intertwingly

Mark Nottingham: Feed Paging and Archiving (nee Feed History) has finally made it to a standards-track RFC.

First thoughts on test cases for adding proper support to this to the Feed Validator.  I’m intending to capture notions of what a “best current practice” would say, so I may be extending a beyond what the spec currently says; this means that some of the potential warnings I list below may be a bit overreaching; any that are identified as such, I intend to retract.  I’ve identified some of the more questionable ones with a question mark.  Feedback welcome.

Updated based on feedback received

first, last, previous, next

• ErrorWarning if any of these occur in an entry
• Warning if any exist without all the others ?
• Exception: if first==self, nextprevious must be omitted
• Exception: if last==self, previousnext must be omitted
• Error if fh:complete is also present

prev-archive, next-archive

• ErrorWarning if any of these occur within an entry
• Error if fh:complete is also present
• Warning if any exist and current does not exist
• Warning if fh:archive is not also present

• ErrorWarning if this occurs within an entry
• Warning if fh:archive is not also present
• Warning if fh:complete is present, and current link doesn't match self

• Error if this occurs within an entry
• Warning if current link is not present
• Warning if at least one of prev-archive, next-archive, or fh:complete does not exist

• Error if this occurs within an entry
• Warning if current link is not present ?

Potentially the DuplicateEntries warning may need to be modified if fh:archive or fh:complete elements are also present?  If so, it isn’t clear to me how.