Uploaded image for project: 'Request For Comments'
  1. Request For Comments
  2. RFC-482

Retire use of Boost.Serialization persistence before Gen3 Butler transition

    XMLWordPrintable

Details

    • RFC
    • Status: Implemented
    • Resolution: Done
    • DM
    • None

    Description

      Boost.Serialization-based persistence currently lives on in the stack only in serialization of PropertyList and PropertySet, and in the pipelines I know of, only when saving Task metadata.

      Several people have proposed switching to YAML-based persistence for these datasets, and (I believe) some partial implementations are underway.

      Because Boost-based serialization is closely tied to daf_persistence code we're hoping to fully replace, I would like to transition to YAML for these datasets before we get to Gen3, and for Gen3 to not to support reading Task metadata persisted in that format (unlike all other Gen2 datasets, which we do intend to support reading via Gen3).

       

      Attachments

        Issue Links

          Activity

            Parejkoj John Parejko added a comment -

            +1 That sounds like a great idea.

            Parejkoj John Parejko added a comment - +1 That sounds like a great idea.
            ktl Kian-Tat Lim added a comment - One partial implementation: https://github.com/lsst/daf_persistence/commit/6ed767db014fbcec7478ff7d616de3b0fc691b14
            ctslater Colin Slater added a comment -

            Yes please.

            ctslater Colin Slater added a comment - Yes please.
            jbosch Jim Bosch added a comment - - edited

            Adopted.

            T/CAMs and free agents: please feel free to help us get you a Gen3 Butler sooner by finding someone to do this work.  If you don't, one of us will do it instead of working on the Gen3 Butler for a week or so .

            jbosch Jim Bosch added a comment - - edited Adopted. T/CAMs and free agents: please feel free to help us get you a Gen3 Butler sooner by finding someone to do this work.  If you don't, one of us will do it instead of working on the Gen3 Butler for a week or so  .
            tjenness Tim Jenness added a comment -

            jbosch all triggered tickets are completed. Does this mean we can mark this RFC as implemented? Also, does this mean we can remove the tests of boost::serialization in daf_persistence that are causing the problems in DM-15593?

            tjenness Tim Jenness added a comment - jbosch all triggered tickets are completed. Does this mean we can mark this RFC as implemented? Also, does this mean we can remove the tests of boost::serialization in daf_persistence that are causing the problems in DM-15593 ?
            jbosch Jim Bosch added a comment -

            Just marked as implemented.  I think those tests are probably gone already; moving further inspection and conversation to that ticket.

            jbosch Jim Bosch added a comment - Just marked as implemented.  I think those tests are probably gone already; moving further inspection and conversation to that ticket.

            People

              jbosch Jim Bosch
              jbosch Jim Bosch
              Colin Slater, Jim Bosch, John Parejko, John Swinbank, Kian-Tat Lim, Tim Jenness
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved:
                Planned End:

                Jenkins

                  No builds found.