[00:00] <wgrant> Has it appeared in your PPA yet?
[00:02] <smo__> no
[00:02] <smo__> i just finished sending my 4 signed deb...
[00:03] <smo__> waiting now
[00:04] <wgrant> You uploaded .debs as well?
[00:04] <wgrant> You'll get an email of rejection in a moment, then.
[00:06] <smo__> ah
[00:06] <smo__> received somemails
[00:06] <smo__> i check
[00:06] <smo__> Rejected:
[00:06] <smo__> Unable to find distroseries: stable
[00:06] <smo__> ok
[00:06] <wgrant> That's not what I was expecting.
[00:06] <smo__> deb not well done "normal"
[00:06] <wgrant> Pardon?
[00:07] <smo__> this was my first debian packages
[00:07] <wgrant> The error about the distroseries is because 'stable' is a Debian series. You need to upload to an Ubuntu one - 'jaunty', for example.
[00:07] <wgrant> And how did you build the package? debuild?
[00:08] <smo__> hum, dpkg-buildpackage -rfakeroot -D -us -uc
[00:08] <smo__> ...
[00:08] <wgrant> Ah.
[00:08] <wgrant> '-us -uc' means "don't sign this!"
[00:08] <wgrant> so omit those.
[00:09] <wgrant> Omit -D too; it's a default option.
[00:09] <smo__> best is debuild?
[00:09] <wgrant> It doesn't really matter.
[00:09] <wgrant> I build with 'debuild -S -sa'
[00:09] <smo__> ok
[00:09] <wgrant> That does a source-only build, and signs it.
[00:09] <smo__> ok i try
[00:10] <wgrant> You might need -kKEYID as well, if your key doesn't match your package.
[00:10] <smo__> ok
[00:57] <smo__> wgrant, a small question, when i use debuild -S -sa i dont have .deb
[00:57] <smo__> normal?
[00:58] <wgrant> smo__: Correct. You can only upload source packages to Launchpad - it will build the .debs there.
[00:59] <smo__> ah nice
[00:59] <smo__> i try
[01:03] <smo__> Successfully uploaded packages.
[01:04] <smo__> i think this one we ll be good ^^
[01:05] <wgrant> Got the email yet?
[01:05] <smo__> no, waiting
[01:05] <smo__> [PPA s-lagui] [ubuntu/jaunty] gstyle 0.1 (Accepted)
[01:05] <smo__> ^^
[01:05] <smo__> thx a lot wgrant
[01:06] <wgrant> Excellent!
[01:06] <wgrant> Now to see if it builds.
[01:06] <smo__> i have to upload and redo sources for missing perl libraries missing in ubuntu...
[01:07] <smo__> (sorry for my english..)
[01:07] <wgrant> What do you mean?
[01:07] <wgrant> What's missing?
[01:07] <wgrant> Aha, it built.
[01:07] <smo__> some perl libraries i use are not in ubuntu repository
[01:07] <wgrant> Ah.
[03:36] <jml> what's the deal with all the bug mail?
[03:37] <SamB> jml: I only sent one today...
[03:38] <wgrant> jml: I got annoyed with some of my bugs waiting in 'launchpad' for 6 months. So I redirected mine and a few others, as CHR people don't seem to be doing it for months at a time.
[03:38] <wgrant> Nobody triages those bugs :(
[03:38] <SamB> ... and that was a bzr bug
[03:39] <SamB> wgrant: you need more medical personel, maybe?
[03:39] <jml> wgrant, it's not listed as a CHR task -- which goes some way to explaining why CHR people haven't been doing it :)
[03:41] <wgrant> I'm sure somebody told me it was.
[03:41] <jml> wgrant, anyway, thanks for JFDIing it. I'll mail the dev list to see what's going on & avoid a backlog in future.
[03:42] <wgrant> jml: It takes about 10 seconds a bug, and it seems similar to processing Answers, so IMO it would fit well with CHR.
[03:44] <jml> agreed.
[03:45] <SamB> what's CHR?
[03:45] <SamB> some kind of synonym for RTS?
[03:45] <SamB> or do I mean RSD?
[03:46] <lifeless> community help roster
[03:46] <wgrant> Rotation, actually.
[03:46] <wgrant> https://help.launchpad.net/HelpRotation
[04:56] <mwhudson> nearly at 400k bugs ...
[04:56] <wgrant> Yep.
[04:57] <wgrant> Late tomorrowish, I suspect.
[05:22] <ajmitch> wgrant: mass bug-filing time?
[05:25] <wgrant> ajmitch: I've already filed ten bugs today :(
[05:28] <MT-> nice..
[05:28] <MT-> how many bugs are in launchpad total?
[05:29] <ajmitch> bug numbers are approaching 400k
[05:30] <ajmitch> which includes the hundreds of duplicates people file, etc
[05:30] <wgrant> And only a few of the intermediate numbers are missing.
[05:31] <wgrant> Bug #1000 was hit 4 years ago yesterday.
[05:32] <wgrant> Oh.
[05:32] <wgrant> I'm a month off.
[05:32]  * wgrant can't count, see.
[05:33] <ajmitch> what worries me is that I remember mpt filing that bug at the time
[05:39]  * wgrant is glad to see all of the bugs he retargeted into sinzui's domain this morning already triaged.
[05:51] <micahg> hmmm
[05:52] <micahg> almost 400k
[07:18] <micahg> are the karma recalculators not working?
[07:20] <spm> wgrant: you'll be pleased to note we have successfully updated edge :-)
[07:22] <wgrant> spm: Excellent. Thanks.
[07:23] <wgrant> Inline bugtask editing... shiny.
[07:26] <wgrant> And my scripts no longer crash. This is good.
[07:30] <spm> micahg: they should be working. the latest runs should be finishing any minute now.
[07:31] <micahg> ok
[07:31] <micahg> thanks
[07:31] <micahg> I noticed the new update, status is AJAXy
[07:37] <wgrant> Why do the status/importance popups not have the usual header styles?
[07:37] <wgrant> They look pretty strange now with the unbold title and large gap at the top.
[07:41] <poolie> !!!! inline status omg
[07:41] <wgrant> poolie: Yes! It is awesome!
[07:41] <poolie> sadly it comes to an abrupt end when you want to target a milestone but anyhow
[07:41] <wgrant> Although it'll stuff up the inline activity log :(
[07:42] <poolie> it won't update til you reload?
[07:42] <poolie> oh and it doesn't seem to group them
[07:42] <wgrant> Right.
[07:42] <wgrant> On both.
[07:42] <wgrant> But the latter is the main problem.
[07:42] <wgrant> It seems to only group changes and comments that happened at the same instant.
[07:42] <poolie> i think there's a separate bug saying it should group as much as possible?
[07:43] <wgrant> There is.
[07:43] <poolie> ooh, fail
[07:43] <wgrant> Ideally using the same algorithm as email batching, I suppose...
[07:43] <poolie> if you then unfold the thingy to set the milestone it undoes your status changes
[07:43] <poolie> so sad
[07:44] <wgrant> Ah, the good old midair collision bug. But colliding with yourself.
[07:44] <wgrant> Although I suppose this one is fairly easily fixable by just setting that widget as well.
[07:44] <wgrant> And I guess the expandy bit might go away entirely once everything's inline-editable.
[07:45] <spiv> The email batching is pretty unsophisticated, really :)
[07:45] <spiv> Off the top of my head I would think the web presentation of updates could do better than that.
[07:45] <wgrant> True.
[07:49] <maco> thank you launchpad people for adding the ajaxy bug status stuff!
[08:22] <mvo> the new bug status change is really nice, well done!
[08:22] <maco> ooooh im gonna start counting to see how many come in to say that.
[08:25] <wgrant> maco: You missed three from before you arrived.
[08:25] <maco> wow, popular feature
[08:25] <maco> so 5 so far then
[09:00] <yurikoles> how report a bug in ppa package?
[09:01] <maco> yurikoles, can't
[09:37] <mpt> bug 179873
[09:39] <svqyqb> http://tinyurl.com/nkypfa
[10:05] <bigjools> james_w, wgrant: is getPublishedSources() behaving today?
[10:06] <james_w> bigjools: it appears to be, yes
[10:06] <james_w> thank you
[10:08] <bigjools> cool
[10:09] <bigjools> james_w: sorry it took so long to push out, edge updates were stopped for a security fix cherry pick.  I'm going to see if we can get a more stable api service put on lp.net as well.
[10:09] <wgrant> bigjools: Working fine for me too.
[10:09] <bigjools> good good
[10:10] <wgrant> Is it actually defined when the API will be un-beta?
[10:10] <bigjools> no
[10:10] <wgrant> Presumably it will never actually be stable.
[10:10] <bigjools> there's a not insignificant amount of work to do to support multiple API versions
[10:10] <bigjools> but we can put an api service on lp.net for now which will prevent you from having to QA our changes for us :)
[10:11] <wgrant> By 'when' I don't mean time, but other criteria.
[10:11] <bigjools> not my area, but I think it's known what needs to be done, it's just not scheduled
[10:11] <james_w> bigjools: I think I would still run against edge, but having lp.net as a fallback would be useful
[10:12] <james_w> if we just run against production then we have no recourse when the changes land there
[10:12] <bigjools> james_w: there's a slight risk that you'll start using edge features before they are ready on lp.net
[10:12] <james_w> so, I support your proposal, but I for one will continue to QA :-)
[10:13] <bigjools> james_w: awesome, can you fill out my QA wiki as well :D
[10:13] <james_w> presumably there will be a lp.net/+apidoc to go along with edge.lp.net/+apidoc
[10:13] <james_w> heh
[10:13] <bigjools> yep
[10:20] <geser> how hard would it be to add to +apidoc a TOC with a list of all objects? I've to use the Firefox page search for navigation
[10:20] <wgrant> I generally edit the URL, so +1
[10:21] <intellectronica> geser: not hard. if you are not afraid of XSLT you'll even be able to do it yourself soon ;)
[10:22] <bigjools> thanks for filling my inbox with bugmail today wgrant :)
[10:22] <wgrant> intellectronica: We can do it ourselves now, it seems. It's in lp:launchpadlib
[10:23] <intellectronica> oh of course, forgot that it's there
[10:23] <wgrant> It seems like a very strange place to have it.
[10:23] <wgrant> bigjools: Sorry.
[10:24] <bigjools> wgrant: heh, it's fine, I appreciate you taking the time
[10:24] <intellectronica> i actually rather it didn't change. i also use firefox search-as-you-type, and if every item will be mentioned twice on the page i'll have more to type before i find the real definition
[10:24] <intellectronica> wgrant: and yes, thanks for the help with moving all those bugs. reading through them now
[10:24] <wgrant> intellectronica: But they're already mentioned many times, as methods say something like "Returns a collection of _foo_bars_"
[10:25] <intellectronica> sure, that will be just one more time. anyway, i don't really care that much, and i appreciate many users do prefer more traditional navigation :)
[10:42] <wgrant> Remind me never to go near XSLT again.
[10:42] <geser> wgrant: that bad?
[10:42] <intellectronica> wgrant: it is one of the more perverse languages out there, indeed
[10:43] <wgrant> It was easy enough to work out, but it's horrendously ugly and verbose.
[10:44] <intellectronica> wgrant: bear in mind, the resulting code is probably less verbose than it would be had you wanted to do the same xml->xml transformation using other languages
[10:44] <wgrant> And I'm not sure I like how the for-each looks at its first children for stuff to us in its calculations.
[10:44] <wgrant> True.
[12:19]  * mpt goes on a bug-reporting bender
[12:29] <wgrant> mpt: Have you been double-clicking on statuses of some of my bugs?
[12:30] <mpt> wgrant, only two, I'm sorry, the change wasn't on staging yet
[12:30] <mpt> reported bug 399236 about it
[12:30] <wgrant> ... and there's the email about that bug.
[12:31] <wgrant> Hopefully staging will update tonight.
[12:34] <mpt> ah, bother, did it by accident that time :-/
[12:35] <wgrant> Heh.
[12:53] <wgrant> cprov: Is depwait retrying really in launchpad-buildd territory?
[12:55] <cprov> wgrant: as I understood the bug, it would be easier if the buildd identifies the situation itself and return the appropriate depwait value.
[12:55] <wgrant> cprov: That wouldn't work. If the buildd knew what the solution was, there would be no problem.
[12:55] <cprov> wgrant: maybe it has implications in both sides, I'm not entirely sure.
[12:56] <wgrant> cprov: Nothing provides the virtual package at that point, so the buildd can't know what will in future.
[12:56] <wgrant> The build log there is from a successful build.
[12:59] <cprov> wgrant: I was under the impression that 'apt' would know better than soyuz how to deal with virtual packages.
[13:00] <wgrant> cprov: Well, of course. Soyuz doesn't know how to do it at all.
[13:00] <wgrant> But the problem here is that apt doesn't know it's a virtual package at the time of the build.
[13:00] <wgrant> It just knows the package doesn't exist.
[13:00] <cprov> wgrant: no, the model doesn't know about 'provided' packages
[13:01] <wgrant> Right. That's the problem.
[13:01] <wgrant> It cannot be solved in launchpad-buildd.
[13:09] <wgrant> cprov: How does the code work now? The cron job just checks each depwait build's missing name and version against published binaries?
[13:10] <cprov> wgrant: yes, it should also check for virtual packages
[13:10] <wgrant> cprov: And I think you do have that information at least somewhere, because BPR pages show lots of fields.
[13:12] <cprov> wgrant: right, BPR.provides would list virtual packages
[13:12] <cprov> wgrant: published BPR providing X ... probably
[13:13] <wgrant> cprov: Right, it needn't be any more complicated than that, because it's for apt to work out.
[13:13] <wgrant> And there's no versioning to worry about with Provides.
[13:14] <cprov> wgrant: yes, if a matching 'provided' package was found we re-dispatch the build and let apt work again, if there is any version-issue it will come back.
[13:16] <wgrant> cprov: No, better than that. If the dependency is versioned, a virtual package cannot satisfy it.
[13:18] <cprov> wgrant: good point, because Provides: elments are not versioned (or are they? let me check debian policy)
[13:18] <wgrant> cprov: Debian Policy 7.5
[13:18] <wgrant> cprov: Provides are unversioned.
[13:19] <cprov> http://www.debian.org/doc/debian-policy/ch-relationships.html#s-virtual
[13:19] <cprov> righto, thank you.
[17:06] <apw> if i am not going mad, i think ppa uploads have stopped being processed
[17:06] <apw> or at least taking over an hour
[17:10] <bigjools> apw: I will check
[17:10] <apw> bigjools, thanks
[17:11] <bigjools> apw: it looks ok, did your upload take a long time to get processed?
[17:11] <apw> i still don't have either a yes or a no in email from it
[17:12] <bigjools> apw: did you sign it with a key known to Launchpad?
[17:12] <apw> in theory yes
[17:12] <apw> but perhaps i am stupid will go check
[17:12] <bigjools> you might want to double check :)
[17:15] <apw> bigjools, somehow i managed to produce it unsigned at all, which i don't even know how to do
[17:15] <bigjools> oops
[17:16] <apw> that goes in history as a mad thing, same tools same keypresses, no signature
[17:16] <apw> hmmm no idea what i did there different from normal, thanks for humoring me
[17:17] <bigjools> np, we're all human
[17:20] <apw> some of us more so than others :(
[17:22] <MT-> What's this? https://code.edge.launchpad.net/~vcs-imports/ubuntu-drupal-planet/trunk-6x
[17:55] <maxb> bigjools: In reference to the build-score increasing with time, and your most recent comment mentioning the source urgency header - the fact that time-on-queue rescoring is now active means the utility of the source urgency header is decreased, since urgency=emergency is approximately the same score boost as 1 hours's on-queue time. What do you think about the possibility of reopening the bug, noting that time-on-queue rescoring only accomplish
[17:55] <maxb> es making the urgency header less useful, and proposing to fix it by turning rescoring off again and amending the BuildScores wiki page appropriately?
[17:56] <bigjools> maxb: good point
[17:57] <bigjools> it depends on exactly what we want to achieve, does urgency always trump time in the queue, the opposite, or a mix?
[17:59] <maxb> I don't think many people know about / use urgency
[18:02] <maxb> About the only usecase being relevant that I can think of would be using urgency=medium for a build that a real person was interactively waiting to finish, to bypass daily-build queues
[18:28] <RenatoSilva> bug 399398, bug 399392
[18:53] <SamB> hmm ... I'm getting an FOAF parsing error:
[18:53] <SamB> http://sparql.captsolo.net/browser/browser.py?url=https%3A//launchpad.net/%257Elaunchpad-beta-testers/+rdf
[19:25] <salsaman> hey, what happened to the "last translation time" in launchpad ?
[19:26] <salsaman> is somebody working on a fix ?
[19:27] <kiko> danilos, henninge?
[19:28] <MT-> kiko: you're my hero
[19:28] <kiko> MT-!
[19:29] <henninge> salsaman: I am not sure what you are referring to. Can you please give me an URL?
[19:30] <MT-> So.. when Launchpad is open sourced, will it be available on launchpad itself? that'd be pretty cool
[19:34] <henninge> MT-: It'll be on github, actually.
[19:34] <MT-> where's that?
[19:34] <henninge> ;-)
[19:34] <henninge> MT-: I am kidding.
[19:34] <henninge> :-P
[19:34] <MT-> oh
[19:35] <MT-> and now I know what you're talking about
[19:35] <henninge> MT-: https://edge.launchpad.net/launchpad-project/
[19:35] <henninge> is that page private? I don't think so ...
[19:36] <MT-> no
[19:36] <MT-> that's a whole lot of karma
[19:36] <henninge> where, who, me?
[19:36] <henninge> oh, you mean danilos!
[19:37] <MT-> all over
[19:37] <henninge> comes with the job.
[19:37] <MT-> A lot handed out in specs, but not overall
[19:46] <henninge> I'm outta here ...
[19:55] <thekorn> congrats everybody for the ajax way of changing task status and importance
[19:56] <micahg> is someone still keeping count on the complements? :)
[20:00] <CaptTofu> hi all!
[20:01] <CaptTofu> anyone here handle the launchpad website? I have a problem with 'create milestone' - clicking on it, nothing happens
[20:01] <joshuahoover> anyone know if it's possible to find bugs assigned to "Nobody" using launchpadlib?
[20:09] <salsaman> henninge: https://translations.launchpad.net/lives
[20:09] <salsaman> for example
[20:09] <salsaman> or anybody
[20:09] <salsaman> what happened to the "last translation time" in launchpad ?
[20:16] <Ampelbein> joshuahoover: searchTasks(assignee=None)? (untested)
[20:17] <salsaman> am i looking at the wrong url ?
[20:23] <joshuahoover> Ampelbein: that'll do it!
[20:23] <joshuahoover> Ampelbein: thank you!
[20:26] <salsaman> hello
[20:26] <salsaman> last translation time seems to be missing in translations
[20:34]  * jblount hugs inline triage prettiness
[20:47] <salsaman> is nobody going to even take a look ?
[20:49] <salsaman> thats real sloppy
[20:49] <salsaman> thanks for nothing
[20:49] <qball> file a bug on the propper launchpad page. you cannot expect the right person be on 24/7 call in the channel
[21:06] <simon-o> Hi, when I set a Remote Bug Watch to an Email address, are there any automated emails going to be sent to that address?
[21:07] <simon-o> because it says: The information about this bug in Launchpad is automatically pulled daily from the remote bug. This information hasn't been pulled yet.
[23:52] <dsprenkels> hello?
[23:58] <poolie> wgrant: http://launchpadlibrarian.net/28872946/run-awaaaaay.png <---- the grease-stain of quality? :)