recur_expansion.js: fix iterator with RDATE-only recurrence

[dilyanpalauzov]

When iterating with

  e is an event-component
  if (e.isRecurring() {
     let i = e.iterator();
     while (n = i.next()) {
       o = g.getOccurrenceDetails(obj)
       …
     }
  }

on the first iteration the DTSTART instance shall be returned. It is returned, when there is RRULE. In the lack of this change, on the first iteration, when RDATE is present without RRULE, the first RDATE instance is returned.

@darktrojan how does Thunderbird handle VEVENTS with RDATE but without RRULE. These are presented correct in the UI. But the instances cannot be obtained by getting an iteratior of the event and then calling .next(). The reason is, that the very first iteration skips then the DTSTART-instance. The DTSTART instance is not skipped by the iterator.next() when the VEVENT has RRULE.

[coveralls]

Pull Request Test Coverage Report for Build 3076542534

• 5 of 5 (100.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.03%) to 95.473%

---

Totals
Change from base Build 3012316465:	0.03%
Covered Lines:	2932
Relevant Lines:	3023

---

💛 - Coveralls

[darktrojan]

@darktrojan how does Thunderbird handle VEVENTS with RDATE but without RRULE. These are presented correct in the UI. But the instances cannot be obtained by getting an iteratior of the event and then calling .next(). The reason is, that the very first iteration skips then the DTSTART-instance. The DTSTART instance is not skipped by the iterator.next() when the VEVENT has RRULE.

Thunderbird doesn't use this code for RDATEs, it maintains a list of them and does iteration itself. So I'm not familiar with how this bit works at all.

[darktrojan]

In my very limited understanding of how this particular bit works, this looks okay to me. But it needs updating for ES6.

[kewisch]

@dilyanpalauzov looks like you are actively working on this one, anything I can do to help get this merged?

[dilyanpalauzov]

This change for proposal is one and a half years old. I do not know what is missing to get it merged. I see now it contains a test, which is failing. Sorry, I do not know anymore what was going on.

[kewisch]

Ah no problem, I had seen a recently referenced commit which made me assume. If you do feel like picking this up again I'm happy to give reviews some priority.