About the bug in SlideShow class

------=_Part_9725_22853957.1170383384490
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

When I use the POI-HLSF to extract the text from the slides.
My code is very easy:
   PowerPointExtractor ppe = new PowerPointExtractor("c:\\4.ppt");
   String s = ppe.getText(true,true);
   System.out.println(s);
   ppe.close();

But there is a error:
   java.lang.ClassCastException: org.apache.poi.hslf.record.Slide cannot be
cast to org.apache.poi.hslf.record.MainMaster
I trace the bug, I find that:
   For the ppt which has more than 2(include 2) MainMasters, the SlideShow
Constructor could not correctly to get MasterRecords:


 SlideListWithText masterSLWT =
_documentRecord.getMasterSlideListWithText();
 SlideListWithText slidesSLWT = _documentRecord.getSlideSlideListWithText();
 SlideListWithText notesSLWT  = _documentRecord.getNotesSlideListWithText();

    //find master slides
    SlideAtomsSet[] masterSets = new SlideAtomsSet[0];
    org.apache.poi.hslf.record.MainMaster[] masterRecords = null;
    if (masterSLWT != null){

        masterSets = masterSLWT.getSlideAtomsSets();
        masterRecords = new org.apache.poi.hslf.record.MainMaster[
masterSets.length];
        for(int i=0; i<masterRecords.length; i++) {
            masterRecords[i] = (org.apache.poi.hslf.record.MainMaster
)getCoreRecordForSAS(masterSets[i]);
        }
    }

Eg. The masterSets.length =2, but there is only 1 master records in the
MostRecentCoreRecords, so the there will be a casting error.

And I also find that the test data given in the Src dir, all ppt file are
simple, all of them have only 1 master slide,

Did anyone test the ones which have more than 2 master slides?
-- 
With respects,
   Mingfan.Lu

------=_Part_9725_22853957.1170383384490--