Hi Sean, Thank you for the pointer! Looks very interesting!
We did discuss this type of testing and, hopefully, will do it at some point. Creating a corrupted HDF5 file is not difficult. We will need to traverse a valid HDF5 file, find metadata and modify it to corrupt it. We will need to make sure that we go through all types of HDF5 metadata. It is not difficult, we just need time and focus. We will be more than happy to get help from the community :-) Thanks again! Elena Sent from my iPhone > On Feb 18, 2016, at 6:16 PM, Sean McBride <[email protected]> wrote: > > On Thu, 18 Feb 2016 23:34:17 +0000, Elena Pourmal said: > >> If you have a corrupted HDF5 file that causes HDF5 library to crash, we >> would love to have it! >> >> We have been fixing the library to fail gracefully on the corrupted >> files (yes, long overdue ;-( ). Some partial fix is coming in 1.10.0 and >> 1.8.17 releases. >> >> To complete the work we will need to generate corrupted HDF5 files, fix >> the problems they expose and use the files in our regression testing. >> This will take some time. Meanwhile, it will help us to move forward >> with the fixes if we can have a collection of the corrupted files (right >> now we have only one). > > Elena, > > Have you thought about using AFL:? > <http://lcamtuf.coredump.cx/afl/> > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng [email protected] > Rogue Research www.rogue-research.com > Mac Software Developer Montréal, Québec, Canada > > _______________________________________________ Hdf-forum is for HDF software users discussion. [email protected] http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org Twitter: https://twitter.com/hdf5
