Tuesday, 2019-06-25

*** zbenjamin_ is now known as zbenjamin01:46
*** frinring_ is now known as frinring02:06
*** Renault_ is now known as Renault05:26
chriadamdcaliste: sorry I'm late, am in meeting with jpetrell and lost trakc of time07:16
dcalisteHello chriadam: no problem I understand you can be busy!07:17
chriadamI hope your week went well!07:17
chriadamsaw that you investigated already the exdate issue - thanks for that.  I haven't had a chance to review your PR yet07:17
dcalisteYour investigations about the storage of rdates were right, the date should not be saved with a time zone.07:18
dcalisteI've explained in the MR discussion why it is currently like that. It's not that trivial.07:19
dcalisteThe correction can be done in two places:07:19
dcaliste* in nemo-qml-plugin-calendar, just before calling addExDateTime(recid), by expurging recid of its timezone when the incidence is full day;07:20
dcaliste* or in mkcal when saving exdatetimes by removing the time zone also under the same circumstances.07:20
dcalisteI'm proposing the later, because like that if it happens somewhere ezlse that addExDateTime() is called with a timezone, it is catched anyway, but the patch itself is ugly in mkcal.07:21
chriadamI see07:28
chriadamI guess we (pvuorela and I) need to investigate this in more detail07:29
chriadambut I agree that the more general solution is probably correct07:29
chriadamI wonder if there are use-cases where you definitely DO want to store the tz for an all-day exdate?  hmm.07:30
chriadame.g. show some all-day event in all timezones except some specific timezone (e.g. to handle some regional holiday exemption or something)07:30
dcalisteA full-day event with a time zone would mean that this event is schedule at a specific time and thus is fuill-day only for this tz. In that case, it is not a full-day event in the sense of date only, but an event that starts at 00:00+xx and finishes at 23:59:59+xx07:32
dcalisteI think, it's easier to think in term of iCal format.07:32
dcalisteIn iCal a full-day event is given by DTSTART:20190625 with no time.07:33
dcalisteSo a full-day event would not appends at the same time depending on your tz, but only at the same date.07:34
chriadamright, I agree.  just wondering how you might say "this recurring holiday occurs every year on day XYZ, but if you're in timezone ABC don't show it in calendar".  I guess that's not exdates anyway, so I'm probably just confusing the issue.07:34
chriadamI poked jpetrell about your jolla-settings-accounts PR for email folder sync again07:36
chriadamhe said that he's super busy this week with contract discussions, and then goes on vacation for two weeks, so no promises that he will get a chance to look at that one before he gets back from vacation07:36
chriadambut definitely when he gets back we'll get that one properly reviewed and integrated.  sorry for the delay there07:37
dcalisteAbout the folder sync MR, thank you. It's not in a hurry, no problem.07:37
dcalisteI've more or less the same time schedule. I'll be here next Tuesday, but after that I'm leaving for summer vacation for three weeks.07:38
chriadamok, sounds good :-)  I hope you have a great vacation07:38
chriadamhopefully we can get your mkcal PR integrated by next week07:39
chriadamand can discuss again the trigger-sync-per-account one with pekka07:39
dcalisteAh nice!07:39
dcalisteAbout the small fix for GPG missing public key you merged last week, do you plan to push it in a upgrade-3.1 branch?07:40
chriadamdcaliste: did I not already?07:42
chriadamlet me double check07:42
chriadamit's in devel07:43
chriadamI guess I webhooked it without creating an upgrade-3.1.0 branch, but it's definitely in the appropriate 3.1.0 release07:43
dcalisteAh good, I didn't see the upgrade something branch, that's why. Thank you very much indeed.07:44
dcalisteI have one question also about this TJC bug: https://together.jolla.com/question/208236/bug-e-mail-no-recipient-on-reply-after-editing/ it is related to an issue in sailfish-contacts RecipientField.qml07:45
dcalisteThe summary property is not updated before there is a focus out, which makes typed in addresses not taken into account when pulling down.07:46
dcalistepvuorela dealed with such PR in jolla-email recently I think, but it was not considered a proper fix but just a work around.07:46
dcalistesee https://bitbucket.org/jolla/ui-jolla-email/pull-requests/44007:47
chriadamwe should fix that properly, by emitting change signals when appropriate etc07:47
dcalisteMay I get access to sailfish-component-contacts so I can give a look to this?07:48
dcalisteI think that the fix indeed should go to RecipientField.qml and not be a work around in jolla-email.07:49
chriadamdcaliste: I think so, please ping jpetrell about that one07:49
dcalisteOk, as soon as I can find him around. I know he's super busy!07:50
chriadamI asked him in my other meeting, but he might not get around to that tonight07:50
dcalisteSure, thank you.07:51
chriadamnot sure if anyone has had a chance to check your sailfish-office change yet either07:55
chriadamagain I suspect that one might have to wait until jpetrell gets back from vacation, unfortunately07:55
dcalistejpetrell indeed looked at the small MR in office and he agreed. But it was not merged already, maybe waiting for pvuorela agreement also.07:59
chriadamah true08:00
chriadamwere there any other PRs outstanding which we need to review or discuss?08:00
chriadamobviously there is the caldav ones which are in my queue08:00
chriadambut won't get a chance to look at those this week unfortunately08:01
dcalisteNo, it's ok, I'm still working on the KF5 move. The fact that QDateTime is not storing anymore ClockTime makes many things to change.08:01
chriadamtrue, no doubt lots of parts affected08:03
chriadamis QDateTime sufficient, according to your current understanding?08:03
chriadamdoes it support e.g. the ClockTime spec from KDateTime?08:04
dcalisteIndeed no, but KCalCore has switched to QDateTime so I'm reading how they handle this.08:05
chriadamok, I'm interested to know that also08:05
chriadamthanks for your effort again to help with this work08:05
chriadamand indeed drive it08:06
dcalisteI'm going to try to add tests corresponding to flypig md document also. I need to discuss a bit with him to understand exactly some point of his investigations.08:07
dcalisteBut it's a good job he did, quite many interesting corner cases indeed ;)08:07
chriadamok, I had nothing else - is there anythign else you'd like to discuss?08:14
dcalisteNo that's great. Have a good week.08:15
chriadamthanks - you too :-)08:15
*** vals_ is now known as tango_14:17
bionade24Does someone know the right way to import a TLS/SSL cert. I tried two ways but none worked and I don't know what I did wrong https://blog.jupyter.org/ros-jupyter-b7e82b5e120215:04
bionade24Sorry, URL mistake, I ment this TJC thread: https://together.jolla.com/question/208387/proper-way-to-add-certs/15:05
kimmolido you want it to work with what? browser?16:08
kimmolibionade24: ^16:09
bionade24kimmoli: With the  IRC App from Thaodan. He said it uses System's certificate storage18:33
kimmolithat was what i used to get it working with my app and browser. it is a selfsigned root ca18:37
bionade24kimmoli: Thanks, will try it.18:45
bionade24kimmoli: I alread tried this method, but I have a simple .pem cert. Does it work with those, too?18:53
kimmolidunno, i have the ss root ca which against my server cert is signed19:04
bionade24kimmoli: Do I have to reboot to get it working?19:17
bionade24kimmoli: The cert isn't listed in the settings, is it?19:22
bionade24As I said, the method kimmoli mentioned didn't work for me. The IRC Client and curl are both not recognizing the cert.19:47
kimmolimy ca is listed in settings19:54
bionade24kimmoli: Mine isn't so I`ll maybe doing something wrong. http://paste.debian.net/1089222/20:02
bionade24Has anyone other ideas? What do I wrong or is this caused by the cert? The openssl test just works fine.20:31

Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!