| uvos | Wizzup: lel died | 00:15 |
|---|---|---|
| uvos | Wizzup: But what if the contact is renamed? @name from db | 00:16 |
| uvos | i gues the issue you take with sphone looking up the contacts is that it can be a bit slow | 00:17 |
| uvos | ideal solution for that would be for sphone to show the list with the db name right away and then update it as the lookups come in. | 00:17 |
| uvos | sicelo: :\ | 00:18 |
| inky | > At last I found reason for unexpected power-offs with Pinephone Keyboard, reason is simple phone side battery can go empty in whole day use, although connected with charger. | 00:45 |
| inky | i had to fix hw of pinephone. | 00:45 |
| inky | shorten the diod. | 00:45 |
| Wizzup | uvos: then the remote_name will be updated, like on maemo | 00:45 |
| inky | here https://xnux.eu/pinephone-keyboard/faq.html | 00:50 |
| inky | section: | 00:50 |
| inky | Keyboard battery is charging slowly or not at all | 00:50 |
| inky | somewhere was a tutorial with photos. | 00:50 |
| uvos | Wizzup: dosent happen atm | 00:50 |
| uvos | also not sure how you invision that working | 00:51 |
| inky | https://forum.pine64.org/showthread.php?tid=16979&pid=111414#pid111414#ts | 00:52 |
| uvos | since eds is a shared database any random application can decide to edit. So the only way i see this working is if eds gets a plugin to edit rtcom-el's database | 00:52 |
| inky | oh kiva is not here. | 00:52 |
| doc|home | is there a way to reset the lock screen on an n900? For the life of me I can't remember the code | 05:42 |
| Wizzup | I think there is some rescue image to reset it, you could probably also remove the xsession that waits for the lock code from leste and then try to recover from there | 05:45 |
| doc | Wizzup: is this it by any chance: https://github.com/quitesimpleorg/N900_RescueOS ? | 05:58 |
| Wizzup | that could be, sorry, I don't know | 05:59 |
| doc | ok, thanks | 05:59 |
| Wizzup | I think one of them came wiht john, I am not sure | 06:00 |
| Wizzup | I think this was brought up in the last month or so here | 06:00 |
| inky | in n900 repos there is jack the ripper, software that will find/brutforre key for you. | 11:30 |
| inky | if you have ssh to device you can find the code if a second. | 11:30 |
| inky | i once forgot it and recovered that way. | 11:31 |
| sicelo | uvos: why :-/ | 12:30 |
| uvos | sicelo: n900 dtmf bitrot | 12:42 |
| uvos | Wizzup: irc log/lel is still broken | 12:42 |
| sicelo | ok. thought it was in connection with the sphone feature request to show call direction in the log | 12:57 |
| sicelo | the dtmf issue . | 12:58 |
| uvos | sicelo: no i think thats a good idea | 12:58 |
| uvos | i just have to think about how to do it | 12:58 |
| uvos | in portrait there is little space | 12:58 |
| sicelo | i could look into n900 dtmf issue someday (but list of things i want/wish to do is growing) | 12:59 |
| sicelo | uvos: yes i noticed @ portrait. if push comes to shove, just a thin column, that fits a small emoji (or icon) ⬆️⬇️ would already make a big difference | 13:01 |
| arno11 | sicelo: uvos: dtmf still works on n900 :) | 14:44 |
| arno11 | able to control my voice mail | 14:44 |
| arno11 | (i made a mistake with busctl when i previously tested dtmf...) | 14:46 |
| arno11 | so 'busctl call org.ofono /n900_2 org.ofono.VoiceCallManager SendTones s 3' works for example | 14:49 |
| arno11 | there is no audible sound generated in speaker, but the command is well recognised by my voice mail | 14:53 |
| arno11 | the problem with n900 is that the keypad is not visible in sphone in landscape. | 14:59 |
| arno11 | but i found a solution last year iirc, i have to check | 15:01 |
| uvos | arno11: sphone only shows the keypad in call when the backend supports dtmf | 15:08 |
| uvos | currently no sphone backend besides the dummy commtest backend supports dtmf | 15:08 |
| arno11 | ok | 15:09 |
| sicelo | arno11: yay!!!! | 15:54 |
| sicelo | @ dtmf still working, that is. i think n900 side - fremantle had something like tonegend or similar. fmg will know best what's what | 15:54 |
| arno11 | ok, thx again for the busctl command | 15:56 |
| sicelo | :-) | 15:58 |
| * sicelo realized he does miss the fremantle feature of vol-keys changing direction according to orientation ... or must blame muscle memory (even though using android for years as well) | 16:30 | |
| arno11 | vol-keys changing direction works with audacious btw, not with orientation but when you close the keyboard. it is like a 'walkman mode' | 16:35 |
| sicelo | oh, interesting. it's their builtin feature? | 16:38 |
| arno11 | yes (with screen off btw) | 16:54 |
| Wizzup | 12:58 < uvos> in portrait there is little space | 17:22 |
| Wizzup | maybe look at fremantle | 17:22 |
| Wizzup | they have a dialer, and then in call, you can get a dialog | 17:22 |
| uvos | ? | 17:23 |
| uvos | you mean for dtmf? | 17:23 |
| Wizzup | yes, for buttons | 17:23 |
| uvos | thats how android and sphone works | 17:23 |
| uvos | allready | 17:23 |
| uvos | this was about adding more info to recents | 17:23 |
| Wizzup | k | 17:23 |
| uvos | sicelo: the vol keys do change direction according to orientation | 17:25 |
| uvos | i gues you turned it off | 17:26 |
| uvos | since i did add a option for this in the volume applets ini | 17:26 |
| uvos | but its on by default | 17:26 |
| uvos | and i just confirmed it dose still work absolutly fine | 17:30 |
| uvos | (when you turn it on) | 17:30 |
| uvos | i think we should turn it off by default, since this feature is bad for muscle memory even if you can get used to it and its pretty broken (also on freemantle) since it sometimes works but sometimes doset | 17:31 |
| uvos | since its related to h-ds rotation, but some apps rotate themselves (like esp egregious tklock but also plenty of others) | 17:33 |
| uvos | and in those apps you end up in portrait but with the volume buttons not swapped | 17:33 |
| uvos | and in apps where the volume buttons do something other than volume, its up to those apps to swap the direction, which generally they do not | 17:34 |
| uvos | all in all this makes which of the volume buttons being + at any given moment to be essentaly random | 17:34 |
| uvos | and then you have the issue that when the display is off, you need to remember what oritentation the device was in when you turned the display off to figure out what the volume buttons will do | 17:37 |
| uvos | over all this feature is just broken and terrible ux in general. | 17:37 |
| uvos | early versions of android also had this feature btw, with simmilar problems - they eventually gave up on it with imo good reason | 17:40 |
| kiva | Wizzup: screen brightness control in settings works with PP, but change is smooth and takes some seconds. | 17:54 |
| Wizzup | are you on -devel | 17:57 |
| kiva | no I am on stable | 17:58 |
| Wizzup | ok, maybe that is the difference | 17:59 |
| kiva | inky: that diode remove for from Pinephone Keyboard to get faster charge between batterys is little bit risky, because I think that remove over voltage protection. | 18:03 |
| kiva | My temporary solution is check charge of both batterys with command: grep -E 'CAPACITY|STATUS' /sys/class/power_supply/*/uevent" | 18:08 |
| kiva | without that "-char in end :) | 18:08 |
| sicelo | kiva: so the keyboard's battery does show up in /sys/class/power_supply/ ? | 18:11 |
| kiva | yes | 18:11 |
| kiva | that command shows like this: /sys/class/power_supply/axp20x-battery/uevent:POWER_SUPPLY_STATUS=Charging | 18:13 |
| kiva | "/sys/class/power_supply/axp20x-battery/uevent:POWER_SUPPLY_CAPACITY=54 | 18:14 |
| kiva | "/sys/class/power_supply/ip5xxx-battery/uevent:POWER_SUPPLY_STATUS=Full | 18:14 |
| kiva | "/sys/class/power_supply/ip5xxx-battery/uevent:POWER_SUPPLY_CAPACITY=100" | 18:14 |
| sicelo | cool. if you don't mind, paste output of `upower -d` on some pastebin. | 18:15 |
| kiva | here you are: https://pastebin.com/FdGEhFVr | 18:23 |
| kiva | and sudden power off comes when axp20x (phone side battery) gets 0 and maemo battery indicator near clock says full. | 18:28 |
| kiva | I just measured that charger draws 5.7 W when Keyboard side battery is full and phone side not...could that be near charging power between batterys? | 18:30 |
| buZz | in power_supply/* somewhere you can find how hard its charging too | 18:32 |
| buZz | i made https://space.nurdspace.nl/~buzz/maemo/batmon.py which isnt very finished | 18:32 |
| buZz | but can show how much power goes into device or battery | 18:32 |
| buZz | https://space.nurdspace.nl/~buzz/maemo/2023-01-13-044313_960x540_scrot.png <- green is 'total coming in from USB' , but the purple part is 'gets removed to operate the phone' | 18:33 |
| buZz | the red is 'discharging from battery | 18:33 |
| buZz | ' | 18:33 |
| buZz | the darkgreen in background is battery voltage | 18:33 |
| buZz | also, i made this on droid4 , you might need to edit some things for pp or n900, but not a lot | 18:34 |
| kiva | "/sys/class/power_supply/axp20x-battery/uevent:POWER_SUPPLY_VOLTAGE_NOW=3811000" | 18:37 |
| kiva | "/sys/class/power_supply/axp20x-battery/uevent:POWER_SUPPLY_CURRENT_NOW=81000" | 18:37 |
| sicelo | kiva: the battery icon uses upower, hence the request for `upower -d` | 18:38 |
| kiva | and it shows full right now | 18:40 |
| Wizzup | buZz: does that use upower? | 18:40 |
| Wizzup | ah, no | 18:41 |
| buZz | it parses the /sys/class files | 18:42 |
| buZz | also, there's still a weird bug that makes it use a lot of cpu all the time, while not doing anything | 18:42 |
| buZz | i must misunderstand the QT event loop :P but this was my first qt python thingy | 18:43 |
| buZz | the high cpu isnt really noticable though, i mean, on droid4 i barely notice it from other applications | 18:43 |
| sicelo | kiva: thanks for that output. i guess battery_ip5xxx is keyboard's charger? needs fixing since it currently shows up as a battery, not charger/line-power | 18:50 |
| kiva | I think battery_ip5xxx is battery of keyboard. | 18:52 |
| kiva | and charger is Device: /org/freedesktop/UPower/devices/line_power_ip5xxx_boost | 18:55 |
| sicelo | sorry, i meant evices/battery_ip5xxx_usb native-path: | 18:56 |
| sicelo | ip5xxx-usb in sysfs ... that sounds like a charger to me | 18:57 |
| kiva | grep -E '=' /sys/class/power_supply/ip5xxx-usb/uevent | 19:04 |
| kiva | POWER_SUPPLY_NAME=ip5xxx-usb | 19:04 |
| kiva | POWER_SUPPLY_TYPE=USB | 19:04 |
| kiva | POWER_SUPPLY_PRESENT=1 | 19:04 |
| kiva | if I take charger out from keyboardside USB port PRESENT=0 | 19:05 |
| sicelo | exactly :-) | 19:06 |
| sicelo | so yeah, that driver is reporting it as battery, not charger. it's fixable - n900 had same. or someone already fixed it upstream by now | 19:08 |
| sicelo | i don't like that 153% in DisplayDevice ... | 19:09 |
| kiva | but Maemo does not show that like PostmarketOS do (or did, I have not used it at all anymore) | 19:10 |
| kiva | Maemo shows keyboard side battery near clock, not phone battery itself, so that is reason why it can go 0 without warnings. | 19:13 |
| buZz | might just need some method added to show two batteries | 19:17 |
| kiva | if keyboard side battery is empty and you restart then Maemo shows phone side battery. | 19:17 |
| kiva | buZz: tha would be best. | 19:17 |
| buZz | right? there is plenty of space for another percentage i think | 19:17 |
| buZz | kiva: maybe someone can point you to where in our sourcecode that stuff exists, or you can find it yourself | 19:18 |
| buZz | improve it, send a patch :) | 19:18 |
| kiva | sorry I cannot C++ that much, I am old school basic+assembler guy. | 19:19 |
| sicelo | https://github.com/maemo-leste/status-area-applet-battery | 19:20 |
| sicelo | it currently takes first battery reported by upower. we could make it parse all of them | 19:21 |
| kiva | would it help also Droid4 users that have Lapdock connected? | 19:27 |
| sicelo | no | 19:30 |
| sicelo | https://github.com/maemo-leste/status-area-applet-battery | 19:30 |
| sicelo | the lapdock just acts like a power bank | 19:31 |
| sicelo | afaik | 19:31 |
| Wizzup | I think it's exposed as a second battery at least | 19:31 |
| Wizzup | oh lapdock, sorry | 19:32 |
| Wizzup | got I'm hazy today | 19:32 |
| Wizzup | god* | 19:32 |
| buZz | sicelo: quite sure thats what kiva wants :) (to see all batteries) | 19:41 |
| buZz | no idea if we support any other device where >1 battery exists | 19:41 |
| sicelo | kiva: is last comment in https://github.com/maemo-leste/bugtracker/issues/726 actually true for ppkb? i.e. when ppkb has > 0%, internal battery is guaranteed to also report > 0%, or at least be in charging state | 19:41 |
| buZz | maybe some x86 laptops i guess | 19:41 |
| sicelo | buZz: yes i understand. i've brought up this topic here before (leading up to that issue 726). at the time, i was fixing the battery reporting wrt N900, which also has two batteries according to kernel (we have two fuel gauges for the same battery) | 19:44 |
| buZz | ooo that makes that all a bit more difficult | 19:46 |
| buZz | are the two gauges even in agreement on n900? :D | 19:47 |
| kiva | sicelo: at least it does no power off before internal battery goes 0% | 19:48 |
| sicelo | buZz: yes, although each provides slightly different properties. e.g. rx51-battery reports charge design full, while bq27200 doesn't (it can't). | 19:50 |
| kiva | buzz: it would it be cool splitted hardware battery, N900 always on, only change half of battery ;) | 19:50 |
| sicelo | kiva: ok. maybe someday i can work on my #2 idea from that issue. not anytime soon though | 19:51 |
| kiva | that sounds good. | 19:54 |
| sicelo | but i think i have some basic clue how to at least start. i think a major issue would be - how to know which icon is for which battery :-P | 19:54 |
| buZz | kiva: you can subscribe to that github issue to get email notified when stuff happens ;) | 19:55 |
| buZz | sicelo: maybe some simple gravicon alike setup? just turn whatever device name into some basic image | 19:56 |
| kiva | sicelo: if there is two icons, order would be always same? | 20:00 |
| sicelo | upower asks udev for devices in power supply ... no idea if they're reported in any special order | 20:04 |
| kiva | how abou color number 8 first bits based to device name ascii-number. | 20:06 |
| kiva | assuming that display uses at least 16-bit colors. | 20:09 |
| kiva | btw..old Apple Powerbook G3 had possibility for two batterys, if take CD-ROM drive out...but Debian had drop support for 32-bit PowerPC..so no Leste for that :( | 20:18 |
| kiva | sicelo: actually it is easy know which icon is for which battery, phone side battery always charging and keyboard side battery always discharging when PPK is without charger. | 20:54 |
| sicelo | the applet needs to be generic :-) | 20:57 |
| kiva | then something color difference based to device name. | 21:11 |
| buZz | gravitar, thats the algorithm i ment for 'colored shape from namestring' | 21:38 |
| sicelo | link? | 21:54 |
| buZz | it seems closed https://docs.gravatar.com/api/avatars/hash/ | 22:04 |
| buZz | ah https://www.libravatar.org/ | 22:05 |
| buZz | maybe something like https://robohash.org/ https://github.com/e1ven/Robohash would be more fun ;) | 22:07 |
| buZz | also handy for the addressbook :P | 22:07 |
| arno11 | Wizzup: i fixed the sound remixing issue on n900 locally: it was not possible to take/make a call (properly) while playing music or something else playing any sound | 23:01 |
| arno11 | remixing is deactivated by default in daemon.conf | 23:01 |
| inky | kiva, are you still here? there is a debian fork for ppc32. and gentoo. i can find you link to ppc32 unofficial debian. | 23:10 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
