| *** zbenjamin is now known as Guest3018 | 02:26 | |
| *** zbenjamin_ is now known as zbenjamin | 02:26 | |
| *** moralrec1rdings is now known as moralrecordings | 05:31 | |
| erani | looks like the theme and component cheat sheet links on this page are broken. https://sailfishos.org/wiki/Application_Development#API_Documentation does anyone happen to have a working URL or the material somewhere? :) unfortunately could not find while searching online | 07:14 |
|---|---|---|
| chriadam | erani: thanks, I've pinged internally about that. hopefully it will be fixed soon. | 07:31 |
| dcaliste | Hello chriadam, good evening. | 07:53 |
| chriadam | hi dcaliste | 07:55 |
| chriadam | I hope you had a great week | 07:55 |
| dcaliste | Yes, thank you, it was fine. | 07:55 |
| chriadam | that's good | 07:56 |
| dcaliste | I wish we can see the end of the all day + local time debugging session ;) At least for the moment. | 07:56 |
| chriadam | I am looking forward to merging those mkcal/kcalcore PRs hopefully tomorrow if flypig agrees ;-) | 07:56 |
| flypig | They both look good to me. Great work both of you :) | 07:57 |
| chriadam | then I should look into your MR#17 in mkcal - if you think it's ready for review now? I guess it will need to be rebased | 07:57 |
| chriadam | flypig: thanks for your review - your thoroughness is always very appreciated! | 07:57 |
| dcaliste | Yes, it can be reviewed. I didn't try how it manages with mkcal!31. | 07:58 |
| chriadam | I will take a look tomorrow, let's see :-) | 07:58 |
| dcaliste | You can merge first your MR and I'll rebase and test it before you can review it again. | 07:58 |
| chriadam | sounds good | 07:58 |
| dcaliste | Indeed thank you flypig for your thorough reviews. It's great. | 07:59 |
| flypig | You're both welcome. Sorry for being annoying. It's a difficult area for me to get to grips with. | 07:59 |
| chriadam | I have an unrelated question: I'm currently looking into improving support for invitations with caldav plugin. from my understanding and brief testing, if the caldav server supports iTIP/iMIP then it will automatically take the appropriate action (i.e., either send iMIP email, or create INBOX schedule event for iTIP if the invited attendee is on the same server) | 08:00 |
| chriadam | flypig: not annoying at all. it is complex, and I am very glad that you took the time and effort to dig into the details. | 08:00 |
| chriadam | dcaliste: from my testing, it seems like the server will do that IF the event is uploaded with the appropriate iTIP header, i.e. METHOD:REQUEST just above VCALENDAR:BEGIN etc | 08:01 |
| chriadam | was wondering if you had any experience with invitation support with caldav, or whether you have any objections to me trying to add codepath to "reformat .ics as iTIP request in the case where the organizer is the device-user and there are attendees" | 08:01 |
| chriadam | just prior to upsync | 08:02 |
| chriadam | of an event addition. | 08:02 |
| chriadam | (so, that would be basically "support server-side scheduling of invitations with caldav") | 08:02 |
| dcaliste | No sorry, I've no knowledge in that area, currently I'm sharing my calendars with myself only :) But, I've just made a "grep -r request kcalcore" in kcalcore and obtained interesting matches. | 08:03 |
| chriadam | then next step, I will look at adding a mkcal invitation plugin which creates iMIP emails for the case where the server doesn't support scheduling, so we can do "client side scheduling with caldav" - but that's trickier, will require linking email account to caldav account, etc) | 08:03 |
| dcaliste | Particularly the schedulemessage header looks interesting. | 08:04 |
| chriadam | I will investigate that also | 08:04 |
| chriadam | I had been looking at mkcal mostly, not kcalcore - seems I might be enlightened by looking there! thanks | 08:05 |
| chriadam | well, icalformat is used to generate the iTIP request, which is from kcalcore I guess. | 08:05 |
| dcaliste | Yes, the SCheduleMessage class looks promising and it is used in icalformat.cpp, whichj is a good omen too. | 08:05 |
| chriadam | great :-) | 08:06 |
| dcaliste | I'll review your MR with interest. | 08:06 |
| chriadam | probably won't be anythign concrete for a week or so, let's see. | 08:06 |
| dcaliste | Yeh, look at icalformat.cpp#380 | 08:07 |
| chriadam | unfortunately I didn't get a chance to look at the "delete/purge calendar events" PR properly yet | 08:07 |
| dcaliste | There is a createScheduleMessage() method returning a QString taking a iTIPMethod and an incidence as argument. | 08:08 |
| chriadam | yep - I believe that will mostly just return the VCALENDAR data properly encoded, but with the METHOD:(REQUEST/REPLY/...) header added, and perhaps some scheduling id | 08:09 |
| dcaliste | Exactly, calling in the plugin instead of the toString() method on the calendar may work (or not). | 08:12 |
| chriadam | the list of PRs I need to review this week: mkcal MR#30 + MR#17, kcalcore MR#13 (that one will require some minor fixes to google and exchange and caldav plugins, to remove the "subtract extra day from all-day end dt" thing) | 08:12 |
| chriadam | was there anything I missed from that list? | 08:13 |
| chriadam | also I saw pvuorela started looking at the email account folder sync PR - and saw also that you mentioned the other PRs for that were ready afaict | 08:13 |
| dcaliste | I've rebased this one also : buteo-sync-plugin-caldav!52 | 08:13 |
| chriadam | oh, let me check that | 08:13 |
| chriadam | LGTM. I will do a sanity check tomorrow then merge+tag | 08:14 |
| dcaliste | Ok, thank you. | 08:15 |
| chriadam | no, thank you | 08:15 |
| chriadam | for the email account folder sync thing - what concrete things can I do to push that along? I would prefer if pvuorela and jpetrell could test and merge that one | 08:16 |
| chriadam | are there clear testing instructions in the PR? | 08:16 |
| dcaliste | Not sure, pvuorela knows how to test it, you need three MRs: | 08:17 |
| dcaliste | - nemo-qml-plugin-email!52 to export the synchronisation flag per folder to QML | 08:17 |
| dcaliste | - buteo-sync-plugins-email!3 that actually implement the sync policy depending on user choice (inbox only, selected folders…) | 08:18 |
| *** frinring_ is now known as frinring | 08:18 | |
| dcaliste | - and the MR in setting account to have a UI for this. | 08:18 |
| dcaliste | Then, you can test by visiting your favorite email account page, see the synchronisation section and the new combobox there to choose the synchronisation policy for folders. | 08:19 |
| chriadam | great | 08:20 |
| chriadam | I will poke again internally. | 08:20 |
| dcaliste | Thanks. I'm sure it's on the todo list of pvuorela, but he may be busy at the moment. | 08:21 |
| dcaliste | I'm looking at https://together.jolla.com/question/216183/caldav-401-problem/ also. I'm convinced that it's a server issue triggered by the calendar-query request on a faulty event server-side. But Thunderbird and Android are working nicely, so I would like to find a solution client side also. | 08:22 |
| dcaliste | Looking at Thunderbird DAV-4-TbSync CalDAV sync source code (https://github.com/jobisoft/DAV-4-TbSync/blob/master/content/includes/sync.js), they are not doing any calendar-queryto get the etags on CalDAV entries or even the entries themselves, as adviced in http://sabre.io/dav/building-a-caldav-client/. It seems to me that they are using plain multiget requests. | 08:23 |
| dcaliste | From this knowledge, I'm going to try to modify the code of the CalDAV plugin here to use the same scheme as fallback when request is returning error 500. | 08:23 |
| chriadam | dcaliste: do we use basic authorization for caldav, or do we just set username+password via QUrl::setUsername()/setPassword()? | 08:23 |
| dcaliste | The second in that case. | 08:23 |
| dcaliste | But both are possible. | 08:24 |
| chriadam | had some cases recently when developing nextcloud account where we had to manually build the basic auth header, or we were getting 401 errors for some webdav requests | 08:24 |
| chriadam | maybe baikal has similar requirement? not sure | 08:24 |
| chriadam | but might be worth trying, if you have a server you can reproduce on. otherwise, yes, the multiget fallback certainly makes sense. | 08:25 |
| chriadam | ah, the 500 internal server error is ... curious. | 08:26 |
| dcaliste | Yes, the mixing of error code between buteo and qt is … fun | 08:28 |
| dcaliste | But here it's internal server error. | 08:28 |
| dcaliste | As we can see from the content of the report. | 08:29 |
| dcaliste | The text message is clear in that respect. | 08:29 |
| dcaliste | What is strange is that there is nothing in the server log, beside a time out… | 08:29 |
| dcaliste | But reducing the sync window solve the issue, so I'm definitely thinking about a server bug triggered by a specific event that may create an infinite loop in a thread of the server that trigger the time-out and the error 500 answer. | 08:30 |
| dcaliste | So I'm working on a fallback to multiget of etags in case the calendar-query is failing, as Thunderbird plugin is doing directly. | 08:31 |
| chriadam | right... so some event which has some complex recurrence rule, and it's trying to "expand it infinitely" to see if it matches the fetch window, or something | 08:31 |
| dcaliste | I'll ping with the user on this and may come back with a MR in buteo plugin later. | 08:31 |
| chriadam | anyway, thanks for looking into that - I agree that the multiget fallback sounds appropriate. | 08:32 |
| dcaliste | Don't puting the window is triggering the error 500 also… I've sent a patched plugin to allow this to the user without success. | 08:33 |
| dcaliste | I may send a MR for this patch too also. The patch is simple and allow to zap the window or part of the window if not specified in the profile. | 08:35 |
| chriadam | exposing window parameters used for sync to account UI sounds sensible. is that what you meant? | 08:36 |
| dcaliste | No, here is a very simple patch in the plugin not to use default value if parameters are not in the profile file explicitely. | 08:37 |
| chriadam | ah! | 08:37 |
| chriadam | I see | 08:37 |
| dcaliste | (they are there by default, because copied from a schema where they are). | 08:37 |
| dcaliste | But having this setting in the UI would be valuable anyway, I agree. | 08:38 |
| chriadam | yep, sounds good. let the profile control that, and not hard-code a window into the plugin .cpp | 08:38 |
| chriadam | but yes eventually some UI support would eb good too. but low priority I guess. | 08:38 |
| chriadam | was there anything else we should discuss, or which need my attention, or things you'd like me to poke people about? | 08:39 |
| pvuorela | pong. yes, planning on getting the folder sync things merged. | 08:43 |
| chriadam | excellent! thanks | 08:43 |
| chriadam | dcaliste: thanks again for your work and effort. I will follow up on the PRs mentioned earlier. | 08:44 |
| chriadam | I have to head off now - have a good week! | 08:44 |
| dcaliste | chriadam: sorry main line power issue in the building. | 08:50 |
| dcaliste | It was late, you did well not to wait for the power to come back here. | 08:51 |
| *** SpeedEvil is now known as Guest22473 | 10:46 | |
| x2s | Since the last update resetting the baseband modem doesn't work anymore reliably by setting the device into flight mode :/ | 11:16 |
| x2s | Looks like I need to order my XA2 soon | 11:16 |
| x2s | (and is there a way to sync both phones somehow?) | 11:17 |
| *** BitEvil is now known as SpeedEvil | 13:01 | |
| Mister_Magister | soo… what was that system qml file i have to edit to add offset to apps windows | 19:46 |
| Redfoxmoon | rrrr, bought an xperia 10 - everything works except connecting to 2.4 GHz wifi, android 9 did find and connect before I flashed SFOS | 19:52 |
| Redfoxmoon | 5 GHz ac APs appear almost immediately and work fine | 19:52 |
| Redfoxmoon | Is this a known issue or should I blame my router | 19:56 |
| abranson | Redfoxmoon: not heard of that before. is your 2.4 network on one of those channels that's restricted in certain countries? | 20:51 |
| Redfoxmoon | I am in Norway so only channel 14 is not available | 20:53 |
| Redfoxmoon | I've had issues with this router's wifi before though, wrt1900acs with openwrt, heh. | 20:54 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!