2005.06.23 15:43 "Re: [Tiff] Fix for corrupted multipage TIFF", by Frank Warmerdam

If we were to incorporate such handling, I think it might be best to control whether it is used based on an extra open flag indicating that an error should not cause complete failure.

I was not familiar with the LibTiff code so I didn't think of that but it sounds like the right way to go so I attempted to implement it. Here are my changes. I don't know if they are all correct, and I'm not sure how to "officially" submit code changes, so feel free to let me know.

Steve,

The approach seems acceptable to me, though I'll leave final judgement up to Andrey. I would suggest you submit the changes via bugzilla. I like your approach because it restores the symmetry of error handling for the first directory as with subsequent directories.

Best regards,

---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam@pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent