Missing .fc-event-past and isPast is false on events ending yesterday

[r20]

Reduced Test Case

Bug can be seen at https://codepen.io/r20-the-styleful/pen/RwVMgey

Bug Description

With the dayGridMonth view, an event on the day before today does not have the .fc-event-past css class.
See the codepen reproduction, on "month" view. A css rule reduces opacity for fc-event-past and you can see "Yesterday event" does not have that class.
As a workaround, I thought of using .fc-day.past, but that won't work for cases like the event titled "Yesterday and today" because although it's within the td.fc-day-past it spans to today so it should NOT have .fc-event-past.

I noticed this in the listMonth and listWeek views as well. The timeGridWeek view seems to work as expected.

[acerix]

It seems like a bug, the comparison for isPast should use <= since end dates are exclusive:

fullcalendar/packages/common/src/component/event-rendering.ts

Line 279 in 6509050

isPast: segRange.end < (nowDate || todayRange.start),

As a workaround you can set a class from eventClassNames:

https://codepen.io/acerix/pen/gOWzbPQ?editors=0010

[camsteffen]

Related or same issue: #6120

[fnagel]

I can confirm this bug in v5.11.4 and the workaround provided by @acerix. Any chance for a bugfix in the actual code?

If you need the internal variables to be fixed too, try this:

eventClassNames: function(arg) {
    const end = arg.event.end ?? arg.event.start;
    
    if (end <= todayStart) {
      arg.isPast = true;
      arg.event.isPast = true;
      return 'fc-event-past';
    }
  },    

[simoncodearcs]

+1 this is still an issue.
Recently upgraded to v6.1.10 and decided to leverage the isPast boolean instead of calculating whether an event was in the past. Think I'll revert to previous calculations.
Did notice that the issue only occurs on the month view and list view, the week grid and day grid render correctly.

[arshaw]

This has been fixed in v6.1.11

Updated repro:
https://codepen.io/arshaw/pen/ZEPZovW

@acerix , your assessment of the bug was correct, thanks!