El mié, 03-04-2019 a las 09:55 +0200, Milan Crha via evolution-list escribió: > On Tue, 2019-04-02 at 22:12 +0200, aguador wrote: > > > repeating Andre's above question: what is the calendar type, > > > please? > > > > Daniel is correct (and of later and I assume Evo creation); > > It has set: > DTSTART;TZID=/freeassociation.sourceforge.net/Europe/Madrid: > 20190406T090000 > DTEND;TZID=/freeassociation.sourceforge.net/Europe/Madrid: > 20190406T100000 > thus the timezone is specified there. > > > Luis_Ignacio has an hour added. > > It has set: > DTSTART:20190402T090000 > DTEND:20190402T100000 > without any timezone information. These are so called floating times, > which are shown at the same time in whatever timezone the user uses. > Evolution doesn't support floating times for editing yet [1]. > Evolution > supports them when showing, only the editor shows a wrong timezone in > those cases, which I believe is the reason why you thought the > timezone > is set properly - because Evolution's editor lied to you. A > workaround > is to edit the series in Evolution, add a letter into the Summary, > then > delete it (which lets you save the series) and save it and propagate > the change into the all instances. It'll add the correct timezone > information into the event. > > > The second, which is wrong, shows a start time for 9.00, while the > > one from the week before shows the correct start time of 8.00. > > Right, the above DTSTART/DTEND is from the later .ics, showing the > 9:00 > start, while the former has: > DTSTART:20190326T080000 > DTEND:20190326T090000 > > > So the Android app is likely the root of this evil? > > Well, saving times with a floating time can be useful in some cases, > but when one needs to setup a meeting with people in different > timezones then the floating time is useless, thus yes, it's safer to > specify the time zone. I do not know whom to blame. > > There is one interesting thing about the broken event. Usually > generated recurrence doesn't change the floating time part, as far as > I > know, by the timezone offset (as I tested it here), thus I'm not sure > what made the two events split, even they look exactly the > same otherwise. Maybe they are detached instances and the server, or > some other client, created them. I do not know, I'm only guessing > here. > Bye, > Milan > > [1] https://gitlab.gnome.org/GNOME/evolution/issues/385 >
Milan, Thank you so much for the detailed explanations. I understand what is going on now and will fix it. As you probably saw from the appointments themselves, the calendar is "on this computer". However, that, in a way, is perhaps deceptive as the "erroneous" ones were not created by Evo so did not have the TZ information. Best, Roy _______________________________________________ evolution-list mailing list evolution-list@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-list
