| Summary | All day event end date should be on the next day, not same day |
| Queue | Kronolith |
| Queue Version | 2.1.4 |
| Type | Bug |
| State | Duplicate |
| Priority | 2. Medium |
| Owners | |
| Requester | bedgar (at) desasecurity (dot) com |
| Created | 02/13/2007 (6752 days ago) |
| Due | |
| Updated | 02/15/2007 (6750 days ago) |
| Assigned | |
| Resolved | 02/15/2007 (6750 days ago) |
| Github Issue Link | |
| Github Pull Request | |
| Milestone | |
| Patch | No |
State ⇒ Duplicate
ticket #4617State ⇒ Unconfirmed
Priority ⇒ 2. Medium
Type ⇒ Bug
Summary ⇒ All day event end date should be on the next day, not same day
Queue ⇒ Kronolith
non-inclusive end of the event." When Kronolith creates an all day
event, it sets the DTSTART and DTEND to the same date. It should set
the DTEND to the following date to obey the RFC. Mozilla Lightning
(and I presume Sunbird) behave the correct way. When Lightning
creates an all day event, the DTEND is the value of the next day.
Lightning is forgiving, though, of DTEND having the same value as
DTSTART and figures out that an all day event is intended. However,
if Lightning is set to publish an event and the kronolith/ics.php is
modified (as I have done) to accept events created in Lightning, the
Kronolith display of all day events created *from* Lightning is
incorrect due to this bug.
It seems that Outlook may have the same behavior Kronolith currently
has as to the value of DTEND for all day events, and so maybe this was
not broken in earlier versions of Kronolith but was broken as a patch
to fix Outlook handling. I cannot verify that, though I saw old bug
reports to that effect. In my opinion, Kronolith should obey the RFC
and perhaps handle Outlook in some other way (another URL, user-agent
sniffing if Outlook's user-agent is unique enough, etc.)