[12:12] <kiko> bradb, yeah, but I think I still would prefer seeing createBug modified.
[12:12] <kiko> that way this comes across as less of a hack
[12:12] <kiko> (for one, you would only need one if clause)
[12:15] <bradb> kiko: hm. it seems to me that getting that right shouldn't up to the caller though. i.e. the API should abstract away the concern of making sure someone (the .bugcontact or .owner) is subscribed to private bugs.
[12:15] <bradb> my other branch does that
[12:16] <bradb> what do you think? should the API rely on passing the correct subscribers for what the UI presents as a "built-in" part of the model?
[12:17] <kiko> bradb, well, hmmm.
[12:18] <kiko> so you're saying that the back-end will know who to subscribe depending on privacy/security settings?
[12:18] <bradb> kiko: yep
[12:19] <bradb> I made an API change to createBug in my other branch, collapsing security_related and private into just security_related, just like the UI presents that model, that I think will bite me now, but I can fix that on that branch. (I'll have to, to keep my tests passing.)
[12:27] <kiko> bradb, I think the safest way forward
[12:27] <kiko> is to not modify createBug other than allowing specifying subscribers
[12:27] <kiko> and then do everything in the browser code
[12:31] <bradb> sorry, phone
[12:31] <kiko> bradb, I think it's best not to
[12:31] <kiko> collapse
[12:31] <kiko> sure
[12:34] <bradb> kiko: is there a need to specify subscribers though, for our immediate use cases? the way it is now (even after uncollapsing the createBug flag back into two) is that you only have to worry about passing it the right values of security/privacy and the API will do the work for you.
[12:34] <bradb> that way it's consistent wherever it's used, e.g., email, xmlrpc, etc.
[12:35] <kiko> hmmm.
[12:35] <kiko> you actually have a good point there
[12:36] <kiko> the multiple interfaces
[12:36] <kiko> so tell you what -- I can't answer this today, my head is busy with my own code
[12:36] <kiko> but tomorrow first-thing I'll look at both patches
[12:36] <bradb> fair enough
[12:37] <bradb> but, maybe i should ping you on that tomorrow then, cause i'll need about 30 mins to uncollapse the flags to .createBug on the other branch
[12:44] <bradb> kiko: i moved my other branch into your queue, and put the landscape hack in there too. i'll followup tomorrow morning. thanks!
[12:45] <kiko> cool
[01:09] <kiko> man bzrlib tests hate me
[04:05] <lifeless> jamesh: so yeah, asterisk
[04:06] <jamesh> lifeless: what about it?
[04:06] <lifeless> do you have a local server ?
[04:06] <jamesh> no.
[04:06] <jamesh> Ekiga was able to talk to voip.canonical.com through my NAT firewall okay
[04:07] <jamesh> (an linux iptables firewall allowing outgoing connections)
[04:12] <lifeless> I can connect to 6701 ok, but I only get static back
[04:13] <lifeless> also, my registration do-dad fails
[04:15] <lifeless> is your sip running ?
[04:15] <jamesh> It is now
[04:15] <lifeless> I'll try calling
[04:16] <jamesh> sip:7543@canonical.com
[04:16] <lifeless> not available
[04:16] <jamesh> it crashed
[04:17] <lifeless> ha!
[04:17] <jamesh> try again
[04:18] <lifeless> are we connected ?
[04:19] <jamesh> I heard about .5 seconds that might be you
[04:19] <lifeless> I've just said a few sentences
[04:19] <jamesh> I could hear you saying something but couldn't make out what it was
[04:19] <lifeless> I heard 'can you hear me speak'
[04:20] <lifeless> and possibly keys
[04:20] <lifeless> but just static for the rest
[04:20] <lifeless> I just said hello four times
[04:20] <jamesh> I'm hearing "pulses" of audio
[04:21] <lifeless> I just badly quoted the quick brown fox thing
[04:21] <lifeless> did you hear any of it ?
[04:21] <jamesh> it's like I get every second half second of your audio (or something like that)
[04:21] <lifeless> in fact, I know, I'll read the voip setup page to you
[04:22] <lifeless> I just read the first section
[04:23] <lifeless> have you turned silence detection off ?
[04:23] <lifeless> I heard some of what you were saying
[04:23] <jamesh> it wasn't on when I checked
[04:23] <lifeless> but its like it fades out and then I get static
[04:25] <lifeless> I'm not hearing any speech at the moment
[04:25] <jamesh> I'
[04:25] <jamesh> m not speaking
[04:26] <jamesh> hang up and I'll try calling you
[04:27] <lifeless> I cannot register properly :(
[04:27] <lifeless> but try anyhow
[04:27] <jamesh> both spiv and me tried shtoom too, without success (complains about some unexpected packet or something)
[04:29] <lifeless> did you try to ring ?
[04:29] <jamesh> yeah.  It went through to your voice mail
[04:29] <lifeless> can you do me a favour
[04:29] <lifeless> go to accounts, properties of your canonical account, more options
[04:30] <lifeless> and tell me the realm
[04:30] <jamesh> sip:7543@canonical.com
[04:30] <jamesh> it got filled in automatically when I connected the first time
[04:30] <lifeless> yes
[04:30] <lifeless> the 'realm' is your sip account ?
[04:30] <jamesh> seems to be
[04:31] <kiko-zzz> stub, was the rollout today postponed?
[04:31] <mpt_> Gooooooooooooooooood afternoon Launchpadders!
[04:32] <mpt_> kiko-zzz, [01:20]  <stub> I got distracted and missed todays rollout. Anyone going to cry if I put it off until tomorrow quiet time?
[04:33] <stub> kiko-zzz: Rollout my yesterday was postponed, rollout today is being kicked off right now
[04:33] <stub> Get with the timezone, you are living in the past ya hippie!
[04:34] <lifeless> jamesh: so, I've filed an RT request on my login.
[04:35] <lifeless> trying a different sound card
[04:35] <stub> kiko-zzz: r3758 before you ask, as discussed with matsubara
[04:36] <lifeless> rtp drops packets failry regularly when ou overcommit
[04:36] <lifeless> so bandwidth - as long as we are 'in the ballpark' is usually fine
[04:37] <jamesh> the GSM codec seems to be doing a better job than the PCMU codec (that was picked before)
[04:37] <lifeless> I've also changed setup
[04:37] <lifeless> I'm using a builting sound card rather than an external
[04:38] <lifeless> but its not good enough yet :)
[04:38] <lifeless> so, lets swap cards back, no other changes, see if it stays better/worse
[04:38] <jamesh> can you try saying something again?
[04:39] <lifeless> did it crash ?
[04:39] <lifeless> ok, my usb headset is art of the problem for incoming audio for me
[04:40] <jamesh> no.  I was trying to get a recording so you could hear what you sounded like
[04:40] <lifeless> ah
[04:40] <stub> lifeless: Is there some magic on balleny making new branches and commits to the rocketfuel archive on balleny automatically get pushed to chinstrap?
[04:40] <lifeless> stub: the push is done by pqm
[04:41] <stub> ok. One less step to remember. I was wondering why pushing a fresh production branch pushed 0 revisions ;)
[04:42] <lifeless> stub: if you want to do it by hand, just do it by hand - bzr push sftp://chinstrap/home/warthogs/archives/rocketfuel/....
[04:42] <lifeless> stub: how are you making the production branches ?
[04:42] <stub> cd ~pqm/archives/rocketfuel/launchpad/production; bzr branch -r xxxx ../devel 1.xx
[04:43] <lifeless> you just made 1.68
[04:43] <lifeless> ?
[04:43] <stub> Yes
[04:43] <lifeless> bzr info | grep publish
[04:43] <lifeless> will tell you where it is set to push to
[04:43] <lifeless> the reason it says '0 revisions pushed' is because of a bug in bzr
[04:44] <lifeless> which is fixed in recent bzr
[04:44] <stub> ok. So I need to manually push production branch updates to chinstrap? Or not?
[04:45] <lifeless> if you do them by hand, push them by hand
[04:45] <lifeless> but the only thing that you need to do by hand is the branch creation
[04:45] <lifeless> after that, pqm can handle it all for you
[04:45] <stub> pqm can handle cherry picks now?
[04:46] <lifeless> erm, no.
[04:46] <lifeless> the only TWO things you need to do by hand ...
[04:46] <stub> ok :)
[04:46] <lifeless> (unless you make a short branch with the cherry pick, and ask pqm to merge the entire branch)
[04:46] <lifeless> which is what I would do
[04:48] <stub> Which means I won't need to disable pqm, which means I won't forget to reenable it ;)
[04:48] <lifeless> :)
[04:48] <lifeless> jamesh: so, how about this
[04:49] <lifeless> I ring you on ekiga
[04:49] <lifeless> I ring you on your landline
[04:49] <lifeless> then talk into one, listen on the other ;)
[04:49] <lifeless> and I can let you hear how you sound
[04:50] <jamesh> okay
[04:50] <lifeless> whats the best landline to ring you on?
[04:52] <jamesh> I /msg'd you the number
[04:52] <lifeless> not all that bad
[04:52] <lifeless> listen to this thought
[04:53] <lifeless> I'm going to put mylandline against hte headset
[04:53] <lifeless> you speak anytime
[04:53] <lifeless> can you hear the ripple?
[04:53] <lifeless> yup
[04:54] <lifeless> any more tests you can think of?
[04:54] <jamesh> actually, I do have one more test
[04:54] <jamesh> conf call
[04:54] <lifeless> oh, hang on
[04:55] <mpt_> Ekiga appears to be a timesink :-/
[04:58] <lifeless> I've tried to get in to the conf call
[04:58] <lifeless> cant tell if I mad eit
[04:59] <jamesh> I heard the "new participant tone", and can hear a similar ripple static effect
[04:59] <lifeless> sounds like I'm in :(
[04:59] <lifeless> does that help the ripple?
[05:00] <jamesh> not really.
[05:00] <jamesh> the conference call line was silent til you joined
[05:00] <lifeless> better ?
[05:00] <jamesh> yep
[05:00] <jamesh> can you try speaking?
[05:00] <lifeless> ok, thats gain on the microphone
[05:01] <lifeless> I am saying testing testing 1 2 3
[05:01] <lifeless> if you are sayin anything I cant hear it
[05:01] <lifeless> I cannot hear myself even
[05:02] <lifeless> ok I'm hearing testing 1,2,3
[05:02] <jamesh> I can hear that but the audio is very bad (about the same as the non-conf-call)
[05:02] <lifeless> with extra ripple
[05:02] <jamesh> can you hear me now?
[05:02] <lifeless> I think the ripple is distortion
[05:02] <lifeless> I can just make it out
[05:03] <lifeless> I am guessing that any noise is transmitted and the ripple introduced by $something
[05:03] <jamesh> I was getting much better audio with steve/stub/spiv on monday, so I wonder if it is something with your link?
[05:03] <jamesh> (or maybe something else has changed?
[05:03] <lifeless> jamesh: who knows :(
[05:03] <lifeless> I'm going to try linphone now
[05:04] <lifeless> can you keep ekiga running ?
[05:04] <jamesh> yep.
[05:04] <jamesh> I'm still in the conference call
[05:24] <lifeless> jamesh: can you hang up on that
[05:25] <jamesh> okay.  I've hung up
[05:25] <lifeless> is it any better ?
[05:25] <jamesh> yes
[05:25] <lifeless> cool
[05:25] <lifeless> you sound like shit still :)
[05:26] <jamesh> I can understand you now (which is good), but the sound quality isn't that great
[05:26] <lifeless> clearly better than before :)
[05:26] <lifeless> you are rippling badly
[05:26] <lifeless> I can *just* make it out
[05:26] <jamesh> what did you change at your end?
[05:26] <lifeless> kphone rather than ekiga
[05:26] <lifeless> took a little guesswork to get it running
[05:27] <lifeless> could you ring me
[05:27] <lifeless> rotfl
[05:27] <lifeless> ok, it answered
[05:27] <lifeless> then blew up
[05:27] <lifeless> hmm, this is interesting
[05:28] <lifeless> I had an asoundrc file
[05:28] <jamesh> asoundrc files seem to be evil
[05:29] <lifeless> entirely
[05:29] <jamesh> but I might just be superstitious
[05:29] <lifeless> try ringing me ?
[05:29] <lifeless> this is ekiga agaon
[05:29] <lifeless> *again*
[05:29] <jamesh> crap again
[05:29] <lifeless> ditto
[05:30] <lifeless> yeah 
[05:30] <lifeless> but if we can get better it wont cause headaches
[05:31] <lifeless> can you head me ?
[05:31] <jamesh> I heard something and then nothing
[05:31] <lifeless> can you speak ?
[05:31] <jamesh> I am now
[05:31] <lifeless> I'm hearing nothing :(
[05:31] <lifeless> run gme ?
[05:31] <jamesh> gme?
[05:32] <lifeless> ring me :)
[05:32] <lifeless> I can you hear you
[05:32] <lifeless> can you hear me ?
[05:33] <jamesh> it is like it was earlier: too broken up to understand
[05:33] <lifeless> argh
[05:33] <lifeless> well, at least ekiga has decided to register now
[05:34] <lifeless> hhhhaaa
[05:34] <lifeless> my bandwitdh use is at 100%
[05:34] <lifeless> 1349.449356 82.211.81.194 -> 192.168.1.5  SIP Status: 401 Unauthorized
[05:34] <lifeless> 1349.450587  192.168.1.5 -> 82.211.81.194 SIP Request: CANCEL sip:RobertCollins@voip.canonical.com
[05:34] <lifeless> 1349.452665  192.168.1.5 -> 82.211.81.194 SIP Request: SUBSCRIBE sip:RobertCollins@voip.canonical.com
[05:34] <lifeless> 1349.507049 82.211.81.194 -> 192.168.1.5  SIP Status: 200 OK
[05:34] <lifeless> 1349.575274 82.211.81.194 -> 192.168.1.5  SIP Status: 401 Unauthorized
[05:34] <lifeless> 1349.576474  192.168.1.5 -> 82.211.81.194 SIP Request: CANCEL sip:RobertCollins@voip.canonical.com
[05:34] <lifeless> 1349.579509  192.168.1.5 -> 82.211.81.194 SIP Request: SUBSCRIBE sip:RobertCollins@voip.canonical.com
[05:34] <lifeless> 1349.634694 82.211.81.194 -> 192.168.1.5  SIP Status: 200 OK
[05:35] <lifeless> and now ekiga sulks again
[05:35] <jamesh> I was able to register with shtoom today
[05:36] <jamesh> argh.  "[shtoom.rtp.protocol.RTPProtocol (UDP)]  received packet with unknown PT 1" errors still
[05:36] <lifeless> heh
[05:37] <lifeless> well, thats enough of my life on this
[05:37] <lifeless> I'll wait for the admins response
[05:37] <lifeless> lunchtime
[09:05] <carlos> morning
[09:07] <SteveA> hi
[09:18] <sivang> morning
[09:51] <stub> BjornT: https://launchpad.net/products/malone/+ticket/1179
[09:51] <stub> BjornT: It seems I should retarget the bugtasks associated with Ubuntu Edgy, but I'm not sure about bugtasks associated with just Ubuntu.
[09:55] <stub> Which I guess is important, as there are 466 associated with just Ubuntu and none associated explicitly with Edgy.
[09:56] <stub> I suppose they should all be reassigned.
[09:57] <BjornT> stub: right, that's a bit tricky since we don't have good release targeting yet. but as you say, they should all be reassigned since atm we use mostly non-release-specific tasks to track bugs.
[09:58] <stub> Only those that are not Rejected or Fix Released?
[09:58] <stub> Or those statuses too?
[10:02] <BjornT> well, it really depends on when they were rejected or fixed... maybe you should start with retargeting only the open bugs and see if someone complains?
[10:04] <carlos> It's just me or chinstrap is ignoring any ssh certificate?
[10:14] <ddaa> BjornT: ping, please review bzr-native today
[10:14] <ddaa> otherwise, I won't be able to handle it before tuesday
[10:18] <BjornT> ddaa: sure, i was planning to do it today.
[10:37] <BjornT> stub: do have time to take a look at https://launchpad.canonical.com/SimpleBugKeywords? (there's an XXX for you in the Implementation section)
[10:41] <stub> BjornT: We only want to exact matching on tags? So searching for 'laptop' will not match the 'laptops' tag?
[10:43] <stub> If so, the best implementation would be a three column BugTag table containing just (id serial, bug int, tag text)
[10:44] <stub> Searching would involve adding this table into the existing disgusting search query as an outer join using simple string matching.
[10:44] <stub> Tags should be automatically converted to lowercase and stored that way.
[10:44] <BjornT> stub: well, there are two cases. we want to be able to limit the search on a specific tag, then i'd say we want exact matches only. but if you type in 'laptop' on the +bugs pages, matching the 'laptops' tag is ok, but not necessary.
[10:45] <stub> I think we want exact matching on tags
[10:46] <BjornT> yeah, that makes sense.
[10:46] <stub> esp. as the UI might end up being a select box 
[10:48] <stub> I'll add the relevant indexes when reviewing the db patch. I think we will need to but might be able to get away with one.
[10:48] <stub> c/to/two
[10:49] <stub> BjornT: That all you need from me?
[10:50] <BjornT> stub: yes, i think that was all, thanks.
[10:50] <stub> last time I looked at the search code it seemed to be straining at the seams and about to explode (I was looking at adding in comment searching). It might need some refactoring :-(
[10:51] <BjornT> yeah, it's quite icky.... how would you refactor it?
[10:51] <stub> It might be worth constructing a view and searching that for distinct bug id's.
[10:52] <stub> Nah... the reason it is 'icky at the moment is to avoid joining tables that are not needed for the current query.
[10:53] <stub> I have to head of for a bit anyway
[10:53] <BjornT> ddaa: could you send me a diff of the bzr-native branch? the pending-reviews page is down.
[10:54] <ddaa> sure
[10:54] <BjornT> thanks
[10:59] <ddaa> BjornT: sent
[10:59] <mpool> SteveA: hi, got a sec?
[11:01] <SteveA> mpool: hello
[11:03] <SteveA> mpool: want to talk using the new canonical sip thing/
[11:03] <SteveA> ?
[11:07] <mpool> SteveA: ah, i haven't managed to successfully register yet
[11:07] <mpool> not sure if it's a problem on the server or with my network
[11:08] <SteveA> then skype?
[11:10] <ogra> hi, i got a little problem with a bzr checkout from LP
[11:10] <ogra> i ran bzr checkout sftp://ogra@bazaar.launchpad.net/~ubuntu-core-dev/ubuntu-seeds/edubuntu.edgy which worked fine
[11:11] <SteveA> mpool: ping
[11:11] <ogra> after that i ran bzr checkout sftp://ogra@bazaar.launchpad.net/~ubuntu-core-dev/ubuntu-seeds/ubuntu.edgy
[11:11] <mpool> SteveA: hi, just talking to robertc 
[11:11] <mpool> maybe tomorrow or later?
[11:11] <ogra> which now results in: ssh: connect to host bazaar.launchpad.net port 22: Connection refused
[11:12] <SteveA> mpool: okay, tomorrow is fine with me
[11:12] <mpool> ogra: there was recently a bug where sftp urls need a trailing slash
[11:12] <mpool> i think that's fixed in .dev but you might still have it
[11:12] <ogra> i'll try that ... 
[11:13] <ogra> mpool, i just heard we have general ssh probs ... i'll come back if that doesnt solve it 
[11:13] <SteveA> telsteve@einheit:~$ telnet bazaar.launchpad.net 22
[11:13] <SteveA> Trying 82.211.81.254...
[11:13] <SteveA> telnet: Unable to connect to remote host: Connection refused
[11:13] <SteveA> 
[11:13] <SteveA> it's a lower level problem
[11:13] <mpool> Someone Else's Problem
[11:16] <elmo> spiv/lifeless/stub: ?
[11:16] <elmo> never mind
[11:20] <carlos> stub: I think I found and fix teh problem with posubmission duplicates
[11:20] <carlos> stub: but I cannot test it until I get access to our DC, I will do a full run on staging and try to set the 'unique' restriction to be 100% sure it's working now
[11:21] <elmo> ok, change my mind, I definitely need spiv/lifeless/stub
[11:27] <ddaa> BjornT: if you have some time this afternoon, I'd like to have a call about https://launchpad.net/products/launchpad-bazaar/+spec/vcs-imports-ui-test-plan
[11:27] <ddaa> Leave for family lunch now
[11:27] <ddaa> * Leaving
[11:32] <elmo> ddaa: gone already?
[11:32] <ddaa> Almost
[11:32] <ddaa> got to catch a bus
[11:36] <elmo> blah
[11:38] <lifeless> elmo: ?
[11:39] <lifeless> elmo: if you add ':' it highlights the channel :)
[11:39] <elmo> lifeless: I rebooted vostok, restarted the push sftp server, but it's listening on the wrong port
[11:39] <lifeless> elmo: wrong IP or wrong socket 
[11:39] <lifeless> or btoh ?
[11:39] <elmo> well, kind of both
[11:39] <elmo> it's listening on 0.0.0.0:5022
[11:39] <elmo> should be on 82.211.81.254:22
[11:39] <lifeless> funky do-dah
[11:40] <lifeless> and ssh in there is still a bit whacky right ?
[11:40] <lifeless> nevermind, just my link going spastic
[11:41] <elmo> ssh to vostok itself is about the only working ssh in the DC right now ;-)
[11:41] <lifeless> meepification
[11:41] <lifeless> how did you start it ?
[11:42] <elmo> sudo su - supermirror, ./start-bzr-sftp
[11:43] <lifeless> config.supermirrorsftp.port is the magic, lets see
[11:44] <jamesh> you might have to set LPCONFIG to something
[11:44] <lifeless> jamesh: indeed, but its meant to be set by the script
[11:44] <lifeless> to avoid forcing huge amounts of cache into the sysadmins heads
[11:44] <lifeless> and indeed, that would appear to be AWOL
[11:46] <lifeless> elmo: look better ?
[11:46] <elmo> tcp        0      0 0.0.0.0:5022            0.0.0.0:*               LISTEN     7768/python
[11:46] <elmo> nope
[11:47] <lifeless> oh, foo
[11:48] <lifeless> are you sure there is not an init script ? I swore that we made one
[11:48] <lifeless> anyhow, fixing for real I think
[11:48] <lifeless> fixed I think
[11:49] <lifeless> that script could not have been working though, so there must have been another way of starting it :)
[11:49] <elmo> oh, crap there is
[11:49] <elmo> sorry, I thought I looked for one
[11:49] <elmo> shall I see if it works?
[11:49] <lifeless> lets do that
[11:49] <elmo> (I have no idea why it didn't work on boot)
[11:49] <lifeless> and if it does rm that start-bzr-sftp
[11:49] <lifeless> you drive, shout if you want me to peek
[11:50] <elmo> yeah, script works fine
[11:50] <elmo> removed that file
[11:50] <lifeless> heh
[11:50] <elmo> I'll investigate later why it didn't work after boot
[11:50] <elmo> sorry
[11:50] <lifeless> np
[11:50] <lifeless> whats the script name ?
[11:50] <elmo> /etc/init.d/bzr-sftp
[11:51] <lifeless> (stub and I need to run it to bounce the server during upgrades)
[11:51] <lifeless> danke
[12:03] <elmo> carlos: ?
[12:05] <carlos> elmo: hi
[12:09] <elmo> carlos: I need to kill mawson
[12:09] <elmo> can I kill your cronjob ,or is it life-endingly important?
[12:10] <carlos> elmo: kill it, I will execute it later
[12:10] <elmo> thanks
[12:11] <BjornT> ddaa: sure, i should have time for a call later.
[12:51] <SteveA> I'm going to be offline for a while, to get some coding done.  Send me a txt message or phone if you need something urgently.
[12:55] <elmo> LP is going down in 15 minutes, ETD is 10 mins
[01:16] <elmo> LP is back, yell if stuff doesn't work
[01:36] <elmo> *.ubuntu.com, launchpad.net etc. are disappearing for a couple of minutes for some essential maintenance
[01:41] <carlos> danilos: hi
[01:41] <danilos> carlos: hi
[01:41] <carlos> how's going?
[01:41] <danilos> carlos: I am having some problems with launchpad schemas
[01:42] <danilos> I can't set the database up
[01:42] <carlos> danilos: what's the error you get?
[03:06] <salgado> stub, around?
[03:08] <rodarvus> good morning
[03:09] <rodarvus> there is a LP account called 'rodarvus', which apparently was automatically created long ago, when translations were imported into Rosetta
[03:09] <rodarvus> (these translations were done by me, many years ago)
[03:10] <rodarvus> I wonder if it is possible to merge this account into my current LP account 'rodrigonovo'
[03:10] <rodarvus> I mean, in other words, if it is possible for me to use 'rodarvus' as LP account, inheriting all history my current account already has
[03:11] <carlos> rodarvus: yes, that's possible
[03:11] <carlos> rodarvus: I guess you have control over the email addresses in both accounts, right?
[03:11] <rodarvus> no, I don't have control over 'rodarvus'
[03:12] <rodarvus> it was apparently automatically created
[03:12] <carlos> rodarvus: but it has an email address associated
[03:12] <rodarvus> oh?
[03:13] <rodarvus> https://launchpad.net/people/rodarvus says nothing about email addresses
[03:13] <salgado> rodarvus, I think I can find the email address associated with it.
[03:14] <salgado> all launchpad accounts have an email associated with them. sometimes they're not validated, and thus won't be shown
[03:14] <carlos> rodarvus: you can also try to merge it from: https://launchpad.net/people/+requestmerge
[03:15] <salgado> rodarvus, it's your email @conectiva that is associated with that account
[03:15] <carlos> but that will migrate all info from rodarvus into rodrigonovo
[03:15] <rodarvus> salgado: right, as I said
[03:15] <rodarvus> I last owned this email 6 years ago
[03:15] <salgado> you don't have access to that anymore?
[03:15] <carlos> instead of rodrigonovo into rodarvus
[03:15] <rodarvus> so I surely didn't created this account myself :)
[03:15] <salgado> in that case we need a launchpad admin
[03:15] <rodarvus> salgado: this email doesn't exists anymore - since six years :)
[03:16] <rodarvus> long before LaunchPad was created, I think :)
[03:16] <rodarvus> salgado: who should I contact for that?
[03:17] <salgado> rodarvus, any of the members of https://launchpad.net/people/admins
[03:17] <carlos> rodarvus: https://launchpad.net/people/admins
[03:17] <salgado> rodarvus, kiko, SteveA or stub
[03:17] <rodarvus> salgado, carlos: thanks!
[03:17] <salgado> preferably then, I mean
[03:17] <rodarvus> *nods*
[03:21] <stub> salgado: yo
[03:21] <salgado> hi stub. I wanted to check with you when your shipit constraints branch went to production, if it already went
[03:21] <stub> I was about to ask the channel....
[03:22] <stub> Is anyone going to cry if the production rollout is delayed a second day until tomorrow? Todays rollout didn't happen during quiet time due to problems on the LAN.
[03:22] <stub> salgado: No export has been run yet today btw.
[03:22] <stub> (and no complaints in the mailbox!)
[03:25] <salgado> stub, can you run https://chinstrap.ubuntu.com/~dsilvers/paste/fileBuwnLM.html on production, to see if the postcode problem will bite us today?
[03:27] <stub> slow query....
[03:28] <salgado> spiv, I guess my real-karma-context branch is approved?
[03:30] <salgado> stub, anyway, I think we can fire the export script. there shouldn't be many postcodes causing problems and we can easily fix the csv after the export, if needed
[03:30] <stub> Once we know what they are
[03:30] <salgado> we can even find them in the CSV
[03:30] <spiv> salgado: yes :)
[03:31] <salgado> the problem happens when importing the CSV in OpenOffice
[03:32] <stub> ahh... ok. Export running now anyway.
[03:33] <salgado> stub, cool
[03:41] <kiko-zzz> morning!
[03:42] <kiko> stub, how does my makefile patch look?
[03:45] <stub> I'll have a look now, but I suck on Makefiles
[03:45] <stub> I don't know if spiv did any work on that today - I assigned him to the spec this morning.
[03:46] <stub> erm... yesterday. forget time.
[03:46] <kiko> stub, the changes are pretty simple
[03:46] <stub> looks fine
[03:47] <stub> r=stub
[03:47] <kiko> okidok
[03:47] <kiko> let me try and land that then
[03:47] <stub> lifeless gave me permission to update pqm.conf too (although he might want to do it if he is still around)
[03:47] <kiko> stub, I'll ping or email you when it lands so you can click it then
[03:54] <ddaa> BjornT: can we have a call soon?
[03:54] <kiko> man bzr is fast
[03:55] <ddaa> kiko: no, it's just less slow than it used to be ;)
[03:55] <kiko> no, it's blazingly fast
[03:55] <ddaa> BjornT: I would like some guidance and sanity checking on https://launchpad.canonical.com/VcsImportUiTestPlan
[03:55] <ddaa> kiko: there was an upgrade recently?
[03:55] <kiko> not sure
[03:56] <kiko> but working with sqlobject is just so trivial with it now
[03:56] <ddaa> oh right
[03:56] <kiko> it used to be fast
[03:56] <ddaa> this sort of small branch are fast
[03:56] <kiko> but it is now VERY fast
[03:56] <stub> salgado: export is done if you want to check it
[03:56] <salgado> stub, thanks, I'll do that
[03:56] <ddaa> for a lot of ops, a significant factor is the ~0.5s it takes to start python and load the modules...
[03:57] <ddaa> kiko: you could get more snappiness using bzr service or bzr shell
[03:57] <kiko> what's that?
[03:57] <ddaa> plugins
[03:58] <kiko> what do they do?
[03:58] <kiko> ah, run bzr as a service?
[03:58] <kiko> avoiding startup time?
[03:58] <ddaa> bzr service is an horribly insecure prototype that sets up a bzrlib runner listening to a socket
[03:58] <kiko> heh
[03:58] <ddaa> and an ultra-thin commandline client in C
[03:59] <ddaa> bzr shell is the spiritual son of aba, tlash, etc.
[03:59] <salgado> stub, did that query finish or you stopped it?
[03:59] <stub> salgado: still running
[03:59] <stub> (told you it was slow!)
[03:59] <ddaa> a simple shell that runs a REPL in python, so you do not have to type "bzr" and so you do not have to spawn python for every command
[03:59] <salgado> wow, that is really slow. :-(
[04:00] <ddaa> non-trivial shell commands are handed down do the system shell
[04:00] <stub> I'll kill it anyway
[04:01] <danilos> hey guys, do I have to restart launchpad whenever I change anything in the code? or is there an easier way to refresh stuff? (removing .pyc didn't help)
[04:02] <kiko> danilos, you have to restart.
[04:02] <kiko> there is no easier way
[04:02] <danilos> kiko: ok, thanks
[04:02] <kiko> danilos, reevaluating the code that was modified is a non-trivial enterprise
[04:02] <kiko> stub, what of https://launchpad.canonical.com/LaunchpadProductionStatus ?
[04:03] <ddaa> python is not yet quite OO enough to allow on comfortable on-the-fly modification of classes
[04:03] <kiko> ddaa, it doesn't have to do exclusively with OOness though
[04:03] <danilos> kiko: yeah, but I was thinking of some slower-non-production-debug-mode or something ;)
[04:03] <BjornT> ddaa: sure, we can have a call. how about in 15 minutes?
[04:03] <kiko> ddaa, because I once did implement something like this using twisted's reload functionality
[04:03] <ddaa> hu
[04:03] <kiko> ddaa, but there's a lot of complexity 
[04:04] <BjornT> kiko: would you have time for a call today to discuss https://launchpad.canonical.com/SimpleBugKeywords?
[04:04] <kiko> in particular in swapping running instance's classes, etc.
[04:04] <kiko> BjornT, sure!
[04:04] <ddaa> I do not know the issue deeply enough to argue, but I expect a lot of the functionality comes from quirks in the python object model
[04:04] <stub> kiko: That is all still valid except for the scheduled update time. Skipped rollout today again due to problems on the LAN. I can update now if there is urgent stuff, but nobody appears to be in tears.
[04:04] <ddaa> a lot of the _complexity_
[04:04] <kiko> stub, no, I was just curious
[04:04] <doko> cprov: can you or anybody else than infinity hand-build a package on a buildd?
[04:04] <kiko> ddaa, few OO languages allow you to swap an instance's class in runtime :)
[04:04] <BjornT> kiko: thanks! i'll ping you after the call with ddaa.
[04:05] <kiko> sure thing
[04:05] <ddaa> gotta learn smalltalk to talk like a real oo programmer :)
[04:05] <cprov> doko: uhm, what do you mean by hand-build, a package not submitted via soyuz ? 
[04:05] <ddaa> So far, in my naive world view, Smalltalk is still "as much OO as can be, and maybe still a bit more"
[04:06] <ddaa> a bit like forth and Scheme, take a simple idea, and push it to extremes
[04:06] <kiko> ddaa, that indeed is naive, because every time I used smalltalk it came across as a toy language :-P
[04:07] <ddaa> kiko: some guy who made a keynote at EP would disagree with you ;)
[04:08] <ddaa> BjornT: okay
[04:09] <doko> cprov: need to build a new gcc-4.1 using binaries which I built myself. so what needs to be done: a) install my binaries on the buildd, b) build gcc-4.1, c) install the just built binaries on the buildd, d) build gcc-4.1 again, e) upload gcc-4.1
[04:09] <doko> cprov: we only need that on powerpc
[04:09] <kiko> doko, why using these specific binaries?
[04:10] <doko> kiko: because the binaries in the archive are broken
[04:11] <cprov> doko: it looks scary, even if I can do this I should not, can we wait infinity ?
[04:12] <kiko> doko, and the chroot is busted?
[04:13] <doko> cprov: sure, we can ... but I would like to have exception handling again
[04:13] <doko> s/again/working again/
[04:14] <doko> kiko: gnat-4.1 is busted
[04:14] <doko> gcc-4.1, libgcc1
[04:15] <kiko> doko, and they are unable to bootstrap a new gcc compile, is that it?
[04:16] <cprov> doko: how do you mean ? are you suggesting we should support this kind of workflow ? can't you solve this by having a special chroot with your binaries installed ?
[04:17] <kiko> cprov, I think the problem isn't in the chroot, but in the archive itself
[04:18] <doko> kiko: yes
[04:18] <cprov> kiko: the chroot would have the newer/correct version of the broken binaries
[04:18] <kiko> cprov, does the chroot contain gcc binaries?
[04:19] <cprov> kiko: no, but we can install them ;) ... okay maybe I'm suggesting blue-sky stuff, nevermind
[04:20] <doko> cprov: well, if you do want to define a workflow for such situations ... it's basically: use untrusted binaries, then build the new binaries (which are then trusted), install them, then upload. not sure if it's worth to automate that
[04:21] <cprov> doko: is there something we can do to fix the archive w/o rebuilding the package ?  was it a soyuz failure in you POV ?
[04:22] <malcc> kiko: Smalltalk == toy will get you a lot of agreement, although not from me, but it doesn't contradict Smalltalk == Very very very OO, which is true whichever way you slice the toy thing
[04:22] <doko> cprov: no, wrong package; but I need gnat-4.1 to build gnat.4,1 ...
[04:23] <cprov> doko: not sure about how we would install the untrusted binaries if not manually in the chroot itself, but the rest is supported.
[04:27] <BjornT> ddaa: i'm ready now, should we have a call?
[04:27] <ddaa> sure
[04:27] <BjornT> sip, skype or normal phone?
[04:28] <ddaa> grah
[04:28] <ddaa> ekiga is one of those goddamn braindead app that confuses my ctrl-A for ctrl-Q...
[04:29] <ddaa> BjornT: let's give sip a try, I was able to have a successful conversation with the echo server and the voicemail before.
[04:29] <ddaa> I just need to refrain from ctrl-A...
[04:29] <BjornT> ddaa: cool. can you call me? (7270)
[04:50] <kiko> bradb, do you have a copy of the no-RHS-portlets CSS somewhere?
[04:50] <bradb> kiko: if you mean scott's stuff, it's:
[04:50] <bradb> http://people.ubuntu.com/~scott/launchpad.user.css
[04:50] <bradb> and http://people.ubuntu.com/~scott/launchpad.user.js
[04:56] <kiko> bradb, Keybuk: hmmm. unfortunate that the RHS portlets get hidden -- can they be displayed under the LHS ones?
[04:56] <Keybuk> kiko: they don't get hidden
[04:57] <Keybuk> that's what the greasemonkey javascript is for
[04:57] <kiko> ah I see.
[04:57] <kiko> you can't do that in CSS alone?
[04:57] <Keybuk> no, you can't
[04:57] <kiko> not even if I modified the main template slightly?
[04:57] <Keybuk> in CSS alone, you can place column one and two above each other
[04:57] <Keybuk> but then the order is almost exactly the opposite to what's desired
[04:58] <kiko> how would you do that?
[04:58] <Keybuk> float the body on the right, and make the portlets floated on the left
[04:58] <Keybuk> they stack
[04:58] <Keybuk> but then you get "portlets, facets, more portlets, more facets"
[04:58] <Keybuk> when I wanted "portlets, facets" only
[04:59] <kiko> hmm, can't seem to get it right.
[05:00] <Keybuk> of course, if you modified the LP css so everything just used the two column layout, that'd be *perfect* :p
[05:00] <kiko> I could modify the main template to make such customizations easier.
[05:01] <Keybuk> the basic problem is that the ordering of the template is such that it provides the order of the portlets and factets
[05:01] <Keybuk> it lists those down the LHS, then those down the RHS
[05:01] <Keybuk> with CSS alone, you can't reorder things
[05:05] <mpt> kiko, by putting an extra div around all the LHS portlets but not the facets/menu?
[05:05] <kiko> maybe
[05:06] <mpt> (by golly this is a depressing conversation)
[05:07] <ddaa> mpt: don't be depressed
[05:07] <ddaa> eventually, everybody will be using that greasemonkey stuff :)
[05:08] <Keybuk> I'm still not entirely happy with it though
[05:08] <Keybuk> the expanding portlets get in the way of ordinary mouse movement
[05:09] <flacoste> mpt: did you have time to take a look at https://launchpad.canonical.com/SupportTrackerWorkflowSpec?
[05:10] <mpt> flacoste, I've been doing non-LP stuff the past couple of days
[05:10] <mpt> I'll have a Launchpad day today and look at it
[05:10] <Keybuk> kiko: it'd be really nice if every portlet and facet had an id, btw
[05:10] <flacoste> mpt: ok, thanks!
[05:10] <mpt> (where by "today" I mean "after I wake up")
[05:11] <flacoste> kiko: i agree with Keybuk, we could improve our HTML id/class tagging. It would make it easier to customize the layout using CSS/Javascript
[05:12] <flacoste> flacoste: in general, i see that there is too much presentation markup in the HTML and not enough semantic markup
[05:15] <mpt> Well, we don't have any <font>s
[05:15] <mpt> Most of our <i>s don't mean anything in particular
[05:16] <mpt> Most of our <b>s are inline headings, but CSS doesn't do inline headings so we'd still need the <br>s before them
[05:17] <mpt> I remove <hr>s whenever I come across them, though some people like inserting them
[05:17] <flacoste> mpt: what do you call inline headings?
[05:17] <mpt> and the Plone CSS was horrific.
[05:18] <mpt> flacoste, such as you might find in a dictionary
[05:18] <flacoste> mpt: when was the last time you checked it?
[05:18] <flacoste> mpt: something like what dt/dd is intended for?
[05:18] <mpt> yes
[05:18] <mpt> but you can't get that presentation with dt/dd either
[05:18] <mpt> (amusingly)
[05:19] <mpt> flacoste, iirc the copy we had was from April-ish 2005
[05:19] <mpt> and when I nuked it circa October 2005, it dropped our CSS bandwidth by half
[05:20] <mpt> There's still remnants of it in launchpad.css, which I will remove one day
[05:20] <flacoste> mpt: indeed the Plone CSS was big, they somewhat improved that with 2.1 though, but still it is big
[05:21] <Keybuk> mpt: why not just use <h6> for inline headings and adjust it in CSS to look right?
[05:21] <Keybuk> or <label> ?
[05:22] <mpt> Keybuk, there's no way to achieve that in CSS
[05:22] <Keybuk> why not?
[05:23] <Keybuk> you can modify any tag to have the same effect as <b> ?
[05:23] <Keybuk> you can also modify any tag to have <b> with a following line break
[05:23] <mpt> yes, but then you still need the <br /> beforehand
[05:23] <mpt> so you haven't gained anything
[05:23] <Keybuk> no, you don't
[05:23] <mpt> except extra browsability in mostly non-existent outlining tools
[05:23] <mpt> oh, and screenreaders
[05:23] <Keybuk> just make it a display: block
[05:24] <mpt> That would be quite different from the current display
[05:24] <Keybuk> why?
[05:24] <Keybuk> or am I misinterpreting what you're attempting?
[05:24] <mpt> https://launchpad.net/distros/ubuntu/+bug/1/+portlet-details
[05:24] <Ubugtu> Malone bug 1 in Ubuntu Dapper "Microsoft has a majority market share" [Critical,Confirmed]  
[05:24] <mpt> Visibility: Public
[05:25] <mpt> Security related? No
[05:25] <mpt> So you want a line break before the bold stuff, but not after
[05:25] <mpt> {display: block} gives you (effectively) a line break both before and after
[05:25] <mpt> {display: inline} gives you neither
[05:27] <Keybuk> <div class="pair"><span class="key">Foo</span><span class="value">Bar</span></div>
[05:27] <mpt> Well, sure
[05:27] <SteveA> Kinnison, malcc: I'm still seeing this:
[05:27] <SteveA> There were 1 imports of names not appearing in the __all__.
[05:27] <SteveA> You should not import NeedsSymlinkInPool from canonical.archivepublisher.pool:
[05:27] <SteveA>     canonical.archivepublisher.publishing
[05:27] <SteveA> 
[05:28] <mpt> so for "you still need the <br /> beforehand", read "you still need either the <br /> beforehand, or a <div> before and a </div> after"
[05:28] <Keybuk> mpt: that's actually better, because now you can id the pair
[05:28] <Keybuk> <div class="pair" id="username">
[05:28] <mpt> #ubuntero {display: none}
[05:29] <Keybuk> now I can BeautifulSoup...("div", { "id": "username" })("span", { "class": "value"})[0] .string
[05:29] <Keybuk> :p
[05:29] <Keybuk> exactly
[05:30] <mpt> fair enough
[05:30] <malcc> SteveA: I've been fixing those as I go along, which means at least one fix is in one of my slow-to-land branches. Do you want me to bump the priority of these fixes?
[05:31] <mpt> though (number of people who could use it) * (amount it would improve Launchpad for them) / (implementation effort required) is pretty small
[05:32] <mpt> #mpt {consciousness: smaller;}
[05:32] <SteveA> malcc: this in particular gives everyone extra noise when writing tests, and prevents me from making that warning into an error.  Also, I mentioned this about three weeks ago, and was told it would be fixed in the next couple of days then.
[05:33] <SteveA> so, yes, please fix it soon
[05:33] <SteveA> if we have a warning being output all the time, then people are less likely to notice additional ones
[05:34] <malcc> SteveA: Ok, willdo
[05:34] <SteveA> ta
[05:40] <kiko> matsubara, can you confirm that https://launchpad.net/products/malone/+bug/6751 is fixed?
[05:40] <Ubugtu> Malone bug 6751 in malone "Forbidden error when viewing a milestone with private bugs" [Medium,Fix committed]  
[05:40] <kiko> spiv, ping?
[05:47] <jordi> danilos: samo proba!
[05:47] <kiko> matsubara?
[05:47] <matsubara> kiko: yes, it's fixed. 
[05:48] <kiko> thanks
[05:48] <kiko> matsubara, would you have time to do the simple fix for https://launchpad.net/products/malone/+bug/4779 -- ?
[05:48] <Ubugtu> Malone bug 4779 in malone "It's impossible to change the bugtracker's type after it is created." [Medium,Confirmed]  
[05:49] <matsubara> kiko: assign it to me.
[05:49] <kiko> cool.
[05:49] <matsubara> kiko: by simple fix you mean 'make the +edit BT type editable' and not the 'infer BT type automatically', right?
[05:50] <kiko> matsubara, correct.
[05:50] <matsubara> s/+edit//
[05:56] <kiko> matsubara, do you know who fixed bug 5542? was it mpt?
[05:56] <Ubugtu> Malone bug 5542 in malone "Malone shouldn't say "No matching results found" (inaccurate and imprecise)" [Medium,In progress]  http://launchpad.net/bugs/5542
[05:56] <matsubara> kiko: I don't remember from the top of my head, but I can find out.
[05:56] <salgado> I think I fixed that
[05:57] <matsubara> kiko: there you go
[05:57] <kiko> salgado, really? the current text is quite cool -- There are currently no open bugs. -- No results for search <b>dede</b>
[05:57] <kiko> salgado, just a period lacking after the </b> there
[06:00] <salgado> you re-assigned it to me and then I got a bugmail from mpt, as if this was a new bug
[06:00] <salgado> this is annoying
[06:01] <kiko> that's weird. BjornT?
[06:01] <salgado> kiko, btw, you said you want to improve the wording but you didn't say what exactly you want to change.
[06:02] <salgado> I think there's a bug open for that already. (the bugmail problem)
[06:02] <kiko> no, the wording is now fine salgado
[06:02] <kiko> salgado, only missing a period at the end of the line.
[06:04] <BjornT> salgado, kiko: yeah, there's a bug open on that
[06:04] <kiko> okay.
[06:04] <kiko> BjornT, why do we send that email to salgado? Ah, do we assume he was never notified of it?
[06:05] <danilos> jordi: that was supposed to go into my local launchpad, and it's serbian for "just a test" ;)
[06:05] <kiko> matsubara, and bug 44986?
[06:05] <Ubugtu> Malone bug 44986 in launchpad-bazaar "vcs-imports cannot post +sourceadmin" [High,Confirmed]  http://launchpad.net/bugs/44986
[06:06] <BjornT> kiko: yeah. i think the idea is that a newly notified should get a summary of the bug.
[06:06] <kiko> matsubara, could you handle that one? it's bugging ddaa..
[06:06] <jordi> danilos: :)
[06:06] <kiko> BjornT, so is the fix improving the text on the message, or checking whether salgado (because he is a launchpad-developer) already got the email?
[06:06] <kiko> perhaps the former
[06:07] <ddaa> actually, I think it might be something else than that ATM
[06:07] <jordi> danilos: are you in the rosetta@ alias?
[06:07] <kiko> BjornT, I need to have lunch, bw, but ping me when you're back.
[06:07] <ddaa> I'm about to give +sourceadmin another try, I'll let you know
[06:07] <danilos> jordi: no, not that I know of
[06:07] <danilos> jordi: how can I get there? :)
[06:07] <jordi> ok. I guess you should at some point.
[06:07] <ddaa> Mh
[06:07] <jordi> Ask Carlos. Who will then tell you to ask admin :)
[06:07] <danilos> heh, ok, np
[06:08] <ddaa> matsubara: I confirm, even without making any change to the form I cannot post it
[06:08] <danilos> jordi: btw, I need to upload guadec pics, there's plenty of you :)
[06:08] <BjornT> kiko: yeah, the former.
[06:08] <jordi> really? :D
[06:08] <flacoste> I need help on SQLObject: anyone can explain to me what the prejoins= parameter is for?
[06:08] <matsubara> ddaa: ok, I'll assign it to me.
[06:08] <kiko> ddaa, matsubara: I can't post either, and I'm an admin!
[06:08] <jordi> reminds me I want to shave today.
[06:08] <danilos> yeah
[06:09] <danilos> jordi: if you shave, pictures won't correspond to the real you anymore :P
[06:09] <ddaa> kiko: that's normal, this form is reserved to vcs-imports superheros
[06:09] <carlos> danilos: send it to the rt@ address where you requested your account
[06:09] <danilos> carlos: ok
[06:09] <jordi> what did I say! :P
[06:09] <kiko> ddaa, frown
[06:09] <danilos> :)
[06:09] <carlos> danilos: also, I'm going to add you to the rosetta-admins team
[06:09] <ddaa> seriously, anything would normally want to do should be accessible from +source, and that should be what the admin can access
[06:10] <jordi> carlos: sigh this xaralx stuff is getting hairy. maybe a query showing if those 8 people have any "accepted" translation would help
[06:10] <danilos> carlos: yeah, will that give me better privileges for rosetta bugs?
[06:10] <carlos> danilos: so you get Rosetta super powers 
[06:10] <danilos> carlos: oh my, that's what I want! :)
[06:10] <carlos> danilos: no changes on bugs, just Rosetta UI
[06:10] <ddaa> kiko: +sourceadmin is meant to allow things that require non-trivial manual poking in the product systems, that launchpad admins are not qualified to do.
[06:10] <ddaa> * in the production systems
[06:10] <jordi> danilos: you'll be able to edit templates, etc. Real fun
[06:11] <danilos> aah, ok, ok, that's good enough for start :)
[06:11] <danilos> jordi: yeah, I'll probably be able to translate to Valencian (which is totally unlike Catalan, of course :) as well? :)
[06:11] <kiko> ddaa, that sounds at least silly. admins wouldn't futz with that page anyway unless they knew what they were doing.
[06:11] <kiko> ddaa, the idea that "you can't trust an admin" is broken
[06:11] <LarstiQ> hi ddaa
[06:11] <ddaa> kiko: I can buy that argument
[06:11] <kiko> ddaa, because, effectively the group should be of people you can trust
[06:11] <jordi> danilos: only when you fix the UI for locale variants :)
[06:12] <flacoste> i get a weird error after running a Functional test: Exception exceptions.TypeError: "'NoneType' object is not callable" in <bound method Transaction.__del__ of <sqlobject.dbconnection.Transaction object at 0x30e9cbd0>> ignored
[06:12] <ddaa> kiko: but I can imagine some admins making too many assumptions...
[06:12] <kiko> ddaa, also note that stub, lifeless, elmo and znarl can in theory do the same thing by running wild SQL! :)
[06:12] <kiko> ddaa, there are not a lot of admins...
[06:12] <ddaa> kiko: but if you do that, you KNOW that you are doing something dangerous
[06:12] <danilos> jordi: heh, noooo, you broke even that? ;)
[06:13] <ddaa> LarstiQ: hello
[06:13] <carlos> danilos: well, we didn't implement variants (yet).... 
[06:13] <carlos> danilos: you have super cow powers now ;-)
[06:13] <jordi> we were waiting for your arrival to make it happen. :p
[06:13] <danilos> well, carlos, the trick is something I learned at guadec: if anything is missing, wrong or broken, blame jordi
[06:14] <jordi> kiko: oh no
[06:14] <danilos> it's easy... just look at me...
[06:14] <danilos> kiko: blame jordi, it's his fault
[06:14] <ddaa> kiko-fud: does it fuck blazingly fast now?
[06:14] <carlos> cool
[06:14] <carlos> so any bug in GNOME/GTK is seb128's fault
[06:15] <carlos> and now, any bug in Rosetta is jordi's fault
[06:15] <carlos> :-D
[06:15] <BjornT> flacoste: you can ignore that error. iirc, it has something to do with sqlobject caches and that it uses __del__
[06:15] <danilos> yeah, all problems resolved :)
[06:15] <danilos> when I get enough privileges for rosetta in malone, I'll go reassign all bugs to jordi with a message "it's your fault" :)
[06:15] <flacoste> BjornT: ignored is what I did, glad you can confirm me it was the right attitude :-)
[06:16] <flacoste> BjornT: any explication on the prejoin SQLObject parameter?
[06:16] <carlos> danilos: hmm, I think you need to joing launchpad developers team for that
[06:17] <carlos> danilos: https://launchpad.net/people/launchpad
[06:17] <danilos> carlos: I think I already tried that and got notified that only an admin can add me (or something like that)
[06:18] <danilos> btw, why not create a rosetta devels team which is then included into launchpad devels team? ;)
[06:18] <BjornT> flacoste: right. it's a way of pre-fetching an attribute of an object at the same time you fetch the object from the database.
[06:20] <flacoste> BjornT: is it kind of a trick to fill SQLObject cache so that traversing to linked object afterwards is faster?
[06:20] <BjornT> flacoste: for example if you loop through all the tickets and use .product, normall a db query will be issued for each ticket. if you use prejoin the product can be fetched in the same query as the tickets, resulting in one db query instead of many
[06:20] <danilos> wow, and look at the rosetta-admins team: it's all the cool guys! :)
[06:20] <BjornT> flacoste: exactly
[06:21] <flacoste> BjornT: ok, thanks!
[06:21] <carlos> danilos: let me add you, I'm an admin ;-)
[06:21] <danilos> carlos: sure, and thanks btw :)
[06:23] <carlos> done
[06:23] <carlos> danilos: well, we need to split rosetta-admins
[06:23] <carlos> to have rosetta-admins and rosetta-experts
[06:23] <carlos> then, we will be able to do that setup
[06:23] <carlos> For instance, Daf is a Rosetta Expert, but he's not a Launchpad developer
[06:23] <carlos> at least he's not it anymore
[06:23] <danilos> carlos: yeah, I know
[06:23] <danilos> but he's still cool :)
[06:24] <carlos> sure ;-)
[06:24] <carlos> I was just arguing about why rosetta-admins is not just a subteam of launchpad team
[06:26] <danilos> ah, yeah :)
[06:32] <jordi> fantastic, 2.6.17 doesn't like my wireless
[06:46] <sivang> carlos: that's the difference between a rosetta expert and a launchpad developer ?
[06:47] <carlos> sivang: s/that/what/ ?
[06:47] <sivang> carlos: right :)
[06:47] <sivang> carlos: you have a good psychic powers ;)
[06:48] <carlos> sivang: a rosetta expert only gets super powers on Rosetta and a launchpad developer is the owner of launchpad products so we get all 'spam' from rosetta/malone/launchpad/etc...
[06:48] <carlos> a launchpad developer will not get rights to translate anything on Rosetta, but a Rosetta Expert will get those rights
[06:49] <sivang> carlos: I see
[06:52] <danilos> carlos: while I am at it, do you think it would be better to also move "checked" (preferred) languages in front of all the others?
[06:53] <danilos> it's a simple template change, but I think it's much more useful ;)
[06:53] <carlos> hmmm
[06:53] <carlos> yeah, sounds like a good idea
[07:00] <danilos> carlos: bug #1788 fix at https://chinstrap.ubuntu.com/~dsilvers/paste/filewpiI4v.html
[07:00] <Ubugtu> Malone bug 1788 in rosetta "Saving preferred languages looks like it does nothing" [Medium,Confirmed]  http://launchpad.net/bugs/1788
[07:04] <carlos> danilos: hmmm, I don't think you need to use '%(language)s' and a dictionary there, it's just one argument
[07:05] <carlos> danilos: other than that, I think the patch is good ;-)
[07:05] <danilos> carlos: well, I do that since I anticipate that one day we'll have localised launchpad ;)
[07:05] <danilos> carlos: and %(launchpad)s is (dubiously) clearer than just %s
[07:06] <danilos> for translators, of course ;)
[07:06] <carlos> danilos: ok, that's a good argument
[07:06] <carlos> ;-)
[07:06] <carlos> danilos: ok, next step (in fact, it should be the first step...)
[07:06] <danilos> is...?
[07:07] <danilos> assign bug to myself? I'd do that if I knew how :)
[07:07] <carlos> danilos: you need to add a test to be sure that we don't have regressions
[07:07] <bradb> kiko-fud: I updated my malone-smallfixes branch (with the security/privacy collapse) this morning. Probably best to first review that, then later review the landscape-hack branch, after I've tweaked it from the results of the sec/priv review.
[07:07] <kiko-fud> bradb, good man. have a URL handy?
[07:07] <bradb> sure, one sec
[07:07] <danilos> ah, ok, some xp techniques :)
[07:08] <bradb> kiko-fud: it's not so small though, fwiw
[07:08] <carlos> danilos: you should have rights now to assign the bug to you, please tell me if you don't have such rights (and also, change its status to In Progress)
[07:08] <kiko-fud> bradb, that's fine
[07:09] <bradb> kiko-fud: https://chinstrap.ubuntu.com/~dsilvers/paste/filemwZdCZ.html
[07:10] <kiko-fud> bradb, oh, I thought you were going to give me the link to pending-reviews
[07:10] <kiko-fud> this is fine too
[07:10] <danilos> carlos: ah, I do, it's not very clear that I need to click on "Affects" field
[07:10] <kiko-fud> bradb, did you move the xx-filebug-shows-security contact stuff to another test?
[07:11] <bradb> kiko-fud: yeah, the distro- and upstream-bugprivacy ones
[07:11] <kiko-fud> ah cool
[07:11] <carlos> danilos: hmm I guess that's a UI issue, talk with mpt (our UI expert) or bradb or BjornT (malone developers)
[07:14] <danilos> carlos: it's already bug #1095
[07:14] <Ubugtu> Malone bug 1095 in malone "Unnecessarily difficult to find how to change status or reassign a bug" [Medium,Confirmed]  http://launchpad.net/bugs/1095
[07:14] <carlos> ok
[07:36] <ddaa> Is there any actual non-doctest non-pagetest unit test code for things like database and browser classes?
[07:37] <ddaa> I can see tests for specific non-web subsystems, like published, sftp server etc
[07:37] <ddaa> but nothing for things like "I have a page with hairy logic and I want unit tests for the brower and database classes"
[07:38] <kiko-fud> man pqm is annoying
[07:38] <kiko-fud> BjornT, wanna do that phone call?
[07:38] <ddaa> bradb: any insight?
[07:41] <salgado> ddaa, we have some unit tests in launchpad/database/ftests/
[07:41] <ddaa> salgado: wouldn't that be a bad place to put unit tests for browser code?
[07:41] <salgado> definitely. that's for database code
[07:41] <ddaa> besides, it's only adapter code
[07:41] <ddaa> no actual tests there
[07:42] <salgado> for browser code I doubt we have any unit tests
[07:42] <ddaa> wow! there's an actual unit test in test_ro_user.py!
[07:42] <ddaa> and test_shipit_constraints
[07:43] <ddaa> salgado: do you think I'm eccentric in thinking that it would a good thing to have some transversal unittest for browser and database code used for the +source and +sourceadmin forms?
[07:43] <ddaa> in addition to the pagetests and doctests
[07:44] <salgado> ddaa, I guess you'd like to have that because the page is quite complex?
[07:45] <ddaa> yes, there's some pretty hairy logic dependent on database state and the class of user changes in the form
[07:45] <ddaa> that should be spread in various appropriate classes involved in those pages
[07:46] <salgado> we have some doctests, usually called <foo>-pages.txt, whose initial purpose was to test these hairy pages
[07:47] <salgado> but that's not unit test
[07:47] <ddaa> well, I'd like if doctests could be restricted to documentation
[07:47] <ddaa> for non-trivial amount of testing, the accumulation of state they cause make them very hairy to maintain
[07:47] <salgado> agreed
[07:48] <ddaa> salgado: anyway, I'm out for an extended week-end
[07:48] <salgado> so, although we don't have any unit tests for browser code, I think it'd be reasonable to have
[07:49] <ddaa> salgado: I'd like to if when I come back, something has been decided about that... 
[07:50] <salgado> ddaa, maybe you can add it as a topic to the next meeting
[07:50] <ddaa> I'm not here for tomorrow's meeting
[07:50] <salgado> even though you won't be here, I guess I can explain it and answer any questions people make
[07:50] <ddaa> thanks
[07:51] <salgado> in fact, I don't think there's much to be discussed about this. I think it should just be a matter of creating launchpad/browser/tests/ and start writing unit tests and placing them there
[07:52] <ddaa> well, it's quite surprising that there's apparently _no_ browser unittest whatsoever so far...
[07:52] <ddaa> it's not like it's a groundbreaking idea...
[07:53] <salgado> it's not, definitely.  I guess it's because since the beginning we've benn writing only pagetests and doctests
[07:53] <salgado> so, it's much easier to stick something else in one of them than to do the right thing and write unit tests in a separate file
[07:53] <salgado> s/else//
[07:54] <ddaa> fortunately, I'm in a situation of writing a spec for somebody else to write tests and code, so it's easier to _ask_ the right thing :)
[07:55] <BjornT> kiko-fud: sure, i want to do that phone call now.
[07:55] <kiko-fud> ok 5 minutes turkish. should I call you?
[07:55] <BjornT> sounds good.
[08:03] <ddaa> Goodbye, see you tuesday.
[08:14] <msgh> hi
[08:15] <msgh> how can i upload my files into branch by sftp?
[08:16] <msgh> is there any one there?
[08:16] <msgh> :d
[08:16] <msgh> :D
[08:22] <Keybuk> bzr push --create-prefix sftp://$USERID@bazaar.launchpad.net/~$USERID/$PRODUCT/$BRANCH
[08:22] <Keybuk> where USERID = your launchpad user name
[08:22] <Keybuk> PRODUCT = the name of the "product" you want to push to
[08:22] <Keybuk> BRANCH = unique name for your branch ("mainline"/"trunk"/"bugfixes"/etc.)
[08:27] <msgh> i connect to account at launchpad by sftp , how can i upload my codes there?
[08:28] <msgh> can any one help me ?
 bzr push --create-prefix sftp://$USERID@bazaar.launchpad.net/~$USERID/$PRODUCT/$BRANCH
 where USERID = your launchpad user name
 PRODUCT = the name of the "product" you want to push to
 bzr push --create-prefix sftp://$USERID@bazaar.launchpad.net/~$USERID/$PRODUCT/$BRANCH
 where USERID = your launchpad user name
 PRODUCT = the name of the "product" you want to push to
 BRANCH = unique name for your branch ("mainline"/"trunk"/"bugfixes"/etc.)<Keybuk> BRANCH = unique name for your branch ("mainline"/"trunk"/"bugfixes"/etc.)
[08:28] <Keybuk> oops
[08:28] <Keybuk> double paste
[08:33] <msgh> sftp> bzr push --create-prefix sftp://socg2006@bazaar.launchpad.net/~socg2006/bzr/saj
[08:33] <msgh> Invalid command.
[08:34] <msgh> I try it but in terminal It's shown this msg : Invalid command
[08:35] <salgado> msgh, you need to do it outside of the sftp prompt
[08:35] <salgado> IOW, first exit the sftp session and then issue the "bzr push ..." command
[08:35] <kiko-fud> BjornT, oops, plop. calling back in
[08:37] <msgh>  bzr push --create-prefix sftp://socg2006@bazaar.launchpad.net/~socg2006/bzr/saj
[08:37] <msgh> bzr: ERROR: Not a branch: /root/
[08:37] <Keybuk> msgh: you're not in a bzr branch?
[08:38] <msgh> but in out of sftp this error has been shown : bzr:Error:Not a branch :/root
[08:38] <salgado> msgh, $ cd /path/to/bzr-branch
[08:39] <msgh> where are my bzr branch? 
[08:39] <Keybuk> what are you actually trying to do?!
[08:41] <msgh> I want to upload files in my accout in launchpad , but i don't know howto?
[08:42] <msgh> in terminal where must i be,im now at root
[08:43] <kiko-fud> BjornT, argh. okay, I give
[08:43] <kiko-fud> up
[08:43] <kiko-fud> BjornT, I'll produce a mockup of the keywords html and you tell me what you think ok?
[08:44] <BjornT> kiko-fud: sounds perfect
[08:44] <msgh> ok
[08:44] <jenda> Hello - what do I put as URL for a bzr branch?
[08:44] <msgh> can you plz explain cd /path/to/bzr?
[08:44] <jenda> It's for Spreadubuntu
[08:45] <salgado> msgh, this service is not for uploading random files. it can be used only to upload bzr branches
[08:46] <LarstiQ> jenda: bzr branches can live at any url provided you have access, can you give a bit more context?
[08:46] <salgado> msgh, it assumes you've used bzr before and you have a bzr branch with (open source) code on it in your computer
[08:46] <salgado> msgh, do you have that?
[08:46] <jenda> Ah
[08:47] <jenda> I was told that launchpad would host a bzr repo for an Ubuntu project.
[08:47] <jenda> Is that true?
[08:47] <danilos> jenda: what is ubuntu? :?
[08:48] <danilos> :P
[08:48] <jenda> gah ;)
[08:49] <salgado> kiko-fud, http://async.com.br/~salgado/shipit-download.jpg
[08:50] <salgado> (no more accessing remote launchpad instances)
[08:51] <kiko-fud> salgado, I wouldn't say ", which will take only a few hours"
[08:51] <LarstiQ> jenda: bazaar.launchpad.net can do hosting of arbitrary branches
[08:51] <LarstiQ> jenda: I guess that Ubuntu people have a specific workflow though
[08:51] <kiko-fud> salgado, apart from that it looks good.
[08:52] <jenda> So would you know how to get hosting there?
[08:52] <LarstiQ> jenda: so perhaps they can give you better advice on how to fit in, or you could roll with me and do something generic ;)
[08:52] <salgado> kiko-fud, to the code, then: https://chinstrap.ubuntu.com/~dsilvers/paste/file27OXKn.html :)
[08:53] <LarstiQ> jenda: http://ddaa.net/blog/ explains it fairly well. 
[08:53] <jenda> LarstiQ: what exactly would generic mean?
[08:53] <jenda> OK
[08:53] <kiko-fud> salgado, "and burn them yourself" was okay..
[08:53] <kiko-fud> salgado, r=kiko on that
[08:54] <salgado> oh, the screenshot was old
[08:54] <salgado> okay
[08:54] <jenda> LarstiQ: looks good. Thanks a bunch, I'll be back with more questions, most prolly :)
[08:54] <LarstiQ> jenda: generic would mean getting a branch hosted on launchpad without any ubuntu specific knowledge (as I don't have any)
[08:55] <salgado> kiko-fud, are you sure people won't be confused with just "burn them yourself." ?
[08:55] <kiko-fud> salgado, hmmm. and record them yourself?
[08:55] <jenda> LarstiQ: Ah - I think that would work OK for me, as it's the first branch for the Marketing Team, and every Team works on it's own.
[08:55] <kiko-fud> salgado, I don't like the repetition of the word "CD" much
[08:55] <LarstiQ> jenda: http://bazaar.launchpad.net/~larstiq/bzrk/bzrk.lq for instance is my bzrk branch
[08:55] <jenda> LarstiQ: What the branch would cantain is artwork, webdesign and text. No code apart from xml and html, I'm quite sure ;)
[08:56] <jenda> hello raphink
[08:56] <raphink> hi jenda
[08:57] <salgado> kiko-fud, neither do I. I think record is fine
[08:58] <stub> burn is a common term now. I think my inlaws use it, and they don't even have a CD burner.
[08:58] <stub> (at least for native speakers)
[08:58] <kiko-fud> "burn them yourself" has connotations
[09:02] <salgado> kiko-fud, what about that shipit discussion?
[09:05] <kiko-fud> salgado, riiiight.
[09:07] <salgado> flacoste, me too, and I loved it even more when I discovered bzr shelf 
[09:09] <flacoste> salgado: thanks! i didn't know that one
[09:17] <kiko-fud> BjornT, http://async.com.br/~kiko/mockups/keywords.html
[09:17] <kiko-fud> BjornT, so that's my initial idea.
[09:18] <kiko-fud> BjornT, I'm not sure how well that works with portlets..
[09:26] <kiko-fud> bradb, how does that look to you?
[09:27] <BjornT> kiko-fud: it looks quite good. i'm not so sure about the box around the description, but maybe it's not too bad.
[09:27] <kiko-fud> BjornT, I'm not so sure myself, but without it.. 
[09:27] <BjornT> yeah, i know...
[09:29] <kiko-fud> bradb, do you have an issue with having keywords on bugs instead of on tasks
[09:31] <bradb> kiko-fud: In practice, I don't think it'll make a big difference either way. (If there were 10 distros heavily using Malone, I might answer differently.)
[09:32] <bradb> also, there's another dimension to consider: distro or product
[09:33] <bradb> and another: user
[09:33] <bradb> i think distro or product would make sense before task
[09:33] <bradb> and that user is pure science fiction, for now
[09:36] <SteveA> i think keywords should be on bugs not tasks
[09:36] <SteveA> it is something that helps people share the work of understanding the bug
[09:36] <SteveA> even if they come to it from different tasks, different contexts
[09:37] <SteveA> we should work to make malone a place where distros and upstreams cooperate on bugs
[09:37] <SteveA> and shared tags is one way to do this
[09:37] <jenda> LarstiQ: Can you help me create a branch? Can't seem to get it done.
[09:37] <bradb> kiko-fud:  I think the UI you propose looks pretty good. The shortcomings that I do see, like maybe having a "No, I meant [      ] " type workflow, can probably emerge over time (my brain is not fully into thinking about this problem right now.)
[09:41] <LarstiQ> jenda: sure
[09:42] <LarstiQ> jenda: what seems to be the problem?
[09:42] <jenda> LarstiQ: I have a SSH key in LP, a directory on my PC with files the way I want them there... now what next? There's the init command and the push command...
[09:42] <LarstiQ> ah :)
[09:42] <jenda> (I also have a product on LP)
[09:42] <LarstiQ> jenda: so, you don't have a bzr branch yet?
[09:42] <jenda> no
[09:43] <LarstiQ> ok, we'll start there then.
[09:43] <LarstiQ> (if anyone objects to this taking place here, we can take it to #bzr)
[09:43] <jenda> The beginning - what a great place to start ;)
[09:43] <jenda> (or PM - or #ubuntu-marketing...)
[09:44] <LarstiQ> jenda: #bzr has the benefit of being my home base ;)
[09:44] <jenda> #ubuntu-marketing is my home base :-D
[09:44] <jenda> but sure
[09:44] <LarstiQ> jenda: I take it you have a directory structure ready, cd to the top of that, and run 'bzr init'. After that, 'bzr add'
[09:45] <jenda> done
[09:45] <LarstiQ> jenda: it should tell you it has added some files. See 'bzr status' to confirm it has what you want.
[09:45] <jenda> Yep
[09:46] <jenda> Spaces in filenames aren't a problem, are they?
[09:46] <kiko-fud> BjornT, want more input, want a phone call, or are you all good?
[09:46] <LarstiQ> jenda: time to commit then, 'bzr commit' throws up an editor, or 'bzr commit -m "commit message"' 
[09:46] <LarstiQ> jenda: not for bzr no, wreaks havoc with my naive shell scripting though ;P
[09:46] <jenda> hehe
[09:47] <jenda> (BTW, I've use svn before)
[09:47] <jenda> Looks done...
[09:48] <jenda> LarstiQ: What next?
[09:49] <LarstiQ> jenda: publish it. 'bzr push sftp://bazaar.launchpad.net/~user/project/branch'
[09:50] <BjornT> kiko-fud: i should be good to go for an initial implementation, thanks! i'll update the spec tomorrow.
[09:50] <kiko-fud> BjornT, feel free to link to that mockup. thanks!
[09:50] <LarstiQ> s/project/product/
[09:50] <jenda> LarstiQ: permission denied...
[09:51] <LarstiQ> jenda: can you connect with a different sftp client to confirm LP knows about your key?
[09:52] <jenda> Never tried :( Nautilus good enough? (I'm afraid I registered the key in LP wrong)
[09:53] <jenda> LarstiQ: how can I try that?
[09:53] <LarstiQ> jenda: ssh comens with 'sftp'
[09:53] <LarstiQ> jenda: so, try 'sftp bazaar.launchpad.net'
[09:53] <LarstiQ> jenda: no need to blush, I only found that out when I started to toy with bazaar.launchpad.net myself ;)
[09:53] <jenda> yeah, permission denied.
[09:54] <jenda> hehe ;)
[09:56] <LarstiQ> jenda: have you used ssh-add on the lp key?
[09:56] <LarstiQ> or specified it to be used in ~/.ssh/config
[09:56] <jenda> nope
[09:57] <jenda> the https://help.ubuntu.com/community/SSHHowto doesn't say so. I guess they don't expect too many people to be doing so. I'll bring it up to the doc team.
[09:58] <jenda> Argh, there's something wrong with the key. the passphrase doesn't work.
[09:58] <salgado> kiko-fud, https://chinstrap.ubuntu.com/~dsilvers/paste/filent4KID.html should do the job
[09:58] <LarstiQ> jenda: it is a bit specific to bazaar.launchpad.net though 
[09:58] <LarstiQ> jenda: ah.
[09:59] <salgado> kiko-fud, actually, wait a second
[09:59] <kiko-fud> sure.
[09:59] <jenda> And besides it throws an error about the private key being too unprotected.
[10:00] <salgado> kiko-fud, https://chinstrap.ubuntu.com/~dsilvers/paste/fileslmshZ.html
[10:02] <kiko-fud> salgado, TypeError: unsupported operand type(s) for +: 'SelectResults' and 'int'
[10:03] <kiko-fud> I think you are missing a count() 
[10:03] <salgado> yeah
[10:03] <salgado> kiko-fud, total_dapper_requests = previous_dapper_requests.count() + 1
[10:03] <salgado> or https://chinstrap.ubuntu.com/~dsilvers/paste/filesv3slu.html
[10:04] <kiko-fud> running...
[10:04] <LarstiQ> jenda: do you need help with the key?
[10:05] <jenda> I'm working on it LarstiQ ;) I'll brb with the key alright - thanks
[10:05] <LarstiQ> k
[10:06] <kiko-fud> > Shipit requests continue to be out of control (currently 1500 pending)..
[10:07] <jenda> wow
[10:09] <LarstiQ> kiko-fud: wuh, what makes that so backlog sensitive?
[10:09] <kiko-fud> ...
[10:09] <kiko> guess I ate too much.
[10:10] <jenda> LarstiQ: done.
[10:10] <jenda> LarstiQ: Now it tells me the parent directory doesn't exist.
[10:10] <LarstiQ> jenda: ah good, use push --create-prefix instead
[10:11] <jenda> Probably because the branch isn't regged on ... ah
[10:11] <jenda> Ah, working :)
[10:12] <LarstiQ> I'm not sure how long it will take for it to be mirrored publicly (on the https side)
[10:13] <LarstiQ> but with teams, it is possible for team members to access the sftp uri
[10:13] <jenda> OK, great ;)
[10:14] <jenda> BTW, can anyone checkout, or only team members?
[10:15] <LarstiQ> jenda: from the public side, everyone, for sftp, they won't even see it
[10:16] <jenda> OK
[10:16] <jenda> the public side is a mirrored, https accessible bzr repo type thing? ;)
[10:17] <LarstiQ> jenda: yup.
[10:17] <jenda> So no-one but me can checkout the intire branch, if it's under my name in launchpad?
[10:17] <LarstiQ> jenda: not entirely sure about that
[10:18] <LarstiQ> https://launchpad.net/people/jenda/+branch/spreadubuntu/spreadubuntu lists the http url already
[10:18] <LarstiQ> will take some time to mirror
[10:19] <LarstiQ> jenda: if it doesn't work as is, try it like https://launchpad.net/people/bzrk/+branch/bzrk/dev
[10:19] <LarstiQ> jenda: that branch is pushed to sftp://bazaar.launchpad.net/~bzrk/bzrk/dev
[10:19] <LarstiQ> jenda: the first bzrk being the team
[10:20] <jenda> eh
[10:20] <LarstiQ> https://launchpad.net/people/bzrk
[10:20] <jenda> I don't understand what that is.
[10:21] <LarstiQ> jenda: so that would be sftp://bazaar.launchpad.net/~ubuntu-marketing/spreadubuntu/spreadubuntu for you
[10:21] <jenda> AH
[10:21] <jenda> I see ;)
[10:21] <jenda> Meaning re-register it to the Marketing Team instead.
[10:22] <LarstiQ> jenda: bzrk is a plugin for bzr to visualise merges, it is registered as a product in lp by the 'bzrk developers' 
[10:22] <LarstiQ> jenda: just pushing would do the registering
[10:22] <jenda> That was meant as an example, though, right? OK.
[10:22] <LarstiQ> yup
[10:22] <LarstiQ> as it is the only team-shared branch I have experience with ;)
[10:23] <jenda> And that would give commit access to all 43 members of the MT.
[10:23] <LarstiQ> jenda: yup.
[10:23] <jenda> I'll have to think about that :)
[10:23] <jenda> But it's quite probably that the people on there that I don't want to have access wouldn't be capable of commiting anyway...
[10:23] <LarstiQ> jenda: I don't know if you could do finer grained control than that, other than setting up a new team
[10:23] <LarstiQ> or that.
[10:26] <jenda> LarstiQ: I have the feeling that nothing got uploaded though.
[10:27] <jenda> Any way to check, or re-try?
[10:27] <LarstiQ> jenda: that is easy to check, can you branch from it?
[10:27] <jenda> commit says no changes
[10:27] <LarstiQ> (or use bzr info on it)
[10:28] <jenda> http://paste.ubuntu-nl.org/17858
[10:28] <jenda> Is it just me or is there something wrong with that, LarstiQ?
[10:29] <LarstiQ> jenda: that checks the local branch, not the one you just pushed to
[10:29] <jenda> ah...
[10:30] <LarstiQ> jenda: the way I would test this is by bluntly doing 'cd /tmp; bzr branch sftp://bazaar.launchpad.net/~jenda/spreadubuntu/spreadubuntu'
[10:30] <LarstiQ> jenda: but, you could also use any other command that handles remote branches, 'bzr log -r -1  sftp://bazaar.launchpad.net/~jenda/spreadubuntu/spreadubuntu; bzr info  sftp://bazaar.launchpad.net/~jenda/spreadubuntu/spreadubuntu', you get the idea
[10:33] <jenda> OK, thanks
[10:34] <jenda> Seems it's all there :)
[10:34] <msgh> i fuck you
[10:36] <jenda> heh
[10:37] <LarstiQ> jenda: great. Anything else I can help with?
[10:37] <jenda> yep :)
[10:37] <jenda> When I try to push... ah wait...
[10:38] <jenda> nevermind - I typed marketing-team instead of ubuntu-marketing :)
[10:38] <LarstiQ> you need not specify the target every time, it will remember so you can just do 'bzr push'
[10:38] <jenda> OK
[10:39] <LarstiQ> jenda: if you want to remember a different location, you can pass the full url and --remember
[10:39] <jenda> great ;)
[10:39] <LarstiQ> jenda: and check bzr info to see what bzr thinks about your branch
[10:39] <jenda> There'll probabyl only be one
[10:40] <jenda> OK, now I need to know the command that one should use to checkout
[10:42] <jenda> nvm ;)
[10:43] <LarstiQ> well, you can do a checkout or a full branch
[10:44] <jenda> what's the diff?
[10:45] <LarstiQ> jenda: a checkout is bound to it's master branch, a commit in the checkout will also try to do a commit in the master. Sorta like the centralized style of svn/cvs
[10:46] <jenda> OK, whereas branch is only a specific branch...
[10:46] <LarstiQ> jenda: if you do 'bzr branch' though, you will get a independent branch
[10:46] <jenda> so when it's a simple branch like mine, there's no difference?
[10:46] <LarstiQ> which you would need to push to a public location, or be merged
[10:47] <jenda> argh... it's not _all_ there.
[10:47] <LarstiQ> jenda: what is missing?
[10:47] <jenda> I added some things after the first commit, and those aren't there.
[10:48] <LarstiQ> just bzr add, or committed those also?
[10:48] <jenda> Let's see.. when I type bzr commit, these things show up as unknown in the editor
[10:49] <jenda> ah!
[10:49] <jenda> So it's bzr add, then bzr commit for each change, right?
[10:58] <LarstiQ> jenda: 'bzr add <file>' will make <file> versioned. a bare 'bzr add' adds all unknown files recursivly (that looks wrongly spelled, hmm)
[10:58] <LarstiQ> jenda: and a commit is needed to make a new revision
[10:58] <LarstiQ> jenda: then after several commits, you might want to publish it with push (or use a checkout, which will do the push after every commit)
[10:59] <jenda> 'recursively' ;)
[10:59] <jenda> OK thanks.
[11:00] <LarstiQ> thanks, that's it :)
[11:00] <jenda> What happens when i 'push' instead of 'commiting'? Is that version vs. revision?
[11:00] <LarstiQ> push will transfer any revisions you have in your branch that are not at the target location
[11:01] <LarstiQ> anything you haven't committed yet is only in your working tree, and not a revision yet, so that will not be pushed.
[11:02] <jenda> OK
[11:02] <jenda> I see.
[11:02] <jenda> So commit does'nt upload at all. Now I get it :)
[11:02] <LarstiQ> exactly.
[11:03] <LarstiQ> jenda: welcome to the world of DVCS ;)
[11:03] <jenda> muhaha ;)
[11:03] <jenda> I did use svn before - but it was a little easier, really :)
[11:04] <LarstiQ> jenda: I'll have to disagree with you on that :p
[11:13] <jenda> LarstiQ: let's agree to disagree ;)
[11:13] <jenda> How would I remove the branch I pushed before as ~jenda ?
[11:14] <jenda> (I want to keep the ubuntu-marketing one only)
[11:14] <LarstiQ> jenda: I think that is launchpad-admin terrain, since afaik lp doesn't have a way to delete things atm
[11:14] <jenda> OK
[11:15] <jenda> I marked it abandoned then.