import org.apache.poi.hslf.HSLFSlideShow; //导入依赖的package包/类
private void extractImageText(XHTMLContentHandler xhtml,
    HSLFSlideShow document) {
  if (Config.inst().getProp(ConfigBool.ENABLE_IMAGE_OCR)) {
    TikaImageHelper helper = new TikaImageHelper(metadata);
    try {
      PictureData[] pictures = document.getPictures();
      for (PictureData picture : pictures) {
        ByteArrayInputStream imageData = new ByteArrayInputStream(
            picture.getData());
        helper.addImage(ImageIO.read(imageData));
      }
      // TODO: find out page number
      helper.addTextToHandler(xhtml);
    } catch (Exception e) {
      e.printStackTrace();
    } finally {
      if (helper != null) {
        helper.close();
      }
    }
  }
}
 

import org.apache.poi.hslf.HSLFSlideShow; //导入依赖的package包/类
/**
 * Constructs a Powerpoint document from the underlying
 * HSLFSlideShow object. Finds the model stuff from this
 *
 * @param hslfSlideShow the HSLFSlideShow to base on
 */
public SlideShow(HSLFSlideShow hslfSlideShow) {
    // Get useful things from our base slideshow
    _hslfSlideShow = hslfSlideShow;
	_records = _hslfSlideShow.getRecords();

	// Handle Parent-aware Records
	for (Record record : _records) {
		if(record instanceof RecordContainer){
               RecordContainer.handleParentAwareRecords((RecordContainer)record);
           }
	}

	// Find the versions of the core records we'll want to use
	findMostRecentCoreRecords();

	// Build up the model level Slides and Notes
	buildSlidesAndNotes();
}
 

import org.apache.poi.hslf.HSLFSlideShow; //导入依赖的package包/类
/**
 * Creates a PowerPointExtractor, from a HSLFSlideShow
 *
 * @param ss the HSLFSlideShow to extract text from
 */
public PowerPointExtractor(HSLFSlideShow ss) {
	super(ss);
	_hslfshow = ss;
	_show = new SlideShow(_hslfshow);
	_slides = _show.getSlides();
}
 

import org.apache.poi.hslf.HSLFSlideShow; //导入依赖的package包/类
/**
 * Extract text from a power point 97-2003 document.
 * @throws Exception 
 * 
 * @see edu.ur.ir.index.FileTextExtractor#getText(java.io.File)
 */
public String getText(File f) throws Exception {
	String text = null;
	if( isFileTooLarge(f) || f.length() <= 0l)
	{
		return text;
	}
	
	FileInputStream inputStream = null;
	try
	{
		inputStream = new FileInputStream(f);
		HSLFSlideShow powerPointDocument = new HSLFSlideShow(inputStream);
		PowerPointExtractor pptExtractor = new PowerPointExtractor(powerPointDocument);
		
		String myText = pptExtractor.getText(true,true);
		if( myText != null && !myText.trim().equals(""))
	    {
	    	text = myText;
	    }
	}
	catch(OutOfMemoryError oome)
	{
		text = null;
		log.error("could not extract text", oome);
		throw(oome);
	}
	catch(Exception e)
	{
		text = null;
		log.error("could not get text for power point document " + f.getAbsolutePath(), e);
		throw(e);
	}
	
	finally
	{
		closeInputStream(inputStream);
		inputStream = null;
	}
	return text;

}
 

import org.apache.poi.hslf.HSLFSlideShow; //导入依赖的package包/类
public static void main(String[] args) throws Exception {
	if(args.length < 1) {
		System.err.println("Need to give a filename");
		System.exit(1);
	}

	HSLFSlideShow ss = new HSLFSlideShow(args[0]);

	// Find the documents, and then their SLWT
	Record[] records = ss.getRecords();
	for(int i=0; i<records.length; i++) {
		if(records[i] instanceof Document) {
			Document doc = (Document)records[i];
			SlideListWithText[] slwts = doc.getSlideListWithTexts();

			System.out.println("Document at " + i + " had " + slwts.length + " SlideListWithTexts");
			if(slwts.length == 0) {
				System.err.println("** Warning: Should have had at least 1! **");
			}
			if(slwts.length > 3) {
				System.err.println("** Warning: Shouldn't have more than 3!");
			}

			// Check the SLWTs contain what we'd expect
			for(int j=0; j<slwts.length; j++) {
				SlideListWithText slwt = slwts[j];
				Record[] children = slwt.getChildRecords();

				System.out.println(" - SLWT at " + j + " had " + children.length + " children:");

				// Should only have SlideAtomSets if the second one
				int numSAS = slwt.getSlideAtomsSets().length;
				if(j == 1) {
					if(numSAS == 0) {
						System.err.println("  ** 2nd SLWT didn't have any SlideAtomSets!");
					} else {
						System.out.println("  - Contains " + numSAS + " SlideAtomSets");
					}
				} else {
					if(numSAS > 0) {
						System.err.println("  ** SLWT " + j + " had " + numSAS + " SlideAtomSets! (expected 0)");
					}
				}

				// Report the first 5 children, to give a flavour
				int upTo = 5;
				if(children.length < 5) { upTo = children.length; }
				for(int k=0; k<upTo; k++) {
					Record r = children[k];
					int typeID = (int)r.getRecordType();
					String typeName = RecordTypes.recordName(typeID);
					System.out.println("   - " + typeID + " (" + typeName + ")");
				}
			}
		}
	}
}
 

import org.apache.poi.hslf.HSLFSlideShow; //导入依赖的package包/类
public static void main(String args[]) throws IOException {
    if (args.length < 1) {
        System.err.println("Usage:");
        System.err.println("\tImageExtractor <file>");
        return;
    }
    SlideShow ppt = new SlideShow(new HSLFSlideShow(args[0]));

    //extract all pictures contained in the presentation
    PictureData[] pdata = ppt.getPictureData();
    for (int i = 0; i < pdata.length; i++) {
        PictureData pict = pdata[i];

        // picture data
        byte[] data = pict.getData();

        int type = pict.getType();
        String ext;
        switch (type) {
            case Picture.JPEG:
                ext = ".jpg";
                break;
            case Picture.PNG:
                ext = ".png";
                break;
            case Picture.WMF:
                ext = ".wmf";
                break;
            case Picture.EMF:
                ext = ".emf";
                break;
            case Picture.PICT:
                ext = ".pict";
                break;
            case Picture.DIB:
                ext = ".dib";
                break;
            default:
                continue;
        }
        FileOutputStream out = new FileOutputStream("pict_" + i + ext);
        out.write(data);
        out.close();
    }
}
 

import org.apache.poi.hslf.HSLFSlideShow; //导入依赖的package包/类
/**
  * Constructs a Powerpoint dump from fileName. Parses the document
  * and dumps out the contents
  *
  * @param fileName The name of the file to read.
  * @throws IOException if there is a problem while parsing the document.
  */
 public SlideShowRecordDumper(String fileName, boolean verbose, boolean escher) throws IOException
 {
optVerbose = verbose;
 	optEscher = escher;
doc = new HSLFSlideShow(fileName);
 }