Recover an original file from 3 different corrupted copies

Main function iterates through 3 corrupted copies and simply compares them byte-by-byte. If at least 2 corrupted versions have the equal value for the same byte location, then we keep the value for reconsturcting the original file as we can have high level of confidence that it is same value as in the original file.

In case the byte value is corrupted at the same location in 2 seperate copies, we get 3 different values for the same byte and we can no longer determine which one we should keep. Therefore we create 3 different new branches for the recovered bytes and temporarily keep all of the 3 values in these new branches.

At the end of bytes reconstruction phase we may end up with many versions based on branches.

And finally we look at the checksums of these temporary recovery branches and look for a match with the original file hash to verify that we have a recovery branch 100% same as the original file.

Branching level increases resource intensity exponentially

Cannot recover files under these conditions:

• Bit-level corruption happened simultaneosly in 2 seperate copies, which is extremely unlikely.

• If corrupted file lengths are different then the current byte comparison algorithm won't work properly.