使用java為pdf添加書簽的方法(pdf書簽制作)

2019-11-26 15:40:39

字體：大中小

來源：轉載

供稿：網友

由于我經常下載一些pdf格式的電子書，有的時候一些好書下載下來沒有書簽，讀起來感覺沒有整體的感覺，所以決定自己寫一個小工具，將特定格式的文本解析成為書簽，然后保存到pdf格式中。
整體思路是從豆瓣啊、京東啊、當當啊、亞馬遜下面的介紹中可以copy出目錄信息，拿《HTTP權威指南》為例：
目錄的結構如：

復制代碼代碼如下:

第1章 HTTP 概述 3
1 HTTP――因特網的多媒體信使 4
2 Web 客戶端和服務器 4
3 資源 5
3.1 媒體類型 6
3.2 URI 7
3.3 URL 7
3.4 URN 8
4 事務 9
4.1 方法 9
4.2 狀態碼 10
4.3 Web 頁面中可以包含多個對象 10
5 報文 11
6 連接 13

每一行后面都有頁碼，而且是用空格分開的。
處理之后，結果為：

主要的邏輯為：

復制代碼代碼如下:

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;
        // 讀入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, 那么是同一層的，這個時候，就加到ArrayList中,繼續往下面讀取
            if (level == preLevel) {
                outlines.add(map);
                addBookmarks(bufRead, outlines, map, level);
            }
            // 如果n>m,那么可以肯定，該行是上一行的孩子，, new 一個kids的arraylist,并且加入到這個arraylist中
            else if (level > preLevel) {
                List<HashMap<String, Object>> kids = new ArrayList<HashMap<String, Object>>();
                kids.add(map);
                preOutline.put("Kids", kids);
                // 記錄有孩子的outline信息
                parentOutlineStack.push(new OutlineInfo(preOutline, outlines,
                        preLevel));
                addBookmarks(bufRead, kids, map, level);
            }
            // 如果n<m , 那么就是說孩子增加完了，退回到上層，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 = "";

        // 當沒有頁碼在字符串結尾的時候，一般就是書的名字，如果格式正確的話。
        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("http://.").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));
        // 沒有書簽
        if (obj == null || !obj.isDictionary())
            return;
        PdfDictionary outlinesDictionary = (PdfDictionary) obj;
        // 得到第一個書簽
        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();
    }
}