07:00:19 #startmeeting Sailfish OS, open source, collaboration -- 11th April 2024 07:00:19 Meeting started Thu Apr 11 07:00:19 2024 UTC. The chair is rainemak. Information about MeetBot at http://wiki.debian.org/MeetBot. 07:00:19 Useful Commands: #action #agreed #help #info #idea #link #topic. 07:00:20 #info Meeting information and agenda can be found here: 07:00:20 #link https://forum.sailfishos.org/t/community-meeting-on-11th-april-2024/18359 07:00:20 I am the meeting's chairperson today, and will be doing my best to keep time and order. Please respect the timings and bee-hive. 07:00:20 #topic Brief introduction (5 min). Please prefix your name/handle with #info 07:02:02 #info David Llewellyn-Jones, community 07:02:51 #info Raine Mäkeläinen, sailor @ Jolla, chairperson today 07:03:19 #info Björn b 07:03:22 #info Björn Bidar , sailor @ Jolla 07:03:59 #info Pekka Vuorela, Jolla 07:04:26 Jolla sailors leading still :-) 07:04:30 #info Andrew Branson, sailing by 07:04:53 #info Sebix, community 07:05:14 No topic proposals for today's meeting. 07:05:23 Only bug discussion 07:05:42 #info Damien Caliste, community 07:05:47 #topic Open PR discussion (5 mins -- asked by Jolla) 07:06:18 #info Any open PRs to discuss? 07:06:34 #info Otto Mäkelä, community 07:07:13 dcaliste, flypig : do you have some? 07:07:29 maybe pvuorela knows? 07:07:36 Not from me today. 07:07:51 Hello rainemak, I've some open PRs at the moment, but pvuorela is already taking care of them, thank you ! 07:08:46 alright, let's move on 07:08:47 #topic Untracked bug reports (5 min -- asked by thigg) 07:08:47 #info Untracked bug reports 07:08:47 #link https://forum.sailfishos.org/t/community-meeting-on-11th-april-2024/18359/2 07:08:47 #info The Community Bug Coordination Team is back from well deserved back. 07:08:48 #info As a result of their work, we now have: 07:08:49 #info 8 high quality bug reports which are mostly recorded internally 07:09:25 there's no substitute option... well deserved break 07:10:01 Good to see you back as well! 07:10:14 Great work, and great to see the bugs being classified nicely again. 07:11:18 One was an opinion / thought / idea 07:11:37 flypig, +1 07:12:20 Which one was the suggestion? 07:12:30 this one https://forum.sailfishos.org/t/4-character-security-code-is-partially-possible/18279 07:13:00 it's not a bug per se 07:13:08 Ah, yes, the last in the list. 07:15:00 Any objections if we move forward already 07:15:04 (I noted that the last SFOS update to 4.5.0.25 made the security code entry fully alphanumeric) 07:15:28 .25 was only eula 07:15:50 I assume he means 4.5.0.x 07:15:54 (ok, so it happened already before this) 07:16:03 Are you using it ExTechOp? 07:16:45 It's really nice functionality and I'm curious to know how many people are using it (I don't expect there's any way to find out definitively). 07:17:27 I still have a numeric security code from updating from an earlier version where it could only be numbers. It's fun finding reuse for old phone numbers permanently etched into your mind. 07:17:39 :D 07:18:24 I'll steal some time from this... let's move to general discussion 07:18:39 #topic General discussion (20 min) 07:19:18 I don't have any topic for this but I'd like to hear esr91 update from flypig 07:20:06 How's the webview looking like? (haven't read blog yesterday) 07:20:24 Oh, I'd be happy to give one. I'm currently still stuck on offscreen rendering. I've narrowed down the problem to the fact that although the site is being rendered to the surface, the surface isn't being rendered to the screen. 07:21:03 So I'm still trying to figure out why not. I think all the pieces are there, but like the original rendering, just one faulty flag can blow everything else up. 07:21:06 And you have right glcontext created and bound 07:21:38 right == only one 07:21:42 I believe so, yes. But there is obviously some break in the link that I'm missing. 07:21:55 Sorry, some break in the *chain*. 07:22:37 But I'll try to double check that the bind is happening correctly tonight. 07:23:04 it has to be so that create gl context creates glcontext properly (that I saw in earlier blogs already) 07:23:29 Awesome dedication flypig, to make it work after such a long time ! 07:23:37 The glContext definitly works, in that I can see the rendering is happening to the texture. But I'm not sure how that ends up on the screen :( 07:23:48 Does anyone use the Danske Bank mobile application? It used to work, but I recently noticed that it just exits without doing anything. Dunno why, I've installed MicroG (because of MobilePay) and some other changes. 07:23:57 I'm referring that CompositorOGL method that create gl context 07:23:58 Thanks dcaliste, there are some parts I wish I could have tackled more quickly :) 07:24:11 when will we get 4.6 image for 10 V? 07:24:22 #info thigg, community 07:24:46 Flypig: you have a milestone in gecko after which you have a big party planned? ;) 07:24:55 ExTechOp, next sauna release candidate has some banking app fixes 07:24:58 ExTechOp: Maybe some kind of "protection" function against tampering? The s-pankki used to do something similar 07:25:33 rainemak, in CompositorOGL::Initialize()? "mGLContext = CreateContext()"? 07:26:02 flypig, don't recall by heart but that sounds familiar 07:26:15 thilo[m], once the offscreen render is working, I'll generate patches, and then it will be time to celebrate (for me at least!). 07:26:43 Cool, i'll prepare pictures ;) 07:26:44 rainemak, I'll check that part out carefully tonight. Thanks for the suggestion, appreciated! 07:26:50 :-) 07:27:00 thilo[m], yes please! 07:27:10 flypig, I doubt it's there 07:27:27 rainemak, every piece is worth checking. 07:28:07 Also, any of the developers here feel like making a mosh package? https://forum.sailfishos.org/t/easy-mosh-installation-on-sfos/14118 07:28:16 flypig, and you got that front and backbuffer thingie sorted out 07:29:16 rainemak, I'm hesitant to say anything is sorted out until it's rendering properly, but I *think* so. 07:29:33 flypig, I feel you 07:29:42 Every time I do another pass of it I find something that's not quite right :O 07:29:57 It keeps me out of trouble :) 07:30:28 ExTechOp, I've never used mosh, but heard such good things about it. 07:30:41 flypig, and you see that rasterizing step takes place 07:31:27 rainemak, yes, when I check the texture (e.g. using fReadPixels) I can see that the rendering is happening. That's going to be in the blog tonight. 07:32:01 So I think most of it is working. 07:32:03 flypig, fReadPixels is good 07:32:22 Indeed! Wish I'd done that earlier! (although earlier I might not have had happy results). 07:33:07 fReadPixels is the way how qmozembed grabs to pixels as well 07:33:43 flypig, sounds like something wrong in the swap 07:34:11 flypig, have you tried to change gl clear color? 07:34:15 Okay, maybe I should take a look at the join between qmozembed and gecko then? 07:34:44 Yes, I tried glClearColor (that's in the blog tonight as well). The change shows up with fReadPixels, but no on screen. 07:35:01 * but *not* on screen. 07:35:17 has to be that off screen is not there 07:35:26 Could you elaborate? 07:36:02 Sorry for everyone else in the meeting. I don't want to monopolise the discussion, but this is super helpful for me :) 07:36:37 we didn't have any other topics 07:36:50 I'd say... don't worry 07:36:51 No problem flypig, that's an interesting discussion, I think. 07:37:33 Thank you dcaliste :) 07:37:33 i'd be happy to see a whole meeting dedicated you helping you with your gecko effort :) 07:38:45 abr, be careful what you wish for, I've not got to video rendering properly yet ;) 07:39:26 flypig, one thing worth looking is that gecko doesn't create any headless things in addition to what you're expecting it create 07:40:02 rainemak, how would that manifest itself? I've been careful to ensure there's only one display value created. Is there anything else to check for this you can think of? 07:40:10 flypig, also gl makeCurrent and swap I'd check 07:40:13 wow I know it was a large task but when you see one person tackle it in their spare time you really get the scale of it. So impressive that you're still at it. I hope you're still enjoying it! 07:40:35 DeclarativeWebContainer::clearWindowSurface 07:40:45 ^ is that being called properly 07:40:52 abr, it's only big because I'm incompetent haha! If I was competent it would have been a much quicker task :D 07:41:43 flypig, no you're not imcompotent! It's just a bug task. 07:41:43 Don't blame yourself flypig, "you're our only hope" ;op 07:41:52 Haha :) 07:42:02 While the rest of us are just reading your blogs looking for words we might know 07:42:14 That's in sailfish-browser rainemak? I've not checked that; I'll do so. 07:42:40 flypig, I've done it once alone and once with you... and you know, it's a big task 07:42:47 (I didn't really change anything in sailfish-browser, but maybe there is a link missing somewhere in between). 07:43:27 The fact you're reading is very motivating. 07:43:32 it seems like this one needs more changes in the layers above than usual too 07:43:57 abr, specifically for offscreen rendering, it seems so. They completely removed the normal flow we use. 07:44:20 yes that sort of thing is rewrite time 07:45:09 flypig, I guess you have tried it so that you have been pushing foreground app (browser) to background and raising something else to foreground and then browser back to foreground 07:45:10 sorry don't mean to interrupt. please carry on! just wanted to agree that this is a meeting well spent if it's all on browser upgrade. 07:45:34 rainemak, I have, since that was causing hangs before, so I've been doing that a lot. 07:45:40 Alright, I'm going to check on the clearWindowSurface side of things and everything between that and CompositorOGL::EndFrame(). 07:45:40 abr, +1 for that 07:46:17 flypig, that's at least scheduling glclear 07:46:37 Okay, that's useful to know. Then maybe I can put lots of readPixels in between the two to find out exactly which step is failing. 07:47:15 This is really helpful. Thanks so much! 07:47:29 alright, I guess that is about it 07:47:49 next meeting scheduling? 07:48:02 Thank you for the great help and for indulging me. 07:49:13 #topic Next meeting time and date (2 mins) 07:49:13 Proposing Thursday 25th April at 07:00am UTC 07:49:29 Works for me! 07:49:53 Same here. 07:52:38 flypig, sorry for poking still 07:52:49 Not at all. 07:53:07 it has to be before https://www.khronos.org/opengl/wiki/Framebuffer#Blitting as you're not even seeing clear color 07:53:33 #info Next meeting will be held on Thursday 25th April 2024 at 07:00am UTC: 2024-04-25T0700Z 07:55:15 thank you everybody 07:55:23 Thank you, everybody! Please feel free to continue your development discussion. 07:55:31 Thank you rainemak and all! 07:55:34 I'll do so 07:55:36 #endmeeting