al.
007 *
008 * This program is free software; you can redistribute it and/or modify
009 * it under the terms of the GNU Affero General Public License version 3
010 * as published by the Free Software Foundation with the addition of the
011 * following permission added to Section 15 as permitted in Section 7(a):
012 * FOR ANY PART OF THE COVERED WORK IN WHICH THE COPYRIGHT IS OWNED BY 1T3XT,
013 * 1T3XT DISCLAIMS THE WARRANTY OF NON INFRINGEMENT OF THIRD PARTY RIGHTS.
014 *
015 * This program is distributed in the hope that it will be useful, but
016 * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
017 * or FITNESS FOR A PARTICULAR PURPOSE.
018 * See the GNU Affero General Public License for more details.
019 * You should have received a copy of the GNU Affero General Public License
020 * along with this program; if not, see http://www.gnu.org/licenses or write to
021 * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
022 * Boston, MA, 02110-1301 USA, or download the license from the following URL:
023 * http://itextpdf.com/terms-of-use/
024 *
025 * The interactive user interfaces in modified source and object code versions
026 * of this program must display Appropriate Legal Notices, as required under
027 * Section 5 of the GNU Affero General Public License.
028 *
029 * In accordance with Section 7(b) of the GNU Affero General Public License,
030 * a covered work must retain the producer line in every PDF that is created
031 * or manipulated using iText.
032 *
033 * You can be released from the requirements of the license by purchasing
034 * a commercial license. Buying such a license is mandatory as soon as you
035 * develop commercial activities involving the iText software without
036 * disclosing the source code of your own applications.
037 * These activities include: offering paid services to customers as an ASP,
038 * serving PDFs on the fly in a web application, shipping iText with a closed
039 * source product.
040 *
041 * For more information, please contact iText Software Corp. at this
042 * address: sales@itextpdf.com
043 */
044package com.itextpdf.text.pdf.parser;
045
046import java.io.IOException;
047
048import com.itextpdf.text.pdf.PdfReader;
049
050/**
051 * Extracts text from a PDF file.
052 * @since       2.1.4
053 */
054public final class PdfTextExtractor {
055
056        /**
057         * This class only contains static methods.
058         */
059        private PdfTextExtractor()  {
060        }
061        
062    /**
063     * Extract text from a specified page using an extraction strategy.
064     * @param reader the reader to extract text from
065     * @param pageNumber the page to extract text from
066     * @param strategy the strategy to use for extracting text
067     * @return the extracted text
068     * @throws IOException if any operation fails while reading from the provided PdfReader
069     * @since 5.0.2
070     */
071    public static String getTextFromPage(PdfReader reader, int pageNumber, TextExtractionStrategy strategy) throws IOException{
072        PdfReaderContentParser parser = new PdfReaderContentParser(reader);
073        return parser.processContent(pageNumber, strategy).getResultantText();
074        
075    }
076    
077    /**
078     * Extract text from a specified page using the default strategy.
079     * <p><strong>Note:</strong> the default strategy is subject to change.  If using a specific strategy
080     * is important, use {@link PdfTextExtractor#getTextFromPage(PdfReader, int, TextExtractionStrategy)}
081     * @param reader the reader to extract text from
082     * @param pageNumber the page to extract text from
083     * @return the extracted text
084     * @throws IOException if any operation fails while reading from the provided PdfReader
085     * @since 5.0.2
086     */
087    public static String getTextFromPage(PdfReader reader, int pageNumber) throws IOException{
088        return getTextFromPage(reader, pageNumber, new LocationTextExtractionStrategy());
089    }
090
091}