| *** spiiroin_ is now known as spiiroin | 07:08 | |
| poetaster | piggz[m], this the F(x)tec Pro¹-X rootfs was shown to successfully boot on a Samsung Galaxy Z Fold2! is a neat trick with a knife! | 08:33 |
|---|---|---|
| direc85[m] | That *is* neat! | 08:59 |
| nephros | cool, that means I can order one now? ;) | 09:11 |
| poetaster | I always found the idea of 'folding' a phone slightly 'ewww'. my friend tuna has one. his kids use it to play fortnite ;-> | 11:10 |
| inz | I hope he doesn't use it to play poker. | 11:16 |
| *** attah_ is now known as attah | 11:20 | |
| poetaster | piggz[m], before I start making tickets, if the browser on both GS5/GS290/volla get's the correct DPI, then the issue with scaling webviews can't be port related, can it? | 12:29 |
| poetaster | piggz[m], I had tried adjusting /var/lib/environment/compositor/droid-hal-device.conf (QT_QPA_EGLFS_PHYSICAL_WIDTH, etc) values, but those are or no account | 12:30 |
| *** attah_ is now known as attah | 12:45 | |
| rubdos[m] | Is there an easy way to temporarily disable screen blanking and locking? | 13:05 |
| rubdos[m] | Just from CLI would be okay | 13:05 |
| mal | rubdos[m]: check mcetool | 13:24 |
| mal | options -P and -v | 13:25 |
| mal | in mce-tools package | 13:25 |
| rubdos[m] | thanks! That seems to be exactly what I was after :-) | 13:31 |
| rubdos[m] | Oh there's also an option to keep the screen on while charging | 13:34 |
| rubdos[m] | that's probably even better | 13:34 |
| rubdos[m] | I'm running some benchmarks via USB ... | 13:34 |
| poetaster | piggz[m], I think it's not hardware/adaption after all: https://forum.sailfishos.org/t/webview-produces-incorrect-screen-height-window-innerheight-values/13328/14?u=poetaster | 14:02 |
| poetaster | piggz[m], flypig mentioned that the calculation of pixelRatio might not work well on the vollas: https://github.com/sailfishos/sailfish-components-webview/blob/8c0fefae1ba56ed351a570341df48e91d974928b/lib/webenginesettings.cpp#L178 | 14:33 |
| flypig | poetaster, I'm also a little concerned about why that value wasn't set in your ~/.cache/org.myorg/webview-screensize/.mozilla/prefs.js file. | 14:35 |
| flypig | It might be worth you reverting your QML change to set the value, removing the ~/.cache/org.myorg/webview-screensize folder, and (crucially) also deleting the ~/.local/share/org.myorg/webview-screensize/__PREFS_WRITTEN__ file to see whether it gets generated properly on a fresh init. | 14:36 |
| poetaster | yeah, it's the same case in all apps that are using webview. | 14:36 |
| flypig | It could be that this is the problem, rather than the algorithm generating the wrong value. | 14:37 |
| poetaster | ah, sec. | 14:37 |
| poetaster | and right you are, it seems. | 14:39 |
| poetaster | ah, perhaps this is residual config? | 14:39 |
| flypig | I'm not sure. It's a bit confusing, because the setting of the value wasn't a new change, so it would have to be quite an old config for it to be missing. | 14:41 |
| poetaster | Hmmm. Well, I've been dragging the configs around since those apps were created, BUT, the more recent case of Tidings, where I just recently added a new webview, can't be residual. but has the same issue. | 14:43 |
| poetaster | Ah, I noticed that the issue does not appear on volla 4.4.0.64 using the same backup .config / .local but .cache is not coming ouf of backups | 14:44 |
| poetaster | flypig, that remove __PREFS_WRITTEN method works, in all cases tested (7 apps). | 14:50 |
| flypig | The __PREFS_WRITTEN__ lock was added to avoid user-configured preferences from getting trashed each time the app was run, if I recall correctly. | 14:53 |
| poetaster | flypig, makes perfect sense. you also didn't have the setting in ~/.cache/org.myorg/webview-screensize/.mozilla/prefs.js on the 10iii? | 14:54 |
| flypig | The setting was there on my 10 III. | 14:55 |
| flypig | Sorry, 10 II. | 14:56 |
| poetaster | ah, neato. an ever more interesting array of issues. Ah, progress :) | 14:56 |
| poetaster | I had done a bunch of checks on changes from .64 -> .68 on the vollas and the only thing that stuck out was qtscenegraph which doesn't seem pertinent. | 14:58 |
| poetaster | In any case, newly installing the same packages on a fresh .64 volla/gs290 leads to the prefs being written properly. On the fresh install of the GS5 with .72 the inits are not written correctly. | 14:59 |
| poetaster | flypig, I just checked the updated tidings on the x10ii and the setting was correctly set. | 15:05 |
| flypig | So, only on .72 Volla the setting is skipped? | 15:05 |
| flypig | Just this setting, or the entire prefs.js file? | 15:06 |
| poetaster | tricky question. I believe so, but it could have started as of .64 -> .68. I just noticed it when doing a clean install which on the gs5 starts with .72. | 15:09 |
| poetaster | on a parallel (my main device got wrecked) gs290 I saw the same issue, but that was an install .64 and upgrade, upgrade + re-install. | 15:10 |
| flypig | I'm confused, I must admit. I don't see why the device type would make a difference to these values being set and/or saved out. | 15:11 |
| poetaster | flypig, I think it must be residual config. I just checked the backup I used to recover my tidings values and the __PREFS_WRITTEN file is there. | 15:13 |
| poetaster | flypig, so that if I introduced the WebView after that, the setting would not have been updated, correct? | 15:13 |
| flypig | The __PREFS_WRITTEN__ file is application-specific though, so a Tidings __PREFS_WRITTEN__ shouldn't affect a webview-screensize __PREFS_WRITTEN__, say. | 15:15 |
| flypig | But what you describe could happen if you renamed the tidings folder in ~/.local/share to the webview-screensize folder in ~/.local/share. Is that possible (it would have had to be a very deliberate action, I think). | 15:17 |
| poetaster | flypig, obviously, but it's a general problem if __PREFS_WRITTEN is copied along with backups of .local/share application data (like sqlite files, &co) | 15:18 |
| flypig | Ah, yes, that is true. | 15:18 |
| poetaster | I think it's just residual config. damn it. | 15:18 |
| poetaster | that's great to know cause I can stop bothering you and piggz and figure out how to deal with it :) | 15:18 |
| poetaster | THANKS! damn it. | 15:19 |
| flypig | Okay, this is good :) I feel your frustration at it being a config problem and not something more fundamental. But it's good to get to the bottom of it, | 15:20 |
| flypig | And it indicates something brittle... I'm not sure what the solution to that is, but it's worth knowing about. | 15:20 |
| poetaster | flypig, it's mostly an issue for apps that are QML only. There is no simple programmatic way to deal with this case. Not easy to guide the user through. | 15:21 |
| flypig | Yes, that's tricky. | 15:22 |
| poetaster | but no show stopper. once your orientation fix gets into the next release, which it will :) | 15:23 |
| flypig | I hope so :) And it's good that you found that orientation problem, that was nasty (from a user's point of view). | 15:23 |
| poetaster | Worse, since my initial discovery was while working with updated upstream code (the mahjong app) I initially filed a bug report in the wrong universe! | 15:25 |
| poetaster | flypig, did you go play https://openrepos.net/content/poetaster/molecules yet (I'm working on more levels for stuff :) | 16:44 |
| flypig | Oh, nice, I've not played it. I've played Osmos though (I'm not very good at it; maybe I'll be better with a touchscreen!). | 17:10 |
| nephros | poetaster: balldrop doesn't seem to take any input on my 10iii, is that a kown issue? | 18:16 |
| nephros | launching it I get small dots floating, but cant draw lines or anything. | 18:17 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!