Tuesday, 2021-04-13

*** zbenjamin_ is now known as zbenjamin01:35
*** frinring_ is now known as frinring02:02
*** nyov is now known as Guest2406003:23
dcalisteHello chriadam, would you agree to start our meeting now or earlier than the 7.00 UTC ?06:29
chriadamhi dcaliste, sure can have the meeting now if it suits you06:29
chriadamI hope you had a good week?06:29
dcalisteGood, thank you. I wish the same applies to you.06:30
dcalisteThanks for the meeting with 30 minutes in advance.06:30
chriadamYes, my week was good thanks06:30
chriadamno problem at all, of course.  not sure if flypig or pvuorela are in the office yet, is the only thing06:30
chriadamso the QMF patches all integrated upstream!  huge thanks for your work there.  glad we finally got those in ;-)06:31
dcalisteWe can start with our business then ;) I've replied to your comment on the caldav MR about the detection of the user principal for server in a subdir.06:31
chriadamI will check, thanks06:31
dcalisteOh, yeh, thanks a lot also to flypig and yourself. It's good that the project still goes on.06:32
flypigI'm around in case you need me :)06:32
dcalisteI'm going to prepare a patch reverting pieces that are not compatible with Qt5.6.06:32
chriadamthere are still a couple of leftover commits in the upstream queue which need to be resolved, in particular: https://codereview.qt-project.org/c/qt-labs/messagingframework/+/308665 is fine IMO just needs review06:33
dcalisteBut trying to keep as many changes as possible (like the transition to QRegularExpression instead of QRegExp)06:33
chriadambut then this one needs a bit of work from our side first https://codereview.qt-project.org/c/qt-labs/messagingframework/+/28704706:34
chriadamMatt was hoping that we'd add some more comprehensive hooks for user input in the SSL errors case, but maybe that can happen in a separate PR I think06:35
chriadam(it seems kind of separate to the "ignore all SSL errors" case which that PR enables)06:35
dcalisteI didn't look much at this later one yet. I will in the coming week.06:36
chriadamfor the Qt-5.6-compatible thing: that sounds like it might be quite a bit of work.. is there much point?  hopefully in future we will be updating our Qt to more recent version, so it might make sense to do that work then instead (e.g. some 5.12 or 5.15 compatible thing, rather than 5.6)06:36
chriadamalthough Qt version upgrade is always unicorn thing, never quite gets accepted by management..06:37
dcalisteFor 5.6, as you think it's better. But developping patches for the SailfishOS "fork" will make difficult the upstreaming of them due to different code base. I'm afraid so. But yes, it would mean more or less the same amount of work to revert selectively that to upgrade to Qt6...06:39
chriadamhmm, I see what you mean06:39
chriadamouch, I hadn't considered...06:40
Nico[m]The world would be so much nicer, if we only had to care about 5.15 and 6 compat at this point .-.06:40
dcalisteThe bright side of it is that it would allow to test part of the migration changes "live", and not all of them at once with a Qt transition at the same time on device.06:41
chriadamthat's a very good point06:41
dcaliste(sorry for users as beta testers).06:41
dcalisteI think I will do the revert patching, but at a slow pace. It may not come before next month.06:44
dcalisteIf you agree.06:44
chriadamwell, they would be beta testers at the future point when qt version upgraded anyway, and like you said, at least this way we'd be testing just the effect of our qt6-transition changes (well, those which don't specifically require qt6) rather than combination of our qt6-transition changes + the qt version update as well..06:45
chriadamif you are willing to do the work, I definitely see that there is value (mostly: that any changes/fixes we make to jolla qmf will be more easily upstreamable, won't need a complicated rebase before upstream)06:45
chriadamand of course I'm willing to help with that effort if I can, although not sure how much time I will be able to spend on it (I will need to speak to pvuorela I guess)06:46
chriadamas for slow pace: not a problem at all, of course!  I mean, the alternative is not doing it at all ;-)  thank you.06:47
dcalisteOk, thanks, I'll ping when I have something ready for git.sailfishos.org.06:47
chriadamtyvm.  and if there's anything I can do, please yell06:48
chriadamflypig: ooi do you know if any of our current work is likely to result in changes to our QMF in the near future?  e.g. fixes or features planned?06:48
chriadamI can't think of anything off the top of my head, but I guess there's always bugs etc06:49
flypigNothing immediately springs to mind either.06:49
dcalisteOk, that sounds good, it means we have time to slowly import the changes.06:50
dcalisteI've another topic to discuss, related to this bug report : https://forum.sailfishos.org/t/dont-go-to-event-view-on-left-swipe/06:50
chriadamlet me check06:50
dcalisteI know it's not your area, but I've found the issue.06:51
dcalisteIt's in in lockscreen/LockScreen.qml06:51
dcalisteWhat is the best ? Give me access to lipstick-jolla-home repository so I can make a PR there or simply put the patch the bug report ?06:51
chriadampvuorela: ping ^06:53
chriadammeanwhile, apparently I don't have permission on the forum to view that report :-P06:54
dcalisteIt's cbeta restricted !06:54
dcalisteIf you want to test, disable the new "view event on lock screen option", enable "left swipe to event view" and set no lock code.06:55
dcalisteWhen you wake up the device (tap-tap gesture), swipe right, you arrive to the event view. Swipe left, you arrive to the last opened application or the home screen.06:56
dcalisteSwipe action is reversed on the lock screen.06:57
dcalisteIt's due to line 476 in LockScreen.qml06:57
chriadamwhen pvuorela gets into office I will ask him what the best way forward here is.  hopefully can just get you access to l-j-h06:57
dcalistethe x position and the overshoot are not taking the same values in 4.1.0 as previously in 4.0.106:58
dcalisteMaking the test not behaving as expected before.06:59
dcalisteThanks for talking to pvuorela about this.06:59
chriadamthank you for investigating06:59
dcalisteWith an access to the repo, I can also check if the code changes in LockItem.qml not to use anchors anymore for the EdgeIndicator were intentional.07:00
dcalisteThe result is that edge indicators on the lock screen are not animated anymore when swiping there.07:00
dcalisteInvestigation of this kind of things in lipstcik itself is a bit tricky. I prefer patching the calendar ;)07:01
chriadamyeah, I understand entirely07:02
chriadamoh, I've poked MartinS again about the jolla-calendar dst transition PR, but afaik he hasn't had a chance to look at it yet.  he's been super busy with something recently, but I think that activity will be finished soon, so he should soon have an opportunity to take a look at that one07:04
dcalisteGreat thanks to him. We'll seen if he has better ideas than us ! But that's probable !07:09
dcalisteWith the per tiem logging in CalDAV, I'm restarting to work on log viewing. I've come up to a simpler solution than the one I had before, with less changes to the core of Buteo. I'll submitted it soon now I think.07:10
chriadamI saw that there was an update to the "expose SyncProfile to QML" PR but I didn't get a chance to look at it in any detail07:12
dcalisteSo don't rush, I will deprecate this one in favour of a new one which provide the same things but much simpler, with almost no changes in Buteo core, only additions.07:14
dcalisteI rebased it on the recent changes and noticed that I could achieve the same result in a much simpler way.07:15
chriadamah great07:15
dcalisteI was yesterday, so I need a bit more days to check that it's actually providing the same capabilities.07:15
chriadamflypig: I guess the next community news letter will be next thursday - would be great to mention dcaliste's contributions to QMF etc there I think07:18
pvuorelanot in the office these days, but i could check the repo access. maybe easiest that way.07:18
chriadampvuorela: tyvm07:18
dcalisteHello pvuorela, thank you. Like that we can discuss about the fix in a more tracable way.07:19
flypigTalking about the QMF upgrade would be a great thing to include. Do either of you fancy writing something about it? It could even be quite detailed (why was it needed, what does it mean for future patches, etc.).07:20
dcalisteand pvuorela, sorry again for the additional work on the Buteo plugins, but as you said, the good point is a cleaner usage there of the error values.07:20
dcalisteFlypig, I can write something and send it to you (CC chriadam if he would like to amend it), would next Friday be ok for it ?07:22
chriadamsounds good, thanks07:22
flypigThat would be great, and next Friday would be perfect.07:22
flypigIt'd be a good opportunity to show work going on behind the scenes that is essentially hidden for most people.07:23
chriadamwere there any other topics to discuss today?  anything I have forgotten or need to follow up on?07:24
pvuoreladcaliste: np, luckily the fixes were mostly simple even if they were needed in many places :)07:24
flypigdcaliste, is there anything else that could/should be included that you've been working on, or that you're aware of?07:25
dcalisteflypig, I don't think I've got anything ready yet. I need to finish the various things that I have started and that are stale in the middle at the moment ;007:29
flypigOkay, no problem. We can take another look in the future.07:30
dcalistechriadam, I think, we've addressed all the points I had in mind today. Do you want to speak about something else I would have forgotten ?07:34
chriadamI had no other topics to discuss.  I will create a bug for the caldav PR and get it merged.  I just looked at https://forum.sailfishos.org/t/nextcloud-account-creation-failed-calendar-make-it-fail/1655/17 and I wonder if that's related?07:35
chriadamwe may need to resolve something in jolla-settings-accounts ?  not sure07:36
chriadamIt sounds like in this case it is failing at account creation time07:38
dcalisteYeh, the discovery code is duplicated and different in CalDAV and account creation.07:39
dcalisteI can see if I can find a way to reproduce the faulty behaviour... I'm adding a bookmark for this one.07:41
chriadamI have created JB#53905 to track that one and will ping Bea about it also07:42
chriadamwell, nothing else from me.  flypig or pvuorela did you have anything to discuss?07:44
flypigI don't think so.07:44
pvuorelai don't have.07:45
chriadamok, if nothing else, thanks for your time and effort as always dcaliste, and flypig and pvuorela too of course ;-)07:46
chriadamhave a great week!07:46
* chriadam -> away, gnight07:46
flypigSame from me07:46
dcalisteThanks all !07:46
MalinuxHi. Why is it in the calendar app, the line: calendar Personal is grayed out? I want to move my events to the nextcloud calendar, but it's possible only to choose if I make a new event10:15
ggabrielMalinux: good question, I did export mine somehow in the past, then imported them in nextcloud (I used something else in fact, but that was the workflow that I used to migrate to different providers)12:05
ggabrielyeah, I forgot about all the CLI tools... I would look in sailfishos.org if I were you, or search/post in the forums12:10
Malinuxyeah, It's not that I can't use cli tools, but I think there should be an easy way to do it within the gui and also to change what calendar to put an event in after it is created in another calendar :)12:11
ggabrielso, I read that you may be able to change the provider in future versions, yes12:11
ggabrielbut, still, that would be on an event-per-event basis12:12
ggabrielif you wish you migrate everything that you have in personal to nextcloud, that would be a bit cumbersome12:12
ggabrielI admit I forgot what I did and when I did it... maybe jolla didn't exist when I finally decided to use some caldav backend for my calendars12:14
Malinuxmaybe :)12:15
Malinuxlike in manage calendars, there should be an option. export calenadar as (and then support some formats) But gonna find the cli-tools for now then :)12:16
* ggabriel just noticed 2 personal calendars switched off in manage calendars lol12:17
ggabrielMalinux: good luck - can't promise there's such a tool now, try to make a backup and see if the format is vcalendar, maybe you're in luck12:18
ggabrielthat would be trivial to import into nextcloud12:18
MalinuxThink I can't make backups with the backup-client iirc. It always hang on the backup calendar part13:09
Malinuxhehe, switched off :) I can easly turn off or enable calendars there, but I just don't understand why I can't either switch calendar pr event, or even the entire caleandar easly, though, I hope as mentioned before, there will be an option for it13:10
ggabrielMalinux: it doesn't do it because it isn't implemented. Expect to be able to do it per event, but certainly not per calendar. Nobody does that13:56
MalinuxI can't even do per events :p14:30
*** vilpan is now known as Guest7030514:39
*** Guest70305 is now known as vilpan14:39
ggabrielMalinux: when I say "expect to", I mean in the future - there is mention of this in the forums16:09
Malinuxah, I haven't seen. sorry :L)17:13
Malinuxeh, :) I mean17:13
ggabrielyeah, no problem. how many events do you have anyway? I was thinking that maybe it's worth just enabling personal and personal[nextcloud], put the new ones in the latter, and let the former ones die a natural death as time itself makes them irrelevant17:36
*** SpeedEvil is now known as Guest3188121:00

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