Closed GoogleCodeExporter closed 9 years ago
Thanks for the report. Will have a look now. Am attaching the file to this
issue as a permanent record.
Original comment by drewnoakes
on 29 May 2011 at 12:02
Attachments:
This was a great find. I've created a unit test that reproduced the issue
using the data from this file. I also created a small unit test that caused
this error with the BufferReader class I created to make all byte[] reads safe.
Seems this integer overflow case was missed. I also updated IccReader to use
the BufferReader in this case (as all code should, for this very reason.) This
improvement covers reading of all metadata types, not just ICC.
Thanks again for the report! Fixed in r106. Please review if you have a
chance.
Original comment by drewnoakes
on 29 May 2011 at 2:44
Make that r107 -- I missed some files in the last commit :)
Original comment by drewnoakes
on 29 May 2011 at 2:47
Thanks! Really fast fix!
I parsed 500K+ images without errors.
Original comment by gud...@gmail.com
on 30 May 2011 at 2:35
That's great news. Thanks for getting back to me.
Original comment by drewnoakes
on 30 May 2011 at 2:41
Original issue reported on code.google.com by
gud...@gmail.com
on 29 May 2011 at 11:22