/srv/irclogs.ubuntu.com/2024/12/02/#ubuntu-devel.txt

=== tsimonq2_ is now known as tsimonq2
=== cpaelzer_ is now known as cpaelzer
sudiplooks like Britney is stuck09:29
tsimonq2sudip: It was just restarted09:29
tsimonq2Although I'm hoping the frontend sync gets somewhat quicker at some point09:29
sudipyeah, just saw in #ubuntu-release09:29
=== guiverc2 is now known as guiverc
waveform@pilot in12:01
=== ChanServ changed the topic of #ubuntu-devel to: Archive: Open | Devel of Ubuntu (not support) | Build failures: http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of Focal-Oracular | Patch Pilots: waveform
seb128cjwatson, hey. I hope you are doing alright! Sorry to bother you but maybe you would have time to check a patch and tell me if you see any reason why we shouldn't do that change? I asked some weeks ago on this channel without luck and it is a topic where I would trust you input, I just want a sanity check that it's not something stupid we would do, https://launchpadlibrarian.net/757804399/firefox-u6-u7.diff15:09
seb128it's on https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/188641415:09
-ubottu:#ubuntu-devel- Launchpad bug 1886414 in chromium-browser (Ubuntu) "the chromium snap takes a long time to install without visible user feedback, seems stuck" [Medium, Triaged]15:09
seb128basically adding 'exec 0> /dev/tty' to the preinst after we are done using debconf to allow snapd to display progress while it's installing the snap15:10
cjwatsonseb128: hm, I can see why you want this, but is there guaranteed to be a useful /dev/tty?  I'm not sure whether that'll necessarily work if you're in a GUI upgrade tool or similar15:27
cjwatsonseb128: I almost wonder whether you need a helper that creates a pty or something15:28
cjwatsonseb128: like https://salsa.debian.org/cpython-team/python3/-/blob/master/debian/script.py only with a bunch of details changed15:29
cjwatsonseb128: (also shouldn't it be 0< rather than 0> ?)15:30
cjwatsonseb128: Has anyone tried asking the snapd team for a command-line option to force progress output even if stdin isn't a terminal?  That seems like it might end up being a lot less code overall, and easier to maintain15:36
cjwatsonseb128: Or is the intent to show progress but only if the preinst actually has a controlling terminal (bypassing debconf)?  If so, I think that code is in principle OK, but (a) should probably be 0</dev/tty rather than 0>/dev/tty (b) needs somebody to explicitly test it in an environment without a controlling terminal to see what happens, since I'm not sure15:40
seb128cjwatson, thanks for the review comments, on 0> vs 0< you are probably right (in practice the current syntax seems to work but you are right that < makes more sense)15:40
seb128and I will check on the snapd side, that's a good point, that would be cleaner indeed15:40
seb128and yes, that solution doesn't help us for feedback on graphical installations15:40
cjwatsonseb128: well, I sort of had second thoughts on the snapd side - it only makes sense if you'd want to unconditionally show progress here, rather than only showing it if you had a controlling terminal to start with15:41
cjwatsonI guess nobody has appetite for trying to plumb "snap install" progress information through to a debconf progress bar :)15:41
seb128to be honest I didn't step back to try to see what the ideal solution would be, I was rather focussing on what the patch is doing which is to make the snapd progress show if you do the update from a command line15:42
seb128I guess not :)15:42
seb128I really dislike those deb-installing-a-snap-from-maintainer-scripts but we don't really have a better option atm...15:43
cjwatsonSo anyway I don't know what the right answer is but hopefully me wittering is slightly helpful :)15:43
seb128it is, thanks! :)15:43
seb128nteodosi, ^ FYI15:44
nteodosiThank you, (a) looks right and I'll try (b), that means installing from any GUI I suppose.15:51
cjwatsonProbably but I'm not sure of the subtleties.  IIRC [ -t 0 ] will test whether you have a controlling terminal so you could use that to test whether your environment is the sort of environment that might be problematic15:53
cjwatson(Not after debconf has started up and taken over stdin though!)15:53
=== JanC is now known as Guest4566
ahasenackjuliank: is there a quick tool to convert a .list to a .sources file?21:37
ahasenackdeb8ss21:37
ahasenackdeb82221:37

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