使用Apache POI將HSSF（excel）嵌入到HSLF（ppt）中

Question

我想使用apache poi將Excel工作表嵌入到presentation（PPT）中。 我們應該怎么做？ 如果有人知道，請幫助我。

Answer 1

這花了我一段時間來弄清楚零件是如何在一起的...

嵌入可以通過兩種方式完成：

• 通過更新已經嵌入的工作表
  • Pro：只需調用ObjectData.get/setData()即可完成
  • 缺點：如果要嵌入多個OLE對象怎么辦？
• 或者您可以從頭開始嵌入元素（請參見下文）

和往常一樣，當我試圖弄清楚如何實現某些POI功能時，我將結果與Libre Office文件進行了比較，在這種情況下，必須創建/修改一些部分：

• 在Powerpoint對象中...
  • 嵌入對象的二進制數據存儲為根級記錄。 大多數根記錄都與位置有關 ，因此在創建新記錄（例如幻燈片）時，您需要重新計算其所有偏移量
  • 通過將記錄嵌入Document記錄中來引用二進制數據記錄
  • ...並對其進行進一步的混淆，實際形狀對象再次引用了此文檔參考
• 在嵌入式工作表的POIF中...
  • 需要創建一個Ole Stream條目
  • 並且根節點必須具有嵌入式文檔類型的class-id
  • 除此之外，嵌入式工作簿對象沒有必要的更改，數據本身是一個自包含的excel文件

此外，我使用了兩個實用的信息類： BiffViewer和POIFSLister 。

由於這只是概念證明，因此還遠遠不夠。 要進一步修改嵌入元素的表示形式，您需要參考規范 。

為嵌入式對象創建預覽圖像仍然沒有解決的問題。 您可能希望使用中性圖像，只要用戶激活（雙擊）ole對象，該圖像就會被替換。另一種方法是使用jodconverter ，但比POI方法有點無意義。 。

（已通過POI3.9 / Libre Office 4.0 / MS Excel Viewer / MS Office 2003測試）

import java.awt.geom.Rectangle2D;
import java.io.*;
import java.lang.reflect.Field;

import org.apache.poi.POIDocument;
import org.apache.poi.ddf.*;
import org.apache.poi.hpsf.ClassID;
import org.apache.poi.hslf.HSLFSlideShow;
import org.apache.poi.hslf.exceptions.HSLFException;
import org.apache.poi.hslf.model.*;
import org.apache.poi.hslf.model.Picture;
import org.apache.poi.hslf.model.Slide;
import org.apache.poi.hslf.record.*;
import org.apache.poi.hslf.usermodel.*;
import org.apache.poi.hssf.usermodel.*;
import org.apache.poi.hwpf.HWPFDocument;
import org.apache.poi.hwpf.usermodel.*;
import org.apache.poi.poifs.filesystem.*;
import org.apache.poi.util.*;

public class PoiOleXlsInPpt {
    static final OleType EXCEL97      = new OleType("{00020820-0000-0000-C000-000000000046}");
    static final OleType EXCEL95      = new OleType("{00020810-0000-0000-C000-000000000046}");
    static final OleType WORD97       = new OleType("{00020906-0000-0000-C000-000000000046}");
    static final OleType WORD95       = new OleType("{00020900-0000-0000-C000-000000000046}");
    static final OleType POWERPOINT97 = new OleType("{64818D10-4F9B-11CF-86EA-00AA00B929E8}");
    static final OleType POWERPOINT95 = new OleType("{EA7BAE70-FB3B-11CD-A903-00AA00510EA3}");

    static class OleType {
        final String classId;
        OleType(String classId) {
            this.classId = classId;
        }
        ClassID getClassID() {
            ClassID cls = new ClassID();
            byte clsBytes[] = cls.getBytes();
            String clsStr = classId.replaceAll("[{}-]", "");
            for (int i=0; i<clsStr.length(); i+=2) {
                clsBytes[i/2] = (byte)Integer.parseInt(clsStr.substring(i, i+2), 16);
            }
            return cls;
        }
    }

    public static void main(String[] args) throws Exception {
        HSLFSlideShow _hslfSlideShow = HSLFSlideShow.create(); 
        SlideShow ppt = new SlideShow(_hslfSlideShow);

        OLEShape oleShape1 = createOLEShape(getSampleWorkbook1(), ppt, _hslfSlideShow, EXCEL97);
        oleShape1.setAnchor(new Rectangle2D.Double(100,100,100,100));
        OLEShape oleShape2 = createOLEShape(getSampleWorkbook2(), ppt, _hslfSlideShow, EXCEL97);
        oleShape2.setAnchor(new Rectangle2D.Double(300,300,100,100));
        OLEShape oleShape3 = createOLEShape(getSampleDocument(), ppt, _hslfSlideShow, WORD97);
        oleShape3.setAnchor(new Rectangle2D.Double(300,100,100,100));

        // create and link visuals to the ole data
        Slide slide = ppt.createSlide();
        slide.addShape(oleShape1);
        slide.addShape(oleShape2);
        slide.addShape(oleShape3);

        FileOutputStream fos = new FileOutputStream("ole_xls_in_ppt_out2.ppt");
        ppt.write(fos);
        fos.close();
    }

    static OLEShape createOLEShape(
          POIDocument sample
        , SlideShow ppt
        , HSLFSlideShow _hslfSlideShow
        , OleType oleType
    ) throws IOException {
        // generate a preview image
        int prevIdx = generatePreview(ppt, sample);

        // add the data to the SlideShow
        ExEmbed eeEmbed = addOleDataToDocumentRecord(ppt);
        ExOleObjStg exOleObjStg = addOleDataToRootRecords(_hslfSlideShow, sample, oleType);
        eeEmbed.getExOleObjAtom().setObjStgDataRef(exOleObjStg.getPersistId());

        OLEShape oleShape = new OLEShape(prevIdx);
        linkOleDataToShape(oleShape, eeEmbed);

        return oleShape;
    }

    static POIDocument getSampleWorkbook1() {
        HSSFWorkbook wb = new HSSFWorkbook();
        HSSFSheet sheet = wb.createSheet();
        sheet.createRow(1).createCell(1).setCellValue("First Workbook");
        return wb;
    }

    static POIDocument getSampleWorkbook2() {
        HSSFWorkbook wb = new HSSFWorkbook();
        HSSFSheet sheet = wb.createSheet();
        sheet.createRow(1).createCell(1).setCellValue("Second Workbook");
        return wb;
    }

    // the sample document has apparently a problem,
    // i.e. word inside ms powerpoint crashed, and libre office doesn't display the text
    // it was just a test, if embedding elements != Excel works
    // in case HWPF is interesting to you, you probably know anyway, where the error below is ...
    static POIDocument getSampleDocument() throws IOException {
        FileInputStream fis = new FileInputStream("src/test/resources/empty.doc");
        HWPFDocument doc = new HWPFDocument(fis);
        fis.close();
        Range range = doc.getRange();
        CharacterRun run1 = range.insertAfter("Sample text");    
        run1.setFontSize(11);       
        return doc;
    }

    /**
     * Generates a modified version of the sample element, which
     * contains embedding informations
     */
    static byte[] wrapOleData(POIDocument oleData, OleType oleType) {
        try {
            ByteArrayOutputStream bos = new ByteArrayOutputStream();
            oleData.write(bos);

            ByteArrayInputStream bis = new ByteArrayInputStream(bos.toByteArray());
            bos.reset();
            POIFSFileSystem poifs = new POIFSFileSystem(bis);

            final String OLESTREAM_NAME = "\u0001Ole";
            DirectoryNode root = poifs.getRoot();
            if (!root.hasEntry(OLESTREAM_NAME)) {
                // the following data was taken from an example libre office document
                // beside this "\u0001Ole" record there were several other records, e.g. CompObj,
                // OlePresXXX, but it seems, that they aren't neccessary
                byte oleBytes[] = { 1, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
                poifs.createDocument(new ByteArrayInputStream(oleBytes), OLESTREAM_NAME);
            }

            // need to set storage clsid, otherwise embedded object is not recognized
            root.setStorageClsid(oleType.getClassID());

            poifs.writeFilesystem(bos);
            return bos.toByteArray();
        } catch (IOException e) {
            throw new RuntimeException("wth?!", e);
        }
    }


    /**
     * to be defined, how to create a preview image
     * for a start, I've taken just a dummy image, which will be
     * replaced, when the user activates the ole object
     *
     * not really an alternativ:
     * http://stackoverflow.com/questions/16704624/how-to-print-a-workbook-file-made-using-apache-poi-and-java
     * 
     * @return image index of the preview image 
     */
    static int generatePreview(SlideShow ppt, POIDocument oleData) {
        try {
            FileInputStream fis = new FileInputStream("src/test/resources/dilbert-2011-09-28-powerpoint.jpg");
            byte previewImg[] = IOUtils.toByteArray(fis);
            fis.close();
            return ppt.addPicture(previewImg, Picture.JPEG);
        } catch (IOException e) {
            throw new RuntimeException("not really?", e);
        }
    }

    static ExEmbed addOleDataToDocumentRecord(SlideShow ppt) {
        // taken from SlideShow.addControl()
        Document _documentRecord = ppt.getDocumentRecord();
        ExObjList lst = _documentRecord.getExObjList();
        if (lst == null) {
            lst = new ExObjList();
            _documentRecord.addChildAfter(lst, _documentRecord.getDocumentAtom());
            try {
                Field f = Document.class.getDeclaredField("exObjList");
                f.setAccessible(true);
                f.set(_documentRecord, lst);
            } catch (Exception e) {
                throw new RuntimeException("not here", e);
            }
        }
        ExObjListAtom objAtom = lst.getExObjListAtom();
        // increment the object ID seed
        int objectId = (int) objAtom.getObjectIDSeed() + 1;
        objAtom.setObjectIDSeed(objectId);

        ExEmbed exEmbed = new ExEmbed();
        // remove unneccessary infos, so we don't need to specify the type
        // of the ole object multiple times
        Record children[] = exEmbed.getChildRecords();
        exEmbed.removeChild(children[2]);
        exEmbed.removeChild(children[3]);
        exEmbed.removeChild(children[4]);

        ExEmbedAtom eeEmbed = exEmbed.getExEmbedAtom();
        try {
            Field f = ExEmbedAtom.class.getDeclaredField("_data");
            f.setAccessible(true);
            f.set(eeEmbed, new byte[]{0,0,0,0,1/*CantLockServerB*/,0,0,0});
            // oops, there seems to be an error in the default constructor ...
            // should be 8 and not 7 bytes
            setRecordLength(eeEmbed, 8);
        } catch (Exception e) {
            throw new RuntimeException("trust me ;)", e);
        }

        ExOleObjAtom eeAtom = exEmbed.getExOleObjAtom();
        eeAtom.setObjID(objectId);
        eeAtom.setDrawAspect(ExOleObjAtom.DRAW_ASPECT_VISIBLE);
        eeAtom.setType(ExOleObjAtom.TYPE_EMBEDDED);
        // eeAtom.setSubType(ExOleObjAtom.SUBTYPE_EXCEL);
        // should be ignored?!?, see MS-PPT ExOleObjAtom, but Libre Office sets it ...
        eeAtom.setOptions(1226240);

        lst.addChildAfter(exEmbed, objAtom);

        return exEmbed;
    }

    static ExOleObjStg addOleDataToRootRecords(
          HSLFSlideShow _hslfSlideShow
        , POIDocument oleData
        , OleType oleType
    ) throws IOException {
        ExOleObjStg exOleObjStg = new ExOleObjStg();
        int slideRecordPos = _hslfSlideShow.appendRootLevelRecord(exOleObjStg);
        exOleObjStg.setPersistId(slideRecordPos);
        exOleObjStg.setData(wrapOleData(oleData, oleType));

        // taken from SlideShow.createSlide
        Record _records[] = _hslfSlideShow.getRecords();

        // Add the new OLE record into the PersistPtr stuff
        int offset = 0;
        int slideOffset = 0;
        PersistPtrHolder ptr = null;
        UserEditAtom usr = null;
        for (int i = 0; i < _records.length; i++) {
            Record record = _records[i];
            ByteArrayOutputStream out = new ByteArrayOutputStream();
            try {
                record.writeOut(out);
            } catch (IOException e) {
                throw new HSLFException(e);
            }

            // Grab interesting records as they come past
            if (_records[i].getRecordType() == RecordTypes.PersistPtrIncrementalBlock.typeID) {
                ptr = (PersistPtrHolder) _records[i];
            }
            if (_records[i].getRecordType() == RecordTypes.UserEditAtom.typeID) {
                usr = (UserEditAtom) _records[i];
            }

            if (i == slideRecordPos) {
                slideOffset = offset;
            }
            offset += out.size();
        }

        // the ole objects needs to know its position within
        // the root records, because it will be later accessed
        // via its index from the shape
        int psrId = usr.getMaxPersistWritten() + 1;
        exOleObjStg.setPersistId(psrId);

        // Last view is now of the slide
        usr.setLastViewType((short) UserEditAtom.LAST_VIEW_SLIDE_VIEW);
        usr.setMaxPersistWritten(psrId); // increment the number of persit objects

        // Add the new slide into the last PersistPtr
        // (Also need to tell it where it is)
        exOleObjStg.setLastOnDiskOffset(slideOffset);
        ptr.addSlideLookup(psrId, slideOffset);

        return exOleObjStg;
    }

    static void linkOleDataToShape(OLEShape oleShape, ExEmbed exEmbed) {
        oleShape.setEscherProperty(EscherProperties.BLIP__PICTUREID, exEmbed.getExOleObjAtom().getObjID());

        EscherSpRecord spRecord = oleShape.getSpContainer().getChildById(EscherSpRecord.RECORD_ID);
        spRecord.setFlags(spRecord.getFlags()|EscherSpRecord.FLAG_OLESHAPE);

        // ExObjRefAtom is not set in OLEShape 
        UnknownEscherRecord uer = new UnknownEscherRecord();
        byte uerData[] = new byte[12];
        LittleEndian.putShort( uerData, 0, (short)0 ); // options = 0
        LittleEndian.putShort( uerData, 2, (short)RecordTypes.ExObjRefAtom.typeID); // recordId
        LittleEndian.putInt( uerData, 4, 4 ); // remaining bytes
        LittleEndian.putInt( uerData, 8, exEmbed.getExOleObjAtom().getObjID() ); // the data
        uer.fillFields(uerData, 0, null);

        EscherContainerRecord uerCont = new EscherContainerRecord();
        uerCont.setRecordId((short)RecordTypes.EscherClientData);
        uerCont.setVersion((short)0x000F); // yes, we are still a container ...
        uerCont.addChildRecord(uer);

        oleShape.getSpContainer().addChildRecord(uerCont);
    }

    static void setRecordLength(Record record, int len) throws NoSuchFieldException, IllegalAccessException {
        Field f = record.getClass().getDeclaredField("_header");
        f.setAccessible(true);
        byte _header[] = (byte[])f.get(record);
        LittleEndian.putInt(_header, 4, len);
        f.set(record, _header);
    }
}