Re: Java error when updating existing Excel spreadsheets using HSSF

Hi Greg,

All of the error messages I've seen so far indicate that the error is a null object reference (i.e. an RPG O=Object field that's set to *NULL -- or zero). Would you concur with this?

Then when I step into the "wwPOIFS = new_POIFSFileSystem(wwFile);"
statement the program terminates.

If the error occurs on that line.... well, wwFile is the only input into new_POIFSFileSystem. So if you're getting a null object reference via JNI, then there's really only two possibilities... either the prototype for new_POIFSFileSystem() is wrong, or the wwFile variable is set to NULL.

Trouble is... I know there are quite a few people using this routine -- including myself -- and I don't get the error your describing. That implies that the prototype is correct!

But, you say that when you debug it, wwFile is not null... So I'm at a loss.

Just to be absolutely sure, you might try changing the monitor block in hssf_open() to look like this:

monitor;
wwStr = new_String(peFilename);
if wwStr <> *null;
wwFile = new_FileInputStream(wwStr);
if wwFile <> *null;
wwPOIFS = new_POIFSFileSystem(wwFile);
if wwPOIFS <> *null;
wwBook = new_HSSFWorkbookFromPOIFS(wwPOIFS);
endif;
endif;
endif;
on-error 301;
// catch error and proceed.
endmon;

That should guarantee that none of the inputs are null. The other possibility is that POFSFileSystem() is crashing due to a bug in the actual Java code. If that's the case, you should be able to send me your spreadsheet (off-list) and I should be able to reproduce the same problem you're having.

Regarding debugging the Java code... I guess you could do that. I would suspect that you'd need to download the source code for POI and re-compile it with a debugging view. Then you should be able to debug it via WDSC. Barbara Morris has posted information on how to debug RPG calling JAva to this mailing list in the past. You might try searching the archives for tips.