Michaël Cadilhac <mich...@cadilhac.name> writes: > I have a task that was recurring, which I KILLED a few weeks ago. It > now looks like: > > ** KILLED Do the right thing > SCHEDULED: <2023-07-14 Fri +1w> > > ox-icalendar still exports it every week, because I have > 'event-if-not-todo in org-icalendar-use-scheduled (which is the > behavior I want for some other headers). > > BEGIN:VEVENT > DTSTAMP:20240622T163042Z > UID:SC-671b3d13-f985-472a-be33-b4eeb298f2cd > DTSTART;VALUE=DATE:20230714 > DTEND;VALUE=DATE:20230715 > RRULE:FREQ=WEEKLY;INTERVAL=1 > SUMMARY:S: KILLED Do the right thing : > CATEGORIES:todos > END:VEVENT > > Would it be acceptable to add a variable to filter todo-types, e.g., > with a variable org-icalendar-excluded-todo-types? More generally, > one could think of introducing a variable: > org-icalendar-entry-filter > which receives the ENTRY argument of org-icalendar-entry, and would > return non-nil if the entry is to be treated. (This is basically what > I do using an advice.)
I agree ox-icalendar should have an option to exclude certain todo-keywords from export. However I think the option should also be flexible to allow exporting todo-keywords as different STATUS: https://datatracker.ietf.org/doc/html/rfc5545#section-3.8.1.11 For example, in your case it might make more sense to export the task as CANCELLED in iCalendar. I think I'd rather have a customization option for this than a filter function. That way, other Lisp programs that import ICS to Org (such as org-caldav) could refer to this option to interoperate with ox-icalendar for 2-way sync. Perhaps we could have an option ox-icalendar-export-todo-keywords-as which could determine whether a keyword should be exported, and if so with what status, e.g.: (defcustom ox-icalendar-export-todo-keywords-as '((no-export . ("SKIP")) (cancelled . ("KILLED" "CANCELLED")) (in-process . ("PROG")))) By default, todo-keywords not in the list would be exported with status "NEEDS-ACTION" or "COMPLETED", depending on whether the keyword is a done state. What do you think?
