Question

I am develop a web app (using Java EE 6 with GF 3.1) that allow user to upload a pdf file. Since this is a closed small community, there are lots of chances that the uploaded file is already in the system. I cant just check name for duplication since it is clearly not enough. I was thinking about hash the entire file and store the entry inside a database. Is this feasible and how to achieve this? If not, then what is a better way.

Answer 1

Consider using checksum.

This is from http://www.exampledepot.com/egs/java.util.zip/CalculateChecksum.html

byte[] bytes = "some data".getBytes();

// Compute Adler-32 checksum
Checksum checksumEngine = new Adler32();
checksumEngine.update(bytes, 0, bytes.length);
long checksum = checksumEngine.getValue();

// Compute CRC-32 checksum
checksumEngine = new CRC32();
checksumEngine.update(bytes, 0, bytes.length);
checksum = checksumEngine.getValue();

// The checksum engine can be reused again for a different byte array by calling reset()
checksumEngine.reset();

Edit:

Be aware that checksums cannot absolutely tell you if two files are different, but they’re very useful.

If two files are the same, they will have the same checksum.

So if the checksums differ, you know absolutely that the files differ.

But two different files can also sometimes have the same checksum.

So the way to use this is to calculate the checksums first – if they differ, the files are different. If they’re the same, you’ll have to do a byte-by-byte comparison. That’s slower, of course, but it won’t happen often.

Note also all of this applies to hashcodes as well.