[00:03] <rockstar> mwhudson, just sent mine to you.
[00:26] <rockstar> mwhudson, https://code.edge.launchpad.net/~rockstar/launchpad/assert-needs-upgraded/+merge/14125
[01:17] <mwhudson> review pls: https://code.edge.launchpad.net/~mwhudson/launchpad/dont-oops-sftp-unlink-directory-bug-372506/+merge/14127
[01:34] <rockstar> mwhudson, r=me - I actually wanted to fix that bug.
[02:11] <mwhudson> rockstar: oh!
[02:11] <rockstar> mwhudson, ?
[02:12] <mwhudson> rockstar: sorry for toe-stepping
[02:12] <mwhudson> rockstar: it took me too long to fix it, i kept seeing TransportError and reading TransportNotPossible :/
[02:12] <rockstar> mwhudson, no, it was pie in the sky wanting.  I was thinking "you know, I could fix this rather simple bug, and maybe learn some more about codehosting while I'm at it."
[02:12] <rockstar> mwhudson, one of my big concerns is that I don't know enough about codehosting.
[02:12] <mwhudson> rockstar: oh ok
[02:13] <mwhudson> rockstar: fair enough
[02:13] <rockstar> Although really, the upgrade_branches stuff has started curing that.
[02:13] <rockstar> Now I know I just want you to fix it all.
[02:13] <rockstar> :)
[02:13] <mwhudson> :)
[02:14] <mwhudson> rockstar: i know very little about js in launchpad, so maybe we're even
[02:14] <rockstar> mwhudson, yea, I can take that tradeoff.
[12:26] <noodles775> al-maisan: did you start abentley's branch? if not, I'll start it now.
[12:26] <al-maisan> noodles775: please do.
[12:45] <jtv> noodles775: thanks for the review btw
[12:46] <noodles775> np.
[12:46] <noodles775> jtv: why was it that your original conf for the test_runner didn't work?
[12:46] <noodles775> al-maisan or anyone else, MP here if you have time: https://code.launchpad.net/~michael.nelson/launchpad/copy-binaries-timeout/+merge/14157
[12:46] <jtv> noodles775: it worked, but I'm documenting the procedure for oops-enabling a script, and this turned out to be the "slightly cleaner" way.  I wanted to validate my documentation.
[12:47] <noodles775> jtv: but why is the test_runner using the development conf?
[12:47] <al-maisan> noodles775: please give me a bit of time and I'll get to it.
[12:47] <noodles775> (or is that just the way it is)?
[12:47] <noodles775> al-maisan: no rush... someone else might get to it too :)
[12:48] <jtv> noodles775: it inherits that config as a base, apparently because there's a lot of overlap.
[12:48] <noodles775> jtv: ah - ok, that makes sense then.
[12:49] <jtv> noodles775: btw, excuse me if my advertising is too aggressive, but if you know anything about this, please take a moment to review https://dev.launchpad.net/LoggingOopses
[12:49] <noodles775> :)
[12:50] <noodles775> jtv: thanks for documenting that! It'll make it much easier for the next person... I haven't had to add an oops yet.
[12:51] <jtv> noodles775: I'll bet there's a script running somewhere that you _could_ do it for and that _should_ be logging oopses.  :-)
[12:51] <noodles775> jtv: without a doubt.
[12:59] <abentley> noodles775: You're a star.  I'm around if you have any questions.  
[13:00] <noodles775> abentley: Thanks.
[13:03] <BjornT_> jtv: i was looking at your lp:~jtv/launchpad/bug-463097. where are oops recorded when you run the script itself?
[13:06] <jtv> BjornT_: from memory, /var/tmp/poimport.test
[13:06] <jtv> BjornT_: though maybe I didn't fully understand your question... the script itself as opposed to what?
[13:07] <noodles775> jtv: when the real script runs - as opposed to running the tests - you mention in your MP that the configuration for production will be in a separate branch (or something to that effect)?
[13:08] <BjornT_> jtv: as opposed to the test runs. i would have thought that we wanted separate dirs for this, like we usually have
[13:08] <BjornT_> jtv: for example 'make run' produces oopses in /var/tmp/lperr, while 'make check' produces oopses in /var/tmp/lperr.test
[13:08] <jtv> noodles775: the config for production goes into lp-production-configs, not devel/db-devel etc.
[13:09] <BjornT_> jtv: why does poimport need to be different?
[13:09] <jtv> BjornT_: existing practice is mixed, and I haven't found any documentation.
[13:09] <jtv> BjornT_: in fact, you may be just the person to make a ruling on this, right now.  :-)
[13:13] <noodles775> abentley: do you think we'll eventually need an option to modify the prerequisite branch? or just create a new one?
[13:14] <noodles775> abentley: also, I wonder whether the +resubmit page should mention the prerequisite too? What do you think?
[13:14] <abentley> noodles775: I think we won't need to modify the prerequisite branch often.  We don't provide a way to modify the target branch, and I was following that.
[13:14] <noodles775> yep, makes sense.
[13:14] <BjornT_> jtv: i think it makes sense to have separate dirs. the test suite does produce oopses, so it would be nice being able to keep the normal oops dir empty for manual testing.
[13:15] <abentley> noodles775: I think we could mention the prerequisite branch on +resubmit.  I did test that it works.
[13:15] <noodles775> abentley: ok, great. r=me (code/ui*).
[13:16] <abentley> noodles775: So are you asking me to change the resubmit text?
[13:17] <noodles775> abentley: not necessarily, I was just asking you whether you thought it would be worthwhile - up to you.
[13:17] <noodles775> (details in the mp reply coming :) ).
[13:18] <abentley> noodles775: Cool.
[13:20] <jtv> BjornT_: I don't feel particularly strongly about it; I use file timestamps to sort out the "manual" oopses from the "test" ones and I don't like having too many directories in /var/tmp, but I can see your point as well.
[13:34] <abentley> noodles775: We plan to use prerequisite branches for more than just calculating the diff-- we'll also be using it when landing merge proposals.  E.g. if the prerequisite branch hasn't been approved or merged, we might prompt for confirmation before landing the subsequent merge proposal.
[13:35] <noodles775> abentley: nice!
[13:36] <noodles775> abentley: I guess I was just trying to think like someone who's not aware of any of that. I create an MP which has a previous branch included - and all I know is that I'd like the diff to not include those changes. Hmm...
[13:36] <abentley> noodles775: So that's why the text is broader than just being about diffs.
[13:38] <noodles775> abentley: do you think it would be too explicit to say "A branch that should be merged before this one. The changes in this prerequisite branch will not be displayed on the merge proposal."
[13:38] <noodles775> s/explicit/verbose
[13:39] <abentley> noodles775: Wouldn't you want to say "...will not be displayed in the diff"?
[13:40] <noodles775> abentley: yes, that's clearer.
[13:41] <abentley> noodles775: But it still is a bit long.  I wonder if we should provide it as inline help?
[13:41] <noodles775> abentley: one other tiny thing, I don't think that should be a capital 'B' for 'Branch' in that description.
[13:41] <abentley> noodles775: Sure.
[13:41] <noodles775> abentley: yeah, it would be quite long. inline help is a great option.
[13:42] <abentley> noodles775: But is it possible with autogenerated forms?
[13:42] <noodles775> as you could also mention there the other affects later on.
[13:42] <noodles775> hm, not sure, but I'd assume so?
[13:42]  * noodles775 checks out of interest.
[13:43] <abentley> noodles775: Because I think it's normally done with a specially-formatted link...
[13:46] <noodles775> abentley: if that is the case, maybe leave it for the moment.
[13:46] <noodles775> (or use the long description - up to you.)
[13:54] <abentley> noodles775: Here's an incremental diff: http://pastebin.ubuntu.com/304310/
[13:55] <noodles775> abentley: great, thanks.
[13:55] <noodles775> abentley: one small thing: s/and target/target in branchmergeproposal-resubmit.pt?
[13:56] <abentley> noodles775: I actually just did s/and target,/target/
[13:56] <noodles775> yep, I was about to mention the comma :) Excellent, r=me.
[13:57] <abentley> noodles775: Cool.
[14:00] <abentley> noodles775: I plan to add a command to bzr-pipeline to create a branch merge proposal, including the prerequisite branch.
[14:01] <noodles775> abentley: great! I'll be looking forward to that :)
[14:01] <noodles775> abentley: btw, I noticed the other day that I had to keep r130 of pipeline, as otherwise it was requiring a later version of bzrlib?
[14:03] <abentley> noodles775:  lp:bzr-pipeline/stable is compatible with the stable bzr series (2.0.x) and lp:bzr-pipeline is the devel branch.
[14:03] <noodles775> abentley: ah great - I'll switch to that, thanks.
[14:03] <abentley> noodles775: np
[14:08] <leonardr> noodles775, for your queue: https://code.edge.launchpad.net/~leonardr/launchpadlib/trusted-workflow-tests/+merge/14167
[14:09] <noodles775> leonardr: I'll jump on it soon :)
[14:16] <maxb> I would like to append the following one-liner to the queue: https://code.edge.launchpad.net/~maxb/launchpad/remove-stray-ref-to-sourcecode-lazr-js2/+merge/14168
[14:31] <al-maisan> noodles775: I am looking at your branch.
[14:31]  * al-maisan needs a review as well: https://code.edge.launchpad.net/~al-maisan/launchpad/relsets-463195/+merge/14169
[14:37] <noodles775> Thanks al-maisan !
[14:43] <noodles775> al-maisan: i'm going to finish reviewing now, I've pushed leonardrs back into the queue as I didn't get to start it.
[14:43] <al-maisan> noodles775: no problem
[18:07] <abentley> rockstar: Did you get my request to review https://code.edge.launchpad.net/~abentley/launchpad/prerequisite-ui/+merge/14110 ?
[18:20] <rockstar> abentley, I did.  I'm on the phone with jam, but I plan on looking at it right after that.
[18:35] <rockstar> abentley, wasn't this going to land against db-devel?  I'm okay with reviewing against devel, just checking though.
[18:36] <abentley> rockstar: Yes, it is.  That's why I think eventually we'll want to support changing merge proposal target branches.
[18:36] <rockstar> abentley, ack.
[18:37] <rockstar> abentley, ui=me
[18:38] <abentley> rockstar-afk: thanks.