Currently the Web API creates a hash of an uploaded file, then compares that hash to already-present files, and rejects duplicates.
The idea is that this will prevent users from inadvertently adding the same file multiple times, and help them realize when files that might have different names are actually the same file. (Having confused myself with this mistake when using the parser at the command line is what prompted this feature idea.)
[x] Decide whether or not this is actually a good idea.
[x] If it is, change hashing from MD5 to SHA256/512 per CM's suggestion.
[x] If it is, make the hash key the primary key for the relevant database table.
Currently the Web API creates a hash of an uploaded file, then compares that hash to already-present files, and rejects duplicates.
The idea is that this will prevent users from inadvertently adding the same file multiple times, and help them realize when files that might have different names are actually the same file. (Having confused myself with this mistake when using the parser at the command line is what prompted this feature idea.)