Checksum while fetching data from iRODS

[Xophmeister]

An obvious enhancement would be to checksum data in lockstep with fetching it from iRODS, rather than waiting for the whole file to be fetched before starting the checksum job. This would give significantly shorter wait times for verification and responding to checksum requests.

As we can only (currently) shell out to iget, one would have to watch the file as it increased in size and control the checksummer on that basis.