[02:17] <CarlFK> https://bugs.edge.launchpad.net/~carlfk  "Summary ... 302548 302548 302548
[02:17] <CarlFK> should that be listed 3 times?
[02:28] <CarlFK> huh - https://bugs.edge.launchpad.net/mandriva/+bug/254668?comments=all bottom says "Report another bug             about Mandriva"
[02:28] <CarlFK> mandriva?
[02:34] <micahg> CarlFK: madriva's another linux distro
[02:39] <CarlFK> yeah, kinda surprised me, but I gues
[14:13] <fale> hi
[14:13] <fale> I have a problem with dput and soyuz
[14:14] <fale> I'm trying to upload a big (39Mb) package on launchpad. It stops the upload 1kb before the end and hangs. How can I solve?
[14:18] <thekorn> fale: there is already a bugreport about it, let me find it
[14:18] <fale> thekorn: thankyou :)
[14:18] <thekorn> fale: bug 251685
[14:19] <fale> thekorn: than... no solution?
[14:20] <thekorn> fale: looks like no solution, yet
[14:21] <fale> thekorn: :(
[14:22] <bigjools> fale: why is it so big?  is it an orig file that you could re-use from somewhere else perhaps?
[14:24] <fale> bigjools: is the libboost package... and I have found no way to split it
[14:24] <bigjools> fale: what files are you uploading exactly?
[14:26] <fale> bigjools: libboost_1.40
[14:26] <fale> or you means extentions?
[14:26] <bigjools> presumably you have a .dsc, a .orig.tar.gz and a diff?
[14:27] <fale> I have: .dsc, .build, .changes and .tar.gz
[14:27] <bigjools> I can check to see if the orig is already in the librarian so you could re-use it
[14:27] <bigjools> what is the full file name of your tar.gz?
[14:28] <fale> bigjools: boost1.40_1.40.0-2ubuntu0.tar.gz
[14:28] <fale> bigjools: there shouldn't be, because I imported it from debian because in ubuntu we are still at libboost 1.38
[14:29] <bigjools> ok no orig just a native tar
[14:30] <bigjools> something that has worked for others is to upload from a different IP address
[14:40] <Remaille> hi all
[14:40] <Remaille> I have a folder containing source code and i would like to "push" it onto launchpad
[14:40] <Remaille> for the very firt time
[14:41] <beuno> Remaille, so you haven't versioned it with bzr yet?
[14:41] <Remaille> no
[14:41] <maxb> What is so special about the last 1K? Is poppy trying to something specially-clever?
[14:41] <beuno> Remaille, https://help.launchpad.net/BzrHowto
[14:41] <fale> bigjools: thankyou... I'll try it ;)
[14:41] <bigjools> fale: np
[14:41] <Remaille> beuno, thanks, the page i missed :)
[14:42] <bigjools> maxb: last I heard it was a problem in the zope ftp server code
[14:42] <bigjools> cprov started a re-write in twisted
[14:42] <bigjools> gonna have to get him to do some weekend hacking :)
[14:44] <beuno> Remaille, let me know if you run into any problems
[14:44] <Remaille> thanks a lot
[14:48] <Remaille> I have this lp:~linuxedu/+junk/sadms-fr
[14:48] <Remaille> +junk stands for whhat ?
[14:48] <Remaille> am i uploading junk ?! ;-)
[14:49] <beuno> Remaille, you use +junk when you don't have a projeect
[15:48] <Remaille> thanks beuno , but it is a "proper"software
[15:48] <Remaille> working ok
[15:48] <beuno> Remaille, create a projec then
[15:48] <Remaille> but, is it important to vae it in junk or not ?
[15:48] <beuno> and push to it with:  lp:~user/projectname/branchname
[15:49] <Fly-Man-> Morning
[15:49] <Fly-Man-> is there something wrong with the VCS-imports ?
[15:49]  * Fly-Man- keeps seeing either errors or a pending review sign
[15:54] <Remaille> beuno, a "project" is different from a PPA ?
[15:54] <Remaille> I have this for now (thanks to you upload worked properly) : https://code.launchpad.net/~linuxedu
[15:55] <Remaille> this is supoosed to be owned by a team
[15:59] <Remaille> is a "project"="a branch" :-)
[16:01] <noodles775> Remaille: have you taken a look at https://launchpad.net/+tour/branch-hosting-tracking ? It outlines how a single project (say, for Mozilla Firefox) can have many branches.
[16:20] <Remaille> ok noodles775 thanks
[16:20] <Remaille> I knew nothing of LP yesterday :)
[16:32] <CarlFK> "             Displaying first 80             comments.                                             View all 125"
[16:33] <CarlFK> it would be nice to see the first X and the last X comments
[16:33] <CarlFK> where X is about 40 or something
[16:34] <AskHL_> Say I have a PPA package with a pure Python program for intrepid.  I would like the corresponding PPA for, say, jaunty.  As I have understood, this should be possible by just copying the package (as per the button in the Lauchpad interface) and specifying that the copy should be for jaunty.  But when doing this, the Changelog will still specify that the series is 'intrepid', and there will probably be other such problems in the remaining package f
[16:35] <Remaille> hi how could i delete a team ? (i am the owner)
[16:35] <Remaille> thanks
[16:35] <AskHL_> The straightforward way would be to create the jaunty package the same way as I created the intrepid package.  Is it better to do that?  (in general, I presume it is, because it may in fact depend on the distribution version)
[16:38] <Remaille> ok : https://answers.launchpad.net/launchpad/+question/72902
[16:38] <noodles775> AskHL_: https://help.launchpad.net/Packaging/PPA/BuildingASourcePackage#versioning
[16:38] <Remaille> "asking an admin to do so" I will wait abit so
[16:38] <bigjools> AskHL_: it's fine to copy to promote the package to a new series
[16:38] <bigjools> AskHL_: in fact this is what the distro does
[16:41] <AskHL_> noodles775 and bigjools, thank you very much.
[17:13] <phuff> Any launchpad devs around that know how launchpad's openid implementation works?
[17:14] <phuff> We're trying to get sf.net compatible but it keeps saying that the openid urls we give it aren't valid.
[17:14] <lfaraone> jml: see above ^^^
[17:23] <jml> phuff, I think sinzui's the person to speak to.
[17:24] <james_w> phuff: are you giving it login.launchpad.net, your profile page, or your identity url?
[17:27] <phuff> james_w: I'm giving it https://launchpad.net/~phuff
[17:29] <james_w> phuff: I think I remember a year or so ago when trying to sign in yo sf that I had to use the identity url
[17:29] <james_w> but I'm not sure how to find what that is
[17:29] <james_w> and the help says that you can use sf with the url you are, so presumably someone tested that
[17:34] <phuff> Yeah, except we can't get it to work :)
[17:35] <phuff> james_w: This is the message I get:
[17:35] <phuff>  A site identifying as https://sourceforge.net  has asked us for confirmation that https://launchpad.net/~phuff  is your identity URL. However, that is not a valid Launchpad OpenID identity URL, such as https://launchpad.net/~USER
[17:35] <phuff>  
[17:35] <phuff> :)
[17:37] <james_w> phuff: https://launchpad.net/~phuff is you?
[17:37] <james_w> does it say "OpenID login:" on that page for you?
[17:41] <phuff> Yeah
[17:41] <phuff> It does
[17:41] <phuff> That's where I got it from
[17:43] <jml> phuff, is this bug the one you're seeing? https://bugs.edge.launchpad.net/canonical-identity-provider/+bug/319614
[17:44] <james_w> jml: the openid help wiki page still points to an LP bug list for openid bugs, is that no longer correct?
[17:45] <jml> james_w, possibly not. I haven't been following the work, but we've been splitting the openid stuff out of Launchpad proper
[17:47] <phuff> jml: Yeah, that looks like the same message.
[17:48] <phuff> jml: I'll look into that now.
[17:48] <phuff> jml: Thanks
[17:49] <jml> phuff, fwiw, I'm back in the office tomorrow; once there, I'll be able to get explanations & answers more readily.
[17:52] <tim_blechmann> hi ... i am trying to upload a package to a ppa with dput ppa:<user>/<ppa> <source.changes> ... it gives me the error message "Connection failed, aborting. Check your network [Errno 111] Connection refused" ... any idea?
[17:52] <bigjools> tim_blechmann: the ftp server has gone down, give me a moment to get it restarted
[17:52] <tim_blechmann> bigjools, thanks!
[17:54] <phuff> jml: Okay, I'll noodle on this for a bit and if I can't figure anything out I'll come back :)
[17:56] <jml> phuff, cool.
[17:57] <jml> phuff, fwiw, we have launchpad devs all over the world -- emailing launchpad-users or launchpad-dev can be a good way to make sure that all of them know about your suffering :)
[17:57] <phuff> jml: Okay :)
[18:02] <bigjools> tim_blechmann: should be up again
[18:07] <sinzui> phuff: jml: I really do not know much about openid. It (along with login and accounts) are owned by the launchpad foundations team. I have worked on openid, the the foundations team removed my work
[18:07] <jml> sinzui, good news for you then :)
[18:10] <phuff> Hah
[18:10] <phuff> Who's the launchpad foundations team?
[18:11] <sinzui> phuff: When I worked on it last year, your profile page contained an xrds instruction in the head that allowed  you to use that page to initiate login in another site. the hook was really nothing more than an openid approved redirect to https://login.launchpad.net/. The user URL does not work, or at least the one that I recall does not
[18:12] <phuff> ok
[18:13] <sinzui> phuff: I believe ~user should work. The main reason my work was removed was because it users wanted to launchpad as identity rather than security
[18:13] <phuff> ok
[18:14] <sinzui> I think I mean obscurity, not security. You are not anonymous using launchpad since your launchpad id is in the url
[19:27] <phuff> Ah I see
[19:31] <krow> Hi! Any thoughts/how to debug why LP is not downloading binaries from other sites?
[22:11] <wgrant> phuff: It's actually bug #449105, which was incorrectly marked as a duplicate.
[22:12] <sinzui> phuff: wgrant: I am branching for that bug now
[22:12] <lifeless> \o/
[22:12] <wgrant> sinzui: I would, but...
[22:13] <wgrant> Hm, why's it back on -foundations?
[22:13] <wgrant> Ah.
[22:13] <wgrant> What a strange abuse of task reassignment.
[22:13] <sinzui> because login/sso is foundation work
[22:14] <wgrant> But surely it belongs on c-i-p.
[22:14] <sinzui> Foundations are sprinting. Since I guessed the xrds is broken, I get to fix it
[22:14] <wgrant> sinzui: If you look in that bug, you'll see I identified and tested a fix already.
[22:14] <sinzui> wgrant: c-i-p does not use ~user. that is a launchpad social feature
[22:14] <wgrant> sinzui: But the XRDS is served by the SSO server.
[22:15] <wgrant> Although I guess maybe ~user could link to its own.
[22:15] <sinzui> wgrant: 14 months ago the ~user page had a mixin that provided the needed metadata
[22:16] <sinzui> I do not know what is going on now, I just knew that the xrds is broken
[22:16] <wgrant> sinzui: It already does provide its own data, mostly. It just also links to the XRDS.
[22:18] <phuff> wgrant: It's actually a bug on our side
[22:18] <phuff> wgrant: I can fix it.
[22:18] <phuff> :)
[22:19] <phuff> We're not respecting xrds delegation correctly
[22:19] <sinzui> wgrant: I do not know how to fix this at this moment, not even readying your remarks. I have to empty myself of all my dislike of openid and SSO and launchpad account/profile separation before I can write a test to understand what is wrong
[22:19] <phuff> And you guys are doing <LocalId> in your xrds file
[22:19] <phuff> The error message is a bit confusing
[22:19] <phuff> But other than that, you're probably doing things right.
[22:21] <wgrant> phuff: The XRDS I see has no openid:Delegate or LocalId.
[22:24] <phuff> The XRDS for my account has a <LocalId> in it
[22:25] <phuff> Which points to something other than https://launchpad.net/~phuff
[22:25] <phuff> Which is why I assume it's dying.
[22:25] <phuff> But the error message is messed up
[22:26] <wgrant> I do not see the LocalId
[22:26] <phuff> See https://launchpad.net/~phuff/+xrds
[22:27] <wgrant> I didn't even know that existed. That's not the one that is linked to in the page.
[22:27] <wgrant> But that one does look correct.
[22:27] <sosborn> it's linked to in the headers
[22:27] <phuff> Hey there sosborn :)
[22:27] <phuff> sosborn works for sf.net too :)
[22:27] <sosborn> see curl -I https://launchpad.net/~osborn-steven
[22:27] <wgrant> Argh.
[22:27] <wgrant> So it is.
[22:28] <sosborn> X-Xrds-Location: https://launchpad.net/~osborn-steven/+xrds
[22:28] <wgrant> So the XRDS in the headers is different from the XRDS in the document.
[22:28] <wgrant> sinzui: ^^
[22:28] <phuff> Well, it looks like your header one is valid, we're just not parsing it correctly...
[22:28] <sosborn>   <meta http-equiv="X-XRDS-Location"
[22:28] <sosborn>             content="https://login.launchpad.net/+id/H8yy64p/+xrds" />
[22:29] <sosborn> lol I didn't even notice the meta tag was different
[22:29] <wgrant> sosborn: Right. If you look at that one, it omits the openid:Delegate and LocalIds.
[22:29] <sosborn> we're not actually doing discovery on them atm
[22:29] <wgrant> Aha.
[22:29] <sinzui> oh
[22:29] <phuff> Yeah, the spec says to give the http headers priority, right?
[22:30] <phuff> So the next question is who can get a fix out faster ;)
[22:30] <sosborn> Yah, I didn't implement the meta tag because I couldn't find a provider that used it and didn't pass the header along with it
[22:31] <sosborn> With Google your identity url is your XRDS file which is also unusual
[22:31] <sosborn> even with directed identity
[22:31] <lifeless> openid has cruft already?
[22:31] <phuff> openid was _born_ with cruft ;)
[22:31] <phuff> Oh maybe you meant the launchpad implementation :)
[22:32] <sosborn> it's pretty much cruft + superglue and ductape
[22:32] <lifeless> No, the spec ;)
[22:32] <sinzui> we hopped from 1 to 1.0 and maybe 2.0 is to not clear exactly which we support
[22:32] <sinzui> 1.1 should be in that list
[22:32] <lifeless> so I'm guessing I'm not the only user that reported this yesterday? Anyhow, thanks phuff, sosborn, wgrant & sinzui for looking into thise
[22:33] <sinzui> DO not thank me. I am still looking for a test to produce this
[22:33] <lifeless> sinzui: you're looking at it; thats all I'm crediting you with :)
[22:37] <wgrant> sinzui: It looks like all that needs to be changed on the LP end is the X-XRDS-Location http-equiv to point to LP's copy, not c-i-p's. I didn't realise that there *was* an LP copy.
[22:38] <sinzui> wgrant: thanks for the insight
[22:38]  * sinzui thinks we should should shutdown Launchpad's versions
[22:43] <sinzui> wgrant: There is no c-i-p version running at the moment. There is only launchpad with a library answering to many names.
[22:44] <wgrant> sinzui: But doesn't it symlink c-i-p into lib/canonical/signon?
[22:44] <sinzui> it does, it is login.launchpad,net
[22:44] <wgrant> So the code is owned by c-i-p.
[22:44] <wgrant> Right.
[22:45] <sinzui> it subclass the many of the objects in Launchpad
[22:45] <sinzui> so if you want to fix many things, you need to change launchapd's models and views
[22:46] <sinzui> wgrant: which is why all the xrds stuff appears to be in launchpad
[22:59] <sinzui> wgrant: phuff: This is the What do you expect this test to return
[22:59] <sinzui> self.assertEqual(
[22:59] <sinzui>     request.response.getHeader('x-xrds-location'),
[22:59] <sinzui>     'http://login.launchpad.dev/~sinzui/+xrds')
[23:00] <wgrant> sinzui: That bit is right.
[23:00] <wgrant> sinzui: It's the <meta http-equiv="X-XRDS-Location" ...> in the page that's wrong.
[23:01] <wgrant> Nothing should use that, but I suspect some things do, and it's otherwise incredibly confusing.
[23:01] <sinzui> oh, the +id/giberish/+xrds
[23:02] <wgrant> Right.
[23:02]  * sinzui shakes his head
[23:03] <sinzui> wgrant: Yes, that should not exist. That was to be removed along with most of my other openid work because openid on launchpad is intended to be social
[23:04] <wgrant> sinzui: It should exist, but perhaps only for use by the SSO server.
[23:04] <sinzui> wgrant: So the screwy part is is that that tests verify everything is sane, but since the test (in SSO) and code (in launchpad) are separated, you cannot easily see they are not working together.
[23:04] <sinzui> wgrant: it does exists, but is not intended to be visible any more
[23:29] <Jordan_U> The launchpad keyserver is timing out, is there a way to fix this or another way to get the public key for https://edge.launchpad.net/~ubuntu-x-swat/+archive/x-retro ?
[23:37] <wgrant> Jordan_U: Try pool.sks-keyservers.net instead.
[23:38] <Jordan_U> wgrant: Thanks, that works