| bencoh | (say, I wanna run sigstoped) | 00:00 |
|---|---|---|
| parazyd | Maybe check if XDG autostart is supported | 00:00 |
| parazyd | I don't actually know though | 00:00 |
| parazyd | https://wiki.archlinux.org/title/XDG_Autostart | 00:00 |
| stan | do you run sessions independtly from booting the device? | 00:00 |
| parazyd | If not, it's probably something we'd want to support | 00:00 |
| stan | i start things in /etc/rc.local | 00:00 |
| stan | oh right some things need running services and x | 00:01 |
| bencoh | xdg-user-dirs is installed at least | 00:02 |
| Wizzup | bencoh: I'd go for xsession if it's for a package | 00:02 |
| bencoh | Wizzup: in /etc? | 00:02 |
| Wizzup | if it's your own thing, then yeah, .xsession or so | 00:02 |
| Wizzup | if it's a package, yeah | 00:02 |
| bencoh | the thing is ... https://wiki.debian.org/Xsession | 00:03 |
| parazyd | Thing is our code might not read it at all | 00:03 |
| parazyd | The local ones | 00:03 |
| bencoh | does it look for x-session-manager and friends if ~/.xsessionrc is present? | 00:03 |
| parazyd | Don't think so. We aren't running h-d as a session | 00:03 |
| bencoh | ah, got it | 00:04 |
| parazyd | It's just a sub-component of the default session iiuc | 00:04 |
| bencoh | .xsession *replaces* syswide session manager, .xsessionrc is just read on top of it | 00:04 |
| bencoh | so let's try the rc thing | 00:04 |
| _uvos_ | Wizzup: sweet :) ill try it as soon as im back @home | 10:36 |
| _uvos_ | so are you ignoreing the return value or did you figure out why wpa was returning -1? | 10:36 |
| stan | old EB41 battery was reporting only 1040mAh. Tossed in a 3800mAh today and it loves me long time. http://0x0.st/-fL4.mkv | 11:26 |
| Wizzup | uvos: former, I need more time to figure out why, but the code has no debug statements at all, so I'd have to add them and recompile a new .a file | 11:39 |
| lel | MerlijnWajer synchronize a pull request: https://github.com/maemo-leste/connui-internet/pull/2 (WIP: add providers tab to advanced iap settings) | 14:54 |
| sicelo | stan: mine reports 811mAh | 16:02 |
| stan | if you want to buy a replacement battery, the one uvos reccommended looks good | 16:03 |
| stan | i had a bigger one lying around, so i sacrificed a back cover | 16:03 |
| Wizzup | normal aging battery still lasts me 2 days or so | 16:04 |
| Wizzup | usually I find a place to charge it by then | 16:04 |
| sicelo | i just need to find the highest capacity that'll fit - have no plans to host an oversize battery | 16:05 |
| sicelo | Wizzup: what capacity does it report? | 16:05 |
| Wizzup | 1301mAh | 16:06 |
| sicelo | heh, so mine is super finished then :D | 16:07 |
| Wizzup | this is from status applet btw | 16:08 |
| Wizzup | not sure where you look | 16:08 |
| sicelo | yes, same | 16:08 |
| stan | <uvos> https://www.ebay.de/itm/PolarCell-Akku-f%C3%BCr-LG-Google-Nexus-4-E960-Optimus-G-E975-E970-BL-T5-Batterie/331503392013?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2060353.m2749.l2649 | 16:08 |
| stan | i haven't found a bigger one that still fits | 16:08 |
| sicelo | i wonder if there's no site that allows you to search LiPo/LiOn batteries by physical size and capacity | 16:09 |
| sixwheeledbeast | https://www.polarcell.de/en/mobile-phone-batteries/lg/others-polarcell-li-polymer-replacement-battery-for-lg-google-nexus-4-e960.html | 16:11 |
| sixwheeledbeast | direct site not ebay | 16:11 |
| sixwheeledbeast | if that helps | 16:11 |
| bencoh | wow, I didn't think they'd sell directly | 16:11 |
| * sicelo wants/needs a battery that's available at least in South Africa | 16:12 | |
| sixwheeledbeast | wannasee electronic in berlin | 16:12 |
| stan | good advice sicelo | 16:13 |
| stan | sixwheeledbeast: | 16:13 |
| sixwheeledbeast | i have recommended those batteries for N900/5800 for years. | 16:14 |
| stan | had good results with their n900 replacement also | 16:14 |
| stan | yeah that is only 5-6mm shorter than original EB41. unlikely we'll find a bigger one | 16:15 |
| sicelo | mmm, they say they send batteries as a "letter shipment" ... i.e., they put battery in envelope and send? | 16:20 |
| stan | in germany there's a shipment class for small cartons called 'letter' but you'd have to check with them to be sure | 16:21 |
| sicelo | i see their site (polarcell) has Swaziland as a shipping destination ... maybe i should try and buy when things quiet down here | 16:22 |
| stan | my big problem with those is soldering wires to the zinc tabs. i need to spot-weld wires to those. | 16:23 |
| sicelo | well i need a battery for my S7 too - no mods needed. wouldn't be a bad idea to get one for the N900 as well | 16:24 |
| stan | ok i went through all polarcell.de batteries - this is the best for a modded case, 6mm 78x46x6 - with a repositioned controller it's about 74x46x6 https://www.polarcell.de/en/mobile-phone-batteries/samsung/galaxy-s-series-polarcell-li-polymer-replacement-battery-for-samsung-galaxy-s9-plus-sm-g965f.html | 17:58 |
| stan | Siarhei Siamashka seems to have a kernel patch for that verycheap/small USB Otg hub https://linux-sunxi.org/USB_OTG_Charging_Hub | 19:00 |
| freemangordon | Wizzup: BTW, IIRC there is advanced tab for non-wifi connections as well | 21:45 |
| freemangordon | that's provided by default | 21:45 |
| Wizzup | freemangordon: what is that in response to? | 21:46 |
| freemangordon | just a random thought | 21:46 |
| Wizzup | ok | 21:46 |
| Wizzup | so there is a way to load a plugin for connui for advanced tabs | 21:47 |
| Wizzup | but the problem is you can only have one per connection type | 21:47 |
| Wizzup | (so it's not fit for providers) | 21:47 |
| freemangordon | I think maybe it is better to have that providers tab there by default | 21:48 |
| freemangordon | but lets not discuss it now | 21:48 |
| freemangordon | I am too tired and could be that I am talking nonsense | 21:48 |
| Wizzup | freemangordon: yes, I implemented that (as wip) in connui | 21:48 |
| Wizzup | https://github.com/maemo-leste/connui-internet/pull/2 | 21:48 |
| freemangordon | nice | 21:49 |
| freemangordon | but, wlan wizard overrides this, no? | 21:49 |
| Wizzup | no | 21:49 |
| freemangordon | ah, right | 21:49 |
| freemangordon | it just adds "Other" page | 21:49 |
| Wizzup | so connui-internet can load one plugin for connection/iap type, but this is always there, like proxies and ip settings | 21:49 |
| Wizzup | yeah | 21:50 |
| freemangordon | yeah | 21:50 |
| Wizzup | I need to dynamically load some of it from gconf I guess, as opposed to hardcoding the providers | 21:50 |
| freemangordon | sure | 21:50 |
| Wizzup | and that is a bit tricky since the code is spread over two files, but I'll figure it out | 21:50 |
| freemangordon | also, it makes no sense to add the page if tehre are no providers installed | 21:50 |
| Wizzup | right, we could probably skip that, otoh, we might make them part of the core pkg (depending on size and such) | 21:51 |
| Wizzup | s/core pkg/core install/ | 21:51 |
| uvos | Wizzup: looks like the wlan scanning issue is now resolved | 22:10 |
| uvos | however another new issue exists | 22:10 |
| uvos | i think you know this one | 22:10 |
| Wizzup | what is it? | 22:10 |
| uvos | but if i click on some network i dont have psk for and then cancle and reopen the dialog | 22:10 |
| uvos | the networks are scanned and displayed just fine but everything is grey | 22:11 |
| uvos | i have to close the dialog again | 22:11 |
| uvos | and reopen it again to accually connect | 22:11 |
| bencoh | the wifi Hydra | 22:11 |
| Wizzup | uvos: is your modem online? | 22:12 |
| uvos | Wizzup: no | 22:12 |
| uvos | this is with no sim | 22:12 |
| Wizzup | do you have libicd-network-ofono installed? | 22:12 |
| uvos | yes | 22:12 |
| Wizzup | there is some odd interaction with icd2 and plugins that do not finish searching that causes this | 22:12 |
| uvos | ok | 22:12 |
| Wizzup | i.e. you will be seeing it, but most others won't | 22:12 |
| Wizzup | if you online the modem I think the problem will be gone | 22:13 |
| uvos | ok | 22:13 |
| Wizzup | but yeah, it warrants a deeper look | 22:13 |
| uvos | ok | 22:13 |
| uvos | seams sensless to block every connection untill every plugin is ready | 22:13 |
| Wizzup | see 98a6ea87cd5a073ef3352ac86e56a503376593f5 on connui-internet | 22:13 |
| Wizzup | https://github.com/maemo-leste/connui-internet/commit/98a6ea87cd5a073ef3352ac86e56a503376593f5 | 22:14 |
| Wizzup | I guess it needs further work | 22:14 |
| freemangordon | uvos: actually kinda makes sense, otherwise you risk clicking a connection you don;t want to | 22:18 |
| bencoh | freemangordon: I think that's how it works on fremantle | 22:19 |
| freemangordon | i.e. - the results might be delivered the moment you click | 22:19 |
| uvos | true, but some connection types can take a loong time to discover | 22:19 |
| Wizzup | yeah but if we know the ofono plugin isn't going to return anything (modem offline), it should not block | 22:19 |
| uvos | so ux wise its worse if it just unblocks immidatly | 22:19 |
| freemangordon | but, it should return immediately then, no? | 22:19 |
| bencoh | at least that's the end result for the user on fremantle, I dunno how they implemented it | 22:19 |
| Wizzup | freemangordon: yes, but it doesn't atm | 22:20 |
| uvos | i think idealy maybe it would unblock all connections per module | 22:20 |
| uvos | when that module is finished | 22:20 |
| uvos | and show modules in the oder the report ready | 22:20 |
| uvos | then you cant missclick because a new module comes in | 22:20 |
| uvos | and fast modules arnt penalized by slow ones | 22:20 |
| freemangordon | uvos: the problem is that connections are ordered by priority, IIRC | 22:21 |
| freemangordon | saved>GPRS>WLAN etc | 22:21 |
| uvos | freemangordon: saved first certenly makes sense | 22:22 |
| uvos | but we could order per module too | 22:22 |
| uvos | i guess | 22:22 |
| freemangordon | the point is that if WLAN is ready before GPRS, once GPRS becomes ready its connections should go on top of WLAN | 22:22 |
| freemangordon | so the list is reordered | 22:23 |
| uvos | not sure why thats nessecary | 22:23 |
| freemangordon | because of the priority | 22:23 |
| uvos | right thats how it is now | 22:23 |
| freemangordon | GPRS > WLAN | 22:23 |
| bencoh | I'd rather say: because you don't wanna scroll down the list just to reach your gps connection | 22:23 |
| uvos | i dont see why this is important | 22:23 |
| freemangordon | and we want to keep it | 22:23 |
| bencoh | gprs* | 22:23 |
| freemangordon | bencoh: exactly | 22:23 |
| uvos | idk wlan is preferd over gprs no? | 22:23 |
| freemangordon | the same goes for saved connections | 22:24 |
| uvos | what youd rather have gprs or wlan? | 22:24 |
| bencoh | uvos: there might be a dozen unwanted wlan | 22:24 |
| uvos | surely wlan | 22:24 |
| freemangordon | wlan is preffered for auto-connect, but in the list GPRS has higher prio | 22:24 |
| Wizzup | uvos: I think we want to avoid things appearing at the top just when click/select something | 22:24 |
| Wizzup | then it can cause terrible UX | 22:24 |
| uvos | Wizzup: right | 22:24 |
| Wizzup | just when a user* | 22:24 |
| uvos | thats why im saying ditch the priority mechanisum | 22:24 |
| uvos | and just show in reporting order | 22:24 |
| freemangordon | no way | 22:24 |
| uvos | then you can missclick | 22:24 |
| uvos | *cant | 22:24 |
| stan | gprs appear first seems sensible | 22:25 |
| uvos | and the dialog can allways be unblocked | 22:25 |
| freemangordon | not until we have all the results | 22:25 |
| uvos | right because of the priority mecanisum | 22:25 |
| freemangordon | mhm | 22:25 |
| freemangordon | and it is here to stay | 22:25 |
| Wizzup | uvos: I think it's mostly a non issue if we fix the current bug (which it is) | 22:25 |
| uvos | you cant have both the priority and unblock before every module is ready | 22:25 |
| freemangordon | uvos: as Wizzup said - we have a bug to fix in one of the plugins | 22:26 |
| Wizzup | gprs is usually available instantly when the modem is online and the gprs is already provisioned | 22:26 |
| bencoh | uvos: seriously though, are you willing to scrol down everytime you want to connect to gprs? | 22:26 |
| uvos | bencoh: why would i ever dissconect gprs | 22:26 |
| uvos | so why would i care | 22:26 |
| uvos | the dialog is just for rare uses really (for me) | 22:26 |
| freemangordon | because you want to connect to WLAN | 22:26 |
| uvos | wlan autoconnects | 22:26 |
| uvos | so its just when i want to connect to new wlan | 22:27 |
| uvos | so quite rare | 22:27 |
| freemangordon | uvos: only if you ar @ home | 22:27 |
| freemangordon | *are | 22:27 |
| uvos | whenever i have something saved | 22:27 |
| stan | don't connect to any random wlan - e.g in a store or shopping mall | 22:27 |
| bencoh | even @home I don't use autoconnects (neither wlan, nor gprs) | 22:27 |
| bencoh | my battery thanks me :) | 22:27 |
| Wizzup | bencoh: were you planning to submit that wifi pkg to -extras? | 22:27 |
| freemangordon | I use 30 minuts for auto-connect interval | 22:27 |
| uvos | bencoh: well then you dont use any ip messaging | 22:27 |
| freemangordon | uvos: trust those who are experienced with maemo ;) | 22:27 |
| uvos | no | 22:28 |
| bencoh | Wizzup: yeah, I need to finish some cosmetics first | 22:28 |
| Wizzup | I don't really see the point of this discussion ( freemangordon uvos ) we just need to fix the bug and then all problems are gone | 22:28 |
| Wizzup | bencoh: cool | 22:28 |
| freemangordon | mhm | 22:28 |
| uvos | right fine | 22:28 |
| uvos | whatever | 22:28 |
| Wizzup | :p | 22:28 |
| bencoh | and for some reason, autotools insist on building static libs, even though I added AC_DISABLED_STATIC | 22:28 |
| bencoh | and LT_INIT([disable-static]) | 22:28 |
| Wizzup | maybe depending on how it's defined in Makefile.am it still does it? | 22:29 |
| freemangordon | ./configure --shared? | 22:29 |
| freemangordon | or what was the option to disable static... | 22:29 |
| Wizzup | uvos: well pls let me know if most of the other wifi problems are fixed for you as well, then I will push it to stable in 1-2 days | 22:29 |
| bencoh | freemangordon: the configure.ac macro should change default | 22:29 |
| bencoh | Wizzup: do you know if any platform should unload driver instead of just ifconfig down on leste? | 22:31 |
| uvos | no | 22:31 |
| uvos | dont ever do that | 22:31 |
| uvos | if the driver is that broken fix it please | 22:31 |
| bencoh | fine by me :) | 22:31 |
| Wizzup | bencoh: I would just go for disabling the iface yeah | 22:33 |
| uvos | (maybe alos use rfkill interface) | 22:33 |
| uvos | altho any reasonable driver will shutdown the chip with the if disabled and no other users | 22:34 |
| uvos | so why do we have lots of maemo stuff in sbin? | 22:45 |
| uvos | "/sbin contains binaries essential for booting, restoring, recovering, and/or repairing the system " | 22:45 |
| uvos | we have binarys in there that strech this definiton | 22:45 |
| uvos | (im noticing this because im diffing the output of the cmake mce package to the current one) | 22:46 |
| uvos | https://refspecs.linuxfoundation.org/FHS_3.0/fhs/ch03s16.html#purpose16 <--- defenition from here | 22:46 |
| Wizzup | mce is pretty essential for booting maemo | 22:46 |
| uvos | sortof | 22:47 |
| uvos | i think booting here is sysinit | 22:47 |
| uvos | mameo is not in sysinit | 22:47 |
| uvos | *mce | 22:47 |
| Wizzup | mgm, no strong opinions | 22:47 |
| Wizzup | mhm | 22:47 |
| uvos | maemo stuff i mean | 22:47 |
| Wizzup | although I am not sure if there are scripts that have the abs path in them | 22:47 |
| uvos | right | 22:47 |
| uvos | that might be a problem | 22:47 |
| uvos | also the system really works fine wihtout mce (just lacking the fatures you would expect) but you can boot to hildon so its not "essential" | 22:50 |
| uvos | i gues ill move it and see if anything breaks (and look at all scripts i can think of first ofc) | 22:50 |
| uvos | devlock-blocker is deff in the wrong place | 22:53 |
| uvos | its not even run by root | 22:53 |
| uvos | user dosent even have it in PATH per spec | 22:54 |
| Wizzup | uvos: assuming mce will online modem for us, it's quite essential for making phone calls | 22:56 |
| Wizzup | but it's kind of a moot discussion imho | 22:56 |
| uvos | sure | 22:56 |
| uvos | im just trying to do whats right per spec | 22:56 |
| uvos | the real diff is if user has it in PATH | 22:57 |
| Wizzup | mhm | 23:04 |
| freemangordon | Wizzup: I think in iap_advanced_create() there should be another section that checks if "service providers" page shall be added, like, have a separate iap_advanced_sp_pages struct that's processed only if needed | 23:05 |
| freemangordon | like, add similar code like https://github.com/maemo-leste/connui-internet/blob/4b46ef42f7b441c1b8667167543904d5bd58691b/src/settings/advanced.c#L811 and https://github.com/maemo-leste/connui-internet/blob/4b46ef42f7b441c1b8667167543904d5bd58691b/src/settings/advanced.c#L823 | 23:06 |
| uvos | Wizzup: so with jenkins mce would now have lots of recomended dependancies because it now works without any other maemo stuff | 23:07 |
| uvos | Wizzup: but the maemo stuff needs to be there at build time (or the relevant modules are not built) | 23:07 |
| uvos | is that possible to do somehow | 23:07 |
| uvos | or do i need to keep the "fake" hard build time dependacies for now? | 23:08 |
| uvos | (untill i also split mce into mce and mce-maemo-modules or something) | 23:08 |
| Wizzup | uvos: I don't really know how that works in debian packages | 23:13 |
| Wizzup | There is no equivalent to "use flags" on gentoo iiuc | 23:13 |
| uvos | Wizzup: ok | 23:13 |
| uvos | parazyd: ^^^ | 23:13 |
| uvos | parazyd: any ideas? how dose jenkins react to Recommends or Suggests | 23:14 |
| Wizzup | it won't install them for build | 23:14 |
| uvos | ok | 23:15 |
| uvos | hmm :( | 23:15 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!