[09:49] infinity: please merge glibc from unstable, currently blocking the binutils/gcc migration === doko__ is now known as doko [11:10] wow, 5 of my uploads migrated to release today. awesome [13:00] juliank: Implement a new sftp method that connects via openssh and then uses paramiko's sftp support. [13:00] but no dependency on paramiko? [13:01] acheronuk: It's a Suggests [13:01] acheronuk: Just like python-bzrlib was befor [13:01] e [13:02] juliank: well, as our CI docker containers use sftp to upload to launchpad, but do not as per standard apt install 'suggests', that brokoe our CI's ability to upload [13:02] our = Kubuntu [13:03] acheronuk: that's unfortunate, but it's just the way it was before [13:04] juliank: not really, as it didn't fail without paramiko before, now it does [13:04] acheronuk: You must have manually installed python-bzrlib earlier, now you need to manually install python3-paramiko [13:04] juliank: I'm looking into. just having a maon when things break :P [13:05] or well, maybe you pulled in bzr and got python-bzrlib by accident [13:05] in any case, none of the default hosts use sftp so it does not seem wise to Depend on it. Recommends might make more sense, though. [13:06] that is was I was going to suggest (or actually recommend :P) [13:07] on the other hand since there are no preconfigured hosts with sftp, it does not make it a usual configuration, and thus is better suited to suggests [13:07] juliank: this requires the python3 of python3-paramiko, yes, but only before the python-bzrlib? [13:08] I can't parse that [13:08] Basically I did s/python-bzrlib/python3-paramiko/ on debian/control :D [13:09] just I have in deploy_in_container.rake python-paramiko & python-bzrlib already. [13:09] right, now you need python3-paramiko instead of python-paramiko [13:09] switching to python3 was the whole point of the merge [13:10] yep, so maybe just need to update that [13:10] right [13:11] acheronuk: you might want python-paramiko too in case you use bzr [13:11] depending on your bzr needs :) [13:12] yes, as long as they are both there for the xenial and artful as well, I was going to have both to be on the safe side [13:14] I have to do some followup on dput, as the error reporting on gpgme is a bit broken. [13:14] like it shows error messages for success and tracebacks for real failures [13:14] nice [13:15] but: it works, unsigned and wrong signatures are rejected and correctly signed is accepted [13:15] and that's an upstream bug. [13:16] juliank: dput-ng recommends python-paramiko, not sure if that makes a difference [13:16] dput only suggests any non-default transport dependencies. [13:17] I have not heard back from upstream yet on my method :( [13:17] or maybe ubuntu-dev-tools could recommend it? [13:18] a recommends would generally be wrong, though, nothing uses it by default. [13:18] you have to modify the config file to switch to sftp [13:18] So I don't see python3-paramiko having to be installed in all but unusual situations [13:18] it's one less step (change this config and install this other pkg) [13:19] I used to have timeout problems with the default upload method with bigger uploads [13:21] right [13:22] I'm not really convinced yet, but I have to take care of other stuff anyway [13:22] * acheronuk tries to update his containers [13:22] gotta defend my master's thesis on thursday, gotta write a presentation for it [13:23] good job, almost done! :) [13:23] juliank: good luck :) [13:24] thanks. I'll try to do my best [13:24] :) [13:29] juliank: I think it would be awesome if MoM could filter by uploader (or by "some" uploaders). Want some code from me, or can you do it? :) [13:29] tsimonq2: You can filter on any text in the main row [13:29] OH [13:29] Hah :) [13:30] tsimonq2: Excluding the comment, though, that's in an input and does not seem included somehow [13:30] gotta fix that :/ [13:31] tsimonq2: And since it's a regex you can also search stuff like foo|bar [13:31] Right [13:31] Ok [13:31] it's pretty awesome :D [13:32] It is :D [13:32] juliank: looks like my container's upload is now fixed :) [13:33] acheronuk: nice [13:33] but I still think a recommends may be better [13:33] :P [13:33] but not that bothered [13:33] I mostly want to get it merged upstream and do what ever they want :) [13:35] juliank: Also, what's up with the retro Ubuntu logo? ;) [13:35] heh [13:37] The whole page is a bit retro :) [13:38] I like retro. for the logo, anyway [13:39] juliank: Hm, I noticed that the search doesn't search comments [13:39] yeah, I just wrote that above [13:39] [14:30] tsimonq2: Excluding the comment, though, that's in an input and does not seem included somehow [13:39] Oh [13:39] Ok [13:39] I'm matching the row's textContent [13:40] I guess I also need to find the input in there and match against it's value too [13:44] tsimonq2: also inputing text in there and hitting enter reloads the page with your filtering gone and you have to go back [13:45] so it needs some more work on dtaisl [13:45] *details [13:45] Right [13:45] whoa, we got another 10 merges down since I last looked at it 4 hours ago [13:46] 254 now [13:46] it was 265 then I thinmk [13:47] 2018-01-27 09:49 main needs-merge=262 [13:47] hmm ok [13:50] juliank: Hm, what's up with mousepad? [13:50] Am I missing something or is MoM not picking up the merge? [13:50] tsimonq2: It shows a merge on universe.html [13:50] ah, now I understand [13:50] odd [13:50] juliank: Right, meant to say the opposite [13:50] Yeah [13:51] I'm the one who wrote that comment on there [13:51] And it's been like that for a while [13:52] I'm not expert at this either, I entirely rely on cjwatson for everything :) [13:52] It also showsa no last uploader [13:53] probably gotta manually delete the merge or something, I don't know [13:54] xfdesktop4 LP PTS [xubuntu-bugs] [13:54] xfce4-datetime-plugin LP PTS [xubuntu-bugs] [13:54] are also affected [13:54] any row without an uploader name [13:55] xfce4-session [13:55] gcc6-cross-ports [13:55] gcc-5-cross-ports [13:55] xfwm4 [13:55] Hmph, so there's the pattern. [13:56] So mostly xubuntu stuff and two gcc things [14:01] juliank: One more thing... so it says "399 outstanding merges" but that number does not update when filters are applied... it should be I think [14:01] maybe, I don't know. [14:01] It was just a quick hack [14:01] do you also want the graphs created dynamically? [14:02] Eventually it should be 399 outstanding merges (... hidden) [14:02] I guess [14:03] Dynamic graphs would be nice [14:03] And yeah, hidden works [14:04] I'm not sure what all the different states actually mean in the graph [14:04] like needs sync or stuff [14:05] "needs sync" is likely things that are candidates for being autosynced but haven't done so already [14:05] Look at the timeline on it, that amount went up after the autosync was turned off for Artful === Spads_ is now known as Spads [22:45] tsimonq2: I can't decide if 252 outstanding merges (XX shown) or 252 outstanding merges (XX hidden) is better [22:45] But it's going to take some effort anyway to associate these lines with the tables and stuff [22:46] And I think I should move the filter section below the number of merges and text [22:47] and of course I need to add a link to the launchpad page for the version in proposed === Spads_ is now known as Spads [23:23] RAOF: did you see that gnome-do was recently removed from Debian Testing? [23:26] jbicha: I did, yes. It needs to be ported to various new things 🙁 [23:29] Mono GTK apps in Debian are in rough shape since only a couple use gtk3, the rest use really old stuff :( [23:36] juliank: hidden is better imho [23:46] jbicha: yes, curse them for using stable and perfectly working existing code! [23:47] (that, admittedly, doesn't support Wayland, but then again Wayland doesn't support Do either) [23:48] GnomeVFS and gconf? [23:51] I'm not worried about gtk-sharp2 (at least not for a couple more years), but gnome-sharp2 is pretty old [23:56] gconf works fine! [23:57] And we've already ported off gnome VFS to gio