3.6. TC CALDAV
Several topics were discussed in the session.
Server-Info document: The basic outline was presented and discussed. A discussion of where the document should “live” ensued. The consensus is to define a WebDAV property and also use a Link header in the OPTIONS response. Another discussion was about whether to include “constant” properties. The TC will research current client
PROPFIND
s and then see what is best, but probably scope it as close as possible to the impacted resources.Scheduling Drafts: The basic outline of a solution was presented and discussed. The overall feeling was to initially move forward with the approach for new drafts only. The TC will do more work on how best to handle the “lost update” problem when creating drafts from existing resources.
Calendar Principal Searching: All agreed that
DAV:principal-property-search
is too structured and something better is needed. The TC will look into whether the existing report can be augmented with a custom property or a new report should be created (e.g. standardizing thecalendarserverprincipalsearch
report).Discussion of the status of WebDAV Prefer Draft and whether return=representation should be allowed on a precondition failure. The consensus was that it should be allowed to eliminate the need for the client to fetch the new resource in a separate request. The draft will be refreshed, reviewed and submitted to IETF.
Discussion on use of patch/diffs for client updates and whether ETag, Schedule-Tag is sufficient for that. The TC will investigate formalizing a diff format for CalDAV with a preference for a JSON-based solution.