duplicates - 如果 pdf 包含“重复的书签”，如何使用 itext 7 按书签拆分 pdf 文件

Question

我正在尝试使用 itext7 按其书签拆分 pdf。

问题：如果 Pdf 在大纲树的其他地方有相同的书签，它就会被覆盖并且无法拆分。

重现问题的示例代码：

public void walkOutlines(PdfOutline outline, Map<String, PdfObject> names, PdfDocument pdfDocument,List<String>titles,List<Integer>pageNum) { //----------loop traversing all paths
        
    for (PdfOutline child : outline.getAllChildren()){
        if(child.getDestination() != null) {
            prepareIndexFile(child,names,pdfDocument,titles,pageNum,list);
        }
    }
}

//------------Getting pageNumbers from outlines
public void prepareIndexFile(PdfOutline outline, Map<String, PdfObject> names, PdfDocument pdfDocument,List<String>titles,List<Integer>pageNum) {
        
        String title = outline.getTitle();
        
        PdfDestination pdfDestination = outline.getDestination();
        String pdfStr = ((PdfString)pdfDestination.getPdfObject()).toUnicodeString();
        PdfArray array = (PdfArray) names.get(pdfStr);
        PdfObject pdfObj = array != null ? array.get(0) : null;
        
        Integer pageNumber = pdfDocument.getPageNumber((PdfDictionary)pdfObj);
        
        titles.add(title);
        pageNum.add(pageNumber);
        
        
        if(outline.getAllChildren().size() > 0) {
            
            for (PdfOutline child : outline.getAllChildren()){
                prepareIndexFile(child,names,pdfDocument,titles,pageNum);
            }
            
        }
        
}

public boolean splitPdf(String inputFile, final String outputFolder) {

        boolean splitSuccess = true;
        PdfDocument pdfDoc = null;
        try {
            PdfReader pdfReaderNew = new PdfReader(inputFile);
            pdfDoc = new PdfDocument(pdfReaderNew);
            
            final List<String> titles = new ArrayList<String>();
            List<Integer> pageNum = new ArrayList<Integer>();
            
            PdfNameTree destsTree = pdfDoc.getCatalog().getNameTree(PdfName.Dests);
            Map<String, PdfObject> names = destsTree.getNames();//--------------------------------------Core logic for getting names
            PdfOutline root = pdfDoc.getOutlines(false);//--------------------------------------Core logic for getting outlines
            
            walkOutlines(root,names, pdfDoc, titles, pageNum,content);  //------Logic to get bookmarks and pageNumbers
            

            if (titles == null || titles.size()==0) {
                splitSuccess = false;
            }else {                                                             //------Proceed if it has bookmarks
                
                for(int i=0;i<titles.size();i++) {
                     
                     String title = titles.get(i);
                     String startPageNmStr =""+pageNum.get(i);
                     int startPage = Integer.parseInt(startPageNmStr);
                     
                     int endPage = startPage;
                     
                     if(i == titles.size() - 1) {
                         endPage = pdfDoc.getNumberOfPages();
                     }else {
                         int nextPage =  pageNum.get(i+1);
                         if(nextPage > startPage) {
                             endPage = nextPage - 1;
                         }else {
                             endPage = nextPage;
                         }
                     }
                     
                     String outFileName = outputFolder + File.separator + getFileName(title) + ".pdf";
                     PdfWriter pdfWriter = new PdfWriter(outFileName);
                    
                     PdfDocument newDocument = new PdfDocument(pdfWriter, new DocumentProperties().setEventCountingMetaInfo(null));
                     pdfDoc.copyPagesTo(startPage, endPage, newDocument);
                     newDocument.close();
                     pdfWriter.close();
                }
            }
        }catch(Exception e){
            //---log
        }       
}

找到根本原因：在PdfNameTree items.put(name.toUnicodeString(), names.get(k));

如何克服这个问题？

提前致谢

标签： duplicatesitext7bookmarks

Answer 1

这部分代码：

PdfDestination pdfDestination = outline.getDestination();
String pdfStr = ((PdfString)pdfDestination.getPdfObject()).toUnicodeString();
PdfArray array = (PdfArray) names.get(pdfStr);
PdfObject pdfObj = array != null ? array.get(0) : null;

Integer pageNumber = pdfDocument.getPageNumber((PdfDictionary)pdfObj);

不考虑目的地可以不命名并显式引用页面的情况。

所以代码需要改编成如下代码：

PdfDestination pdfDestination = outline.getDestination();
PdfObject pdfObj = null;
if (pdfDestination.getPdfObject().isString()) {
    String pdfStr = ((PdfString) pdfDestination.getPdfObject()).toUnicodeString();
    PdfArray array = (PdfArray) names.get(pdfStr);
    if (array != null) {
        pdfObj = array.get(0);
    }
} else if (pdfDestination.getPdfObject().isArray() && ((PdfArray)pdfDestination.getPdfObject()).get(0).isDictionary()) {
    pdfObj = ((PdfArray)pdfDestination.getPdfObject()).get(0);
}

Integer pageNumber = pdfDocument.getPageNumber((PdfDictionary)pdfObj);

此外，如果要获取包含父链的完整标题名称，则需要替换String title = outline.getTitle();为以下代码：

String title = outline.getTitle();
PdfOutline parentChain = outline.getParent();
while (parentChain != null) {
    title = parentChain.getTitle() + "." + title;
    parentChain = parentChain.getParent();
}

结果，我在输出目录中得到了 6 个文件，其中 5 个文件每个 1 页，一个文件 4 页。

完整代码：

public void walkOutlines(PdfOutline outline, Map<String, PdfObject> names, PdfDocument pdfDocument,
        java.util.List<String>titles,java.util.List<Integer>pageNum) { //----------loop traversing all paths

    for (PdfOutline child : outline.getAllChildren()){
        if(child.getDestination() != null) {
            prepareIndexFile(child,names,pdfDocument,titles,pageNum);
        }
    }
}

//------------Getting pageNumbers from outlines
public void prepareIndexFile(PdfOutline outline, Map<String, PdfObject> names, PdfDocument pdfDocument,
        java.util.List<String>titles,java.util.List<Integer>pageNum) {

    String title = outline.getTitle();
    PdfOutline parentChain = outline.getParent();
    while (parentChain != null) {
        title = parentChain.getTitle() + "." + title;
        parentChain = parentChain.getParent();
    }

    PdfDestination pdfDestination = outline.getDestination();
    PdfObject pdfObj = null;
    if (pdfDestination.getPdfObject().isString()) {
        String pdfStr = ((PdfString) pdfDestination.getPdfObject()).toUnicodeString();
        PdfArray array = (PdfArray) names.get(pdfStr);
        if (array != null) {
            pdfObj = array.get(0);
        }
    } else if (pdfDestination.getPdfObject().isArray() && ((PdfArray)pdfDestination.getPdfObject()).get(0).isDictionary()) {
        pdfObj = ((PdfArray)pdfDestination.getPdfObject()).get(0);
    }

    Integer pageNumber = pdfDocument.getPageNumber((PdfDictionary)pdfObj);

    titles.add(title);
    pageNum.add(pageNumber);


    if(outline.getAllChildren().size() > 0) {

        for (PdfOutline child : outline.getAllChildren()){
            prepareIndexFile(child,names,pdfDocument,titles,pageNum);
        }

    }

}

public void splitPdf(String inputFile, final String outputFolder) {

    boolean splitSuccess = true;
    PdfDocument pdfDoc = null;
    try {
        PdfReader pdfReaderNew = new PdfReader(inputFile);
        pdfDoc = new PdfDocument(pdfReaderNew);

        final java.util.List<String> titles = new ArrayList<String>();
        java.util.List<Integer> pageNum = new ArrayList<Integer>();

        PdfNameTree destsTree = pdfDoc.getCatalog().getNameTree(PdfName.Dests);
        Map<String, PdfObject> names = destsTree.getNames();//--------------------------------------Core logic for getting names
        PdfOutline root = pdfDoc.getOutlines(false);//--------------------------------------Core logic for getting outlines

        walkOutlines(root,names, pdfDoc, titles, pageNum);  //------Logic to get bookmarks and pageNumbers


        if (titles == null || titles.size()==0) {
            splitSuccess = false;
        }else {                                                             //------Proceed if it has bookmarks

            for(int i=0;i<titles.size();i++) {

                String title = titles.get(i);
                String startPageNmStr =""+pageNum.get(i);
                int startPage = Integer.parseInt(startPageNmStr);

                int endPage = startPage;

                if(i == titles.size() - 1) {
                    endPage = pdfDoc.getNumberOfPages();
                }else {
                    int nextPage =  pageNum.get(i+1);
                    if(nextPage > startPage) {
                        endPage = nextPage - 1;
                    }else {
                        endPage = nextPage;
                    }
                }

                String outFileName = outputFolder + File.separator + title + ".pdf";
                PdfWriter pdfWriter = new PdfWriter(outFileName);

                PdfDocument newDocument = new PdfDocument(pdfWriter, new DocumentProperties().setEventCountingMetaInfo(null));
                pdfDoc.copyPagesTo(startPage, endPage, newDocument);
                newDocument.close();
                pdfWriter.close();
            }
        }
    }catch(IOException e){
        System.out.println(e);
    }
}