**** BEGIN LOGGING AT Sun Jun 18 03:00:02 2017 Jun 18 10:55:40 Bug report: with profile General (not Silent), with IM sound set to 0 volume, with music being played in Media Player, IMs beep annoyingly on arrival. If Music is paused, then IMs don't beep (which is the expected behaviour). Jun 18 10:59:30 Wikiwide: I'd tend to think this a feature(tm), but ... :) Jun 18 10:59:37 +is Jun 18 11:07:40 Wikiwide: is this OMP? Jun 18 11:09:25 Nay, just Media Player. And, I think that it's bug inside something audio-profile-related, not Media Player itself. Jun 18 11:10:18 this is *deep* into terra incognita riddled by evil alsaped etc stuff, the unspeakable language blob stuff that Pali once decompiled Jun 18 11:10:21 And, when I am on a talkative IRC channel, and using music to lull myself to sleep, IM beeps are not a welcome feature. Jun 18 11:11:29 if anything bug, then that's a bug in your IM Jun 18 11:11:40 And I am trying to stop myself from using Silent profile, because it makes answering phone calls difficult (if I leave Silent profile for all day long) Jun 18 11:11:55 but probably it's a fsckup audio policy Jun 18 11:12:02 In my IM? It's just rtcom-messaging-ui plus telepathy-idle Jun 18 11:12:25 Yes, audio policy. I recall reading it sometime long ago. Jun 18 11:12:29 aka pidgin, no? Jun 18 11:12:33 that seems to be problem in libplayback library Jun 18 11:12:51 similar problem is also with OMP + bluetooth headset Jun 18 11:12:59 probably also bluez related Jun 18 11:13:05 No pidgin. This violet strangeness... Jun 18 11:13:15 aaa purple Jun 18 11:13:19 No bluetooth here. No OMP here. Jun 18 11:14:34 AIUI basically friggin audio policies work by looking for *startup* of a binary with a certain command name, and then adjusts all mixer settings accordingly. This is prone to fail Jun 18 11:15:01 cron to switch back to general in the morning? Jun 18 11:15:35 ~alsaped Jun 18 11:15:46 thought as much Jun 18 11:16:07 Granted, these beeps aren't heard if conversation window is the one brought to the front. Maybe, effect of "messages window here, hence new messages are already visible, no need to notify" Jun 18 11:18:30 less /usr/share/policy/etc/rx51/alsaped.conf Jun 18 11:19:29 don't expect any rationale in that, it's a flaw Jun 18 11:21:20 when IM on top, the IM audio policy takes effect. When MP on top, music audio policy takes effect. Audio volume of music policy isn't muted, and there are no distinct audio channels in neither hardware nor pukeaudio for different policies and their properties. IOW only one policy at a time takes effect Jun 18 11:22:49 it does NOT really have a per-app volume setting, not even a per-policy volume setting that works concurrently with other setting, I guess Jun 18 11:27:58 less /usr/share/policy/etc/current/pulse/xpolicy.conf Jun 18 11:28:27 check if IM and MP are in the right groups there Jun 18 11:29:06 GOD!!!! this config file sucks dinosaur balls Jun 18 11:29:19 the whole concept does Jun 18 11:32:03 I think that if IM window of channel A is maximised and message arrives to channel B, then the message beeps regardless of IM policy being 0. If message arrives to channel A (maximised), it doesn't beep (I think). Jun 18 11:38:52 #channel ? Jun 18 11:39:09 that's your IM then, would make a weird kind of sense Jun 18 11:39:12 Yes Jun 18 11:39:22 Wouldn't it have been easier to use alsa-mixer directly without pulseaudio? Jun 18 11:39:27 anyway, that might help, as a template: Jun 18 11:39:29 [stream] Jun 18 11:39:31 #exe = hildon-home Jun 18 11:39:32 property = event.id@equals:"message-new-email" Jun 18 11:39:34 group = event Jun 18 11:39:41 from /usr/share/policy/etc/current/pulse/xpolicy.conf Jun 18 11:40:10 exe is commeted out Jun 18 11:40:58 no idea where that crap gets stuff like property=event.id from, and how a process would define/set any such property Jun 18 11:41:49 Quite interesting :-) I doubt it would be fixable, because surely nobody would deliberately decide "ooh, music is on, let's annoy the listener with IM beeps and ignore the policy" Jun 18 11:42:48 anyway /usr/share/policy/etc/current/pulse/xpolicy.conf and /usr/share/policy/etc/current/pulse/xpolicy.conf are the most obscure and undocumented files around maemo audio or even maemo at large Jun 18 11:44:01 Nice... /usr/share/policy/etc/rx51/alsaped.conf right? Jun 18 11:45:43 what pali said, maybe a problem in libplayback. The process needs to define some sort of property or stream or group or whatever, for audio. When it doesn't, then I guess the audio is "anonymous" aka phantom, aka not noticed at all by alsaped, and thus inherits whatever is currently active as audio policy Jun 18 11:46:44 /usr/share/policy/etc/rx51/ and /usr/share/policy/etc/current/ are identical Jun 18 11:46:51 symlink or whatever Jun 18 11:47:19 current is a symlink to rx51 Jun 18 11:47:59 Hmm, interesting. So whatever calls libplayback hasn't defined the property for IM arrival notification, and inherits random-currently-active policy? Jun 18 11:48:52 My current workaround is: stay on one channel, keep it maximised. Then, no IM beeps. Jun 18 11:48:56 that's my guess, yep Jun 18 11:49:54 that whole thing is totally opaque and mysterious/obfuscated/closed Jun 18 11:50:22 ask Pali, he disassembled the Prolog shit, maybe he even understands it Jun 18 11:51:05 Cron-ing profiles could be good, but too finicky. As it is, I would like to be able to switch between cellular and SIP depending on whether WiFi is connected. Jun 18 11:54:10 Great :-) I wish there was some description somewhere about "how to reverse-engineer". I would gladly document rtcom-messaging-ui innards, especially obscure changeThemeSomehow thing (it's totally not exposed in GUI, many humans don't even know that changing Conversations theme is possible) Jun 18 11:54:30 IIRC it is up to application to follow profile settings Jun 18 11:54:59 so if some application play sound even if it should not, then that application does not call needed libplayback functions Jun 18 11:55:21 http://paste.ubuntu.com/24889757/ Jun 18 11:55:59 that package is open, see https://github.com/community-ssu/pulseaudio-policy-enforcement Jun 18 11:56:09 Wikiwide: try OMP Jun 18 11:56:12 if it has same problem Jun 18 11:56:53 Pali: I hope somebody will be able to confirm that this bug isn't experienced by only me. I can try OMP, yes... Not installed atm. Jun 18 11:57:26 as wrote I see same bug, but with bluetooth Jun 18 11:57:38 but maybe it is not same... Jun 18 11:58:46 that looks promising, though REALLY I wish it was formatted to be readable Jun 18 11:58:51 https://github.com/community-ssu/pulseaudio-policy-enforcement/blob/master/doc/lyx/pulseaudio_policy_enforcement.lyx Jun 18 11:59:01 jonwil: thanks! Jun 18 11:59:36 For now, I should contend myself with easier things, like making a bootup video, and a theme. Considering that ffmpeg on N900 doesn't write h.263 video, both of these undertakings require "a computer not a N900" Jun 18 12:00:24 there she went Jun 18 12:00:33 Wikiwide: what happened? Jun 18 12:01:31 DocScrutinizer it looks like you should be able to get docs out of that repo, the repo certainly contains a makefile to turn that stuff into readable docs Jun 18 12:01:35 btw (as I already suspected or even knew) OHM is also involved in that shit Jun 18 12:02:00 jonwil: no idea how Jun 18 12:02:02 Also OMP has some settings that you can tweak RE "interruptions" notifications during playback, so check them out too if you try OMP Jun 18 12:04:31 DocScrutinizer: randomly disconnected WiFi. sixwheeledbeast: will check, but is it possible to write them as "plugins" for stock media player? Jun 18 12:06:45 no, nothing is possible for stock media player Jun 18 12:06:47 Downloading OMP, so far. Ubuntu Humanity theme in App Manager is amazing. Jun 18 12:06:52 it is closed/undocumented Jun 18 12:06:58 reason why OMP was created Jun 18 12:09:14 https://github.com/community-ssu/pulseaudio-policy-enforcement/blob/master/doc/lyx/pulseaudio_policy_enforcement.lyx#L145 to #L159 is pathetic. Some of the most important info parts simply empty stubs Jun 18 12:21:27 Wikiwide: don't we have dsp support for h.263? you could use gstreamer maybe Jun 18 12:34:25 h.263 should, yes. h.264 not, or only partially Jun 18 12:34:40 iirc Jun 18 12:45:54 err?! "/home/christian/n900/cssu-testing/policy-settings-rx51/policy-settings-rx51/rx51/policy/media_groups.pl" in /usr/share/policy/rules/rx51/policy.plc Jun 18 12:52:52 https://github.com/community-ssu/policy-settings-rx51/blob/master/rx51/policy/audio_volume.pl hmm, Prolog gibberish Jun 18 12:55:27 https://github.com/community-ssu/policy-settings-rx51/blob/master/rx51/policy/context.pl ACME Prolog gibberish Jun 18 12:56:28 we have h264 encoding support Jun 18 13:05:25 h.264 baseline afaik Jun 18 13:05:42 could be wrong Jun 18 13:05:54 that h.* stuff is confusing Jun 18 13:08:34 https://wiki.maemo.org/Manual_video_encoding >>The N900 supports AAC and H.264 up to Baseline 3.0 profile<< Jun 18 13:12:12 https://bugs.maemo.org/show_bug.cgi?id=6635 Jun 18 13:12:14 04Bug 6635: OpenMax returns error when trying to play an mp4 file with h264 video Jun 18 13:42:28 Darn Jun 18 13:42:40 Charging when powered off failed again Jun 18 13:42:42 So i reflashed Jun 18 13:42:46 Charging worked Jun 18 13:42:49 Then installed BackupMenu Jun 18 13:42:56 Charging when off stopped working again Jun 18 13:43:18 So now i will have to reinstall my Backup with BackupMenu, then uninstall BackupMenu, and hope charging when off will work again then ☹ Jun 18 13:44:14 maybe you have something in your backup that breaks it? Jun 18 13:44:40 also which version of bm Jun 18 13:45:00 Could be Jun 18 13:45:20 But i did not reinstall Backup just yet Jun 18 13:45:43 I just installed BackupMenu, and before Restoring i tried out Charging Jun 18 13:45:44 And it failed Jun 18 13:46:23 Well, could be that the microUSB-Jack has evil tottering Contact so it works randomly ;( Jun 18 13:47:06 use a sticky tape for testing ;) Jun 18 13:47:10 BackupMenu Version 1.2.0 Jun 18 14:03:53 Restored Backup, removed BackupMenu and „bootmenu“, now it is charging again when off, but it seemed to boot into Charge-Mode like 3 Times Jun 18 14:04:25 Maybe the Flash-Chip is generally having Issues… Jun 18 14:04:36 And/or the microUSB-Jack Jun 18 14:04:42 Whatever Jun 18 14:04:49 I can't think of any possible way BM would break charging Jun 18 14:05:05 Yup, that is probably just Voodoo-Placebo-Magic here 😉 Jun 18 14:05:28 poweredoff-charging* Jun 18 14:05:38 did you install u-boot? Jun 18 14:05:39 aka "actdead" mode, right? Jun 18 14:05:59 ooh bencoh, you got a point there Jun 18 14:06:08 doesnt backupmenu change something in init process? Jun 18 14:06:15 yes Jun 18 14:06:36 blocks init when booting with slider open Jun 18 14:06:38 I had U-Boot for Kernel-Power on my last working Setup Jun 18 14:06:52 i mean *now* Jun 18 14:06:56 blocking init means no BME Jun 18 14:07:07 But now i did not install it again because i do not need Kernel-Power because i have my Nexus-5X for Tethering Jun 18 14:07:09 no BME -> no charging Jun 18 14:07:14 Now i have no U-Boot, no Jun 18 14:07:55 APic\splat: is slider opened or closed when trying to charge from poweroff? Jun 18 14:08:10 Closed Jun 18 14:08:17 s/from poweroff/in off state/ Jun 18 14:08:18 bencoh meant: APic\splat: is slider opened or closed when trying to charge in off state? Jun 18 14:08:18 maybe magnet defect? Jun 18 14:08:30 DocScrutinizer05: doubt it Jun 18 14:08:44 I mean, he would notice it Jun 18 14:08:52 (I guess?) Jun 18 14:09:02 well, easy to tell: when BM menu comes up with slider closed on boot, it's defect Jun 18 14:09:03 Yah, when BackupMenu was installed the Boot-Menu only appeared with Slider open Jun 18 14:09:08 Yup Jun 18 14:09:16 ok Jun 18 14:09:27 then I don't know Jun 18 14:10:00 does it work if you try removing BM? Jun 18 14:10:02 you're positively sure you have not restored any backup (even with the nokia/stock backup)? Jun 18 14:10:07 does it wake up when plugging in the charger? Jun 18 14:10:21 bencoh: I removed BackupMenu and bootmenu, and now it is doing what i want Jun 18 14:10:30 it needs to boot nevertheless, even when in act_dead charging Jun 18 14:10:32 interesting Jun 18 14:11:08 sicelo: Yes, i _did_ restore my Backup Jun 18 14:11:10 you *always* will notice at least the NOKIA screen from bootloader Jun 18 14:11:13 fwiw, i reflahsed my n900 during the week, and installed BM. no problem whatsoever Jun 18 14:11:25 ah .. Jun 18 14:11:26 DocScrutinizer05: Yes, i know that Jun 18 14:11:42 waitamin... bootmenu? isn't this thing deprecated? Jun 18 14:11:45 So now it is doing what i want, so i just hope it will keep to behave 😎 Jun 18 14:11:49 you backup has issues most likely. sixwheeledbeast already mentioned Jun 18 14:12:10 so you plugged in to a powered down device, it booted and then stopped (or didn't start) charging? How did you tell? Jun 18 14:12:21 Man, it works now. Can we just stop this Discussion until something will break again? 😉 Jun 18 14:12:25 bootmenu is not deprecated. BM depends on in. you might be thinking of multiboot .. which is dep Jun 18 14:12:31 APic\splat: ^^ Jun 18 14:12:44 sicelo: maybe Jun 18 14:13:01 DocScrutinizer05: When it was b0rk, it had now orange Light Jun 18 14:13:11 s/now/no/ Jun 18 14:13:12 APic\splat meant: DocScrutinizer05: When it was b0rk, it had no orange Light Jun 18 14:13:42 (Or green when done) Jun 18 14:13:46 you're aware you can adjust that yellow light to your liking, from redblue flashing to no light at all? Jun 18 14:13:57 ~xy Jun 18 14:13:58 extra, extra, read all about it, xy is The XY problem: You want to do X, but don't know how. You think you can solve it using Y, but don't know how to do that, either. You ask about Y, which is a strange thing to want to do. Just ask about X. http://www.catb.org/~esr/faqs/smart-questions.html#goal Jun 18 14:13:58 Yes Jun 18 14:14:18 ~xyz Jun 18 14:14:37 But i doubt the Flash-Chip had a Bit-Rot in exact that Location where the Light-Behaviour is configurend Jun 18 14:14:44 s/configurend/configured/ Jun 18 14:16:02 I doubt a bitrot could get restored to unrotten state by deinstalling any package Jun 18 14:16:46 makes no sense Jun 18 14:17:22 and I'm a big fan of BM, don't like it getting flamed for something it's most likely completely innocent Jun 18 14:18:06 also bug reports that "fixed themsekves" without proper analysis what caused the error are highly annoying Jun 18 14:18:42 next time you reflash, don't restored any backups via BM or osso-backup. install BM. then test for some time Jun 18 14:19:01 -d Jun 18 14:19:25 there's even a fiasco image with BM preinstalled :-) Jun 18 14:20:29 http://maemo.cloud-7.de/maemo5/BM/flashable_rootfs_v0.56-1-20101109.img Jun 18 14:22:04 never tested that though Jun 18 14:22:18 robbiethe1st told me it works Jun 18 14:23:51 sicelo: Yup, will do that, but i hope it will just work without Reflashing ☺ Jun 18 14:24:49 can't recall when I last time reflashed my device Jun 18 14:25:19 I reflashed a bazillion other devices, but not my daily phone Jun 18 14:26:43 i finally reflashed mine last week because i wsn't sure whether i had cruft corrupting my modem or not. turns out that the problem is not related. anyway, enjoying my "new" faster n900 Jun 18 14:27:20 faster? Jun 18 14:27:31 oh yeah, cleaning out stuff like el-v1.db can do wonders Jun 18 14:27:40 ah Jun 18 14:28:34 yes in that sense. Jun 18 14:30:06 my el-v1.db was small because i don't keep conversations as a general rule, but who knows what else i had accumulated. i elected to not restore my backups wholesale. i'm only restoring single files as the need comes up Jun 18 14:30:17 * sicelo should restore keymapping Jun 18 14:33:01 nice approach Jun 18 14:33:22 I'm regularly too lazy for such approach ;-) Jun 18 14:33:46 but we might eventually do a little fancy and write a cleanup script Jun 18 14:42:06 what I'm just doing this very moment though is not reflashing but recalibrating Jun 18 14:42:39 that script is pretty inconvenient, needs a cd path/to/bq27200.sh to work Jun 18 14:42:51 * DocScrutinizer05 blames DocScrutinizer05 Jun 18 15:59:57 only a bad worker blames there tools Jun 18 16:34:39 sometimes tools are really worn/bad quality ;) Jun 18 16:52:17 sometimes the toolmaker should have been more diligent Jun 18 16:53:27 with a tiny bit of effort, I could have made a decent app out of it Jun 18 17:19:48 I always say that "there" is the most important part of that well known saying. Jun 18 17:21:00 If you have bad tools why are you still trying to do the job. Your first job is to get the right tools. An important point IMO Jun 18 17:21:12 :) Jun 18 17:21:45 unless you live in some remote area where there is nothing. not even good rocks Jun 18 17:21:53 The more you think about it the more stuff you end up applying it to. Jun 18 17:23:01 In that case your tools are the best you can get hold of, so why are you blaming them ;) Jun 18 20:23:49 Stock Media Player isn't well documented, true. But there is mafw-grilo-source , for instance. Just, adding a media-source is likely a whole lot simpler than changing audio-policy-things. Jun 18 20:52:41 Some bits are open, there was a WONTFIX bug where community tried to get the closed bits opened. Jun 18 22:12:55 how's the main camera support in the newer kernel? i wonder why don't i see the driver in v4.9-n900 branch Jun 19 00:12:23 anyone know what this model is called? Jun 19 00:12:24 https://i.ebayimg.com/00/s/MTYwMFgxMjAw/z/c6cAAOSwTf9ZRxFT/$_20.PNG Jun 19 00:16:11 6760 Jun 19 00:21:52 Thanks. It's in Australia for $30.50us, $40au. **** ENDING LOGGING AT Mon Jun 19 03:00:03 2017