001/*
002 * $Id: PdfPageLabels.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: 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;
045
046import java.io.IOException;
047import java.util.Arrays;
048import java.util.HashMap;
049
050import com.itextpdf.text.ExceptionConverter;
051import com.itextpdf.text.error_messages.MessageLocalization;
052import com.itextpdf.text.factories.RomanAlphabetFactory;
053import com.itextpdf.text.factories.RomanNumberFactory;
054
055/** Page labels are used to identify each
056 * page visually on the screen or in print.
057 * @author  Paulo Soares
058 */
059public class PdfPageLabels {
060
061    /** Logical pages will have the form 1,2,3,...
062     */
063    public static final int DECIMAL_ARABIC_NUMERALS = 0;
064    /** Logical pages will have the form I,II,III,IV,...
065     */
066    public static final int UPPERCASE_ROMAN_NUMERALS = 1;
067    /** Logical pages will have the form i,ii,iii,iv,...
068     */
069    public static final int LOWERCASE_ROMAN_NUMERALS = 2;
070    /** Logical pages will have the form of uppercase letters
071     * (A to Z for the first 26 pages, AA to ZZ for the next 26, and so on)
072     */
073    public static final int UPPERCASE_LETTERS = 3;
074    /** Logical pages will have the form of uppercase letters
075     * (a to z for the first 26 pages, aa to zz for the next 26, and so on)
076     */
077    public static final int LOWERCASE_LETTERS = 4;
078    /** No logical page numbers are generated but fixed text may
079     * still exist
080     */
081    public static final int EMPTY = 5;
082    /** Dictionary values to set the logical page styles
083     */
084    static PdfName numberingStyle[] = new PdfName[]{PdfName.D, PdfName.R,
085                new PdfName("r"), PdfName.A, new PdfName("a")};
086    /** The sequence of logical pages. Will contain at least a value for page 1
087     */
088    private HashMap<Integer, PdfDictionary> map;
089
090    /** Creates a new PdfPageLabel with a default logical page 1
091     */
092    public PdfPageLabels() {
093        map = new HashMap<Integer, PdfDictionary>();
094        addPageLabel(1, DECIMAL_ARABIC_NUMERALS, null, 1);
095    }
096
097    /** Adds or replaces a page label.
098     * @param page the real page to start the numbering. First page is 1
099     * @param numberStyle the numbering style such as LOWERCASE_ROMAN_NUMERALS
100     * @param text the text to prefix the number. Can be <CODE>null</CODE> or empty
101     * @param firstPage the first logical page number
102     */
103    public void addPageLabel(int page, int numberStyle, String text, int firstPage) {
104        if (page < 1 || firstPage < 1)
105            throw new IllegalArgumentException(MessageLocalization.getComposedMessage("in.a.page.label.the.page.numbers.must.be.greater.or.equal.to.1"));
106        PdfDictionary dic = new PdfDictionary();
107        if (numberStyle >= 0 && numberStyle < numberingStyle.length)
108            dic.put(PdfName.S, numberingStyle[numberStyle]);
109        if (text != null)
110            dic.put(PdfName.P, new PdfString(text, PdfObject.TEXT_UNICODE));
111        if (firstPage != 1)
112            dic.put(PdfName.ST, new PdfNumber(firstPage));
113        map.put(Integer.valueOf(page - 1), dic);
114    }
115
116    /** Adds or replaces a page label. The first logical page has the default
117     * of 1.
118     * @param page the real page to start the numbering. First page is 1
119     * @param numberStyle the numbering style such as LOWERCASE_ROMAN_NUMERALS
120     * @param text the text to prefix the number. Can be <CODE>null</CODE> or empty
121     */
122    public void addPageLabel(int page, int numberStyle, String text) {
123        addPageLabel(page, numberStyle, text, 1);
124    }
125
126    /** Adds or replaces a page label. There is no text prefix and the first
127     * logical page has the default of 1.
128     * @param page the real page to start the numbering. First page is 1
129     * @param numberStyle the numbering style such as LOWERCASE_ROMAN_NUMERALS
130     */
131    public void addPageLabel(int page, int numberStyle) {
132        addPageLabel(page, numberStyle, null, 1);
133    }
134
135    /** Adds or replaces a page label.
136     */
137    public void addPageLabel(PdfPageLabelFormat format) {
138        addPageLabel(format.physicalPage, format.numberStyle, format.prefix, format.logicalPage);
139    }
140
141    /** Removes a page label. The first page label can not be removed, only changed.
142     * @param page the real page to remove
143     */
144    public void removePageLabel(int page) {
145        if (page <= 1)
146            return;
147        map.remove(Integer.valueOf(page - 1));
148    }
149
150    /** Gets the page label dictionary to insert into the document.
151     * @return the page label dictionary
152     */
153    PdfDictionary getDictionary(PdfWriter writer) {
154        try {
155            return PdfNumberTree.writeTree(map, writer);
156        }
157        catch (IOException e) {
158            throw new ExceptionConverter(e);
159        }
160    }
161
162    /**
163     * Retrieves the page labels from a PDF as an array of String objects.
164     * @param reader a PdfReader object that has the page labels you want to retrieve
165     * @return  a String array or <code>null</code> if no page labels are present
166     */
167    public static String[] getPageLabels(PdfReader reader) {
168
169                int n = reader.getNumberOfPages();
170
171        PdfDictionary dict = reader.getCatalog();
172                PdfDictionary labels = (PdfDictionary)PdfReader.getPdfObjectRelease(dict.get(PdfName.PAGELABELS));
173        if (labels == null)
174            return null;
175
176                String[] labelstrings = new String[n];
177
178                HashMap<Integer, PdfObject> numberTree = PdfNumberTree.readTree(labels);
179
180                int pagecount = 1;
181                Integer current;
182                String prefix = "";
183                char type = 'D';
184                for (int i = 0; i < n; i++) {
185                        current = Integer.valueOf(i);
186                        if (numberTree.containsKey(current)) {
187                                PdfDictionary d = (PdfDictionary)PdfReader.getPdfObjectRelease(numberTree.get(current));
188                                if (d.contains(PdfName.ST)) {
189                                        pagecount = ((PdfNumber)d.get(PdfName.ST)).intValue();
190                                }
191                                else {
192                                        pagecount = 1;
193                                }
194                                if (d.contains(PdfName.P)) {
195                                        prefix = ((PdfString)d.get(PdfName.P)).toUnicodeString();
196                                }
197                                if (d.contains(PdfName.S)) {
198                                        type = ((PdfName)d.get(PdfName.S)).toString().charAt(1);
199                                }
200                        }
201                        switch(type) {
202                        default:
203                                labelstrings[i] = prefix + pagecount;
204                                break;
205                        case 'R':
206                                labelstrings[i] = prefix + RomanNumberFactory.getUpperCaseString(pagecount);
207                                break;
208                        case 'r':
209                                labelstrings[i] = prefix + RomanNumberFactory.getLowerCaseString(pagecount);
210                                break;
211                        case 'A':
212                                labelstrings[i] = prefix + RomanAlphabetFactory.getUpperCaseString(pagecount);
213                                break;
214                        case 'a':
215                                labelstrings[i] = prefix + RomanAlphabetFactory.getLowerCaseString(pagecount);
216                                break;
217                        }
218                        pagecount++;
219                }
220                return labelstrings;
221    }
222
223    /**
224     * Retrieves the page labels from a PDF as an array of {@link PdfPageLabelFormat} objects.
225     * @param reader a PdfReader object that has the page labels you want to retrieve
226     * @return  a PdfPageLabelEntry array, containing an entry for each format change
227     * or <code>null</code> if no page labels are present
228     */
229    public static PdfPageLabelFormat[] getPageLabelFormats(PdfReader reader) {
230        PdfDictionary dict = reader.getCatalog();
231        PdfDictionary labels = (PdfDictionary)PdfReader.getPdfObjectRelease(dict.get(PdfName.PAGELABELS));
232        if (labels == null)
233            return null;
234        HashMap<Integer, PdfObject> numberTree = PdfNumberTree.readTree(labels);
235        Integer numbers[] = new Integer[numberTree.size()];
236        numbers = numberTree.keySet().toArray(numbers);
237        Arrays.sort(numbers);
238        PdfPageLabelFormat[] formats = new PdfPageLabelFormat[numberTree.size()];
239        String prefix;
240        int numberStyle;
241        int pagecount;
242        for (int k = 0; k < numbers.length; ++k) {
243            Integer key = numbers[k];
244            PdfDictionary d = (PdfDictionary)PdfReader.getPdfObjectRelease(numberTree.get(key));
245            if (d.contains(PdfName.ST)) {
246                pagecount = ((PdfNumber)d.get(PdfName.ST)).intValue();
247            } else {
248                pagecount = 1;
249            }
250            if (d.contains(PdfName.P)) {
251                prefix = ((PdfString)d.get(PdfName.P)).toUnicodeString();
252            } else {
253                prefix = "";
254            }
255            if (d.contains(PdfName.S)) {
256                char type = ((PdfName)d.get(PdfName.S)).toString().charAt(1);
257                switch(type) {
258                    case 'R': numberStyle = UPPERCASE_ROMAN_NUMERALS; break;
259                    case 'r': numberStyle = LOWERCASE_ROMAN_NUMERALS; break;
260                    case 'A': numberStyle = UPPERCASE_LETTERS; break;
261                    case 'a': numberStyle = LOWERCASE_LETTERS; break;
262                    default: numberStyle = DECIMAL_ARABIC_NUMERALS; break;
263                }
264            } else {
265                numberStyle = EMPTY;
266            }
267            formats[k] = new PdfPageLabelFormat(key.intValue()+1, numberStyle, prefix, pagecount);
268        }
269        return formats;
270    }
271
272    public static class PdfPageLabelFormat {
273
274        public int physicalPage;
275        public int numberStyle;
276        public String prefix;
277        public int logicalPage;
278
279        /** Creates a page label format.
280         * @param physicalPage the real page to start the numbering. First page is 1
281         * @param numberStyle the numbering style such as LOWERCASE_ROMAN_NUMERALS
282         * @param prefix the text to prefix the number. Can be <CODE>null</CODE> or empty
283         * @param logicalPage the first logical page number
284         */
285        public PdfPageLabelFormat(int physicalPage, int numberStyle, String prefix, int logicalPage) {
286            this.physicalPage = physicalPage;
287            this.numberStyle = numberStyle;
288            this.prefix = prefix;
289            this.logicalPage = logicalPage;
290        }
291    }
292}