[09:39] morning all! [09:57] Good morning, everyone! :D [10:54] man.. I hate lightdm! [11:19] is this a good place to talk about Ubuntu SSO? We're making some changes to our LoCo site and I'd like to learn what it would take to make it use Ubuntu SSO for logins. [11:31] oh. For web stuff, it's just like any other OpenID? [11:34] jo-erlend, here is the right place to ask about the ubuntu client for sso, so I might be able to help you :) [11:34] jo-erlend, if not, I can point you to the right guy [11:37] hello everyone! [11:40] good morning! [11:40] @ping [11:40] mandel, well, in this case it's primarily about a website. Is it possible to link to login.ubuntu.com in a way that chooses a particular language other than English by default? From a web-site perspective, are there any differences between using Ubuntu SSO and any other OpenID service? [11:40] gatox, nessita morning! [11:40] mandel, hi [11:41] hello everyone! [11:41] nessita, hi [11:41] nessita, I know you know more about the sso service than I do, got any help for jo-erlend ? [11:41] mandel: sure! the question specifically is...? [11:41] jo-erlend: ^ [11:41] nessita, mandel, well, in this case it's primarily about a website. Is it possible to link to login.ubuntu.com in a way that chooses a particular language other than English by default? From a web-site perspective, are there any differences between using Ubuntu SSO and any other OpenID service? [11:44] jo-erlend: I'd guess the site will pick the language from the web request headers, as far as I know there is no way to create a lang-dependent link [11:44] jo-erlend: just FYI, we have no control over login.ubuntu.com, U1 is just a 'client' of that service [11:44] nessita, ah, so if my locale is Norwegian, then the login page would've used Norwegian by default if it was translated? [11:45] oh, ok. [11:46] jo-erlend: I would guess so, or the web site could also go some geoip location and use that (the same google that does) [11:53] wtf.. bzr broke in my last update.. [11:53] nessita, is there an IRC channel for the login service itself? [11:54] jo-erlend: let me find out [11:54] thank you! I find it a little difficult to get a good overview of all this stuff. :) [11:55] jo-erlend: #canonical-isd is the channel in freenode [11:55] great, thanks. Then I'll go bother them instead. :) [12:02] * mandel back [12:21] nessita, do you know if ralsina is going to be around? [12:22] mandel: he will, as far as I know [12:22] mandel: he usually starts working at 9:30 [12:23] nessita, ok, thx [12:23] :-) [12:32] goodmorning! [12:32] 9:32 to be exact :-) [12:32] mandel: Iknow you are used to see me around 7AM but I fell asleep today [12:33] ralsina, meh, I just wanted to let you know that the windows reviews is ready, nothing important [12:33] mandel: cool, url? [12:33] ralsina, https://code.launchpad.net/~mandel/ubuntu-sso-client/retrieve-proxy-creds/+merge/98828 [12:34] ralsina, I asked alecu to double check, but that is how he is setting the tunnel.. I guess he show the issue there but never made the changed in the webclient [12:34] mandel: ack, will do a IRL after I boot to windows (sigh) [12:35] good morning to all, too! [12:35] ralsina, cool! it should work as expected.. although the dialog is very ugly, but I'll focus on more important buts first [12:35] alecu, morning! [12:35] alecu, can I have a review: https://code.launchpad.net/~mandel/ubuntu-sso-client/retrieve-proxy-creds/+merge/98828 ? [12:35] mandel, sure [12:35] alecu, I had to make some changes to ensure that the webclient used the correct proxy on widows [12:36] nessita, the tests in this branch has been updated: https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/main-moved/+merge/98703 [12:37] gatox: ack! [12:48] nessita: 1:1 ? [12:48] ralsina: can we do it right after the standup? I'm with reviews now :-) (but if you need let's!_ [12:49] ) [12:49] I meant: right before [12:49] :-) [12:49] tiday there is team call [12:49] yes [12:49] ok, then 11:30? [12:49] deal! [12:49] ack [12:49] * ralsina goes to make the dialog button stay small [12:49] but first, windows [12:49] ralsina: saw my last comment in you branch? can I re-review? [12:50] nessita: yes, changed and pushed [12:50] ack [12:56] gatox: can you try running your main_moved tests suite in a clean branch? [12:56] nessita, ok [12:56] gatox: be sure to never run ./setup.py build [13:01] ralsina: approved installer-option [13:09] nessita: thanks! [13:09] briancurtin, mandel: building binaries to test your branches IRL [13:10] ralsina: cool [13:11] ralsina, superb! [13:13] gatox: were you able to run the suite? [13:14] nessita, i'm on it.... just a sec [13:14] ack [13:15] nessita: my installer-option branch failed to merge with a new, different error (not in the tests, I think) [13:16] ralsina: links pliz? [13:16] nessita: oh, wait, segfault in a test [13:16] ralsina, briancurtin in the mean time, I'm fixing the sso broken tests on windows [13:16] nessita: https://pastebin.canonical.com/62826/ [13:17] ralsina: I don't see any error there [13:17] nessita: segfault in line 2 [13:18] ah oh [13:18] uh [13:18] eh! [13:18] :-/ [13:18] ralsina: wanna try sending it to tarmac again? [13:18] nessita: sure [13:19] sent [13:20] nessita: fixing the win32/symlink test with your suggestions [13:21] "you probably foud a bug in bazaar"... like that's going to make me feel better [13:21] briancurtin: thanks! we try to avoid as much as possible to have if sys.platform checks around, we try to "concentrate" them either in python package's init, or using custom skips [13:22] nessita: yeah i should have had a better approach...the symlink tests for the python stdlib are close to what you describe in the comments, so i should have known better :) [13:22] nessita, briancurtin: so the with-simlink is skipped on windows and the other one is skipped on linux, right? [13:23] ralsina: which other one to skip in linux? :-) [13:23] nessita: the one without the symlink [13:23] i would think both tests run on linux, just that the symlink part doesn't run on windows...but i'll see how it works [13:23] nessita: or else, we run most tests twice on linux? [13:23] ralsina: no, we run it in linux as well. We run them twice but with different setups [13:23] so, in words, that would be: [13:24] nessita: what does running it without the symlink test on linux? [13:24] hrmm [13:24] * a test suite that checks that the calculator thread do the right things with files and dirs [13:24] nessita: besides adding another 30 seconds to our patience ;-) [13:24] * a test suite that checks that the calculator thread do the right things with files and dirs and symlinks [13:24] ralsina: I just described the 2 test suites [13:24] nessita: I don't see what the first one does that the second doesn't [13:25] ralsina: the first does not use symlinks [13:25] which is a different scneario that having symlinks [13:25] ralsina: your code may "fail" if it doesn't find a symlink in the disk [13:25] nessita: ok, I don't want to be argumentative so I will shut up [13:25] (yes, that's a weird scneario, but I'm just trying to "show" the goal of the test) [13:26] ok [13:26] hey, installer-option merged! So it was a random segfault, which is quite weird [13:27] nessita, what is the problem running ./run-tests? everything it's ok here [13:28] gatox: you sure is a clean branch? you never run ./setup.py build? [13:28] ahhhh without running that.,, sorry [13:31] nessita, ahhhhh i know what you mean now [13:31] ;-) [13:31] gatox: would you fix please? [13:31] nessita, of course [13:31] what's the point of automating fetching and merging branches if bzr will fail 9 out of every 10 times I try to do it on windows :-( [13:32] ralsina: it was working fine for a while, then this week it has worked like crap so i just do it manually [13:32] briancurtin: it worked fine yesterday for me, so if it fails this time, manual it is [13:32] briancurtin: it was a good idea though :-) === urbanape_ is now known as urbanape [13:43] nessita: https://code.launchpad.net/~brian.curtin/ubuntuone-control-panel/no-symlinks-on-windows/+merge/98736 is updated [13:43] briancurtin: thankssssss [13:44] nessita: for the autostart one where i have 3 dependency branches, one which changed...do bzr pull (that branch) or do another bzr merge (that branch) === yofel_ is now known as yofel [13:46] briancurtin: merge [13:47] briancurtin: if you pull you are going to get a "branches have diverged" error eventually [13:47] so, merge and commit -m "merged latest whateverbranch" [13:47] arg, two conflicts, but it workd otherwise. thanks [13:48] briancurtin: that's for windows-autostart? [13:48] if yes, pig me when you resolve the conflicts because I just got them myself ;-) [13:48] briancurtin: I'm getting 2 conflicts on no-symlinks-on-windows... could you please merge trunk in? [13:48] nessita, fixed.... was really silly.... just needed to add a new path to ignore during gtk tests [13:48] ralsina: windows autostart, merging your installer-option. it looks like an easy fix - just adjusting the start_from_license (since i added a line right after it) [13:49] nessita: will do [13:49] gatox: :-) [13:55] briancurtin: ping [13:55] urbanape: pong [13:55] heya. how's your morning looking? [13:58] urbanape: cleaning up some branches for release, hoping to be done with that shortly. i'll let you know once i'm past that so we can get some mac stuff done [13:58] rock on. [13:59] I've made some progress getting just ubuntu-sso-client testable on its own so we can work on discrete branches. [13:59] ok, lunch time for me! [13:59] * mandel lunch [13:59] urbanape: awesome, looking forward to it [14:02] briancurtin, urbanape: we are getting a mac mini around here soon so I or someone else can start helping you out on that [14:02] briancurtin, urbanape: and we can start getting some mac branches in trunk :-) [14:04] ralsina: is the desktop+ meeting in an hour? [14:04] joshuahoover: yes [14:05] ralsina: cool...my calendar confuses me since the time change here :) [14:05] damn google calendar and DST! [14:06] briancurtin: not pushed autostart yet? [14:06] ralsina: doing it now [14:06] briancurtin: cool [14:07] ralsina: I look forward to that. [14:08] ralsina, nessita: https://code.launchpad.net/~brian.curtin/ubuntuone-control-panel/windows-autostart/+merge/98741 [14:09] briancurtin: ack [14:09] briancurtin: thank [14:31] nessita: can't to the 1:1 now, have the repairman for my fridge here :-( [14:31] will be back fr the weekly call [14:31] ralsina: ok [14:36] what's a little background noise [14:39] brb, reboot [14:55] nessita, alecu: if you wat to IRL the branches that are up for review by mandel and gatox, here is an installer with both applied: http://ubuntuone.com/3m5dJp4BCD7NxfrVJehPUC [14:55] and BTW, briancurtin, mandel, gatox [14:55] and BTW, briancurtin, mandel, gatox, that thing works SMOOTH [14:55] ralsina, roger that [14:55] so my IRL of all your branches is a +1 :-) [14:56] got asked for proxy, it worked fine [14:56] could login, got added to autostart [14:58] ralsina, hurray! [14:58] hopefully this will be a short meeting today [14:58] ralsina, does it work without proxy? [14:58] mandel: have not tested that yet [14:59] mandel: looks like a yes to that [14:59] ralsina, great! [15:00] mandel: but I stopped it before disabling proxy, since I understand that's not supposed to quite work, right? [15:00] errday i'm mumblin, mumblin [15:00] ralsina, what do you mean? [15:00] mumble [15:01] mandel: enabling/disabling proxy with a runing syncdaemon? [15:01] ralsina, no no no.. [15:01] ralsina, although is a very nice feature to have :) [15:01] any time i switched proxy/no-proxy, i rebooted the whole box (probably not necessary, but rebooting a VM is quick) [15:02] alecu: mumble? [15:02] alecu, nessita: mumble [15:02] gatox: and you [15:02] * ralsina needs a team rster somewhere visible on his desk [15:03] connecting [15:35] yay, now i can get lunch [15:35] bbiab [15:45] FYI i'm doing another lunch+doctor combo in a bit...girlfriend needs a ride to get a huge needle stuck in her neck and they moved the appointment up :/ (supposedly they have wireless!) [15:45] ha, that looks so horrible. sad story...wireless! [15:50] * gatox lunch [15:57] ralsina, gatox, briancurtin lets get the sso tests back to pass on windows => https://code.launchpad.net/~mandel/ubuntu-sso-client/fix-broken-tests/+merge/98868 [15:57] that should mean we get green lights on that project :) [15:57] * mandel moves along to the next [15:58] mandel: sweet, will look [16:04] * briancurtin lunch [16:14] mandel: is that supposed to fix *all* tests? [16:15] ralsina, yeah :) [16:15] ralsina, well, in sso, try it out please [16:15] mandel: because it doesn't for me [16:15] ralsina, argh! can I see your output? [16:15] mandel: you know on windows that's not easy, when the run ends I'll do my best :-) [16:16] mandel: https://pastebin.canonical.com/62854/ [16:16] ralsina, oh, true, I usually run the thing with runtests >> test.out [16:17] ralsina, you don't have sso or ubuntu one running, right? [16:17] mandel: no, I stopped it [16:17] * mandel curses! [16:18] ralsina, it could be that your machine is faster than mine, since I'm in a vm, and it has the time to close the connection while in yours it does not.. [16:19] mandel: so, what could I do? [16:19] ralsina, nothing, I'll fix that being smart somehow, is a matter of thinking :) [16:19] * alecu out for lunch, and to get a macmini [16:20] alecu, are you getting a small mac donals burger? [16:20] mandel: DOOMED! We are DOOMED! ;-) [16:20] alecu: take good care of it ;-) [16:20] ralsina, if we depend on me thinking, yes we are :) [16:20] ralsina, ubuntuone-client has probably the same issues.. [16:22] ralsina, we should not use the bloody ipc in tests.. but well, I'll do my best [16:26] ralsina: have a minute for helping me with qt stylesheet? [16:27] nessita: sure! [16:27] ralsina: is there any way to achieve a border for the tabwidget like is requested here https://launchpadlibrarian.net/96895343/u1_desktopClient_colours.png ? [16:27] I'm just gettingt either this: http://ubuntuone.com/0G4ebsD1ECcxfkMRksWpE6 [16:27] or this: http://ubuntuone.com/27plBng2XJmCgCsIPzocrp [16:27] nessita: yes [16:27] ralsina: would you teach me how? [16:28] nessita: set the border-bottom of tabbar::selected to none, I think [16:28] ahhh let's try [16:28] ralsina: is already set to white... [16:28] you think None will work? [16:28] nessita: examples on how to do that thing: http://doc.qt.nokia.com/4.7-snapshot/stylesheet-examples.html#customizing-qtabwidget-and-qtabbar [16:28] nessita: one chance is that you have another selector of the same specificity lower [16:29] so it's overwriting it [16:29] will confirm [16:29] nessita: can you push the branch? [16:29] ralsina: let me play with what you just said and will push if I can't fix it... [16:29] nessita: ack [16:31] mandel: eve with the dirtyreactor, your fix is important because for me sso tests in trunk *hang* [16:31] mandel: so I would accept that branch as a partial fix [16:35] oh my poor grey matter :-/ [16:36] dobey: allergies again? [16:39] of course [16:39] my silver car was green. [16:39] nessita, dobey, briancurtin, urbanape, gatox, mandel, alecu: Tomorrow I have to take my son topediatrician at noon, so will probably not be around from 3PM to 5PM UTC. I will start 1 hour earlier and not take a lunch break, but don't be surprised if I am not around at that time. [16:39] or well, still is green. i didn't wash it during lunch :P [16:40] kk [16:40] ack [16:40] ralsina: i work with mandel. *nothing* surprises me any more. [16:40] dobey: hey, behind you! A three headed monkey! [16:41] his name is kerberos [16:41] lol [16:41] dobey: nice pet! Here's a snack for him: http://feedproxy.google.com/~r/boingboing/iBag/~3/SUc0Bo-E3QA/just-look-at-this-mutant-tripl.html [16:42] * ralsina has a boingboig postfor every occasion [16:44] heh [16:45] dobey: I fear asking but why was your silver car green? [16:45] it was made of copper and rusted? [16:45] ralsina: no success :-/ lp:~nataliabidart/ubuntuone-control-panel/new-colors [16:45] it was made of wood and got wet? [16:45] it was made of potatoes and got old? [16:46] ralsina: because it's covered in pollen [16:46] nessita: will give it a shot [16:46] dobey: I suppose that's like if it was made of kryptonite [16:46] kryptonite is really just a metaphor for 'plants' in Superman [16:46] which was my next guess [16:47] dobey: so superman is the story of a guy and his allergies? Projecting much? [16:47] I mean, dracula? Sure, alergies. [16:49] ralsina: I'm just saying. Nerd that works at the newspaper; kryptonite? Come on. [16:50] briancurtin: so there *was* wifi i the neck-piercing doctor's? [16:51] ralsina: yep! [16:51] ralsina: pushed a minor tweak to also have border on windows for QAbstractItemView [16:52] nessita: ack [16:56] ralsina, running tests in a deterministic manner on sso for windows, will you be free in a couple of mins to try it in your machine? [16:57] mandel: sure [16:57] ralsina, sweet, tests should be a little slower but you can trust them :) [16:57] nessita: trick is QTabBar::tab:selected doesn't seem to be doing anything [16:58] nessita: or maybe not, need to do some more tests ;-) [16:58] mandel: let me know if you need any windows work. i have my win laptop here, not doing mac stuff atm === zyga is now known as zyga-food [16:58] mandel: yes, briancurtin2 can test it [16:59] nessita, briancurtin: i noticed bug #851810 is still in-progress...any update on this one? [16:59] Launchpad bug 851810 in Ubuntu One Client "[FFe] Notify clients when volumes info from server is ready" [Medium,In progress] https://launchpad.net/bugs/851810 [16:59] ralsina: let me know [17:00] joshuahoover: briancurtin needs to apply some changes, and we haven't received an approval yet... so is not released [17:01] the only reason it "needs" an ffe, is because it's a dbus api change [17:06] dobey: actually, an adding [17:06] but yes [17:07] well, adding is a subset of change, yes :) [17:07] ralsina: i'm not sure if this is an artifact of the installer being setup for QA or not, but SD & SSO only run when CP is open. if you close it, everything else closes [17:07] anyway [17:08] briancurtin: that's how it works on windows [17:08] briancurtin: since on windows it should start with --with-icon and live in the tray :-) [17:08] ralsina: ah, that's what i was getting to....i was noticing it wasn't living in the tray [17:09] nessita: ok, I have an aswer about the border now [17:09] ralsina: I'm listening [17:09] and reading [17:09] nessita: the border is being set correctly, but it's BELOW the border of the pane [17:09] so, the only way to fix this would be to have the pane have no top border [17:10] and then it looks like ass [17:10] ralsina: so is one of the options I shown? [17:10] nessita: to see that this is the case, set this on QTabBar::tab:selected {border-bottom: 5px solid red; margin-bottom: -2px;} [17:11] nessita: sorry, don't understand you [17:12] ralsina: when I "showed" you the problem I linked 2 possible solutions: [17:12] (01:27:26 PM) nessita: I'm just gettingt either this: http://ubuntuone.com/0G4ebsD1ECcxfkMRksWpE6 [17:12] (01:27:37 PM) nessita: or this: http://ubuntuone.com/27plBng2XJmCgCsIPzocrp [17:12] ralsina: you're saying the best option is the second one there? [17:13] right, the second one [17:13] right [17:13] no, I say both look ugly :-) [17:13] and I would rather have the border cross the selected tab [17:13] but let me make one final try [17:13] ralsina: I agree. If you consider there is no way to have what design is requesting, I will ask lisette what she prfers [17:16] nessita: I have something that just needs a tweak: http://screencast.com/t/6qyOJ4T5J [17:17] nessita: diff https://pastebin.canonical.com/62865/ [17:17] nessita: you will notice some magic numbers there that need to be corrected to make it look right [17:18] nessita: based in the third example from http://qt-project.org/doc/qt-4.8/stylesheet-examples.html#customizing-qtabwidget-and-qtabbar [17:20] nessita: if you want me to do it, ping me, I am going to prepare mate [17:32] ralsina: I will give it a try, I love learning this [17:32] nessita: it's fun, yes :-) [17:32] nessita: if/when we do a QML frontend, it's like this but worse :-) [17:33] lol [17:34] ralsina, briancurtin can you pull lp:~mandel/ubuntu-sso-client/fix-broken-tests and check that you get clean reactors, my machine is slow I don't get the issue that often [17:34] mandel: sure [17:34] I would pay $5 for a windows version of yakuake. really. [17:35] ralsina, will be revno 935 [17:35] mandel: ack [17:37] mandel: so far so good! [17:37] ralsina, hey! [17:38] * ralsina goes see if fagan wants to +1 that branch for full dejavu [17:38] ralsina, hahaha [17:38] ralsina, I'm sure that on linux is going to complain due to pylint, let me fix that first :) [17:38] oops, 63 errors because I did not stop ussoc [17:38] it complains on windows about pylint too [17:39] but that's probably me having an old devtools or something [17:39] mandel: got some uncleans again, waiting for it to finish [17:39] :-( [17:39] ralsina, hm.. as I suspected, is nearly there, but I'm mockey patch PbServerFactory and PbClientFactory and I fear there is one somewhere puteando.. [17:40] * dobey pokes pylint with very hot poker hoping for it to start burning === zyga-food is now known as zyga [17:40] ralsina, may I have the output when done? [17:40] mandel: sure [17:40] i should just slip in a branch to make pyflakes the default. i bet nobody would notice [17:41] mandel: just 18! https://pastebin.canonical.com/62868/ [17:41] ralsina, but that number is 'random' so.. [17:48] hrmm, would be nice to get storage-protocol knocked down to 0 bugs [17:49] dobey: current score? [17:50] 10 [17:50] but definitely too late to do some of them for precise [17:51] mandel: 1 error - unclean reactor in ubuntu_sso.main.tests.test_clients.RegisterTestCase.test_remote_signals [17:51] dobey: isn't 620558 a dup of 720707 ? [17:52] not sure. but probably [17:52] if it is, then one less to fix :) [17:52] but most of them are "we need to send X over the protocol with a new message" and such [17:53] I need to triage / kill / move to u1cp all the windows-installer bugs [17:53] boooooooooorig [17:53] eh [17:54] dobey: you are pretty close to 0 on ubuntuone-installer ... would be nice to get there before we kill it ;-) [17:55] dobey: and I thik bug #853060 is already fixed? [17:55] Launchpad bug 853060 in ubuntuone-installer (Ubuntu) "ubuntuone-installer crashed with GError in function(): Failed to execute child process "ubuntuone-control-panel-gtk" (No such file or directory)" [High,Incomplete] https://launchpad.net/bugs/853060 [17:55] no [17:55] well [17:55] it's incomplete [17:56] it's "new" [17:56] ;-) [17:56] no it isn't [17:56] ;) [17:58] there's 13 against installer in ubuntu though [17:58] most of them will probably become wontfix though ;) [17:59] * gatox stare at the code, thinking how to fix this without breaking other things [18:00] rm -rf all the things [18:01] anyway [18:02] :P [18:24] hospital duties are over. stopping for lunch on the way home, be back online in a bit [18:28] ralsina, can you try the branch again, I have ran it several times and its ok in my vm [18:30] gatox, got a windows machine? care to test something for me? [18:30] mandel: sure [18:31] mandel, ok [18:31] ralsina, and I'm going to make jenkins swear whenever someone does this again.. [18:31] mandel: that's the whole point of jenkins :-) [18:31] gatox, please, run the tests on windows for lp:~mandel/ubuntu-sso-clitn/fix-broken-tests [18:32] mandel: did tests just become a lot slower? [18:33] mandel: and it still gives unclean reactors [18:33] ralsina, yes, because we are makins sure we do connect the client and the server and we do disconnect correctly.. [18:33] ralsina, .lnads.lasf ;ojs [18:33] cojones!" [18:34] ralsina, and you do not have sso or anything else running, right? [18:34] mandel: no, if I do that, I get 70 failures or so [18:34] mandel: https://pastebin.canonical.com/62872/ [18:35] ralsina, revno 937, right? [18:36] mandel: right [18:36] ralsina, sorry to doubt.. but I want to be sure :) [18:36] mandel: it's ok :-) [18:36] mandel, running tests.... [18:36] ralsina, weird.. I have ran it like 10 times and it aways passes all of them.. [18:37] ralsina, can you get the full stdout? runtests >> test.out works [18:37] mandel: sure [18:37] thx [18:38] mandel, i have a couple of fails.... but i'm restarting the vm..... because sometimes i need to do that to run tests on windows [18:39] gatox, really? [18:39] mandel, yap [18:39] mandel, it's like it doesn't close some process or something [18:40] mandel: https://pastebin.canonical.com/62876/ (2 errors!) [18:41] mandel, running again here [18:46] ralsina, looks like the protocol does not get disconnected.. [18:46] I wish closing things was easier in twistd [18:49] mandel, sorry.... i have to run it again because i leave it running and ends with a lot of pylint isssues and i lose the trace of the errors, but it was a dirty reactor in some place [18:58] mandel, 1 error: http://paste.ubuntu.com/895484/ [19:01] gatox, ralsina I think I found the issue.. can you branch again and cat the stdout? [19:01]