[00:00] <lifeless> dash: the trash can icon in the branch page
[00:00] <dash> but it seems to be incorrect
[00:00] <dash> Hmm!
[00:00] <dash> i guess i am not that observant. thanks.
[00:00] <lifeless> barry: to help people in this situation, remind them that mercurial *does not support branches that aren't forks*
[00:00] <lifeless> barry: hg defines a branch as a fork as a branch
[00:01] <lifeless> barry: hg idiomatic use is to have a separate repository for conceptual branches
[00:04] <lifeless> mtaylor: yeah ask a question please
[00:05] <mtaylor> lifeless: /me not grokking difference between fork and branch in above conversation
[00:06] <mtaylor> lifeless: me also doesn't fully grok how people use hg for work - but that's likely my fault
[00:06] <lifeless> mtaylor: hg defines branch as 'the graph has two heads'
[00:06] <lifeless> mtaylor: *users* of most other systems define branch as 'label I want to tell people about'
[00:06] <mtaylor> ah. gotcha
[00:07] <mtaylor> maybe that's why I haven't grokked hg yet... I've been trying to map it on to #2 there
[00:08] <lifeless> right
[00:08] <lifeless> its, like mtn, a totally different definition of branch to what people what
[00:08] <lifeless> *want8
[00:08] <lifeless> for all that its mathematically correct.
[00:09] <lifeless> svn, bzr, git, cvs all allow arbitrary labels-that-can-be-edited
[00:09] <lifeless> hg does have tags
[00:09] <mtaylor> lifeless: question asked
[00:10] <mtaylor> lifeless: https://answers.edge.launchpad.net/launchpad/+question/77340
[00:12] <lifeless> thanks
[00:12] <lifeless> I have a guess about whats going on
[00:12] <lifeless> the mirror process is opening the mirrored copy
[00:12] <lifeless> which hasn't been upgraded
[00:13] <lifeless> that tries to stack on trunk (upgraded), and fails.
[00:13] <lifeless> it then doesn't reset the mirror and mirror it again
[00:13] <lifeless> mtaylor: so just push your perl stuff to a new branch
[00:14] <mtaylor> lifeless: k
[00:14] <lifeless> and we'll get the mirror code fixed for add-bindings async
[00:14] <lifeless> which reminds me
[00:14]  * mtaylor hides
[00:14] <lifeless> perl bindings, kgo.
[00:16] <lifeless> timing and progress for subunit today I think
[00:16] <lifeless> bbiab
[00:59] <exarkun> What does it mean when a merge proposal has the status "Needs review" and a review with a status of "Disapprove"
[00:59] <exarkun> Are merge proposals active forever one created until the branch is merged?
[01:04] <wgrant> exarkun: There is the merge proposal status, and then there are votes.
[01:04] <wgrant> 'Disapprove' is a bote.
[01:04] <wgrant> *vote
[01:04] <jml> exarkun, whether they are active or not depends on the status of the merge proposal
[01:04]  * wgrant defers to jml.
[01:04] <lifeless> exarkun: there is a state for the proposal, and state for a reviewer
[01:04] <jml> exarkun, consider yourself told :)
[01:04] <exarkun> I don't see the ui for manipulating the state of the proposal
[01:05] <lifeless> exarkun: and lp doesn't currently do much workflow for you, so the states can be incoherent
[01:05] <exarkun> Oh, except that little trash can icon...
[01:05] <lifeless> exarkun: go to the proposal page; at the top there is a state for the proposal, and a pen icon
[01:05] <exarkun> existence vs non-existence is like state
[01:05] <jml> exarkun, there's a pen icon too
[01:05] <lifeless> if you have rights to set it, I think
[01:05] <exarkun> Oh.
[01:05] <exarkun> That's a pen?
[01:05] <lifeless> is pen, rearrange for humour
[01:06] <exarkun> I didn't really think about what it was before.  Upon examination, I see that I interpreted it as a drunken exclamation point.
[01:06] <jml> exarkun, I can see how that could happen
[01:06] <exarkun> Hooray, status rejected.
[01:06] <exarkun> Oh, except that was the wrong merge proposal, neat.
[01:06] <lifeless> \o/
[01:06] <exarkun> Hooray, status needs review.
[01:07] <exarkun> Cool, thank you.
[01:08] <jml> exarkun, bug 400925, just for you.
[01:08] <wgrant> I'm pretty sure that's a dupe.
[01:09] <exarkun> heh :)
[01:09] <exarkun> Wow now I'm seeing them all over the place
[01:09] <wgrant> Bug #287154
[01:10] <lifeless> I'd like the save-this-edit button to be readable to colourblind people
[01:10] <lifeless> as clicking on a grey circle to save things is disconcerting
[01:13] <pasteeater> does the bug report "box" allow HTML or BBcode?  i'm afraid to try because I see no "preview post" and I didn't see any mention of it anywhere.
[01:14] <wgrant> pasteeater: No, but it is monospace and does preserve leading spaces.
[01:16] <pasteeater> wgrant: ok.  thanks.
[01:18] <lifeless> jml: pretty bug pie -
[01:18] <lifeless> https://bugs.edge.launchpad.net/subunit
[01:19] <jml> haha
[01:34] <mtaylor> nice pie
[02:34] <omerm> I just tried packaging for the first time to fix this bug: https://bugs.launchpad.net/launchpad-foundations/+bug/90150 . I applied the patch found in the comments, and followed the packaging guides to create a debdiff. I don't have an AMD64 system, so I can't test it. Should I send it to one of you for approval? Post it in launchpad? Thanks in advance.
[02:35] <omerm> oh whoops, wrong bug
[02:35] <omerm> https://bugs.launchpad.net/ubuntu/+source/vnc4/+bug/119982
[03:12] <omerm> nevermind, got pointed in the right direction by the ubuntu-motu people
[03:34] <peplin> Hello - I'm trying to figure out how to propose a merge for a branch I created, but I can't seem to find the "propose for merging" link that the documentation talks about. Could someone direct me to where I should look?
[03:37] <exarkun> peplin: If you go to the page for the branch, then you should see a green circle with a plus in it below the branch details area, with a link next to it, "Propose for merging into another branch"
[03:38] <exarkun> peplin: Just below the "This branch is not proposed for merging into any others." line (unless it has already been proposed to be merged, then that line will be different)
[03:38] <exarkun> peplin: It's not there if you're not logged in.  I'm not sure if there's some other circumstance in which it is also missing.
[03:38] <peplin> would that not work if the branch was pushed to my +junk folder?
[03:39] <peplin> regardless, where should i be pushing branches that i intend to propose to be merged
[03:39] <peplin> +junk was the only place I could get to work without owning the project
[03:40] <exarkun> You should be able to push to lp:~username/project/branchname even if you don't own the project.
[03:41] <exarkun> I'm not sure if being in +junk precludes proposing for merging, but it sounds like that could be it.
[03:42] <spiv> Yes, I'd assume that you can only propose branches of that project.
[03:42] <peplin> ah yes, that was it
[03:42] <spiv> But anyone can create a branch of any project.
[03:43] <spiv> (Or reassign a +junk branch to a project)
[03:43] <peplin> i pushed again to using the project name and the "propose" link is there
[03:43] <peplin> spiv, could you elaborate on reassigning a branch?
[03:49] <wgrant> spiv, peplin: You can only reassign through the API, now.
[03:50] <wgrant> The UI got too complex with the addition of package branches, so it was dropped (at least for now).
[03:50] <spiv> wgrant: ah, right.
[03:50] <spiv> peplin: wgrant apparently knows more about this than I do :)
[03:51] <peplin> well, both of you combined are enough for me to know I should just fix the original branch
[03:51] <exarkun> Anonymous checkouts are timing out for me.  Local problem or launchpad problem?
[03:52] <jml> exarkun, via http or bzr+ssh?
[03:52] <exarkun> http
[03:52] <jml> exarkun, lifeless was reporting delays earlier over http
[03:52] <exarkun> actually, it happens from three different hosts on three different networks
[03:54] <jml> exarkun, yeah, so it looks like something is going on at our end.
[03:55] <wgrant> Hm, yes, it's not responding to HTTP at all.
[03:55] <wgrant> It just accepts and hangs.
[03:56] <jml> sadly, there aren't any people around with access to the relevant machines.
[03:57] <spiv> jml: possibly loggerhead woes?
[03:57] <exarkun> That's a bit of a bummer.  I can't adequately test my software until it's working.
[03:58] <exarkun> On the other hand, I'm going to bed.
[03:58] <jml> spiv, possible but unlikely.
[03:58] <wgrant> At least a LOSA should be awake...
[03:58] <jml> spiv, if bzr is asking loggerhead for packs, then we have a whole other problem.
[03:58] <spiv> (I wonder if bzr should include some sort of fancy Accept header or something in its requests for branch data that excludes text/html...?)
[03:59] <spiv> jml: well, I think bzr in some cases tries to GET the URL the user gave it in case it is a bundle before trying .bzr/...
[04:00] <spiv> jml: which AIUI would probably cause loggerhead to answer the request with a 'changes' page or something.
[04:00] <jml> spiv, that might be right. I'm not sure what to do about it.
[04:01] <spiv> jml: nothing, it's a Saturday ;)
[04:01] <jml> exarkun, g'night.
[04:02] <spiv> jml: btw, thanks for suggesting that coffee grinder.
[04:02] <jml> wgrant, just because a losa is awake doesn't mean they're around to fix minor outages :)
[04:02] <wgrant> jml: Right, but I wouldn't really classify this as minor.
[04:02] <jml> spiv, you got one?
[04:02] <spiv> Yeah.
[04:03] <spiv> Oh, that reminds me, I'm about due to give it a bit of a clean...
[04:03] <spiv> But first: coffee!
[04:03]  * spiv wanders away from IRC.
[04:03] <jml> wgrant, it's not classified as critical though.
[04:04] <lifeless> wgrant: it will get more urgent the longer it lasts :)
[04:04] <wgrant> jml: Really? Huh.
[04:05] <lifeless> jml: note however, that the process we have is likely underestimating somethings :) :- see the discussion I provoked a couple of weeks back and the follow on decrease in delays
[04:05] <jml> lifeless, that only proves that we used to underestimate some things :)
[04:07] <wgrant> It is a bit concerning that breaking easy access to code isn't critical.
[04:08] <lifeless> wgrant: technically its not broken...just slow, isn't it?
[04:08] <wgrant> lifeless: Oh, you're right. I finally got a response.
[04:09] <wgrant> But my checkout from 15 minutes ago has so far downloaded 10KB
[04:09] <wgrant> == broken
[04:09] <jml> wgrant, yes it is.
[04:10] <lifeless> wgrant: 10kb in 15 minutes I'd call broken too
[04:11] <lifeless> jml: if you want to escalate, I'm +1 on that
[07:43] <jml> yay topic fix
[12:17] <exarkun> Is http access to bzr broken again?
[12:20] <tsimpson> seems to works here
[12:21] <tsimpson> at least with bzr ls
[12:22] <LarstiQ> exarkun: wfm
[12:26] <mantiena-baltix> hello all
[12:26] <mantiena-baltix> cprov: are you online ?
[12:29] <jml> exarkun, we addressed the underlying problem, so it ought not be broken.
[12:34] <mantiena-baltix> SteveA: labas
[12:35] <exarkun> ah, I was not patient enough.  Thanks.
[12:37] <mantiena-baltix> jml: hi, I wanna ask launchpad developers about one important bug - #211223
[12:38] <cprov> mantiena-baltix: yes
[12:38] <mantiena-baltix> cprov: could you help me? it seems bug 211223 still isn't fixed, but launchpad.net/guadalinex distro has registered series and milestones!
[12:39] <mantiena-baltix> I'm main developer of Baltic Ubuntu derivative - launchpad.net/baltix and I'm waiting for ability to register series and milestones for about 3 years :(
[12:41] <cprov> mantiena-baltix: I know ... but I think the guadalinex series were created by lp-admins, let me check.
[12:42] <mantiena-baltix> So, maybe cprov or other launchpad developers know how I can register new Baltix distro series and milestones, because this ability already exists in launchpad.net
[12:42] <cprov> mantiena-baltix: there is only one and it was created by LOSAs -> https://edge.launchpad.net/guadalinex/buho
[12:44] <mantiena-baltix> cprov: We will be happy if someone creates 2 series for Baltix - one for 8.04 LTS (baltix-2008) and one for 9.04 (baltix-2009) ;)
[12:45] <cprov> mantiena-baltix: right, please file a question on launchpad product and the I will find someone to create it for you.
[12:45] <mantiena-baltix> cprov: thank you very much
[12:46] <cprov> mantiena-baltix: np
[12:47] <mantiena-baltix> cprov: should I ask a question at https://answers.edge.launchpad.net/launchpad or at https://answers.edge.launchpad.net/launchpad-project ?
[12:48] <cprov> mantiena-baltix: launchpad-registry, if you want to be precise, but launchpad will do it.
[12:49] <maxb> I have a bunch of teams and PPAs that date from before multiple PPA support, and are now redundant. What should I ask to be done with them?
[12:50] <cprov> maxb: they already have multiple keys, right ?
[12:53] <cprov> maxb: if this is the case we can't easily reassign the PPAs, think you can create new ppas on the account you want to preserve and merge the other accounts in.
[12:54] <cprov> maxb: the ppas in the merged accounts will be purged automatically at some point.
[12:55] <maxb> oh, no this is a separate issue
[12:55] <maxb> teams, not user-accounts
[12:57] <maxb> I already have created new PPAs on the one team that replaces all the others, and have deleted all the packages from the old PPAs, and edited the descriptions of the teams and PPAs to include "[OBSOLETE]", I'm just wondering if there's a way to tidy up a bit more thoroughly
[12:58] <maxb> Is there a concept of deleting teams? deactivating them?
[13:01]  * maxb wonders if launchpad should really be offering the only administrator of a team the option to deactivate their own membership
[13:01] <wgrant> maxb: Note that the team owner has superpowers.
[13:02] <wgrant> You can get a team deactivated, but I'm not sure what that will do to PPAs.
[13:04]  * maxb files a question and will wait and see what happens
[13:50] <Mez> hmm.. is it me, or is edge (dunno about normal) taking it's time to save info?
[13:54] <Daviey> Mez: i was thinking that
[13:57] <Mez> :D
[15:46] <mdz_> is there any way to find out who the bug contact for a package is, through the API?
[17:14] <didrocks> hey guys, is it possible to push to staging ppa (for dry-run app testing) with something similar than ppa:user/ppa ?
[17:24] <fta> didrocks, did you try ppa:user/staging?
[17:25] <fta> or whatever name you used for your staging ppa
[17:26] <fta> wrt the lp api, is it possible to use getBuildRecords() to filter both on build_status and current_source_publication?
[17:36] <didrocks> fta: didn't try. But I just wanted to hurt the staging soyuz (if it exists) for my tests and it to be erased everyday. As staging launchpad
[17:38] <fta> didrocks, oh, then i don't know. i have a ppa called staging (and another called sandbox), but it's a regular ppa
[17:38] <didrocks> fta: I just don't want to upload something which will not be removed on a production server in a testsuite :)
[17:38] <geser> fta: something like bug 369112?
[17:38] <didrocks> ok, let's try to upload something there and see if tomorrow I still have something there
[17:39] <fta> geser, yes
[17:40] <fta> geser, because so far, my requests are far too slow
[17:42] <geser> fta: currently I don't know any better way as to filter on build_state and check every build record if it has a current_source_publication or not
[17:44] <MT-> kiko: WOL
[17:44] <MT-> now... how long does it take kiko to wake on lan??
[17:45] <fta> geser, can i do negative filtering? ie, build_status is not 'Successfully built'?
[17:46] <fta> geser, or multiple filter, build_status is Needs building or Dependency wait or Currently building?
[17:48] <geser> fta: I call the function for every build_state once and put the records I need into a dict
[17:48] <geser> fta: what are you up to?
[17:49] <fta> -daily maintenance
[17:50] <geser> fta: perhaps you can borrow some code from the FTBFS script (http://bazaar.launchpad.net/~geser/%2Bjunk/qa-ftbfs/annotate/head%3A/source/build_status.py)
[17:51] <fta> looking..
[17:54] <maxb> didrocks: I don't think LP staging *has* a soyuz backend
[17:56] <didrocks> maxb: ok. I will create my staging ppa, so
[18:23] <aquarius> I'd like to write a Greasemonkey script for LP which, among other things, can get a list of branches for a given project. How am I best to do this? I could have it call the LP API, but then you have to give the script an OAuth token, which seems rather daft for JavaScript running *on* Launchpad pages.
[18:24] <aquarius> i could have it, er, screenscrape, which is a no-no.
[18:24] <aquarius> Are there other alternatives?
[18:27] <aquarius> presumably there must be a LP JS API, but it's not actually documented anywhere, unless I'm missing something.
[18:29] <james_w> you can find it by fishing about in the HTML source I believe
[18:31] <aquarius> james_w: yeah. I was, er, hoping for something a bit nicer than spending my Saturday reverse-engineering Launchpad ;-)
[18:32] <aquarius> especially since I don't think that the JS API is as comprehensive as the actual API, i.e., if no-one's needed a "query branches and get back JSON" thing for a Launchpad AJAX thingy, it won't exist
[18:32] <aquarius> LP's HTML/JS front end doesn't actually call the proper launchpadlib API, does it?
[18:32] <aquarius> (er, from JS, I mean; it does on the back end in Python, natch)
[18:33] <aquarius> I was hoping that it does, and there's therefore some sort of useful shortcut :)
[18:47] <james_w> aquarius: I heard that it did use the lplib API, or something approaching it, and had a way to query stuff without an OAuth token
[18:47] <james_w> I may be lying of course
[18:47] <aquarius> james_w: that is of course ultra-useful information, assuming you haven't made it up ;-)
[18:47] <james_w> and if you want to query private teams then I don't know how it would work
[18:47]  * james_w -> out for the evening
[18:47] <james_w> good luck :-)
[18:48]  * aquarius waves at any LP developers who happen to be wandering past to see if it's true or not :)
[18:52] <aquarius> why do I always come up with these ideas at the weekend when there aren't LP hackers around who know stuff? gah. :)
[18:52] <intellectronica> aquarius: obviously js doesn't use launchpadlin, which is written in python, but it uses exactly the same webservice api
[18:53] <aquarius> intellectronica: aha! where does it get an oauth token from?
[18:53] <aquarius> intellectronica: and well played for being around at the weekend. You shall have a cookie for dutifulness :)
[18:53] <intellectronica> in fact i know people have been using it to write greasemonkey extensions
[18:54] <aquarius> intellectronica: if you can point me at a particular GM extension which uses it thus that would be well useful. I've been poking around and can't find one.
[18:54] <intellectronica> aquarius: it doesn't use oauth. no need to, since you're already authenticated
[18:54] <jpds> aquarius: https://launchpad.net/launchpad-gm-scripts
[18:54] <intellectronica> aquarius: bdmurray would be a good person to ask. i think he has played with that. i don't remember right now where they keep the scripts
[18:54] <intellectronica> aha :)
[18:55] <aquarius> yeah, I looked at a few of those, it's just slow to eyeball them all ;-)
[18:56] <aquarius> lp_karma_suffix.user.js screenscrapes, which is a bit norty of it.
[18:56] <intellectronica> we should really have a prefab kit for writing gm scripts / xul extensions like that. so many nice things can be done once you get going
[18:57] <intellectronica> aquarius: well, not evertything is available via the api. also, there are many legacy scripts that were written before the api was available
[18:58] <aquarius> intellectronica: I couldn't agree more
[19:00] <aquarius> I cannot find an example script in launchpad-gm-scripts which calls the API
[19:00] <intellectronica> aquarius: b.t.w you have access to the lp sourcecode, don't you? you can see how it's used. being async, the library works a bit different from launchpadlib, but the api is the really the same
[19:00]  * intellectronica can't believe it's days before _everyone_ can have access to the sourcecode :D :D :D
[19:01] <aquarius> intellectronica: the API, if I'm calling it randomly from Python on my desktop, exists on http://api.launchpad.net/beta/ and requires OAuth signed headers. Is there a place where it can be called which requires cookie auth?
[19:02] <aquarius> since that's what I'd need to call it from greasemonkey
[19:02] <intellectronica> aquarius: we have a hack in place to allow you to call it from the launchpad front-end
[19:02] <aquarius> intellectronica: really? what hack? how do I get to use said hack? :)
[19:02] <intellectronica> aquarius: for writing a gm script, what you'll want is to login into launchpad, and start from there
[19:02] <aquarius> ahem!
[19:03] <aquarius> the api exists at https://bugs.edge.launchpad.net/api/beta/ as well!
[19:03] <aquarius> that, I take it, is the hack ;)
[19:03] <intellectronica> yup
[19:03] <MT-> I think a user signed up just to spam.. https://edge.launchpad.net/~cheezytwiggy  -  on bug 77289
[19:03] <aquarius> am I supposed to pretend that I don't know that? Or is it a hack which will live on?
[19:04] <intellectronica> aquarius: it's no secret, obviously
[19:04] <intellectronica> the only reason to use it, is if you try to avoid cross-domain scripting, of course
[19:04] <aquarius> intellectronica: agreed, I just don't want to rely on something that you guys might pull the plug on tomorrow :)
[19:05] <intellectronica> if you don't want to log in to launchpad, you'll have to sort out the authentication. the rest of the library is written, and i'm sure it can extracted into something like launchpadlib
[19:07] <aquarius> nope, being logged into LP is fine
[19:09] <intellectronica> aquarius: i've got a palindromic paste for you, in case you don't have access to the tree: https://pastebin.canonical.com/20002/ look for 'lp_client'
[19:10] <aquarius> intellectronica: ah, nope, I'm flying now, no worries
[19:10] <intellectronica> i'm not sure how to interpret this
[19:22] <aquarius> intellectronica: it's all working, I completely uderstand what I'm doing, and thanks for your help, is what it means :)
[20:00] <aquarius> intellectronica: ping?
[20:01] <dD0T> Hi. I want to switch a project which previously used svn import to git import. How can I remove the old import or edit it? Thanks
[20:10] <savvas> dD0T: without being sure, I think I remember someone said you can't switch the import on the fly
[20:10] <savvas> better wait for someone else though :)
[20:24] <SiDi> Hey, i spotted an user on LP sending links to websites apparently selling contraband pills :|
[20:24] <SiDi> https://bugs.launchpad.net/ubuntu/+source/ubuntu-calendar/+bug/77289
[20:29] <fta> from a getPublishedSources() object, is it possible to go down to the details of the tarball? (date, size, url)?
[20:29] <fta> -the+a
[20:32] <fta> i can get the .changes url, am i supposed to fetch & parse it myself or could the api do that for me?
[20:48] <parolang> Hello.  You need a launchpad login just to download from a repository?
[20:53] <SiDi> no
[21:02] <geser> fta: spph.changes_file_url
[21:02] <geser> sorry, should read more carefully
[21:03] <geser> what changes do you need that aren't in the API? the files itself=
[21:04] <fta> the same details as we see in lp for all src packages: link, filename and size of {dsc, changes, diff.gz, tar.gz}
[21:06] <fta> i can just get the .changes url. it sure contains what i need, but i expected the api to give everything to me
[21:07] <geser> bug 394827 sounds right for it
[21:09] <fta> ok, thanks
[21:09] <fta> i think the api is not ready for me.
[21:10] <fta> after 2h+ and 100 lines of python, i'm still unable to replace a 6 lines shell script written in 2 minutes
[21:10] <fta> ok, i'm a noob in python, but still :P
[21:12] <geser> the LP API still misses some data which is easily available through the web browser
[21:32] <rockstar> fta, what specifically are you wanting to do?
[21:37] <fta> rockstar, it was an experiment to see if it was time for me to start moving all my perl & shell scripts to python, it failed miserably.
[21:42] <rockstar> fta, did you file bugs about missing api bits?
[21:52] <fta> rockstar, geser pointed me to 2 existing bugs
[21:53] <rockstar> fta, great.
[22:36] <BUGabundo> AHHHHHHHHHHHHHhh wth is new ajax on LP edge for Status!?!
[22:36] <BUGabundo> now I can't add a reply with the change :(
[22:38] <kolby> how do I force bzr to push changes?
[22:40] <Ampelbein> kolby: are you looking for --overwrite?
[22:41] <kolby> I'll try that.  Thank you!
[22:42] <Ampelbein> kolby: if its the first time you push to lp and have an empty branch already created there, use --use-existing-dirs
[22:43] <Ampelbein> -dir without "s" ;-)
[22:43] <kolby> I used both.
[22:43] <kolby> ^^  hope it works
[23:37] <RenatoSilva> how about lp supporting hg?