Tuesday, 2021-03-09

*** zbenjamin is now known as Guest7514902:33
*** zbenjamin_ is now known as zbenjamin02:33
*** arza- is now known as arza03:59
*** haasn` is now known as haasn03:59
*** frinring_ is now known as frinring04:02
*** albertuxone is now known as albertux07:23
dcalisteHello chriadam, could you give me 10 more minutes before our meeting please ?08:06
chriadamhi dcaliste, yes of course08:06
chriadamno rush08:07
dcalisteFinally, I'm ready. And reading your comment on buteo-syncfw#59. It's definitely valid. I'll has some sync code on startup. This will also provide a bonus point for the transition : users won't have to visit the account setting page to actually sync the two enable properties.08:17
dcalisteI'll add not I'll has... sorry.08:17
chriadamthanks :-)08:18
chriadamsorry that I didn't think of that case previously08:18
chriadamI know that PR is dragging on a bit08:18
chriadamflypig: have you had a chance to take a look at that one, by any chance ^08:18
dcalisteNo problem, it will be much better with your suggestion anyway.08:19
flypigI'll need to check, but I doubt it.08:19
chriadamflypig: it involves some things you've looked at in the past (e.g. signals from accounts-glib etc) so your opinion would be very helpful, if you get time to take a look08:20
flypigYeah, I've found it now, and I'm afraid I haven't looked at it, although I'm sure you asked.08:21
dcalisteflypig, don't worry, it's not really urgent.08:21
chriadamdcaliste: aside from that one, I see pvuorela_ merged the kf5-calendarcore PR (thanks for handling that one)08:22
flypigI'd be happy to take a look, will do.08:22
chriadamflypig: thanks very much08:22
dcalistechriadam, yes and https://git.sailfishos.org/mer-core/buteo-sync-plugin-caldav/merge_requests/78 is resulting from it.08:22
dcalisteflypig: indeed thank you.08:22
chriadamthanks, I'll check that one08:22
chriadamwas there a bug number associated with https://git.sailfishos.org/mer-core/kf5-calendarcore/merge_requests/3/diffs08:25
chriadamhmm, I can't see one, yet the tag triggered and built in OBS08:27
chriadamanyway, I will create a bz bug for the lastModified thing, sec08:27
pvuorela_think i did an annotated tag.08:27
dcalisteYeh, I'm afraid not. It's not related to a report on the forum or a bug.08:27
dcalisteIf you create one, chriadam, I can add it to the commit message in CalDAV MR.08:28
chriadamcreated JB#53475 for that one08:33
chriadamthanks very much.  we will look into doing similar for google and exchange if needed08:33
chriadamregarding the QMF things: I will do the qmflist_p.h into qmfclient thing tonight08:34
chriadamafter that, I think we're getting really close to trying to integrate all those changes08:35
dcalisteYes, and I'm preparing a MR in CalDAV to migrate the storage of URL and etag from comments to setUrl() and volatile custom property respectively. But doing this noticed that I was doing more and more lastModified(), update, setLastModified() calls.08:35
chriadamI must be mis-remembering, but I thought the setUrl thing was already done08:37
chriadamor maybe that was for notebook and not event08:37
dcalisteGreat about QMF. I still need to do a big grep in the code to catch all the possible deserialisation issues between 32 bit counters and 64 ones. I guess on my JollaC it won't be a big issue, but aarch64 ones are coming, aren't they ?08:37
chriadamI assume so08:37
dcalisteIn CalDAV, the URL for each incidence is still saved as a comment.08:37
chriadamah true.  thanks for looking into that!08:38
dcalisteFor historical reason when custom properties were trashed when an incidence was marked as deleted.08:38
chriadamah, that's right.08:38
dcalisteNow, they are not trashed anymore, except when the incidence is actually purged from the database.08:38
chriadamflypig: any ETA on when you might get a chance to take a look at the upstream QMF PRs?  is it likely to be this week, or next week?08:39
dcalisteBut also, upstream introduced a new incidence property to save URL for shared events. It was not available before the transition to KCalendarCore.08:39
chriadamah, nice08:39
chriadamhmm, interesting thought...08:39
chriadamwhen you say a "shared event" do you mean an event which exists in multiple calendars on the server?08:40
chriadamis the URL property then storing some canonical URL?  or?08:40
dcalisteAnd I've added it to mKCal some MRs ago, so now it should be ready to be used in CalDAV and clean a bit how we save and store incidence properties relqted to sync qctions.08:40
dcalisteUrl for an incidence is single, from the doc, it says that it is used to save a remote location for the incidence.08:41
dcalisteSo myself using shared was a bit of an abuse. I was meaning "shared" between device and a single remote server.08:41
chriadamah cool08:41
chriadamexcellent, thanks.08:42
dcalisteYes, ideally, I would like to arrive to a point where we can call ICalFormat::toString(incidence) without first filtering out CalDAV sync related changes or additions.08:44
chriadamthat would indeed be nice08:44
dcalisteAfter the next MR migrating the url and etag (and some other things), only the UUID changes will remains. But I've no idea at the moment how to change this in a backwrd compatible way. Anyway, let's first clean the url and etag storage.08:45
chriadamthe "real" solution would be to allow storing same event in multiple notebooks, I guess08:45
chriadamthen we wouldn't need the prefixing stuff08:45
chriadamI know this is something that pvuorela_ has considered in the past08:46
dcalisteExact but for this, I need to alter the DB and I would like to avoid this at the moment.08:46
*** albertux is now known as albertuxone08:46
chriadamyes, I agree08:46
chriadamon QMF front: I'll hopefully update the qmflist_p.h PR tonight, then try to review any of the earlier PRs which I haven't yet +2'd later this week (or more likely, early next week).  hopefully flypig gets a chance to review also.  then we can try an integration mid/late next week.08:47
chriadam(if you get a chance to look into the serialisation potential issue further before then, great - but if not, all good, we can address that issue later in any case, I think)08:47
flypigI dropped out for a bit. For info, I plan to look at the QMF PRs today. I'm not sure I'll get through them all though.08:47
chriadamoh, excellent!  thanks08:47
dcalisteGood, good. I'll try to look into the desrialisation case before Friday.08:48
dcalisteThank you flypig.08:48
flypigdcaliste, as an asie thanks for the FingerTerm PR changes. I'll take a look at that too.08:48
chriadamI need to poke blam to +2 my QtPIM PRs also08:48
*** albertuxone is now known as albertux08:48
flypigasie -> aside08:49
chriadamdcaliste: unrelated, but you might be interested to see https://git.sailfishos.org/mer-core/buteo-sync-plugin-carddav/merge_requests/30 related to what I mentioned last week.  seems like just a silly bug by me, fortunately.  easy to fix :-)08:49
dcalisteflypig, thank for the idea and the intent to look at it again. If this ncurses could be solved at last, it is great.08:49
chriadamthe redirect wasn't truly circular after all, as it was a http->https (or https->http, I don't remember which) redirect.08:49
chriadamwas there anything else to discuss to tonight?  have I forgotten anything I needed to follow up on from last week?08:53
dcalistechriadam, ok, I see. You wanted to save the redirected URL. I need to look at the QUrl doc, but could a orig != redir be equivalent to orig.path != redir.path || orig.scheme != redir.scheme ?08:53
dcalistechriadam, no that's all for this week I think. I've a MR to upgrade Calligra to the latest upstream, but I'll discuss this with pvuorela in the comments.08:54
chriadamgood point about the equality check, as what is there wouldn't cover the port redirect case either, I suppose08:55
chriadamI have added a comment, will think more about that08:55
chriadamthe problem is that we want to avoid possibly unsafe redirects (e.g. to different domain)08:56
dcalisteAh yes, redirection to another domain would be nasty, indeed.08:57
chriadamso I'm not sure we want to do a normal equality comparison there, in case the host() changes, not just path() or scheme() or port()08:57
chriadamanyway, thanks for raising that - I will look at what can be done to improve08:58
chriadamdcaliste: one final thing:08:59
chriadamflypig is hopefully going to be doing a bi-weekly information thing09:00
chriadamnot sure if he's spoken to you about this already?  anyway, we were thinking it might be nice if we included that, some brief sentence or two about various contributions during the last two week period etc09:00
chriadamwondering if you'd be ok with that?  (you wouldn't have to write anything, we'll do that)09:01
chriadambut basically just summarising what the recent contributions from the community have been, which community members made the contribution, what areas it affected, what the outcome will be, etc.09:01
dcalisteSure, I've read the two first iterations on the forum with great interest. That's a very nice and welcome initiative in my opinion to difuse information about the OSS parts. It may end up attracting more contributions.09:02
chriadamoh have a couple of them already been posted?  I didn't know haha.09:03
chriadamanyway, great :-)09:03
dcalisteYou proposed to do it, but I can send him some sentences of previous work from last two weeks regularly.09:03
dcalisteWhich day of the week would the best flypig ?09:03
chriadamI mean, that would be even better - not depending on me remembering to do it haha.09:04
chriadammaybe could event just paste something here on Tuesdays, and flypig could copy it09:04
flypigThanks both. It's supposed to go out on Wednesday/Thursday, and I tend to write the first draft over the weekend. So, I'll take whatever I've got at the weekend in the first instance.09:05
flypigIf you could paste it on Tuesday, then I'll add it in. That would be really great.09:06
dcalisteOk, I can prepare something for Tuesdays (when there is something relevant of course). I'll paste here something about what we've done with attah recently about compiling calligraconverter for SeaPrint to converter documents on the fly to PDF for the printer.09:07
flypigThat sounds just the sort of thing. Thank you.09:08
chriadamthanks very much!09:08
chriadamok, if nothing else, let's close the meeting.  thank you again for your time and effort, it is greatly appreciated as always!09:09
* chriadam -> away, gnight09:09
dcalisteflypig, you can modify or repkrase anyting of course :09:35
dcalisteattah who is developping SeaPrint, bringing printing to SailfishOS, is planning to extend document support from PDFs to other available formats like ODT, https://forum.sailfishos.org/t/seaprint-network-printing-for-sailfishos/796/7709:35
dcalisteOne possibility is to reuse the document library that the OS is using : Calligra. It's a KDE project for desktop use but versatile enough to be reused as a backend for the document application in SailfishOS. It is providing a converter but it seems that this tool was suffering issues and was not packaged for SailfishOS.09:35
dcalistedcaliste corrected the issues upstream https://invent.kde.org/office/calligra/-/merge_requests/19 and inclusion inside SailfishOS is under review https://git.sailfishos.org/mer-core/calligra/merge_requests/5609:35
dcalisteStill on document front, ApB provided a document making the application crashes, Leinir (who initially ported Calligra to SailfishOS and created the QML components used by the application) accepted the correction upstream to avoid the crash https://invent.kde.org/office/calligra/-/merge_requests/20 Inclusion into SailfishOS is still pending.09:35
flypigGreat, thank you dcaliste. I've added it to the next update. In future, if you prefer to write it in the first person, then I can include it as a "quote" attributed to you. Otherwise I can just adjust it to flow in with the rest of the text. So, your choice :)10:11
*** albertux is now known as albertuxone10:26
*** albertuxone is now known as albertux10:27
*** albertux is now known as albertuxone11:10
*** albertuxone is now known as albertux11:12
*** albertux is now known as albertuxone11:20
*** albertuxone is now known as albertux11:21
*** vup2 is now known as vup13:06
*** albertux is now known as albertuxone13:06
*** albertuxone is now known as albertux13:06
*** albertux is now known as albertuxone13:59
*** albertuxone is now known as albertux14:00
*** vilpan is now known as Guest1461715:41
*** Guest14617 is now known as vilpan15:42
*** albertux is now known as albertuxone16:26
*** albertuxone is now known as albertux16:27
*** SpeedEvil is now known as Guest5374620:25
*** SpeedEvil is now known as Guest2105122:56

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