[09:34] <davmor2> morning everyone
[09:34]  * davmor2 thinks the wiki still isn't quite right
[09:52] <JonPackard> davmor2: Morning davmor2 :)
[09:55] <davmor2> morning :)
[09:56] <JonPackard> davmor2: do you have the template you were working on ready? I want to add the link to the meeting summary
[09:56] <davmor2> JonPackard: https://wiki.ubuntu.com/Testing/Bug-Report-Layout I posted it to the qa list too :)
[09:57] <JonPackard> great thanks!
[09:58] <persia> davmor2: A couple questions about that template: firstly, does the testing team always know the cause?
[09:59] <persia> Secondly, Do you think the tester shouldn't self-triage the reported bug?  While more eyes are good, I'd think more streamlining would be a faster path to a fix.
[10:00] <davmor2> persia: Sometimes yes, if you read the wiki page you'll see why I added it..  Nine times out of ten before I report a bug I'll go onto the relevant -devel channel and say this is happening most of the time I get an answer as to the cause.  So I think it is useful to have it in there as a reminder to the dev's about the discussion.
[10:01] <persia> davmor2: That makes sense.  I suppose it depends on the class of bug, but many of the ISO testing bugs are certainly explainable.
[10:03] <davmor2> persia: No one should triage there own bug.  However I see no issue with the stuff me and stgraber were doing where by we confirmed each others bug's and then I believe slangasek add priorities etc to all the confirmed bugs :)
[10:04] <persia> davmor2: Hmm.  From a process point of view, that makes sense, although in many cases when one understands a bug well, and knows that all the information is provided, it's tempting.  I think working in pairs sounds like a good solution.  Thanks for the explanation.
[10:05] <davmor2> persia: No Probs :)
[10:10] <davmor2> ara: which channel did you let know about the wiki?
[10:23] <ara> davmor2: i didn't
[10:23] <ara> davmor2: someelse did
[10:23] <davmor2> ara: ah right thanks :)]
[11:16] <stgraber> davmor2: I updated the wiki page layout a bit
[11:16] <davmor2> :)
[11:25] <davmor2> stgraber: how did you get the magic lines to appear ?
[11:27] <stgraber> you put --- instead of [11:27] <davmor2> D'oh I know I'd done something wrong I couldn't think what :)
[11:27] <stgraber> hehe :)
[11:29] <davmor2> Looks good ta stgraber :)
[11:53] <cr3> ara: hi there, would you like to spend a moment with that ldtp integration in hwtest?
[11:53] <ara> cr3: sure
[11:53] <ara> cr3: did you get the chance to try what i sent you?
[11:55] <davmor2> cgregan: bit twirly for you isn't it dude :)
[11:55] <cgregan> yep
[11:55] <cgregan> meeting
[11:56] <cgregan> mobile IRC over on ubuntu-meeting in 5
[11:56] <cr3> ara: nope, I had forgotten about that. give me a little moment
[12:00] <persia> cgregan: Well, no.  still 16:00 UTC this time.
[12:00] <cgregan> persia: I thought dave moved it as of yesterday
[12:01] <cgregan> grrrr
[12:01] <davmor2> I didn't :P
[12:01] <cgregan> hehe
[12:02] <persia> cgregan: It was proposed.  Changing it requires consensus :p
[12:02] <davmor2> http://fridge.ubuntu.com/node/1576 no change here :)
[12:02] <cgregan> persia...well then...I'm off to start my morning in a more traditional way. (30 more minutes of sleep, and breakfast)
[12:03] <davmor2> which is a shame :(
[12:03] <cgregan> :-)
[12:07] <cr3> ara: ok, I've checked your email. first, how do you typically run your ldtp tests?
[12:07] <ara> I use ldtprunner
[12:08] <cr3> ara: right, what's the exact command?
[12:08] <ara> it depends on the test you want to run
[12:08] <ara> for updatesystem, i.e.
[12:08] <ara> https://wiki.ubuntu.com/Testing/Automation/LDTP/UpdateSystem
[12:09] <ara> you can do ldtprunner update_system.xml
[12:09] <cr3> ara: is that one test or a suite of tests for update system?
[12:10] <ara> cr3: in that case it only contains one test. but to be specific, it is a test suite that contains 1 test for the time being :-)
[12:11] <cr3> ara: cool, exactly what I needed to know :)
[12:11] <ara> cr3: if you use openAll, there are test suites with more than one test case :-)
[12:13] <cr3> ara: in your ubuntu-tests project, where do you intend to install your xml files?
[12:13] <cr3> ara: maybe we should start packaging your project...
[12:13] <ara> cr3: yes
[12:14] <cr3> ara: I could help with that, where do you want the files in the respective directories: gedit, openAll and updateSystem?
[12:14] <ara> cr3: but you can suppose any shared folder
[12:14] <cr3> ara: /usr/share/ubuntu-tests?
[12:14] <ara> cr3: yes
[12:15] <cr3> cool, mind if I try it out and then push to your branch?
[12:15] <ara> cr3: and keeping the folder structure I think is a good idea
[12:16] <cr3> ara: do you want the package to be called ubuntu-tests or ubuntu-desktop-testing as we discussed in london?
[12:17] <ara> mmm, ubuntu-desktop-testing includes both ubuntu-tests and testing-library
[12:17] <ara> cr3: and I think they should be different packages
[12:17] <cr3> ara: when you say "includes", you mean "requires" in packaging lingo, right?
[12:18] <cr3> ara: so you're working on three projects right now?
[12:18] <ara> cr3: and maybe ubuntu-desktop-testing being a meta package that requieres both of them
[12:18] <ara> ubuntu-desktop-testing would be a metapkg
[12:18] <cr3> ara: what's the LP project for testing-library?
[12:18] <ara> cr3: same
[12:19] <ara> cr3: but another bzr branch
[12:19] <cr3> ara: evil! :)
[12:19] <davmor2> sbeattie: ping
[12:20] <ara> cr3: :D
[12:20] <cr3> ara: you should either fold the code into the same branch, which could result in two packages anyways. or, you should create another project
[12:21] <ara> cr3: ok. will think about it
[12:21] <cr3> ara: for now, I would lean towards the first option. we could always change our minds later
[12:22] <cr3> basically, keep things self contained as much as possible unless there is a specific use case
[12:24] <ara> ok, then I will fold the code into the same branch
[12:25] <cr3> ara: cool, please do that and I'll continue working on creating two packages from that branch. we could create a meta package afterwards if necessary
[12:25] <cr3> let me know when you've pushed your changes and where the changes reside
[12:25] <ara> cr3: ok. will do. lunch time now :) will continue afterwards
[12:26] <persia> Hrm?  Why is it bad to have multiple branches in a project?
[12:27] <cr3> persia: unrelated branches, not just branches of the same code
[12:27] <persia> cr3: But in the same rough collection of stuff?
[12:27] <cr3> persia: for example, pushing a mozilla branch under the xorg project is uncool
[12:28] <persia> cr3: Certainly.  On the other hand, the ubuntu-dev-tools project has a lot of different little utilities, which are released together, but each could be developed separately.
[12:28] <cr3> persia: not the same to the extent where merges can be done between the branches
[12:29] <persia> I think that for smaller projects, it might make sense to leave them together just to avoid namespacing issues, but perhaps that's just me.
[12:30] <cr3> persia: that's fine, the point of branches is that features can be developed independently. however, once these features are done, they should be merged into the trunk
[12:31] <cr3> persia: furthermore, the concept of series is meant to have one main branch, typically called "trunk", per project
[12:31] <cr3> persia: err, one main branch per release. for example, there might be a 1.0 series for a particular release
[12:31] <persia> Ah, right.  I've misread the backlog.  I perceived the concept of identity between package and branch, but now see that there is one branch for multiple packages as a target, which works.
[12:34] <cr3> I kinda wish LP was a bit more opiniated in terms of project management, it's too easy to get confused with all the features made readily available
[12:52] <cr3> ara: let me know when you come back from lunch, I have another suggestion for you: hwtest integration could be provided as another package developed within the same project
[12:53] <cr3> ara: so, all the code relating to desktop testing could be self contained in a single project and be decoupled appropriately in the form of packages
[13:13] <cr3> schwuk: ping, how does pyrering make use of those generated reports?
[13:16] <schwuk> cr3: AFAICT it doesn't. No reporting framework whatsoever. I suspect they will have something internally (after all can you imagine a Google *not* using a database of some kind?), but as far as Pyreing itself goes the reports are the end result.
[13:19] <ara> cr3: just got back
[13:19] <ara> cr3: that sounds good
[13:58] <cr3> ara: have you got ubuntutesting and the ubuntu-tests under a single branch now?
[13:58] <ara> not, yet. will let you know
[13:58] <ara> cr3: ^
[13:59] <cr3> ara: mind if I do it? I've got packaging done which you could review for me
[13:59] <ara> cr3: sure
[14:01] <cr3> ara: see packaging branch under the ubuntu-testing project. you can run debuild in there to create the ubuntu-tests and testing-library packages
[14:02] <cr3> ara: please try installing those packages and let me know if they look good
[14:03] <ara> cr3: ok
[14:21] <davmor2> So cgregan up for  real this time then :)
[14:22] <cgregan> davmor2: Yes....and feeling much better about it! :-)
[14:22] <cgregan> large cuppa and a little food.
[14:23] <davmor2> :)
[14:59] <davmor2> cody-somerville: Saturday's aren't good for me :(
[15:24] <sbeattie> davmor2: pong
[15:43] <davmor2> sbeattie: How do what were you saying yesterday about the dl-iso-test script?
[15:45] <sbeattie> oh, right. the latest dl-ubuntu-test-iso script in the ubuntu-qa-tools bzr branch supports a "--versions" argument.
[15:45] <sbeattie> This walks your ~/iso/ tree looking for files named *.iso and runs isoinfo on them to pull out and report the contents of /.disk/info which includes the specific build version.
[15:46] <sbeattie> (Though looking now, it doesn't include the type of iso, livecd vs alternate)
[15:47] <davmor2> :) something to perfect then :)
[15:47] <sbeattie> that's useful for after the fact gathering the build information to report in the bug.
[15:49] <davmor2> true.
[17:14] <sbeattie> davmor2: I'd be curious if you can reproduce https://bugs.launchpad.net/ubuntu/+source/memtest86+/+bug/246412
[17:17] <davmor2> That's good it just fails everything then dies
[17:18] <sbeattie> you got the same result? I'm going to nominate it for intrepid then, so that hopefully someone looks at it.
[17:19] <davmor2> I've added a confirm to it anyway :)
[17:19] <sbeattie> heh, thanks. :-)
[18:07] <cgregan> ogasawara: ping
[18:07] <ogasawara> cgregan: pong
[18:07] <cgregan> Hello
[18:07] <ogasawara> cgregan: what's up
[18:08] <cgregan> I just ran your script for the test results.....worked great for the results....but it did something weird with the data it pulled
[18:08] <cgregan> It grabbed the URL and put it as the section header, dropping the section header
[18:09] <cgregan> Not sure why....it was working fine the last time
[18:09] <cgregan> At the sprint
[18:09] <ogasawara> hrm, can you paste bin it for me to look at?
[18:09] <cgregan> Sure
[18:11] <cgregan> ogasawara: https://pastebin.canonical.com/7804/
[18:11] <cgregan> ﻿ogasawara: BTW....the customer is still happy....so thanks for working on this
[18:12] <ogasawara> cgregan: and can you refresh my memory for an example test case wiki to test with
[18:13] <cgregan> ogasawara: https://wiki.ubuntu.com/cgregan/testing/sample-cases
[18:20] <cgregan> ﻿ogasawara: One thing I noticed...not sure if it matters: When the ubuntu wiki translates [18:20] <cgregan> I would think moinmoin would translate the same way everywhere.....but I think we are using the raw text anyway
[18:21] <ogasawara> cgregan: hrm, just testing here really quick with the sample-cases wiki and it worked.  can you maybe grab me the raw wiki text of one of the actual test cases from Canonical site
[18:22] <cgregan> sure
[18:23] <cgregan> ﻿ogasawara: https://pastebin.canonical.com/7805/
[18:27] <ogasawara> cgregan: do you mind if I temporarily edit the sample-cases wiki so I can test?
[18:27] <cgregan> not at all
[19:01] <ogasawara> cgregan: how hard would it be to get me a wiki login for harbour?  I can't seem to reproduce what you guys are seeing.
[19:02] <cgregan> Hmm.....Not that hard. moving to private
[19:02] <ogasawara> ok