A brief discussion took place regarding moving forward with the migration-utils, and preparing of tickets for some focused sprint work.
The following points were made:
- authz migration should probably just entail moving over the XACML as is
- the command line tool should use spring, and possible just be a spring boot application rather than a camel component
- Ben could be convinced to work on service migration
- discussion of audit log
- much of this is redundant for versioned datastreams
- migrate to premis?
- wait for the audit service discussions to come up with something
- we should use the same property name to preserve historic modifcation dates as the hydra effort
- we should consider and design the interaction with the command line tool if that will be our ultimate goal
2 Comments
Andrew Woods
Does the above statement imply that the migration-utils workflow will not tie into a camel flow?
Unknown User (daniel-dgi)
It's possible, depending on the status of a java fcrepo4 client library that's out there. It came up as an option due to the fact that the main benefit of using camel would be the fcrepo-camel component, and if the client library was functional then spring's batch framework could also be used and the whole project could be kept as a single command line tool. No decisions were made, however.
I think if the other client lib is in a less finished state than the fcrepo-camel component, then that makes camel the winner.