[01:35] <lifeless> kiko-zzz: looks ok to me
[05:06] <dilys> Merge to devel/launchpad: [trivial]  Add small script to rebuild data in indexes on an online database (r2995: Stuart Bishop)
[08:37] <mick_linux> hi everyone
[08:37] <mick_linux> i'm just creating a team and i was wondering about the Contact Email Address
[08:37] <mick_linux> my team is already subscribed to a mailinglist
[08:38] <mick_linux> can i have them use the list to get email notifications?
[10:00] <mantiena-baltix> hi all
[10:00] <mantiena-baltix> mick_home, I think yes, you can use any email adress
[10:04] <SteveA_> hi
[10:07] <mick_home> under email, i put the mailinglist
[10:07] <mick_home> mantiena-baltix, i was wondering if that is ok
[10:29] <mantiena-baltix> SteveA, labas rytas ;)
[10:45] <mantiena-baltix> SteveA, it seems launchpad still has many problems :( I can't add Baltix milestone, when I go to http://launchpad.net/distros/baltix/+addmilestone I get an error about permissions :(
[11:57] <\sh> guys, I wanted to say: Great Job, and thanks for your hard work :) of merging bugzilla to malone :)
[12:40] <kiko-zzz> thanks \sh 
[12:40] <kiko> mantiena-baltix, I'll fix that today.. I meant to but forgot.
[12:47] <jordi> kiko: hey
[12:47] <kiko> hey jordi man
[12:47] <jordi> kiko: what about the +translations patch?
[12:47] <jordi> kiko: we need to get that merged or discussed or whatever
[12:47] <kiko> you know that your rosetta docs is one of the only emails in my inbox?
[12:47] <kiko> the one I wrote at ubz?
[12:47] <kiko> I have it stashed away in a branch here
[12:47] <kiko> this week I can land it
[12:48] <jordi> what do you mean with my rosetta docs mnail?
[12:48] <jordi> ok
[12:48] <jordi> because the current page is fucked up
[12:49] <jordi> for example, bum has templates in distro and product, but +translations only shows distro
[12:49] <kiko> oh
[12:49] <kiko> my patch fixes that too?
[12:49] <jordi> 'I can't remember
[12:49] <jordi> I suspect not though :)
[12:50] <kiko> me neither
[12:50] <kiko> but I like the sound of that
[12:50] <jordi> heh
[12:50] <matsubara> good morning!
[12:51] <jordi> hmm
[12:51] <jordi> do you have a lp install with current data, kiko?
[12:51] <kiko> with production data? not me
[12:51] <jordi> hm
[12:51] <jordi> well, the bum thing might be a different issue
[12:52] <jordi> kiko: https://launchpad.net/products/bum/+translations shows only the ubuntu template
[12:53] <kiko> yeah, which is broken broken broken
[12:53] <jordi> the translatable branches portlet does show the other one
[12:53] <jordi> but it should be the other way round, given that bum is an official rosetta product
[01:08] <mantiena-baltix> kiko, thanks
[01:09] <dilys> Merge to devel/launchpad: [trivial]  Fix for bug 5505: Bug nicknames no longer used. Fixes traversal by implementing an IBugSet.getByNameOrID method, and using that in places which traverse to bugs (r2996: kiko)
[01:10] <kiko> rock and roll!
[01:11] <kiko> let's use that tree to fix mantiena-baltix's problem
[01:19] <jamesh> hi kiko
[01:19] <kiko> hello jamesh 
[01:19] <kiko> how are you?
[01:19] <jamesh> good
[01:20] <kiko> congratulations on the amazing work
[01:20] <jamesh> so the migration seems to have gone fairly well
[01:20] <kiko> the transition was smooth
[01:20] <jamesh> thanks for doing the announcement
[01:21] <kiko> no problem
[01:21] <kiko> I added a banner to bugzilla's show_bug page too
[01:21] <jamesh> brilliant
[01:22] <kiko> elmo helped me with the necessary access
[01:23] <kiko> jamesh, bradb had suggested some bugs didn't have watches set?
[01:24] <kiko> I can't remember now which one though, that was a bit hectic
[01:33] <kiko> hey stub 
[01:33] <kiko> how's it going?
[01:33] <kiko> I had a test question if you could help me with it
[01:33] <kiko> shouldn't be too hard
[01:34] <stub> Yo
[01:35] <kiko> stub, I want to understand why doc/milestone.txt has no login() call and still works.
[01:35] <kiko> stub, I want to test permissions there, but apparently the script is running with full privileges
[01:35] <kiko> so login() is a nop
[01:35] <stub> Maybe a previous test logged in and didn't logout in the teardown
[01:36] <kiko> hmm are doctests susceptible to that?
[01:36] <stub> Sure. Nothing really different about doctests
[01:36] <kiko> do you need to manually log out in them then?
[01:37] <stub> I don't know - I havn't looked at the login() stuff that was put together for the test framework
[01:37] <kiko> there's no logout calls in the doc/ directory, so obviously I'm confused :)
[01:38] <stub> logout call would likely be in test_system_documentation
[01:38] <stub> LaunchpadFunctionalTestCase's tear down calls it
[01:38] <kiko> hmmm
[01:38] <stub> What happens if you just add a login call?
[01:38] <kiko> I don't think that's it
[01:38] <kiko> nothing
[01:38] <kiko> I login as anonymous
[01:39] <kiko> and it still works
[01:41] <stub> Is the object you are dealing with security wrapped? ie. you retrieved it from a FooSet Utility, and that utility was a securedutility and not the default insecure one?
[01:41] <kiko> hmmm, good question. I should print it out.
[01:42] <kiko>     + <security proxied canonical.launchpad.database.milestone.MilestoneSet instance at 0x31b01210>
[01:42] <kiko> that's crazy
[01:44] <stub> And the permissions in the content directive for MilestoneSet are correct?
[01:45] <kiko> ah, perhaps not, perhaps the web ui is broken elsewhere
[01:45] <kiko>     <!-- IMilestoneSet -->
[01:45] <kiko>     <securedutility
[01:45] <kiko>         class="canonical.launchpad.database.MilestoneSet"
[01:45] <kiko>         provides="canonical.launchpad.interfaces.IMilestoneSet">
[01:45] <kiko>         <allow interface="canonical.launchpad.interfaces.IMilestoneSet" />
[01:45] <kiko>     </securedutility>
[01:45] <kiko> hmmm
[01:46] <kiko> so if I read that right there's no permissions in place to handle that, right stub?
[01:46] <kiko> it's free-for-all
[01:46] <kiko> so hmm.
[01:46] <stub> Public read, and nobody can set I think (not sure about the default set)
[01:46] <stub> c/set/write for less confusion
[01:47] <kiko> I think any methods on IMilestoneSet are allowed, no?
[01:47] <kiko> anyway
[01:47] <kiko> hmmm
[01:47] <kiko> how do I friggin test this thing without writing a pagetest
[01:48] <stub> Any methods defined in IMilestoneSet
[01:48] <kiko> this is unfortunate
[01:49] <kiko> I thought it would be possible to check the permissions with a doctest
[01:49] <stub> I thought your test was failing because you were expecting something to be protected that wasn't, so fixing the security declaration would do the trick? Or is MilestoneSet supposed to be free-for-all?
[01:49] <kiko> but now I realize we're using page-based permissions to control this distro-milestone addition
[01:49] <kiko> my test wasn't failing, i was trying to get it to fail to create a milestone when logged in as test@canonical.com
[01:50] <kiko> however, what we use to limit creation of milestones is page-based permissions
[01:50] <kiko> so I think I need a pagetest
[01:51] <kiko> First, let's see if Foo Bar can view the page to add a milestone to the
[01:51] <kiko> Ubuntu distribution.
[01:51] <kiko> foo bar is a superuser :)
[01:52] <stub> I still don't understand. Why not fix the permissions on MilestoneSet to launchpad.Edit or something is needed to create Milestones?
[01:53] <kiko> stub, I can do that, but I will still need a pagetest because the actual pages are also protected by permissions, right?
[01:53] <stub> Yes - that is correct.
[01:53] <stub> Well.. they wouldn't have to be if the permissions on the MilestoneSet were correct ;)
[01:53] <kiko> that's what I was driving at -- I didn't want to write a pagetest. I hate them. :)
[01:53] <kiko> hmmm
[01:54] <kiko> I don't know, because it's one specific method in IMilestoneSet that's forbidden
[01:54] <kiko> you can still query them
[01:55] <stub> Sure. You can split the IMilestoneSet interface into two, and use different permissions to protect the different interface components. Or use the attributes attribute to specify permissions for individual attributes rather than by the whole interface.
[01:56] <kiko> that's interesting!
[01:57] <kiko> hmmm. but there's a catch -- you don't want the guy to even see the form.
[01:57] <kiko> so unless I call a checkAllowedToCreate or something on IMilestoneSet..
[01:57] <kiko> (in the browser code)
[01:58] <stub> Sure. So if you want to test that the form redirects to the login page if you don't have permissions (or returns an error code), then you need a page test
[01:59] <kiko> right.
[02:12] <kiko> interesting. if you don't logout in a pagetest it blows up
[02:14] <sivang> stub: is there a way to undo what launchpad/utilities/launchpad-database-setup does? I ran into some sudo problem with it last night, how can I take my pg back to start, and retest with my "fixes" ? (AFAICT there are really minor sudo usage fixes to make ti work)
[02:14] <stub> Page tests need to use basic auth, don't they? Or are you doing non-pagetesty tests in between
[02:15] <kiko> stub, right
[02:15] <kiko> it ain't workin though :)
[02:15] <kiko> oh, it is
[02:15] <kiko> I missed a flush_database_updates
[02:15] <kiko> that's a bug, man
[02:15] <kiko> f_d_u is a wart 
[02:15] <stub> eek... a shell script!
[02:16] <sivang> stub: yeah :)
[02:17] <stub> Doesn't look like there is any way to back out the changes, except by reverse engineering the script and doing it manually
[02:17] <sivang> what if I purge my pg and reinstall?
[02:17] <stub> There is a DatabaseSetup doc I think on the wiki that takes you through things manually I think (might be up to date...)
[02:18] <stub> https://wiki.launchpad.canonical.com/DatabaseSetup
[02:18] <stub> Looks up to date
[02:18] <sivang> stub: ok, stil re: l-d-s do you think it owuld make sense to pythonfie it?
[02:18] <sivang> stub: k, thanks, I will try to follow it and try to make my rocketfuel take off :)
[02:19] <stub> I wouldn't invest time in it unless you are bored - it is rarely used.
[02:19] <sivang> stub: ah I see :) thanks for the note
[02:20] <kiko> stub, can you look at distribution.zcml and help me understand why +addmilestone isn't reachable by the distribution owner?
[02:20] <stub> I guess the shell script is more readable anyway
[02:20] <kiko> oh!
[02:20] <kiko> I see!
[02:20] <kiko> class EditDistribution(AdminByAdminsTeam):
[02:20] <kiko> stub, isn't that crazy?
[02:21] <kiko> shouldn't the distribution owner be allowed to edit it?
[02:21] <stub> I guess so
[02:22] <stub> Or maybe it is considered too scary - dunno
[02:22] <kiko> let me see what editing a distribution gets you
[02:22] <kiko>             permission="launchpad.Edit"
[02:22] <kiko>             set_attributes="displayname title summary description
[02:22] <kiko>                             translationgroup translationpermission
[02:22] <kiko>                             members owner bugcontact"
[02:22] <kiko>               name="+reassign"
[02:22] <kiko>         name="+addmilestone"
[02:23] <kiko> looks pretty harmless.
[02:23] <stub> I mean fallout. eg. can you screw up archives or the build system?
[02:27] <kiko> stub, not really
[02:27] <kiko> those attributes are all safe
[02:27] <kiko> the problem is with distroreleases
[02:27] <kiko> that's a bit more dangerous
[02:27] <kiko> but that's launchpad.Admin still
[02:28] <sivang> hmm,how can I see a bug that I've been CC'd to ? (or rather, search for it)
[02:30] <matsubara> sivang: +subscribedbugs ?
[02:32] <sivang> matsubara: so that means, that if someone wants me to see a bug report he's fileing, he can subscribe me to the bug?
[02:33] <kiko> he has to, sivang 
[02:33] <matsubara> sivang: I think so or assign it to you. :)
[02:35] <sivang> ok, so I'll revert to searching the bug report according the product and text :)
[02:39] <sivang> matsubara: is there a way to search by reporter?
[02:41] <matsubara> sivang: you can search using the advanced search
[02:41] <sivang> matsubara: how do you switch to it?
[02:41] <matsubara> or you can go the person in question page: launchpad.net/people/<person>/+reportedbugs
[02:41] <sivang> oh right, thanks!
[02:45] <matsubara> sivang: btw, here the URL for the advanced search on launchpad bugs. https://launchpad.net/products/launchpad/+bugs-advanced
[02:45] <matsubara> s/here/here is/
[02:46] <sivang> wow, that's nice
[02:53] <kiko> pqm give me one more failure AND THAT"S IT
[02:54] <kiko> you're HISTORY
[03:02] <\sh> OOPS-14B203
[03:02] <\sh> for the advanced search page when I search for MOTU as assignee...
[03:02] <\sh> but this behaviour is already filed as a bug :)] 
[03:03] <kiko> matsubara is slacking on fixing that
[03:05] <matsubara> \sh: do you know what is the bug number?
[03:06] <\sh> matsubara: let me have a look
[03:07] <matsubara> \sh: bug 5389 ?
[03:07] <Ubugtu> Malone bug 5389: "System error if you specify an inexistent assignee in any advanced search for bugs" Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Diogo Matsubara, Status: Confirmed http://launchpad.net/bugs/5389
[03:08] <matsubara> hmm it's not that, actually motu does exist.
[03:09] <\sh> matsubara: well...MOTU is the common name..but universe-bugs@... is resolved...
[03:10] <matsubara> \sh: I tried to reproduce, but it didn't give me an oops
[03:10] <\sh> matsubara: i just tried it again...and I can't reproduce it either...strange
[03:11] <\sh> but I had some oops pages today, when I search for people e.g. but the second try was successful
[03:11] <matsubara> \sh: maybe it's another bug. I'll check the log as soon as it becomes available.
[03:11] <\sh> thx :)
[03:15] <dilys> Merge to devel/launchpad: r=lifeless Fix for bug 1512: Admins creating products should be allowed to set owner and is_reviewed. Does just that: sets fieldNames in the ProductAddView constructor, handles the options in the createAndAdd handler, and tests it is sane. (r2997: kiko)
[03:16] <kiko> violence works
[03:22] <matsubara> kiko: wanna review the fix for bug 2982?
[03:31] <kiko> bug 2982?
[03:31] <Ubugtu> Malone bug 2982: "The bugs link on the user page should show all bugs that a user is involved with" Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Diogo Matsubara, Status: In Progress http://launchpad.net/bugs/2982
[03:31] <kiko> aieee
[03:31] <kiko> matsubara, sounds like a job for SUPER SALGADO
[03:31] <kiko> :)
[03:32] <matsubara> kiko: ok. :)
[03:55] <sivang> heh
[03:58] <sivang> phew, I found that bug I wanted../me hails the advanced search
[03:58] <kiko> you mean google?
[03:58] <sivang> kiko: hehehe
[03:59] <sivang> kiko: actually, using +bugs-advanced search , suprisignly , it worked.
[03:59] <kiko> stub fixed the fti borkage
[03:59] <kiko> so it should be better
[04:01] <sivang> yes, it seems so. what was to my side is that I assumed (correctly) Scott's merg account text bug wasn't still assigned, and I asked for "newest first", and in bug id/text I set "merge accounts". That did the trick :)
[04:03] <sivang> kiko: how do I reach the merge accounts page?
[04:03] <kiko> you need to kill a baby goat
[04:04] <sivang> erm, I'm ideaologically opposed to meet sacrifices :)
[04:04] <sivang> but I can kill an onion.
[04:04] <sivang> s/meet/meat/
[04:05] <kiko> read the bible, it's all about killing baby goats
[04:05] <kiko> you should know, you guys wrote it
[04:05] <kiko> but since you've got some moral problem with it I will do it for you
[04:06] <kiko> https://launchpad.net/people/+requestmerge
[04:06] <sivang> hehe
[04:06] <kiko> it is the best-guarded secret in launchpad
[04:07] <sivang> oh, why so?
[04:13] <sivang> hmm, will we have some sort of mechanism to save "bookmarks" of bugs I'm interested to view in a snap? sort of "My Launchpad Home" content?
[04:14] <sivang> (for instance, I now want to have 3 bugs grouped under such for easy viewing over the next coupld of days)
[04:35] <kiko-afk> suicide bike ride
[04:36] <sivang> kiko-afk: take care
[04:57] <ddaa> sivang: I have been vaguely thinking about something like that
[04:57] <ddaa> at least for branches
[04:58] <dilys> Merge to devel/launchpad: r=SteveA Fix for bug 3289: Can't edit gstreamer-midi bounty. Just add a field to the schema for +admin, and add a trivial test. (r2998: kiko)
[04:58] <ddaa> sivang: I would like if you could write up your use cases for bugs somewhere. For example on the wiki on to the launchpad mailing list.
[05:00] <sivang> ddaa: ah you're here :) Yes, it makes sense to come up with focused use cases for that. I might send ti to the mailing list then.
[05:01] <sivang> ddaa: bugzilla has something like "My Searches" but this could be extended in launchpad to hold not only bugs, as you just noted.
[05:03] <sivang> ddaa: how's you btw?
[05:03] <ddaa> I'm fine.
[05:04] <ddaa> Got the importd2bz transition under way, and back hacking on the branch registration UI.
[05:04] <sivang> oh, cool
[05:04] <ddaa> It's a treat to do some hacking after all the communication and deployment I have done lately.
[05:05] <ddaa> Also, the whole launchpad team (except stub) will be on a sprint about Bazaar integration in Launchpad at the end of march.
[05:05] <sivang> you as well, right?
[05:06] <ddaa> Well, if I understand correctly my job ATM, I'm basically lead Bazaar-Launcpad integrator... so I think I would need to clarify with the mgmt where I will stand in that sprint.
[05:06] <ddaa> At least, I'm excited about it.
[05:07] <ddaa> In particularly interested in putting up some vision thing to guide the work on that sprint. But deep thinking needs time, and I lack time.
[05:08] <ddaa> And right now, it's week-end, so I stay away from that.
[05:08] <sivang> I see, well, I'm pretty sure you can handle it. You're so deep into the inner workings (as demonstrated in person) .... :-)
[05:09] <ddaa> Being deep down in the gears and building a vision are two very different things.
[05:09] <sivang> ddaa: btw, do you kno wanything about the merge accounts dialog? I'm trying to figure out which account you need to be logged in as , when you want to remove the duplicated account? AFAICT , whose the duplicated account is a matter of preception ;)
[05:10] <ddaa> In the use case it was designed for, it's not a matter perception.
[05:10] <ddaa> The use case is something like:
[05:11] <ddaa> "I have created a launchpad account, and found out that one of my email addresses is associated to an automatically generated account. I want to get this email and the objects associated to the autogenerated account."
[05:11] <ddaa> Which ought to be pretty common for debian devels.
[05:12] <sivang> I see, ok, that cleared for me. thanks!
[05:14] <ddaa> Mh... the help text is indeed confusing.
[05:14] <sivang> ddaa: so following that use case, the newly created account by the DD is getting the ${NAME}X (X being a number) suffix ?
[05:14] <sivang> ddaa: that is the NOT auto genereated account
[05:14] <ddaa> sivang: can you file a bug about that help text?
[05:14] <sivang> ddaa: Keybuk already did, I'm trying to find out what to really have there and possibly devise a patch 
[05:15] <sivang> https://launchpad.net/products/launchpad/+bug/28477
[05:15] <Ubugtu> Malone bug 28477: "Merge text is confusing" Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: Unconfirmed
[05:15] <sivang> ddaa: that's why I was asking you these questions.
[05:16] <ddaa> I did not parse your previous question. And honestly I do not know how person names are set in Launchpad.
[05:16] <sivang> ok, thanks indeed for the previous tips.
[05:16] <ddaa> Basically, in the usual duplicate account case, the user can only log in using one account, and want to merge autogerated accounts that refer to the same user.
[05:17] <ddaa> But caveat, I've not been on the team doing that work. It's mostly educated guesses.
[05:18] <sivang> oh, I understand.
[05:24] <sivang> https://wiki.launchpad.canonical.com/PeopleMerge seems to be what I was looking for.
[05:25] <ddaa> yup
[05:28] <sivang> better be waiting for mpt , and ask him about the right text he thinks should be there.
[05:34] <nlindblad> hello
[05:50] <sivang> I'm out for a bit
[06:18] <sivang> jblack: fine, fighting to make my bzr push working again
[06:18] <jblack> What's wrong? 
[06:19] <matsubara> \sh: around?
[06:19] <\sh> matsubara: somehow :)
[06:20] <matsubara> \sh: I checked the logs about that bug you've run into today in the morning, it's a manifestation of the bug 5389
[06:20] <matsubara> \sh: have you reported it?
[06:21] <\sh> no not now...
[06:22] <\sh> but if the oops page is not showing up, it won't find anything, doesn't matter which name I enter (e.g. MOTU motu or the email address of this team)
[06:22] <sivang> jblack: I can't seem to install bzrtools: 
[06:22] <sivang> The following packages have unmet dependencies.
[06:22] <sivang>   bzrtools: Depends: bzr (= 0.7+200512311044)
[06:22] <sivang> E: Broken packages
[06:22] <jblack> I emailed jbailey about that...
[06:22] <jblack> jbailey: Will the real jbailey please stand up? 
[06:22] <\sh> hmm..
[06:23] <\sh> oh those packages are broken, I thought the ubuntu ones
[06:23] <jblack> He's no on irc.
[06:23] <sivang> \sh: would you think removing the dependency may solve this?
[06:24] <\sh> sivang: adjusting the dep would make more sense
[06:24] <sivang> \sh: err, "adjusting" 
[06:24] <sivang> :)
[06:24] <\sh> bzrtools without bzr is not good :)
[06:25] <sivang> yes yes, I know, it happens to me all the time that I say not what I want to do...I need bug fix release :)
[06:25] <\sh> sivang: wait for jbaily or fetch the sources and do it on your local system :)
[06:25] <\sh> I'm just to tired to touch any packages today again...just finished the libXft.la clean up for kubuntu (i hope so)
[06:26] <sivang> \sh: no's ok :) Already did and installed cleanly.
[06:26] <sivang> \sh: have a nice evenin,g takesome time off!
[06:26] <sivang> jblack: ok, back to you ;-) :
[06:26] <sivang> jblack: pooh@tigershark ~/specs/home-user-backup/utilities $ bzr push sivan@mercury.linuxguru.net:/home/sivan/public_html/home-user-backup
[06:26] <sivang> bzr: ERROR: Parent directory of sivan@mercury.linuxguru.net:/home/sivan/public_html/home-user-backup does not exist.
[06:27] <sivang> any idea?
[06:27] <jblack> Hmm.
[06:27] <sivang> (IIRC I didn't change anything from how I used ti push before)
[06:28] <jblack> run "bzr plugins". Is bzrtools listed? 
[06:28] <sivang> let's see
[06:28] <sivang> pooh@tigershark ~/specs/home-user-backup/utilities $ bzr plugins
[06:28] <sivang> /usr/lib/python2.4/site-packages/bzrlib/plugins/bzrtools
[06:28] <sivang>         Various useful plugins for working with bzr.
[06:29] <jblack> cat ~/specs/home-user-backup/utilities/.bzr/x-push-data
[06:30] <sivang> cat: /home/pooh/specs/home-user-backup/utilities/.bzr/x-push-data: No such file or directory
[06:30] <jblack> what? 
[06:30] <sivang> weird
[06:30] <jblack> Try to cd into /home/pooh/specs/home-user-backup
[06:30] <jblack> then try "bzr push"
[06:31] <sivang> bzr: ERROR: No push location known or specified.
[06:31] <sivang> should I try with the location again?
[06:31] <jblack> What is the top directory of your branch? 
[06:32] <sivang> ~/specs
[06:33] <jblack> So you have ~/specs/.bzr  ? 
[06:34] <sivang> no, only ~/specs/home-user-backup/.bzr
[06:34] <jblack> ok. then ~/specs/home-userbackup is the top level dir of your branch.
[06:35] <jblack> cd into that, and run "bzr push sivan@mercury.linuxguru.net:public_html/home-user-backup"
[06:35] <jblack> see if that gives you love
[06:35] <sivang> pooh@tigershark ~/specs/home-user-backup $ bzr push sivan@mercury.linuxguru.net:public_html/home-user-backup
[06:35] <sivang> bzr: ERROR: Parent directory of sivan@mercury.linuxguru.net:public_html/home-user-backup does not exist.
[06:36] <jblack> cd into that, and run "bzr push sivan@mercury.linuxguru.net:/home/sivan/public_html/home-user-backup"
[06:36] <sivang> same..
[06:37] <jblack> I'm looking at the dir. its there.
[06:37] <sivang> weird, aien't it?
[06:37] <jblack> sivan@mercury:~/public_html/home-user-backup$ cd /home/sivan/public_html/home-user-backup
[06:37] <jblack> sivan@mercury:~/public_html/home-user-backup$
[06:38] <jblack> try "bzr push sftp://sivan@mercury.linuxguru.net/home/sivan/public_html/home-user-backup"
[06:38] <jblack> Thats a different protocol
[06:39] <sivang> seems to work now :)
[06:39] <sivang> yes, sftp does support locking when the other side is being changed, as opposed to rsync ? (I think I read that over some launchpad docs)
[06:40] <jblack> Yes
[06:40] <jblack> the catch is that sftp is much slower
[06:40] <sivang> ah, bad :-( It' just waited a couple of seconds, and then gave the same error
[06:40] <jblack> ping -c1 mercury.linuxguru.ent
[06:40] <jblack> ping -c1 mercury.linuxguru.net
[06:41] <sivang> 64 bytes from static-209-158-45-74.scr.east.verizon.net (209.158.45.74): icmp_seq=1 ttl=52 time=205 ms
[06:41] <jblack> I'm going to move your branch out of the way.
[06:41] <jblack> It thinks here that 
[06:42] <jblack> Ok. bzr push sftp://....
[06:42] <sivang> yep, retrying
[06:43] <sivang> same...
[06:44] <sivang> totally funky
[06:44] <jblack> There should be a file named ~/.bzr.log 
[06:44] <jblack> Post the last 50 lines or so to pastebin.com ? 
[06:44] <sivang> yep, just a sec.
[06:47] <sivang> http://mercury.linuxguru.net/~sivan/50_lines.txt
[06:48] <sivang> let me know if I shall use manual rsync for that, after it's saturday and we could continue this next week :)
[06:48] <jblack> I'm starting to consider that.
[06:49] <jblack> You can ssh in directly? 
[06:49] <jblack> Of course you can. you put that file there..
[06:49] <sivang> yeah :)
[06:49] <sivang> You were kind enough to let me have an account on your box :)
[06:49] <jblack> try "bzr push sftp://sivan@mercury.linuxguru.net//home/sivan/public_html/home-user-backup"
[06:50] <jblack> Try that. two slashes between host and home
[06:50] <jblack> oh. Duh
[06:50] <jblack> try "bzr push sftp://sivan@mercury.linuxguru.net/home/sivan/public_html"
[06:50] <jblack> Try with both one slash, and if that doesn't work, two
[06:50] <sivang> try only the last one then?
[06:50] <sivang> (17:50)
[06:50] <jblack> First try with one slash. Note the lacking home-user-backup at the end.
[06:50] <sivang> once with , and without two slashes?
[06:51] <sivang> yes
[06:51] <sivang> ok, one slash, no hub at the end not working
[06:51] <sivang> now:
[06:51] <jblack> its pushing
[06:52] <sivang> man, what's changed? why won't it work the old way?
[06:52] <jblack> I suspect...
[06:52] <sivang> (it also gives me : bzr: WARNING: Unable to update the working tree of: sftp://sivan@mercury.linuxguru.net/%2Fhome/sivan/public_html/home-user-backup/)
[06:52] <jblack> That's still there? Don't worry about that.
[06:57] <sivang> yay, seems finished. 
[06:59] <sivang> jblack: hmm, http://mercury.linuxguru.net/~sivan/home-user-backup/ suddenly gives me nothing, is that normal? (that is empy dir)
[06:59] <jblack> Yes, but you can still branch and merge it fine.
[06:59] <jblack> It just doesn't show the working tree.
[07:00] <sivang> I ssh'd in, ls -la gives only the .bzr dir
[07:00] <jblack> Yes. Thats fine.
[07:00] <sivang> ah ok, could just briefly explain what's happened? :-)
[07:00] <jblack> Bzr now pushes the branch and not the working tree.
[07:01] <sivang> what's the difference ?
[07:01] <jblack> The files you hack on and the rcs data are two different things.
[07:01] <jblack> The RCS data is pushed, but not the files you hack on.
[07:03] <sivang> I see, but what if I want to push to location to serve as a backup repository of the files I hack on? I would then need the original files with the changsets in the RCS to get the latest tree, no?
[07:03] <jblack> Then you start looking at the bzrtools rsync plugin.
[07:05] <sivang> hmm, but when I bzr branch from the remote location, I see all the files do get here in their latest version, so it does pushes changes as it did before. I think I didn't understand what you explained fully.
[07:05] <sivang> (bzr branch to a different new fresh checkout that it)
[07:05] <jblack> Ok. Lets try it a different way.
[07:06] <jblack> When you commit, does it go to the files you edit? 
[07:06] <sivang> what go to the files I edit?
[07:06] <jblack> I.E. if you hack up files you edit.. in your working tree... Then you run "bzr commit", do the files you edited change when you run commit? 
[07:07] <sivang> no, why should they? the new changes are already in.
[07:07] <jblack> Bzr commit must save or do something. But when you run commit, your files don't change, do they? 
[07:07] <jblack> Ok. So when you run commit, something must be saved somewhere, right? 
[07:07] <sivang> right :)
[07:07] <jblack> Ok. Thats the data thats pushed.
[07:08] <jblack> Thats the only stuff we need. For bzr to do its magic, it doesn't need the files you hack on. Just the record keeping that it keeps in the .bzr directory
[07:08] <sivang> ah ok, and that's actually does describe the tree from rev 1 to rev N, so having the files pushed as well as mere convinience for the web branch
[07:09] <jblack> Exactly.
[07:09] <sivang> .bzr already has all files in, in form of timstamped changes ontop of previous ones.
[07:09] <sivang> thanks for putting some clue into me, wrt that :)
[07:09] <jblack> Something like that. The changes are stored in a special way, but you have the right idea.
[07:10] <jblack> Those files that you _do_ edit, the ones not in the .bzr dir, we call that the "working tree"
[07:10] <jblack> There are all sorts of problems that crop up as concerns fixing up working trees.
[07:10] <jblack> during push, that is
[07:10] <sivang> I see.
[07:11] <jblack> They didn't affect you, but they did affect others. The problems were bad enough that we're not pushing working trees right now at all.
[07:11] <sivang> I see. I can imagine launchpad provided enough use cases for that :)
[07:12] <sivang> ok, so if I Want to let people web browse through my code, I will then use the rsync plugin or just rsync the tree to a web folder.
[07:13] <LarstiQ> that, or you can use a web thingy like hgweb
[07:15] <sivang> LarstiQ: I will have to look, but now I Have to go :)
[07:15] <LarstiQ> sivang: ok, ciao!
[07:15] <sivang> jblack: thanks alot for the help :)
[07:15] <jblack> welcome
[07:16] <sivang> jblack: I'll catch you again next week I suppose, I wonder if we still have anything to do together on RFS.
[07:16] <sivang> LarstiQ: thx for the tip.
[07:17] <jblack> It just needs one thing, that I hope to do tonight.
[07:17] <jblack> A way to update copies of launchpad checkouts
[07:19] <LarstiQ> fwiw, https://bazaar.launchpad.net/~lifeless/bzr/integration still gives me a 404
[07:22] <sivang> jblack: refuel does that no?
[08:01] <mantiena-baltix> kiko-afk, I still have problems on http://launchpad.net/distros/baltix/+addmilestone - I get an error about permissions :(
[08:50] <jblack> sigang: Huh. Thought I already answered. No, it doesn't.
[08:50] <jblack> It just updates the local mirror of rocketfuel-built, not the various copies of that, that you have.
[09:49] <jblack> I wonder if a slinky put on a upward moving escalator ever reaches the bottom.
[10:15] <LaserJock> I have a few (maybe stupid) questions about malone
[10:16] <LaserJock> is there a way to get the list of bug #s for a particular source package?
[10:25] <LaserJock> is there a plan/spec on being able to file Intent To Package and Request For Package bugs in Malone?
[11:42] <lamont-away> so how does one actually _close_ a bug, I wonder.
[11:44] <lamont-away> have I mentioned this week that hiding links all over the t)%&*&%*_^(+ page is an absolutely non-intuitive interface?
[11:46] <LaserJock> fix committed means you uploaded and fix released means its in the repo, ready to go?
[11:47] <LaserJock> that's how I interpret it anyway
[11:48] <Burgundavia> LaserJock, yes, https://launchpad.net/products/launchpad/+bug/809
[11:48] <Ubugtu> Malone bug 809: "[WISHLIST]  Easy way to request packages" Fix req. for: launchpad (upstream), Severity: Wishlist, Assigned to: Nobody, Status: Confirmed
[11:50] <lamont-away> LaserJock: I'm reading it as 'committed' --> checked into source control, released--> uploaded
[11:51] <Nafallo> I agree with lamont-away on that :-)
[11:51] <LaserJock> hmm, I'll have to think about that
[11:56] <LaserJock> so MOTU wannabes would change status from "In Progress" to "Fix committed" when they have added a patch that should be reviewed and uploaded and then a MOTU can change status to "Fix Released" when it gets uploaded
[12:00] <Nafallo> fix commited to what? REVU? bzr?
[12:01] <LaserJock> we often add debdiffs to the bug reports
[12:01] <LaserJock> that are then reviewed and uploaded
[12:01] <LaserJock> REVU mostly I guess