Time/Place

This meeting is a hybrid teleconference and slack chat. Anyone is welcome to join...here's the info:

Attendees 

Agenda

  1. Announcements
    1. Beta sprint:  Feb 8-19
  2. Sprint issues/Questions:
    1. Preparing OCFL objects for side-loading
    2. OCFL object validation
      1. Dealing with F3 bug involving reordering/whitespace in inline xml
      2. Updating the RebuildITs - problem is that the F6 Objects are no longer valid
      3. Larger question:  "if in the future Fedora introduces new required fields to the headers, old versions would then be invalid per the latest definitions." - Peter Winckles
  3. State of the board
  4. PRs needing attention:
    1. Ensures that external resources are not exported by default.
    2. https://github.com/fcrepo/fcrepo/pulls
  5. Other topics
    1. bloated directory from migration utility update 


Tickets

  1. In Review


  2. Please squash a bug!


  3. Tickets resolved this week:


  4. Tickets created this week:


Notes

  1. Announcements
    1. Beta sprint:  Feb 8-19
    2. Andrews Departure - Feb 12
      1. Proposed meeting to review the state fedora at this time
  2. Sprint issues/Questions
    1. Preparing objects for sideloading
      1. Who creates the object?
        1. If a client were written in java, there would already be the fcrepo-storage-ocfl maven dependency
        2. Aside from that, there is a python OCFL client.
      2. Create a short document describing the requirements and tools to share with the community
    2. OCFL object validation
      1. Dealing with F3 bug involving reordering/whitespace in inline xml
      2. Updating the RebuildITs
        1. The example OCFL objects used by this integration test were probably valid when originally generated, but there are new required fields, so the objects are no longer valid. Some also had invalid IDs. There were a few hand editing mistakes.
        2. Whenever we change the format/required fields of the headers, we will run into these problems
        3. Proposal to store the version of Fedora used to produce header file, or the version of the header format, so we can determine what rules apply/applied.
        4. Should document requirements for header files
        5. Not going to make any changes to the way headers are handled in the code at this point, until the need to handle multiple versions arises.
      3. Larger question:  "if in the future Fedora introduces new required fields to the headers, old versions would then be invalid per the latest definitions." - 
    3. Fcrepo-3477 - In 5 if when requesting a memento via date negotiation, it got the closest memento in the past, and if none existed it returned a 406 (if requesting before the first memento). At the moment, in fcrepo6 gives the first memento.
      1. We will stick with the 406 behavior. The test suite may be updated to remove this requirement, but not essential at this point
    4. Fcrepo-3480 - all good - Danny will do
    5. Fcrepo-3596 -danny got the confirmation he was looking for
    6. Fcrepo-3334 - "return=minimal" is not from the LDP specification, but is defined for fedora. It is the same thing as the LDP minimal preference + omit server managed triples. Conclusion is to remove it.
      1. On PUT 'received="minimal"' doesn't seem necessary anymore, "handling=lenient" performs the required need here. Will remove this as well
  3. State of the board
  4. PRs needing attention
  5. Other topics
    1. bloated directory from migration utility update - will discuss on slack

Actions