[07:52] so quiet this morning... [07:53] your morning :-) [07:55] good morning! [08:02] hey [08:02] morning Laney [08:03] hi Laney, willcooke [08:03] hey willcooke [08:03] good trip back? [08:03] hi oSoMoN [08:04] morning oSoMoN [08:04] how are you? [08:04] Morning all [08:04] so glad for tab completion with your nick [08:04] I’m good, thanks :) [08:04] Laney, uneventful. I got upgraded on the way there becuase my TV was broken. Alas not on the way back. [08:04] can't win 'em all [08:04] hey davmor2 [08:05] * oSoMoN makes a note to complain next time I travel and the TV is broken [08:05] hey davmor2 [08:09] Morning UK [08:09] and Europe [08:10] hey davmor2 & duflu [08:10] willcooke: Silly question: Is there any reason for having Pulse at all other than mixing and per-app volumne? Seems like a poor choice if it's those features vs decades of bugs [08:11] *volume [08:11] duflu, afaik it does some software sample rate conversions as well, but that might well be a hangover from the bad old days. [08:12] willcooke: Sounds trivial. I just wonder why must we bother ourselves with years of pain and bugs for such a tiny feature gap pulse is filling... [08:13] iiinteresting [08:13] * willcooke reads up [08:13] I mean, by now any software project that age should be perfect, or should be dead [08:14] Looking at the Pulse pages here https://www.freedesktop.org/wiki/Software/PulseAudio/About/ [08:15] Software mixing - I expect ALSA handles that just fine now [08:15] Network - dont think anyone uses this by default [08:15] API abstraction - we only use ALSA, so !care [08:15] Heh. I did around 1997 [08:16] Hardware abstraction - which it says boils down to per app volume... which is... not super exciting [08:16] Does GStreamer rely on Pulse at all? [08:17] duflu, so I can't see any immediate concerns, but let's socialise it a bit and see if anyone has any strong feelings. I don't know enough to make a call. [08:18] Laney, any thoughts on Pulse? [08:18] willcooke: Yeah no problem, me neither. Just started bug bashing but I had to wonder why complicate the architecture in the first place... [08:19] My expectation is that it made sense at the time, and we never revisited [08:20] I'm wondering if it's now something like "because that's what the applets use exclusively" [08:21] What's the actual problem? [08:24] Just something else to maintain [08:24] it's pulse doing the output switching? [08:24] ie, plug in hdmi, earphones [08:25] *doing it badly* [08:25] Feels like a party like it's 2007 Linux hatefest [08:26] :) i've been having issues with it running on gnome. Seemed to work better with u7 for some reason. [08:26] loosing all sound sometimes. seems to be getting confused when switching outputs. [08:27] Laney, I don't think so, not really. More a case of a critical look at a component. It makes sense to review it, even if that decision is "keep it and fix bugs". [08:27] plug earphones in/out again and it starts to work. [08:28] Laney, to be fair I've got my back up about non-Ubuntu problems today. But it's still an objective question: Does the functionality provided outweigh the cost/pain of maintenance? [08:31] duflu: I think we don't know what the functionality is, what the cost is, or what the pain is, or even really what the bugs are. [08:32] Laney: Fair point. I'm only about 5% through the Pulse bug list and will learn more. Was just hoping someone might know off the top of their head [08:32] Or what kind of maintenance costs having a different stack from the rest will incur. [08:32] Oh, you're on Ubuntu? They don't have pulse so that's probably your bug. BYE! [08:40] Feeling a bit grumpy this morning, sorry for being short [08:41] "Looks like someone's got a case of the Mondays" [08:41] *hugs* [08:46] Laney, I wish you a pleasant Monday and pleasant week. [08:47] willcooke: I'm still suffering from friday [08:47] Sounds like an excellent Friday [08:47] davmor2, that sucks man. I'm really sorry. :( [08:47] ? [09:19] ah hey Laney, willcooke, oSoMoN e duflu [09:19] :) [09:20] Evening Trevinho [09:20] Wouldn't it be great to sprint in Asia, right after you get home? :) [09:20] evening... it's sitll afternoon :-P [09:20] Getting darkish, so I called it [09:20] yeah... let's do it here.. [09:20] It's cheap :) [09:21] hey Trevinho [09:23] hey Laney... [09:23] Laney: speaking of the devil... Would you mind to give a look to these debdiff https://bugs.launchpad.net/ubuntu/+source/gnome-software/+bug/1689239 ? [09:23] Ubuntu bug 1689239 in nautilus (Ubuntu) "Headerbar used as toolbars in unity are missing proper css classes" [High,In progress] [09:24] Or i'd break things apps soon once my updated themes will be there... [10:04] Hmm, not just dark but also noisy frogs telling me it's night time [10:04] Good night [10:11] Trevinho: hi, yes I saw the bug but hadn't gotten to it yet [10:12] jbicha: ok, good [10:13] if I understand you correctly, your patches are to support a future change to ubuntu-themes? [10:14] jbicha, re: webapp - I've not looked properly yet - but does your suggestion handle the geo-specific stores, like if I run it here in the UK, I get the .co.uk version, in the States I get the .com version etc? [10:14] willcooke: that's up to the website so the answer should be yes [10:15] willcooke: for instance, the BBC News app gives me bbcnews.com instead of bbcnews.co.uk [10:15] Humm, I think that didnt work properly for some reason. I'll look at what we used to do and check [10:15] Maybe we have different referral codes depending on which store you hit. [10:16] just a blind guess, but if you have to do stuff like that, maybe you would need to patch webkit directly? [10:17] Or a wrapper which works out what's what and launches the correct site - probably easier than patching webkit :) [10:19] maybe [10:26] yeah, we have different codes depending on which site is launched: http://bazaar.launchpad.net/~webapps/libunity-webapps/trunk/view/head:/src/runner/unity-webapps-runner-amazon.c [10:26] we could re-use that [10:37] Trevinho: what updated theme? [10:37] are you going to SRU these fixes? [10:41] Laney: I'm fixing ambiance/radiance to work properly in GS (keeping unity compatibility) [10:46] Okay [10:46] Well I'm not sure it's sensible for us to keep carrying UI patches for Unity in applications [10:52] …particularly because we never got around to patching many GNOME apps so it's still going to be inconsistent [10:57] Guess it's okay to update existing patches though [10:57] I spoke to the design team last week about the possibility of a new theme, I need to follow up with them this week and find out if that's going to be possible this cycle. In the meantime though getting Amb/Rad generally working in G-Shell is worth doing. And as Laney said, I dont think keeping U7 working perfectly should be a priority. Getting them working in G-Shell first should be the priority [10:58] * Laney nods willcooke [11:03] willcooke: sure... That's what I'm doing in fact. Just I prefer not breaking things for unity too... As it's just easy changes to apply [11:03] (to the patches we already carry) [11:06] willcooke: jfyi - I had to fiddle with pulseaudio - finally got around to looking into why the dac always said 44.1kHz, then I stopped that and went to alsa so it doesn't muck about at all [11:07] Trevinho, ack, thanks [11:14] Laney: do we have a list of app for which we did patches to remove headerbar? [11:15] doubtful [11:42] Trevinho: epiphany is another one [11:43] jbicha: mh, ok... fix is easy, when we carry patches... it's just to update this [11:43] jbicha: I currently updated only the default installed ones... [11:45] eog & evince [11:52] xnox: so how did you make that list? [12:10] Laney, multiple ways. 1) I used the demote email, from when unity8-desktop-session was demoted from main to universe 2) then I also used http://people.canonical.com/~ubuntu-archive/germinate-output/ubuntu-touch.artful/touch.sources (combined with touch-core.sources, touch-*.sources, sorted and uniquified) [12:11] then i tool all of those source packages and filtered them, to exclude things that are seeded anywhere. [12:11] and then did a check to see if a package exists in debian, and filtered that out as "to sync" [12:12] e.g. we have forked android-tools bits, which we should now sync from debian. [12:12] i think i also looked through it manually, to remove things that are generic. [12:12] ... [12:14] it's mostly complete, apart from a few things that got seeded in other distros, and does not touch anything that is unity7 related (as all of that is still seeded in ubuntu-desktop) [12:47] xnox: ok, commented, thanks [12:47] * Laney goes to eat a burrito [12:47] *nom* *nom* [12:47] awh man [12:48] I would also like a burrito [12:48] Laney, nice review =) [12:48] * xnox thought watershed was only used by upstart jobs, will check [12:49] xnox: while read pkg; do chdist apt-cache artful-amd64 showsrc $pkg; done < removals-take-1.txt | grep-dctrl -sPackage,Version . [12:49] willcooke: it's just a supermarket one, don't be too sad :P [12:49] nice =) [12:49] ah, right. In which case, be lucky. [12:50] Laney, i did ask before, but is there at all any way to grep autopkgtest dependencies? [12:50] afraid not [12:50] wait [12:51] xnox: You can look at Testsuite-Triggers [12:53] Testsuite-Triggers? where is that? [12:53] in the .dsc [12:53] therefore in Sources.gz [12:54] Might not be there for things which haven't been uploaded in ages or whatever, but should be good enough [12:54] nice... can we teach reverse-depends about it? [12:54] probably [12:56] ooh, firefox formats json [12:57] anyway, burrito: back shortly [14:21] hello! is there any way to install Ubuntu on a x86 based Android tablet? (I have an ASUS Zenpad 10 Z300C / P023) === tyhicks` is now known as tyhicks [16:53] dinner, bbl [17:22] https://opensource.googleblog.com/2017/05/oss-fuzz-five-months-later-and.html <- yo guys. [17:51] Thanks Sweetshark, much appreciated [17:52] I can't grok their reports at all. There doesn't seem to be a link back to a fix at LO [17:54] willcooke: They just linked to the list of issues: https://bugs.chromium.org/p/oss-fuzz/issues/list?can=1&q=Proj%3Dlibreoffice+Type%3DBug-Security+status%3ANew%2CAccepted%2CFixed%2CVerified&colspec=ID+Type+Component+Status+Proj+Reported+Owner+Summary&cells=ids [17:54] Sweetshark, right, but each of those issues is unreadable to me [17:55] It says "It's fixed" but doesnt provide any links to where it's fixed afaict [17:55] « ClusterFuzz has detected this issue as fixed in range 201612212121:201612220925. » [17:55] it gives a window in which the fix was committed upstream, I guess? [17:56] a link to an upstream commit would be useful indeed [17:57] oSoMoN: oss-fuzz just autosdetects "build it a day later. issue was gone", there is little human interaction in oss-fuzz there ... [17:58] ah, I see [18:00] FWIW, I dont think libreoffice is the only thing to worry about (or even the most to worry about). I mean: 10 FreeType2 issues, 17 FFmpeg issues, 10 GnuTLS, 25 PCRE2? each of those are use by half the world. [18:00] s/use/used/ [18:01] heh - it's good practice at least :) [20:04] Laney: this is what I meant https://code.launchpad.net/~3v1n0/ubuntu-themes/gs-decorations-fixes/+merge/323760 [20:04] and.... bed time now [20:48] night all === Guest7204 is now known as RAOF