| *** zbenjamin is now known as Guest75538 | 02:24 | |
| *** zbenjamin_ is now known as zbenjamin | 02:24 | |
| *** frinring_ is now known as frinring | 02:45 | |
| *** nyov is now known as Guest2827 | 04:15 | |
| gmc | So, vanganz is gone, but for future reference: i'm using the signal android app, that works fine, including access to contacts.. and it seems whisperfish, the native signal app, is also picking up pace again | 06:52 |
|---|---|---|
| Thaodan | Good Morning. I noticed when I switched to Sailfish OS that closed of cloud services is something that you should move away from as supporting/using them is a cat and mouse game: they won't work relay able you put energy into them again and again. Its a battle that is lost from the beginning. I moved to the least power intensive way of using them for me by moving the hassle of supporting them from the phone to my server by using | 08:00 |
| Thaodan | bitlbee. I can bridge everything to IRC thanks to my bouncer everything is relay able, everything that goes wrong is just on my server and my phone is save. Incidentally this also makes tracking go to waste as my location in their point of view newer changes and I don't provide any data from my phone. | 08:00 |
| Nico[m] | I do the same with Matrix ;p | 08:02 |
| *** gigetoo_ is now known as gigetoo | 08:06 | |
| Thaodan | Yeah matrix is also good but harder to support. | 08:32 |
| Thaodan | Most clients are kinda meh. | 08:32 |
| Nico[m] | Working on it! | 08:52 |
| ninewise | Nico[m], you're working on a sailfish matrix client? | 08:54 |
| ninewise | do link :D | 08:54 |
| Nico[m] | It's konheko on openrepos (since I'm a nheko dev). It is very barebones atm, waiting for OpenSSL 1.1 and university is busy atm... | 08:57 |
| ninewise | understandably; I'm still bookmarking it, though | 09:01 |
| dcaliste | chriadam, sorry to hijack you, do you still have 10 minutes ? | 10:00 |
| chriadam | certainly | 10:00 |
| dcaliste | Thanks a lot, it's about https://forum.sailfishos.org/t/changelog-and-brief-notes-for-4-0-0-rc1/3404/5 | 10:01 |
| dcaliste | Are you aware of this issue and is iut already fixed ? | 10:01 |
| Nico[m] | Oh, the forum has private topics? | 10:02 |
| dcaliste | Lost all contacts in UI (db file is still there but migration somehow failed) which is a bit annoying. | 10:03 |
| dcaliste | Nico[m], it's the cbeta stuff ;) | 10:03 |
| chriadam | I think that is already fixed (hopefully). There is a commit in qtcontacts-sqlite recently which changes the way DROP INDEX is used in some of the migration steps | 10:03 |
| chriadam | I have been unable to reproduce unfortunately | 10:03 |
| ggabriel | dcaliste: it's already fixed in the next RC | 10:03 |
| chriadam | but basically: some devices, who upgraded from some versions to some versions, have a case where they do not have two of the indexes | 10:03 |
| Nico[m] | Yeah, I was just surprised, that you can have topics only a few people can see :3 | 10:03 |
| chriadam | I don't know how... some historical spelunking would be required | 10:04 |
| chriadam | the PR hopefully fixes that - but not sure how to repro / test properly... so not 100% sure | 10:04 |
| dcaliste | I've added the indexes by hand so the drop is not an issue anymore. | 10:04 |
| ggabriel | chriadam: def fixes it :) | 10:04 |
| ggabriel | dcaliste: did you reboot after adding the indexes? | 10:04 |
| Nico[m] | That the next relwase will be 4.0 was obvious from the git branches already :D | 10:04 |
| ggabriel | Nico[m]: well, a 3.10 would be trolling ;) | 10:05 |
| dcaliste | ggabriel : no I didn't reboot, sorry because I didn't see the point ! | 10:05 |
| ggabriel | dcaliste: reboot | 10:05 |
| dcaliste | ;D | 10:05 |
| dcaliste | For my knowledge, which service is renunning that requires to be resetarted from scratch ? | 10:06 |
| Nico[m] | ggabriel: why? Can't version numbers be arbitrary? Is there a big awesome feature to expect? Guess I'll know soon enough :D | 10:06 |
| ggabriel | Nico[m]: it would just break the pattern. Indeed version numbers are arbitrary, as are numbers themselves, and mathematics fwiw | 10:06 |
| Nico[m] | Oh, I didn't know there was a pattern :D | 10:07 |
| ggabriel | we did go from 2.9 to 3.0, right? | 10:07 |
| chriadam | dcaliste: apkd I think | 10:07 |
| Nico[m] | There was some discussion on the forums regarding the version number | 10:07 |
| dcaliste | ggabriel, indeed reboot fixed the problem. | 10:07 |
| chriadam | systemctl --user stop msyncd/contacts/commhistoryd/sailfish-eas + devel-su systemctl stop apkd, should suffice. | 10:08 |
| ggabriel | Nico[m]: oh, no, we didn't :) | 10:08 |
| Nico[m] | And yes, you did, but I thought it was for other reasons :3 | 10:08 |
| dcaliste | There are things I don't understand (which is an obvious sentence by itself). | 10:08 |
| chriadam | but might have to kill the user session also in case | 10:08 |
| Nico[m] | Or you didn't, I'm bad with numbers | 10:08 |
| dcaliste | Ah, I see I should have done a lsof on the db to see services using it. | 10:08 |
| ggabriel | Nico[m]: well, we jumped from 2.2 to 3.0 | 10:09 |
| ggabriel | dcaliste: it isn't that simple, since the db migration opens and then closes it (I think) | 10:09 |
| dcaliste | Anyway, I can reuse my phone, good, good. Thank you ggabriel and chriadam. | 10:10 |
| dcaliste | Still have to report some regressions on websites with gecko 60... | 10:11 |
| chriadam | dcaliste: sorry for the trouble. definitely my bad. | 10:12 |
| chriadam | thanks to everyone for reporting such issues :-) | 10:13 |
| dcaliste | Oh, no problem, too bad I didn't think about other processes accessing the db concurrently otherwise I would have solve it yesterday by myself... | 10:13 |
| ggabriel | chriadam: tbf, I think it was fixed before it was first reported. I would use the "if exists" on every drop, fwiw - I recall doing this on a totally unrelated project for exactly the same reason | 10:13 |
| chriadam | indeed, not sure why we didn't originally. perhaps an old version of sqlite didn't support the syntax? or just an oversight. not sure. | 10:15 |
| ggabriel | s/it was fixed/you fixed it/ (credit where it's due ;) ) | 10:15 |
| chriadam | well, it was spiiroin who provided all the information I needed to fix it | 10:15 |
| chriadam | all I did was monkey press some keys and pray :-D | 10:15 |
| ggabriel | hehe | 10:15 |
| chriadam | but I'm glad that you were able to verify that the fix worked for you also, as I had no way to repro personally | 10:15 |
| chriadam | thanks again | 10:15 |
| ggabriel | reproducing is basically going back to 3.9, dropping those indexes manually and then upgrading, but that's a faff | 10:16 |
| ggabriel | and no worries, I'm glad it was fixed ;) | 10:16 |
| chriadam | well, sure, but why would dropping the indexes be required? e.g. I tried: downgrade. rm -rf the db. restart phone. import vcard. db has old schema. upgrade. | 10:17 |
| chriadam | but in this case, the "old schema" still had those indexes - so it must have been a mismatch earlier on (than just the most recent version) | 10:17 |
| chriadam | e.g. something back when the indexes were first added? not sure. | 10:17 |
| ggabriel | there has certainly been some db upgrade shenanigans | 10:18 |
| ggabriel | *maybe* self inflicted, my db is not only from my first j1, the way I upgraded devices was by tar/untar my home, which isn't technically supported | 10:19 |
| chriadam | there must have been a schema upgrade mutation which wasn't versioned properly. | 10:19 |
| chriadam | no, it was definitely a "code problem" as multiple people have hit it | 10:19 |
| chriadam | i.e. my fault | 10:19 |
| chriadam | but I haven't been able to bisect exactly where / how in history. | 10:19 |
| chriadam | well, sledgehammer solution hopefully works. | 10:19 |
| ggabriel | happy to give you my .schema if you're in the mood for comparing, just shout | 10:20 |
| chriadam | if you could email the broken one to chris dot adams at jolla dot com I'd appreciate it, sure. but that by itself unfortunately isn't enough to determine which version-upgrade-steps occurred to get it to that point haha. | 10:21 |
| chriadam | (e.g. there are stop release upgrades vs normal upgrades, depending on whether you ever did factory reset or not, etc) | 10:22 |
| ggabriel | sorry, I don't have broken one any more :) I meant if you want to compare my "up to date production" .schema with what is supposed to be | 10:22 |
| chriadam | oh. nah, if it works I assume it's correct, now. | 10:22 |
| dcaliste | I have the broken one, I saved the db files before messing with them. | 10:23 |
| chriadam | that'd be helpful if you could send me that schema, dcaliste | 10:23 |
| chriadam | don't need the data, just the schema | 10:23 |
| chriadam | thanks very much | 10:23 |
| chriadam | it's getting late here - anything else to discuss? if not, I will head off :-) | 10:24 |
| dcaliste | No no I'm fine now, thanks for your time. | 10:24 |
| chriadam | no problem of course, thank you :-) | 10:24 |
| chriadam | gnight | 10:24 |
| *** dirkx__ is now known as dirkx | 10:40 | |
| Mister_Magister | Is there way to use DeviceInfo in c++? | 19:54 |
| Mister_Magister | is there some header or something? | 19:54 |
| Mister_Magister | add systemsettings to .pro and inclue "deviceinfo" or "aboutsettings" | 20:10 |
| Mister_Magister | big brain time | 20:12 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!