[11:05] <jtv> hi adeuring!  are you on call?  I put one in the queue and, to make up for it, reviewed one that was already there.  :-)
[11:13] <adeuring> jtv:  it's quite long diff... 
[11:16] <jtv> adeuring: looking it up now... I hadn't realized
[11:18] <jtv> adeuring: argh, the gateway's power supply blew up earlier so I'm on a problematic connection, can't check now.  :(  Never mind then...
[11:19] <adeuring> jtv: can't you send the "real" diff via mail?
[11:20] <jtv> adeuring: don't have the branch handy here either
[13:51] <al-maisan> hello jml, I'd appreciate it if we you could review https://code.edge.launchpad.net/~al-maisan/launchpad/duration-485524 before the sessions start
[13:51] <al-maisan> s/we you/you/
[15:04] <henninge> adeuring: are you working on jtv's branch?
[15:04] <adeuring> henninge: that's 2000 lines... or the diff is wrong
[15:05] <henninge> adeuring: yeah, I just saw that ....
[15:05] <henninge> adeuring: so you could take on one for me?
[15:05] <adeuring> henninge: sure
[15:06] <henninge> adeuring: cool, let me update the mp
[15:08] <jml> al-maisan: it's too late for before now.
[15:09]  * jml suspends
[15:16] <henninge> adeuring: review requested
[15:16] <adeuring> henninge: ok
[15:16] <henninge> adeuring: danke
[15:16] <adeuring> immer grerne :)
[15:21] <jml> al-maisan, hi
[15:21] <jml> al-maisan, doing code review tasks now
[15:28] <al-maisan> jml: thanking you, please let me know if you want to meet and talk (I am currently on level 3 in the Riviere room).
[15:37] <adeuring> henninge: why did you add "pipes" to .bzrignore?
[15:37] <henninge> ;)
[15:38] <henninge> adeuring: that's what "bzr reconfigure-pipeline" does for you ...
[15:38] <adeuring> henninge: ah, ok
[15:38] <henninge> I don't know if that is supposed to go into the tree. What do you say, abentley?
[15:40] <abentley> henninge: It goes into the tree because reconfigure-pipeline is not intended for your use case.
[15:40] <henninge> abentley: ???
[15:40] <abentley> henninge: It's meant for people who don't have a shared repository and want to convert their standalone working tree into a pipeline.
[15:41] <abentley> henninge: since you already have a shared repository, and presumably a lot of location-specific configuration, you should just create a lightweight checkout of your branch.
[15:42] <sinzui> flacoste: ping
[15:42] <henninge> abentley: e.g. of devel etc.?
[15:43] <abentley> henninge: Your branch, not a mirror of upstream.  Well, you have have upstream as your first pipe if you like, but it's not the recommended configuration.
[15:43] <henninge> abentley: do we have this use case documented somewhere?
[15:43] <adeuring> henninge: could you add some short documentation to the new function init_status_choice()?
[15:44] <henninge> adeuring: sure
[15:44] <adeuring> thanks!
[15:44] <abentley> henninge: Probably not.
[15:45] <henninge> abentley: because I have to admin I didn't understand the instructions you just gave me ... :-/
[15:45] <henninge> *admit*
[15:45] <abentley> henninge: Instead of running reconfigure-pipeline in your branch, you should create a lightweight checkout of that branch.
[15:46] <henninge> abentley: in another directory, I assume
[15:46] <henninge> abentley: and when I create a new pipe?
[15:47] <abentley> henninge: Yes, it's not a lightweight checkout if you create it in the same directory.
[15:49] <abentley> henninge: When you create a new pipe, you run "bzr add-pipe".  It will create a new branch/pipe in the same directory as your first branch/pipe.
[15:49] <abentley> henninge: Perhaps this is a "how do I turn my branch into a pipeline?" question?
[15:50] <abentley> henninge: By default, branches are pipelines with a single branch in them.
[15:51] <henninge> abentley: I guess. It's a bit clearer now, I will have to try it out, though.
[15:51] <abentley> henninge: I mean, all branches are part of pipelines.  If you haven't explicitly added them to a pipeline, then it's a single-branch pipeline.
[15:51] <henninge> abentley: I understand!
[15:52] <abentley> henninge: There's no equivalent to "bzr-loom's loomify" because it's not needed.
[15:52] <henninge> abentley: how do I turn a lightweight checkout back into a branch?
[15:53] <abentley> henninge: bzr reconfigure.  I'm not sure whether you want a branch or a branch with a tree in it.  The commands are "reconfigure --branch" or "reconfigure --tree" respectively. 
[15:54] <henninge> abentley: I mean, I have to find a way to get rid of that .bzrignore and the pipes directory
[15:54] <henninge> I will try that
[15:56] <abentley> henninge: I suggest: "bzr branch pipes/$FOO ../$FOO; rmdir pipes; bzr switch --force ../$FOO"
[16:00] <flacoste> hi sinzui
[16:02] <sinzui> flacoste: I have an extraordinary request: I want a review of https://code.launchpad.net/~sinzui/launchpad/parent-packaging-links-bug-484828/+merge/15092 and I want to run it in production today or Monday.
[16:02] <flacoste> sinzui: looking at it now
[16:02] <abentley> henninge: thumper posted a blog about pipelines, though the shelve machinations it describes are thankfully obsolete: http://how-bazaar.blogspot.com/2009/07/breaking-up-work-for-reivew.html
[16:03] <adeuring> henninge: in HasTranslationImportsView.change_status_action(), aren't ou simply trusting user input, when you omit all the permission checks?
[16:04] <henninge> adeuring: I don't
[16:04] <adeuring> so, how/where is the check done?
[16:05] <henninge> adeuring: checks using check_permissions build on what is configured in zcml and enforced in security.py, right?
[16:05] <adeuring> rkght
[16:06] <henninge> This allows to restrict access to attributes and methods.
[16:06] <adeuring> henninge: ah, now I get it...
[16:06] <henninge> but here we are restricting the value of parameters, so we have to do the checking ourselves.
[16:08] <henninge> that is what canSetStatus does, it checks the given user if he's allowed to set the desired status.
[16:08] <henninge> the user is taken from the view
[16:09] <henninge> adeuring: does that explain it?
[16:10] <henninge> adeuring: I spend some time on the last branch which prepared the API for this and thought about these things ... ;-)
[16:13] <adeuring> henninge: what I was looking for is the call of canSetStatus() in setStatus() ;)
[16:13] <henninge> is it not?
[16:14] <adeuring> henninge: no, it _is_ called. Problem was that I asked before I searched....
[16:14] <adeuring> sorry for the noise
[16:14] <henninge> np
[16:25] <adeuring> henninge: r=me. Just one detail (UI stuff, so, nothing ou asked for ;): for example on your screenshot http://people.canonical.com/~henninge/screenshots/status-picker-after.png there are six buttons which pop up a dialog, and the values shown identical for two set of three dialogs. Couldn't you change the dialog title from "Change status to" to "Change status of <queue entry> to" so that it is clear what exactly the dialog is doing?
[16:26] <henninge> adeuring: the problem is that <queue entry> could be quite long. Unless I just use the filename but even that could have a long path.
[16:27] <adeuring> henninge: I see. So, let's leave it as it is.
[16:27] <henninge> adeuring: also be aware that since the screenshots the rightmost edit buttons have been moved to the left, right next to the text of the entry.
[16:28] <henninge> as Curtis had suggested
[16:28] <henninge> adeuring: thank you for the review1
[16:28] <henninge> !
[16:30] <adeuring> henninge: right. But I was more concerned about the _vertical_ ambiguities: Once you click the the "edit" button, you can't see what queue entry will be affected. 
[16:31] <adeuring> henninge: you could perhaps mark the complete row of the entry with a different background colour
[16:31] <adeuring> ...for the time the dialog is shown
[16:34] <al-maisan> hello jml, how are things? Any luck with the schema patch review?
[16:36] <henninge> adeuring: We still have some work for this page in the future, so we could include that idea then.
[16:36] <adeuring> henninge: OK, great
[16:40] <jml> al-maisan, reviewing it now.
[16:43] <jml> al-maisan, I've added a comment. Basically, it would help me a lot if you could give more context on the patch. :)
[16:43]  * jml offline to stalk the corridors.
[16:54] <deryck> rockstar, https://code.edge.launchpad.net/~brian-murray/launchpad/bug-485229/+merge/15040
[17:13] <bac> adeuring: can i add a branch to your queue?  barry are you reviewing today?
[17:14] <adeuring> bac: sure
[17:14] <barry> bac: i am, but probably not until after lunch
[17:14] <bac> adeuring: https://code.edge.launchpad.net/~bac/launchpad/bug-485237/+merge/15095
[17:42] <adeuring> bac: r=me
[17:42] <bac> adeuring: thanks, that was fast!
[17:44] <bac> jml has anyone mentioned lately how much 'ec2 land' rocks?