Itext创建书签

        由于我经常下载一些pdf格式的电子书，有的时候一些好书下载下来没有书签，读起来感觉没有整体的感觉，所以决定自己写一个小工具，将特定格式的文本解析成为书签，然后保存到pdf格式中。

        整体思路是从豆瓣啊、京东啊、当当啊、亚马逊下面的介绍中可以copy出目录信息，拿《HTTP权威指南》为例：

目录的结构如：

第 1章 HTTP 概述 3
1. 1 HTTP——因特网的多媒体信使 4
1. 2 Web 客户端和服务器 4
1. 3 资源 5
1. 3. 1 媒体类型 6
1. 3. 2 URI 7
1. 3. 3 URL 7
1. 3. 4 URN 8
1. 4 事务 9
1. 4. 1 方法 9
1. 4. 2 状态码 10
1. 4. 3 Web 页面中可以包含多个对象 10
1. 5 报文 11
1. 6 连接 13
1. 6. 1 TCP/IP 13
1. 6. 2 连接、IP 地址及端口号 14
1. 6. 3 使用Telnet 实例 16
1. 7 协议版本 18
1. 8 Web 的结构组件 19
1. 8. 1 代理 19
1. 8. 2 缓存 20
1. 8. 3 网关 20
1. 8. 4 隧道 21
1. 8. 5 Agent 代理 21
1. 9 起始部分的结束语 22
1. 10 更多信息 22
1. 10. 1 HTTP 协议信息 22
1. 10. 2 历史透视 23
1. 10. 3 其他万维网信息 23
第 2章 URL 与资源 25
2. 1 浏览因特网资源 26

每一行后面都有页码，而且是用空格分开的。

处理之后，结果为：

 

 

主要的逻辑为：

package org.fra.pdf.bussiness;

import java.io.BufferedReader;
import java.io.FileOutputStream;
import java.io.IOException;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Stack;

import com.itextpdf.text.DocumentException;
import com.itextpdf.text.pdf.IntHashtable;
import com.itextpdf.text.pdf.PdfArray;
import com.itextpdf.text.pdf.PdfDictionary;
import com.itextpdf.text.pdf.PdfIndirectReference;
import com.itextpdf.text.pdf.PdfName;
import com.itextpdf.text.pdf.PdfNumber;
import com.itextpdf.text.pdf.PdfObject;
import com.itextpdf.text.pdf.PdfReader;
import com.itextpdf.text.pdf.PdfStamper;
import com.itextpdf.text.pdf.PdfString;
import com.itextpdf.text.pdf.SimpleBookmark;

public class AddPdfOutLineFromTxt {
     private Stack <OutlineInfo > parentOutlineStack = new Stack <OutlineInfo >();

     public void createPdf(String destPdf, String sourcePdf,
            BufferedReader bufRead, int pattern) throws IOException,
            DocumentException {

         if (pattern != AddBookmarkConstants.RESERVED_OLD_OUTLINE
                 &&  pattern != AddBookmarkConstants.RESERVED_NONE
                 && pattern != AddBookmarkConstants.RESERVED_FIRST_OUTLINE)
             return;
         // 读入pdf文件
        PdfReader reader = new PdfReader(sourcePdf);

        List <HashMap <String, Object >> outlines = new ArrayList <HashMap <String, Object >>();
         if (pattern == AddBookmarkConstants.RESERVED_OLD_OUTLINE) {
            outlines.addAll(SimpleBookmark.getBookmark(reader));
        } else if (pattern == AddBookmarkConstants.RESERVED_FIRST_OUTLINE) {
            addFirstOutlineReservedPdf(outlines, reader);
        }

        addBookmarks(bufRead, outlines, null, 0);
         // 新建stamper
        PdfStamper stamper = new PdfStamper(reader, new FileOutputStream(
                destPdf));

        stamper.setOutlines(outlines);
        stamper.close();
    }

     private void addBookmarks(BufferedReader bufRead,
            List <HashMap <String, Object >> outlines,
            HashMap <String, Object > preOutline, int preLevel)
             throws IOException {
        String contentFormatLine = null;
        bufRead.mark( 1);
         if ((contentFormatLine = bufRead.readLine()) != null) {
            FormattedBookmark bookmark = parseFormmattedText(contentFormatLine);

            HashMap <String, Object > map = parseBookmarkToHashMap(bookmark);

             int level = bookmark.getLevel();
             // 如果n==m, 那么是同一层的，这个时候，就加到ArrayList中,继续往下面读取
             if (level == preLevel) {
                outlines.add(map);
                addBookmarks(bufRead, outlines, map, level);
            }
             // 如果n>m,那么可以肯定，该行是上一行的孩子，, new 一个kids的arraylist,并且加入到这个arraylist中
             else if (level > preLevel) {
                List <HashMap <String, Object >> kids = new ArrayList <HashMap <String, Object >>();
                kids.add(map);
                preOutline.put( "Kids", kids);
                 // 记录有孩子的outline信息
                parentOutlineStack.push( new OutlineInfo(preOutline, outlines,
                        preLevel));
                addBookmarks(bufRead, kids, map, level);
            }
             // 如果n<m , 那么就是说孩子增加完了，退回到上层，bufRead倒退一行
             else if (level < preLevel) {
                bufRead.reset();
                OutlineInfo obj = parentOutlineStack.pop();
                addBookmarks(bufRead, obj.getOutlines(), obj.getPreOutline(),
                        obj.getPreLevel());
            }

        }
    }

     private HashMap <String, Object > parseBookmarkToHashMap(
            FormattedBookmark bookmark) {
        HashMap <String, Object > map = new HashMap <String, Object >();
        map.put( "Title", bookmark.getTitle());
        map.put( "Action", "GoTo");
        map.put( "Page", bookmark.getPage() + " Fit");
         return map;
    }

     private FormattedBookmark parseFormmattedText(String contentFormatLine) {
        FormattedBookmark bookmark = new FormattedBookmark();
        String title = "";
        String destPage = "";

         // 当没有页码在字符串结尾的时候，一般就是书的名字，如果格式正确的话。
         int lastSpaceIndex = contentFormatLine.lastIndexOf( " ");
         if (lastSpaceIndex == - 1) {
            title = contentFormatLine;
            destPage = "1";
        } else {
            title = contentFormatLine.substring( 0, lastSpaceIndex);
            destPage = contentFormatLine.substring(lastSpaceIndex + 1);
        }

        String[] titleSplit = title.split( " ");
         int dotCount = titleSplit[ 0].split( "\\.").length - 1;

        bookmark.setLevel(dotCount);
        bookmark.setPage(destPage);
        bookmark.setTitle(title);
         return bookmark;
    }

     private void addFirstOutlineReservedPdf(
            List <HashMap <String, Object >> outlines, PdfReader reader) {
        PdfDictionary catalog = reader.getCatalog();
        PdfObject obj = PdfReader.getPdfObjectRelease(catalog
                .get(PdfName.OUTLINES));
         // 没有书签
         if (obj == null || !obj.isDictionary())
             return;
        PdfDictionary outlinesDictionary = (PdfDictionary) obj;
         // 得到第一个书签
        PdfDictionary firstOutline = (PdfDictionary) PdfReader
                .getPdfObjectRelease(outlinesDictionary.get(PdfName.FIRST));

        PdfString titleObj = firstOutline.getAsString((PdfName.TITLE));
        String title = titleObj.toUnicodeString();

        PdfArray dest = firstOutline.getAsArray(PdfName.DEST);

         if (dest == null) {
            PdfDictionary action = (PdfDictionary) PdfReader
                    .getPdfObjectRelease(firstOutline.get(PdfName.A));
             if (action != null) {
                 if (PdfName.GOTO.equals(PdfReader.getPdfObjectRelease(action
                        .get(PdfName.S)))) {
                    dest = (PdfArray) PdfReader.getPdfObjectRelease(action
                            .get(PdfName.D));
                }
            }
        }
        String destStr = parseDestString(dest, reader);

        String[] decodeStr = destStr.split( " ");
         int num = Integer.valueOf(decodeStr[ 0]);
        HashMap <String, Object > map = new HashMap <String, Object >();
        map.put( "Title", title);
        map.put( "Action", "GoTo");
        map.put( "Page", num + " Fit");

        outlines.add(map);
    }

     private String parseDestString(PdfArray dest, PdfReader reader) {
        String destStr = "";
         if (dest.isString()) {
            destStr = dest.toString();
        } else if (dest.isName()) {
            destStr = PdfName.decodeName(dest.toString());
        } else if (dest.isArray()) {
            IntHashtable pages = new IntHashtable();
             int numPages = reader.getNumberOfPages();
             for ( int k = 1; k < = numPages; ++k) {
                pages.put(reader.getPageOrigRef(k).getNumber(), k);
                reader.releasePage(k);
            }

            destStr = makeBookmarkParam((PdfArray) dest, pages);
        }
         return destStr;
    }

     private String makeBookmarkParam(PdfArray dest, IntHashtable pages) {
        StringBuffer s = new StringBuffer();
        PdfObject obj = dest.getPdfObject( 0);
         if (obj.isNumber()) {
            s.append(((PdfNumber) obj).intValue() + 1);
        } else {
            s.append(pages.get(getNumber((PdfIndirectReference) obj)));
        }
        s.append( ' ').append(dest.getPdfObject( 1).toString().substring( 1));
         for ( int k = 2; k < dest.size(); ++k) {
            s.append( ' ').append(dest.getPdfObject(k).toString());
        }
         return s.toString();
    }

     private int getNumber(PdfIndirectReference indirect) {
        PdfDictionary pdfObj = (PdfDictionary) PdfReader
                .getPdfObjectRelease(indirect);
         if (pdfObj.contains(PdfName.TYPE)
                 && pdfObj.get(PdfName.TYPE).equals(PdfName.PAGES)
                 && pdfObj.contains(PdfName.KIDS)) {
            PdfArray kids = (PdfArray) pdfObj.get(PdfName.KIDS);
            indirect = (PdfIndirectReference) kids.getPdfObject( 0);
        }
         return indirect.getNumber();
    }
}

 

 

 

转载于:https://www.cnblogs.com/kakafra/archive/2013/03/05/2944747.html