/srv/irclogs.ubuntu.com/2019/02/13/#ubuntu-desktop.txt

mitya57Trevinho: thank you!05:17
mitya57I am going to make a new official compiz release today or tomorrow, hope you don't mind. It would be useful for other distributions like Arch.05:21
didrocksgood morning07:29
oSoMoNgood morning desktoppers07:42
didrockssalut oSoMoN07:44
oSoMoNsalut didrocks08:00
=== pstolowski|afk is now known as pstolowski
ricotzgood morning :)08:41
ricotzoSoMoN, see pm08:41
didrockshey ricotz08:50
Laneyb0b0b009:01
didrockshey Laney09:02
* Laney pats didrocks 09:09
WimpressMorning desktoppers o/09:10
didrockshey Wimpress09:14
* Laney nods gravely towards Wimpress 09:14
Wimpress:-)09:14
Laney("take him away")09:14
seb128good morning desktopers10:11
didrockssalut seb12810:16
seb128lut didrocks, en forme ?10:16
talxmorning folks10:17
talxI'm looking for assitance with preseed installation10:17
didrocksseb128: ça va, et toi ?10:17
talxtrying to figure out what I'm doing wrong for the third way in a raw10:18
seb128ça va :)10:18
=== alan_g is now known as alan_g_
jbichaseb128: do you want to try demoting notification-daemon again? I've added some alternate dependencies so maybe it will stick this time12:57
jbichaalso, we already got the first review of the libhandy MIR LP: #1815483 (I wasn't expecting that so soon, thanks!). There's a question there about whether we want Glade in main13:19
ubot5Launchpad bug 1815483 in libhandy (Ubuntu) "[MIR] libhandy" [Undecided,New] https://launchpad.net/bugs/181548313:19
seb128jbicha, can do for n-d13:36
seb128jbicha, @libhandy, we don't want glade in main no, the -dev can stay in universe, also I'm not sure we want that lib at all in main until it's somewhat stable13:39
jbichathey claim to be releasing a "stable" 0.1.0 next month: https://source.puri.sm/Librem5/libhandy/wikis/home13:40
seb128I'm still not sure why that lib exists/is needed, outside of being a playground with the goal to merge those widgets in gtk13:40
jbichabecause it might not be practical to merge those improvement in until gtk413:40
seb128other experimentals widget libs like egg are usually included as copies13:40
seb128and?13:40
seb128egg has been around for years13:41
seb128still it doesn't need to be an external lib13:41
seb128what's GNOME r-t position on libhandy?13:41
seb128do they plan to add it to the core set?13:41
jbichait's already there (gnome-contacts is part of GNOME Core and uses libhandy without embedding it) https://gitlab.gnome.org/GNOME/gnome-build-meta/tree/master/elements/core-deps13:43
jbicha🤷 some Fedora guys were saying the same thing that they want it to be an embedded library, but that's not really how it's being used right now13:44
seb128mclasen's comment on #gnome-hackers the other days suggested that he didn't think it should be a core lib for GNOME13:44
seb128I'm not sure how they decide to accept or not new depends as part of their platform though13:45
seb128anyway -1 from me to use as a lib until it's somewhat stable, sharing a moving target playground is just asking for problems imho13:46
seb128if some wonder why retracing are failing recently in disco, bug #181577415:23
ubot5bug 1815774 in binutils (Ubuntu) "binutils 2.32 update breaks debug symbols in disco" [High,New] https://launchpad.net/bugs/181577415:23
seb128(I was wondering why the recent report from the g-c-c segfaults failed to retraced and ended up chasing it down to binutils)15:23
seb128we are going to need to rebuild $things once binutils is fixed15:23
Laneygood find15:25
seb128thx15:25
seb128doookooooo15:25
Laneybinutils is crazy stuff15:25
seb128we should have some autopkgtest testing that dbg still work when a binutils lands15:26
* seb128 adds to his would-be-nice-to-add part of his todolist15:26
Laneyusing libunwind or something?15:28
seb128good question, I didn't give it much thinking yet, I was thinking maybe building a small package with a buggy .c example, install it/the dbgsym, start the buggy example under gdb and get a bt/check that it's in a debug format15:32
seb128but looks like libunwind would be good in that scenario, thx for the hint!15:33
Laney:>15:43
seb128does anyone know an example of a package that rebuild itself in an autopkgtest?15:45
seb128d_idrocks suggests doing that for rygel as part of the MIR to ensure that vala updates don't break it15:46
Laneythat's a weird requirement, autopkgtests usually test runtime breakage really and archive rebuilds are used to detect ftbfs15:48
Laneyyou can do it with Restrictions: build-needed and a no-op test but like I say I wouldn't advise that15:48
seb128well, I can see the value of blocking vala updates is they break things15:48
Laneywhy doesn't it apply to any library?15:49
seb128but maybe that should be done fromt he vala side15:49
Laneyautopktest isn't meant to be a general keep-the-archive-buildable service15:49
seb128you have a point there15:49
seb128we also don't fully archive rebuild to validate gcc updates :p15:50
Laneyright, and these almost always do introduce a few failures15:50
seb128that was an optional suggestion on the MIR, I'm going to skip that for now15:50
Laneynow a test in vala itself which builds a few programs (Restrictions: superficial probably) would be a nice thing15:51
Laneysame for all libraries/compliers15:51
seb128right15:52
seb128Laney, thx for the input15:52
Laneyyou're welcome15:52
=== alan_g is now known as alan_g_
andyrockjbicha: https://code.launchpad.net/~azzar1/ubuntu/+source/gnome-online-accounts/+git/gnome-online-accounts/+merge/36314417:16
=== pstolowski is now known as pstolowski|afk
mdeslaurseb128: do you have a curl merge planned?18:42
seb128mdeslaur, no, but I can look at it tomorrow18:42
mdeslaurseb128: awesome, thanks18:42
seb128yw!18:42

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!