=== Edwin is now known as Guest15853 === asac_ is now known as asac [01:43] Could I get a admin to delete https://launchpad.net/netrek-old-clients ? I'm going to never do anything with it [01:45] Hellow: can you ask the same via LP answers? that way we can verify it's a legit request to remove. https://answers.edge.launchpad.net/launchpad/+addquestion [01:45] heh, ok === Edwin is now known as Guest72826 [04:28] i have a question about my registered name on launchpad. i have an account and a ppa linked to a name i would like to change but as i get it my ppa is linked to the name i would like to drop... is it possible to flush the ppa and change launchpad name (not display namne) somehow? [04:28] i'm on edge.launchpad.net by the way [04:28] Philip5: ask a question on answers.launchpad.net/launchpad with the details [04:28] Philip5: it is possible just not trivial [04:29] oki, thanks [04:29] It'll work better if you empty the PPA first, I believe. Moving a populated one would get... messy. But ask a question, and an admin should sort you out. [04:30] wgrant: that was my next question... how do i delete packages or just empty my ppa? [04:33] never mind... found it... :) [04:39] but i can only delete packages and not the whole ppa to start a new one :( [04:39] just create a new [06:23] Does anybody know how fta does his ppa-scripts, apparently he's asleep when I'm awake, and I can't get any answers from him. He has alot of examples on how to use get-orig-source with his bzr packages, but I'm head is hurting trying to figure out how to do the same thing with git. And what the hell is all this debian_tag stuff? [06:26] ripps, His infrastructure is mostly checked into bzr branches, but you probably want to ask on #ubuntu-mozillateam, as others there ought be at least passingly familiar. [06:27] persia: that'd be great if anybody there would talk [06:27] Hrm. Can't help you there. Email maybe? [06:34] Does a team deletion have to be done by a administrator? [06:35] Hellow: yes [06:37] spm, By the way, did you ever hear anything back on how to resolve the merge between a "person" and a team for pkg-games? [06:40] persia: hmmm. I don't believe I did. I shall poke. [06:40] spm, No huge rush, but thanks :) [06:48] persia: just revisiting this. appears *now* that they show up as two teams. naturally it's not all roses, as a trial merge on staging oopsed. [06:50] heh. Progress at least :) [06:50] An oops means a real bug, and something someone can fix, rather than the murky confusion we had previously. [06:50] that was my thinking as well [06:51] I wonder... can we not just "do away" with the old team? ISTR there was something around email addresses tho? [06:52] Right. The two teams have different addresses, and the members of the team would prefer that LP bugmail, etc. not be sent to the address used for tracking uploads to Debian. [06:52] (while the team itself covers both Debian and Ubuntu, there are some members who prefer to ignore Ubuntu entirely) [06:52] :-) [06:53] So, the idea was to have the address used for Debian uploads be a secondary address for the team used for the PPA, code branches, etc. [06:55] persia: so to rephrase it: 1. we have pkg-games-ubuntu@lists... this is good and we're all happy with how it works. 2. we want another address for other stuff (PPA, code branches etc);; yes? [06:56] Err, no. [06:56] We have pkg-games-ubuntu@lists. This is good, and we're all happy with how it works. [06:56] cool [06:56] We upload to Debian with pkg-games@alioth, and we want that to show maintainer credit for the team on the overview page for the team. [06:57] hmmmm [07:00] persia: ok. try #2. you want the contact address for ~pkg-games to become pkg-games@alioth (vs the current pkg-games-ubuntu@lists)? or am I still not getting it? AIUI a team can only have a single contact address, and I suspect you're trying to have two? [07:00] One contact address, but multiple addresses in total. [07:01] like a person has? eg myself with 4 ish. I don't think teams can do that. of it they can, be damned if I can find where... [07:02] Let me see. [07:02] I know it's confusing. [07:02] spm, Yeah. A single contact address fails to work for this case. [07:02] Or, hrm. Am I missing something in the definition of "contact address"? [07:02] heh [07:02] contact address has 2 cases [07:02] It has a dual meaning. [07:03] The primary address. [07:03] We only want to receive email from LP at the pkg-games@lists.lp address. We want to have stuff done by the alioth address recognised as having been done by the team. [07:03] or a team address. [07:03] 1. goes to an external list/alias for example [07:03] 2. goes to a team exploder within launchpad itself [07:03] kinda thing. [07:03] Hmmm. [07:03] There used to be a +editemails [07:03] But there isn't any more. [07:04] It's certainly useful to have multiple addresses for a team... [07:04] +contactaddress now === mneptok is now known as mnepton [07:04] There used to be that and +editemails === mnepton is now known as mneptok [07:04] 'k [07:04] mneptok: you should try for mneptRon ? [07:05] Bug #174452 says it wasn't intentional that it was present. [07:05] Launchpad bug 174452 in launchpad-foundations "Contact addresses impossible to remove from teams" [Undecided,Invalid] https://launchpad.net/bugs/174452 [07:05] spm: just making sure $ALTNICK doesn't get dropped [07:05] But it was there, and worked, so presumably the DB allows multiple addresses. [07:05] persia: wgrant: I wonder.. could we achieve same by a sub team. I suspect no, but throw the thought out there... [07:06] mneptok: for shame. such a boring reason :-) [07:06] spm: Not really, no. [07:06] spm, I don't think a subteam would accomplish the branding goal: I can't imagine it appearing significantly different than the team + the person. [07:07] spm: would you prefer me telling you about my prosthetic breasts and full-head latex Charo mask when i $ALTNICK? [07:07] mneptok: hmmm. tempting... [07:07] CUCHI-CUCHI! [07:08] persia: yeah. tricky... [07:08] spm, Goes back to the concept that people upload stuff and teams don't, which doesn't quite map cleanly for team-maintained stuff (not that it usually matters, as few actual maintainers use LP). [07:09] how about a team list? ... hmmm. no. that works the wrong way. [07:09] right. [07:09] spm, Yeah. At a top level, we want something that looks like a person, yet works like a team :) [07:09] persia: Why does it need to be like a person? [07:12] wgrant, https://launchpad.net/~pkg-games-devel/+maintained-packages vs. https://launchpad.net/~pkg-games/+maintained-packages [07:13] Hrm. That worked better than I thought it would, but the latter isn't UI-accessible. [07:13] persia: Teams can maintain packages, so that the latter isn't linked to is a bug. [07:14] wgrant, Ah, so the bugs are 1) teams don't have secondary addresses, and 2) there's no way to navigate to +related-software (and similar) for teams? [07:15] persia: I think rather 1) teams don't have secondary addresses exposed in the UI. [07:15] But the DB could have changed, I guess. [07:16] I heard that the difference between "person" and "team" was a single binary flag at one point. Dunno if that's still the case. [07:16] They are the same DB table, right. [07:16] But I presume it's at least close, as otherwise pkg-games-devel probably wouldn't have been able to convert from a person to a team. [07:17] I suspect the merge will copy the addresses over, or things would go wrong in normal team merges. [07:18] Well, the merge oopsed, which at least gives something to track. [07:18] If the email address wasn't transferred, the autocreated team would just be recreated fairly quickly when the old address was seen again. [07:18] https://lp-oops.canonical.com/oops.py/?oopsid=OOPS-1258S301 for those that have oops access... [07:18] https://devpad.canonical.com/~jamesh/oops.cgi/1258S301 [07:18] * wgrant shakes a fist or three. [07:19] three? [07:19] Somewhere around that, yes. [07:19] persia: he's from melbourne - close to tasmania. all bets are off at that point. :-P [07:20] spm: Heh. [07:21] At least *my* city wasn't built as a compromise between the two capital candidates. [07:21] wgrant: spm hey guys here in Tassie, were are thinkers but not doers !!! :P [07:21] yeah I love that. half way compromise. 280km from sydney, 800km from melbourne. [07:23] persia: ok, so back to the question. I'll update with a summary of my understanding of what is wanted and where we're at. but I suspect we're headnig to report bug land. [07:24] spm, I think you're right. Are you filing bugs? Shall I? Perhaps wgrant will volunteer as he understands both how LP works AND what I want to do? [07:24] * wgrant doesn't understand how LP works. [07:25] persia: bug? no. better if you file - coming from the user perspective as it were. I'd just be puppeting your wishes and liable to get it wronger. [07:26] wgrant: I don't think anyone really understands it all. too big, too complex, too arcane. [07:26] persia: LP works? [07:26] are you ... certain? [07:26] * spm wanders off to suspend samb's account. citing irrecoverable rudeness. :-P [07:27] SamB, SamB It at least works for some things some of the time, and can usually be wrangled to work for the rest with enough bug reports and enough time. [07:27] spm: it's called joking! [07:27] The average time is actually reducing now. It's a bit strange. [07:27] it does make an okay free bzr hosting service [07:28] SamB: it was taken that way :-) [07:29] irrecoverably rude joking! [07:29] wgrant, Don't complain about a declining average time-to-fix: I'm not sure I want that fixed! [07:29] But anyway, to make sure I understand the bugs I'm filing. I think there are three. [07:29] 1) There doesn't seem to be a way for the UI to control alternate addresses for a team [07:30] 2) There doesn't seem to be a navigation path to +related-software for a team [07:30] 3) There's an oops when trying to merge two specific teams that may be an implementation bug of some sort. [07:31] Agreed. [07:31] yup [07:31] But s/may be/is/ [07:31] wgrant, Well, it could also be DB corruption as a result of something that went wrong when converting pkg-games-devel from a person to a team. I don't know which part has the bug. [07:31] 3. could be tricky - I merge teams a lot; but they're usually both active teams. in this case, the mergee is not activated; the merged-onto is. that may/not be cause & effect. [07:32] or that too [07:32] spm, Would you mind filing #3 ? I think you understand the issues there more. I'm happy to file #1 and #2, which I'll need anyway in order to reach my end state. [07:32] persia: true. [07:32] sure [07:32] Thanks. [07:33] I was thinking I'd need to anyway - if only to capture the what/how. [07:33] What do these go against? Just launchpad, or is there a better target for team stuff? [07:33] generically LP. the wonderful QA folks will adjust from there [07:33] (registry maybe?) [07:33] Registry [07:34] I personally can never remember which team stuff goes against. So tend to stick with the generic. [07:34] Teams/people/projects are Registry. [07:35] Parts of distros are also, to add to the confusion. [07:35] products too, from the looks of things. [07:35] (and some bits of branches, but not much) [07:35] By projects I meant products+projects, but yes. [07:35] What is the difference between a "product" and a "project" in LP-speak? [07:36] isn't "ubuntu" a product, vs eg solitaire as a project? just guessing... [07:37] Aha. The Great Confusion. [07:37] Ubuntu is a distro. That's unambiguous. [07:37] "Product" in the code is "project" in the UI. [07:38] Aha. [07:38] "Project" in the code is "project group" in the UI. [07:38] ah yes. I recall that discussion now. ta. [07:38] Someone ought rationalise the language and branding for those. [07:38] They were renamed in the UI 2007ish, but never in the code. There has been much confusion. [07:38] Indeed. [07:38] Fortunately the webservice API exports have mostly been given the UI names, rather than the code ones. [07:38] But I can see how "product" is poor branding for many of the things registered on LP. [07:39] Right. [07:39] Seems #1 is bug #156340 [07:39] Launchpad bug 156340 in launchpad-registry "Need to be able to associate multiple email addresses with a team" [Low,Triaged] https://launchpad.net/bugs/156340 [07:39] Or maybe even #28688 [07:39] Bug #28688 [07:39] Launchpad bug 28688 in launchpad-foundations "Receive mail about different products/packages/etc (roles/"hats") at different e-mail addresses" [Low,Triaged] https://launchpad.net/bugs/28688 [07:40] Not really. [07:40] Ah, the last paragraph of 156340's description is encouraging. [07:40] The merge should probably DTRT. [07:43] Well, except that it oopsed. Anyway, the comment about 28688 comes from comment 2. [07:43] Not an email-related OOPS, though? [07:44] I'm not familiar enough to know either way. [07:52] #2 is bug #385825 [07:52] Launchpad bug 385825 in launchpad-registry "No obvious UI to navigate to +maintained-packages for a team" [Undecided,New] https://launchpad.net/bugs/385825 [07:52] It should probably be more general than that; a list of related projects is also useful for a team. [07:53] a dashboard of various admin tasks would be awesomeness too :-) [07:53] A dashboard of anything would be awesome. [07:53] A timeline for a project would be awesome. [07:53] But LP doesn't do anything like that yet :( [07:53] A timeline for a team would be nifty, or even for a person. [07:53] Karma almost does that, but it's not linked to the objects. [07:54] one of the things I love about freshmeat is exactly that - the activity graph(s) [07:56] wgrant, For +related-software for teams, what beyond PPA and maintained packages would be relevant? [07:57] Teams don't usually comment on bugs, or do uploads. [07:57] persia: Related Projects [07:57] ie. everything except uploaded packages [07:57] How does one calculate that? [07:57] It's based on the configured driver/owner of projects. [07:57] The only link I can imagine is team branches of projects. [07:57] Nothing to do with activity. [07:58] Interesting. Things show up in my Related Projects semi-randomly. I always presumed it came from bug reporting or branch submissions. [07:58] It's from being a member of ~motu, generally. [08:00] I think I have a number of inputs, but that explains why I keep having stuff I've never heard about show up there. [08:00] And yes, that does become interesting for teams. [08:00] And even "Uploaded Packages" is conceivably meaningful, in the case that someone stuffed the Changed-By: value. [08:01] (this *shouldn't* happen, but it could) [08:01] Well, that should really reject the upload and delete the signer's Launchpad account :P [08:02] Well, at least discoverability would allow us to apply social pressures. Right now, we have no idea. [08:02] Teams can do lots of things that they shouldn't be able to (eg. comment on bugs). [08:02] How can a team comment on a bug? [08:02] Email. [08:03] Doesn't that need a GPG key? [08:03] Not if it's just a comment. [08:03] Ugh. [08:03] Only modifications need GPG. [08:03] Which is questionable. [08:03] But I guess it would be user-hostile otherwise. [08:05] Right. Some people find mail readers to be something other than torture devices. === abentley1 is now known as abentley [08:47] How often do the LOSAs review and act on LP questions assigned to them? [08:48] maxb: daily, at least, usually more often. Which one is bothering you ? [08:49] my gpg key reassignment one :-) [08:49] The unassigned questions, however... [08:49] https://answers.launchpad.net/launchpad/+question/72929 [08:49] Incidentally, is that SQL command seriously implying that LP expects 8-digit keyids to be unique? [08:50] It looks like it... IIRC a conflict has happened in the past. [08:50] But the user recreated the key. [08:51] well, gah. I rather assumed that since users have to enter full fingerprints to import keys, that foresight would apply throughout :-) [08:52] Also, LP thinks my question is in the "Answered" state - should I be always coming back to it and adding a "not fixed yet" comment every time someone else makes any comment on it at all, to keep it in the "Open" state? [08:55] maxb: I suspect it'll be ignored if it remains Answered. 'Needs Information' ones do, at least. [08:57] * maxb wonders if there's a bug for "should be possible to comment on a question without setting it to "answered" [08:59] If somebody other than the asker of the question comments, it gets set to Needs Information. It's a bit restrictive. [09:03] maxb: we don't rely on unique key IDs, only fprs, for the record. [09:03] Hmmm. What did I remember, then... [09:04] the users instructions here and there use it, though. [09:05] cprov: Isn't the SQL that sinzui put in the question buggy, then? [09:06] wgrant: yes, I think it was more like a hint than anything else [09:06] wgrant: we usually don't paste SQL recipes in questions. [09:08] bug 114710 ftr [09:08] Launchpad bug 114710 in launchpad-answers "Allow to post comments which are not information requests or answers" [Low,Triaged] https://launchpad.net/bugs/114710 [09:09] cprov: That's good - Answers comments are easy to impersonate. [09:10] * cprov nods [09:24] can someone on LP *please* update the ancient bzr server! [09:25] trying to download fast-import: $ bzr branch lp:bzr-fastimport fastimport [09:25] Server is too old for streaming pull, reconnecting. (Upgrade the server to Bazaar 1.2 to avoid this [09:25] this is really bad [09:25] Launchpad is running the latest stable bzr. [09:25] why is LP running such an old release of the bzr server!? [09:25] well stable is too old :) [09:25] I suspect your client is too old - there was a bug causing that message to show up. [09:25] oh ok sorry then [09:26] my client is 1.5-1.1 on debian squeeze [09:26] Launchpad is currently using bzr 1.15. [09:26] Right, that's pretty old. [09:26] ok the error message said the *server* [09:26] guess it meant my client heh [09:26] It did. [09:26] Your client thinks it's the server's fault. [09:26] wonder if i should file a bug report [09:26] But it's wrong. [09:26] wgrant: ah :) ok [09:27] sorry guys for blaming LP [09:27] No - that bug was fixed perhaps a year ago, but your old version doesn't have the fix. [09:27] zu22: squeeze has 1.13.1-1, it's lenny that has 1.5-1.1 [09:27] ok gotcha [09:27] well i have some squeeze packages some lenny packages [09:27] i did dist-upgrade but some packages *still* on lenny [09:28] not sure why [09:28] erm, that's rather broken [09:28] yeah :( [09:36] maxb, Just for the record, some packages haven't been updated between lenny and squeeze (for that matter, some packages haven't been updated between warty and karmic) === gmb changed the topic of #launchpad to: https://launchpad.net/ | Help contact: gmb | Join https://launchpad.net/~launchpad-users | Channel logs: http://irclogs.ubuntu.com | Open Sourcing: https://dev.launchpad.net/OpenSourcing [10:17] /quit/quit [10:17] lol [10:17] .. [10:18] ripps, still there? [10:19] fta: hi, asac has been helping me [10:19] I'm checking to see if my new get-orig-source works now [10:19] ripps, your get-orig-source target could be pretty simple, you could ignore all my attributes (tags, dates, etc) [10:20] as long as by default, it returns the freshest tarball you want to build [10:20] fta: can you discuss that in mt channel ;)? [10:20] ok [10:20] i am interested too and cannot follow this ;) [10:20] (e.g. git is used by NM and connman etc.) [10:20] ok, moving to m-t [10:20] cool ;) gratias === abentley1 is now known as abentley [11:06] if a vcs-import is dead in favour of another, what do we have to do to get it switched? [11:07] https://code.edge.launchpad.net/~vcs-imports/kerneloops/trunk is now http://git.infradead.org/?p=kerneloops.git [11:12] question on answer.l.n/launchpad-code [11:21] thanks === mrevell is now known as mrevell-lunch [13:48] cprov: Do malformed .changes file names still result in silent rejection? [13:49] wgrant: yes, unicode handling still having some slips. [13:49] wgrant: do you have a new example ? [13:50] cprov: I mean, does a .changes file with a filename other than NAME_VERSION_source.changes get rejected silently? [13:51] wgrant: no, this one was fixed some time ago [13:51] wgrant: broken .changes filenames result in a rejection upload notification if it's content is correct. [13:51] cprov: That's what I thought - I just saw a warning about it while editing the docs. I'll remove it. [13:52] wgrant: great, thank you. === mrevell-lunch is now known as mrevell === thekorn_ is now known as thekorn === henninge_ is now known as henninge [14:09] my svn project failed to import. now i think i've got the correct url, but i can't find a way to modify the importurl. how can i do that?? [14:11] eagle00789, Ask a question against launchpad-code [14:11] ?? [14:12] go to https://answers.launchpad.net/launchpad-code/ , file a question describing the specifics (and the new URL). Wait for an admin to fix it. [14:13] thx [14:19] question posted. no we will wait :D thx again [15:33] Is it possible/allowed to integrate launchpad bug tracker into your own website layout? Does LP provide an API for it? [15:34] pro-rsoft, you've got some APIs, but I'm not sure they're complete enough to use it as a bug tracker from somewhere else [15:34] BjornT, ^ [15:38] pro-rsoft: pretty much everything that can be done using launchpad's web interface is available as an API. one thing that may prove a problem for you is that the API uses a single user's login. you can't have a web application operate on behalf of other users [15:39] well, that's not completely true. it's possible in theory, but i don't know how easy. i don't think anyone every tried [15:39] pro-rsoft: but as long as you're happy using a single user, yes, you can use the API to integrate very tightly with launchpad [15:41] ok [15:42] but yeah people will need a launchpad login anyways to be able to track their own comments, etc. === noodles776 is now known as noodles === herb__ is now known as herb [16:15] Hi :) [16:15] I think I've done a stupid change in my lp project [16:15] could someone help me ? [16:15] I tried to change the name of a branch, now there's a mess [16:15] it looks like I now have 2 distinct urls for the same branch ... [16:16] file a support request and it will be put back by an admin [16:16] thank you pro-rsoft, I'll do that === matsubara is now known as matsubara-lunch [16:20] pro-rsoft: did ou mean there: https://answers.launchpad.net/launchpad/+addquestion ? [16:20] yes, I think so [16:21] thx [16:55] anyone to check bug 384217? [16:55] Launchpad bug 384217 in launchpad "subscribed to a code branch, but did not receive most updates" [Undecided,New] https://launchpad.net/bugs/384217 [16:56] (it's converted from a question) [16:56] when you subscribed to a code branch you can also alter subscription settings [16:56] where you can change which updates you receive exactly [16:57] I have, I've set it to receive all update without diff [16:57] and are you sure the code branch already got enough time to update itself? [16:57] I get changes a day later (CVS import) [16:57] if not longer. [16:58] I have received revision numbers: 3, 10 [16:58] I have NOT received revision numbers: 4, 5, 6, 7, 8, 9, 11, 12 [16:58] Update since revision 24: [16:58] Received: 24, 30 [16:58] Did not receive: 25, 26, 27, 28, 29 [16:58] no offense, but I think it has had enough time since revision 3 :P [16:58] can the svnimport of launchpad work together with the dyndns service?? [16:58] it's really weird to receive some and lose some [16:59] eagle00789, why not/ [16:59] the log still tells me this: pysvn._pysvn_2_4.ClientError: Can't connect to host 'smartass.webhop.info': Connection refused [16:59] i have someone over on the #svn channel who checked out my link i use and it works for him... [17:02] so why doesn't it work for launchpad?? [17:03] well resolve works, ping works, but for some reason http://smartass.webhop.info doesn't show anything [17:04] what's the svn address? [17:04] no, because http runs on a different port. the svn adress is svn://smartass.webhop.info:3421/share/HDA_DATA/svn/ToolbarKiller [17:05] eagle00789: and the link to the svn import project at launchpad? [17:05] the same as posted above [17:06] here is the link to the project: https://code.launchpad.net/~vcs-imports/tbu/trunk [17:06] that's the one, thanks [17:06] np. sorry :D [17:09] hm.. no idea [17:09] you'll have to stick with paul hummer I guess: https://answers.launchpad.net/launchpad-code/+question/73930 :) [17:09] :D i guess i will indeed [17:10] you were also able to connect to the svn server?? [17:10] rockstar: here? eagle00789 is on IRC and having hard time with https://answers.launchpad.net/launchpad-code/+question/73930 :) [17:10] eagle00789: yes [17:11] ok. thx [17:11] eagle00789, hi [17:11] hey rockstar [17:12] the problem: i get the following errors in the importlog: pysvn._pysvn_2_4.ClientError: Can't connect to host 'smartass.webhop.info': Connection refused but the svn url i use now is correct [17:12] the svn url is: svn://smartass.webhop.info:3421/share/HDA_DATA/svn/ToolbarKiller [17:14] eagle00789, ah yes, I should have noticed this before. We block those ports on our outbound slaves. [17:14] what port should i be using then?? [17:14] eagle00789, port 80? [17:15] can't use that as a special service runs on that... [17:15] http runs on 50000 on my server [17:15] eagle00789, I think 443 is open as well. [17:15] eagle00789, this is on your home server? [17:15] 443 isn't used for anything else?? [17:16] yes it is [17:16] eagle00789, it's used for https [17:16] don't use https [17:16] so i could use that [17:16] eagle00789, are you wanting to do a one-time import of your code, or ongoing? [17:16] ongoing [17:18] eagle00789, so you'll have to hack around and probably put it on 443 [17:18] i basicly could use my router so that you can connect to 443 but internally i still use 3421 [17:19] eagle00789, yeah, however you do that. We try and lock those import slaves down pretty heavily. [17:20] i indeed noticed now yes [17:21] gmb: hullos, I was wondering if you could convert a user to a team and how much work that would be :) [17:22] apachelogger: It's not possible to do that. Why not just create a new team? [17:22] gmb: preserve the PPA url ... would it be possible to rename the user and then create a team with that name? [17:22] apachelogger: Yes, you can do that, certainly. [17:23] apachelogger: In fact, that was going to be my next recommendation :) [17:23] gmb, Are you sure it's not possible? I swear it's happened in the past. [17:23] gmb: hehe, ok, thanks :) [17:23] persia: I've never seen it done. Essentially it's flipping a bit in the DB but I'm not sure if there mightn't be funky consequences. [17:23] apachelogger: However, just htinking about it... [17:23] apachelogger: I'm not sure the PPA for the user would get renamed. [17:24] apachelogger: Let me talk to someone from the Soyuz team about this. [17:24] gmb, I believe there are some funky consequences, based on the experience I've had with the one I've seen, but I'm not sure the funky consequences aren't bugs. [17:24] gmb: okies [17:24] persia: Right. [17:24] Also, the PPA (ought) to get a new key, which may confuse PPA users. [17:24] With switching, there might be other issues, but the key could probably be preserved. [17:24] gmb: you can't rename if you have a PPA [17:25] rockstar: i have changed the port now. the url is: svn://smartass.webhop.info:443/share/HDA_DATA/svn/ToolbarKiller [17:25] bigjools: Ah. So, what's the best way to deal with this problem then, I wonder? [17:28] apachelogger: So, having discussed it with the Soyuz guys, it seems your best course of action is to file a question at http://answers.launchpad.net/launchpad and request that your PPA (and user) be renamed. An admin can take care of that and then you can create your new team with the correct details. [17:29] gmb: I don't think we are going to do it. [17:29] eagle00789, updated. [17:29] thx [17:29] cprov: Can you be more specific? [17:30] gmb: yes, it's not correct to move PPAs like that. [17:30] cprov: So, what's the preferred solution? I'm getting conflicting reports here :) [17:31] Um, Why is there a rename or move? I thought that was just a workaround. Wasn't the original request just a person->team conversion? [17:31] gmb: ideally the user should create the new team, new PPA, new keys and abandon the old one. [17:31] cprov: he wants to preserve the URL [17:31] persia: Yes. However, as I said before, this could lead to odd consequences which I'm keen to avoid. [17:32] bigjools: we can support this temporarily with symlinks on the repository server, it doesn't necessary has to be represented in the model. [17:32] gmb, Ah, I underestimate your keenness :) [17:33] the key breaking is necessary since the the group in control of the PPA will change. [17:33] cprov: I don't really want symlinks all over the PPA root directory though [17:34] bigjools: agreed, it's a favor we can offer to make the transition less painful. [17:34] how about: sneak-in the new URL via a package update of the old team [17:34] eeeeeeew! [17:34] cprov: right, we delete them after a while [17:34] cprov: would avoid symlinking ;-) [17:34] heh [17:36] apachelogger: yes, but I rather avoid going down this path, it's super-wrong security wise. [17:38] cprov: the only other option would be to create an app that tells the user what is going on and how to migrate, which is most preferrable security wise I suppose ... though the one which is the least comfortable for the user [17:39] eventually offer the user to add the new URL automagically, which would make it less of a PITA I suppose [17:42] apachelogger: right, repository EOL isn't an easy problem === rmcbride_ is now known as rmcbride [17:42]