| blr | huh, `ag` can find zope templates with --plone | 02:01 |
|---|---|---|
| blr | wgrant: thank you for the review (imagine that took some time!) | 02:07 |
| wgrant | blr: Yup, mostly tiny things. | 02:07 |
| blr | wgrant: do we want the target ref to default to heads/master? | 02:10 |
| blr | context +register-merge ref picker | 02:10 |
| wgrant | blr: It should default to the target repo's HEAD ref, which is GitRepository.default_branch as of yesterday. | 02:11 |
| blr | ah cool, handy | 02:12 |
| wgrant | I'm not completely sure how it should all work at this stage. I think we eventually want a two-stage picker that selects a repo and then a ref within it. | 02:13 |
| blr | yep, would make sense to be within the same control | 02:13 |
| wgrant | We can't have the target ref input prepopulated with the default_branch, because the user could change the selected repo and invalidate the branch. | 02:13 |
| wgrant | So I think we for now have to go with a picker that asks for a vocab on the selected repo, but for the normal case accept that an empty ref input means a server-side fallback to default_branch. | 02:14 |
| blr | would be easy enough to reset it if the content of the target repository input changes | 02:14 |
| wgrant | Right, but reset to what? :) | 02:14 |
| wgrant | It'd have to be empty. | 02:14 |
| blr | well, it would have to be empty yes | 02:14 |
| blr | unless we could make an api call | 02:15 |
| wgrant | I think blank == HEAD makes sense in general. | 02:15 |
| wgrant | Like, if you clone a repo, you get HEAD. | 02:15 |
| wgrant | HEAD is a sensible thing for a null branch name to mean. | 02:15 |
| blr | ok, perhaps we just need to add a note in that case that HEAD is assumed unless otherwise specified | 02:15 |
| wgrant | I don't think it's worth improving on that until we can have the unified picker. | 02:15 |
| wgrant | Yep | 02:15 |
| mwhudson | wgrant, blr: just curious, have you thought at all about getting go get launchpad.net/foo to work if the foo project uses git? | 02:32 |
| blr | mwhudson: I'm afraid I'm not that familiar with golang, but that certainly sounds like a useful feature given our increasing use of the language... | 02:35 |
| mwhudson | yeah, there is support in the go tool for urls like that, but it assumes bzr | 02:36 |
| mwhudson | would be nice to fix it somehow | 02:36 |
| mwhudson | i'm not sure about all the intricacies of how go get works | 02:36 |
| blr | welp, apparently there's a carshare service in melbourne called GoGet. Thanks google. | 02:39 |
| mwhudson | i think one approach would be to put <meta> tags on the launchpad.net/project page | 02:40 |
| blr | presumably the go source is on github somewhere | 02:40 |
| mwhudson | (and delete the special casing for launchpad in the go tool) | 02:40 |
| mwhudson | blr: docs here https://golang.org/cmd/go/#hdr-Remote_import_paths | 02:41 |
| mwhudson | source is here https://github.com/golang/go/blob/master/src/cmd/go/discovery.go | 02:41 |
| mwhudson | (and get.go, vcs.go) | 02:42 |
| blr | mwhudson: thanks, will have a look this evening | 02:42 |
| blr | have been meaning to spend some time with golang at any rate. | 02:42 |
| mwhudson | blr: would it be useful to file a bug about this? | 02:42 |
| blr | mwhudson: absolutely yes, please tag it 'git' | 02:42 |
| mwhudson | if you do the lp bits, i'll do the go bits :-) | 02:42 |
| wgrant | Yep, that definitely makes sense. | 02:45 |
| wgrant | Inasmuch as go get itself does :) | 02:45 |
| mwhudson | https://bugs.launchpad.net/launchpad/+bug/1465467 | 02:45 |
| mup | Bug #1465467: put <meta name="go-import"> tags on project, series pages <git> <Launchpad itself:New> <https://launchpad.net/bugs/1465467> | 02:45 |
| mwhudson | well yeah | 02:45 |
| mwhudson | i guess testing this against launchpad.dev will actually be easier than testing it against launchpad.net, unless you hack your own build tool | 02:46 |
| mwhudson | er | 02:46 |
| mwhudson | *go tool | 02:46 |
| blr | thanks mwhudson, at a glance it seems easy enough | 02:49 |
| mwhudson | blr: yeah, seems like it should be fairly easy | 02:49 |
| mwhudson | blr: go 1.5 will be released in august, would be nice to be able to get the relevant change into that | 02:50 |
| mwhudson | not completely sure i want to set up a launchpad dev environment just for this, but i guess i could... | 02:51 |
| blr | mwhudson: could you mock it locally somehow and point me at your golang fork on gh perhaps? | 02:51 |
| mwhudson | blr: sorry, i don't really understand | 02:54 |
| blr | mwhudson: I'm not certain you're going to be able to avoid setting up LP heh, although it is reasonably painless these days | 02:54 |
| mwhudson | blr: depends on whether i can con one of you guys into doing it first | 02:55 |
| blr | mwhudson: I can have a poke at it at codecraft this evening | 03:00 |
| mwhudson | blr: awesome | 03:00 |
| mwhudson | i've put it on my todo list | 03:00 |
| mwhudson | but not at the top :-) | 03:00 |
| maozhou_ | Can someone help me? how to init series? | 06:54 |
| cjwatson | maozhou_: Um, what are you trying to do exactly? | 06:55 |
| cjwatson | Project series don't require initialisation, and initialising a distribution series on launchpad.net is something we'd expect to have had prior notice of. | 06:56 |
| maozhou_ | i set up launchpad server on local, then i add a series.but when i init it ,it dosen't work | 07:02 |
| cjwatson | maozhou_: This is not even slightly easy to get going, but https://wiki.ubuntu.com/NewReleaseCycleProcess may provide some clues. | 07:04 |
| maozhou | cjwatson: i do it by this way.(https://dev.launchpad.net/LEP/DerivativeDistributions),but when i init it ,it dosen't work. | 07:09 |
| cjwatson | You'll probably need to run 'cronscripts/process-job-source.py IInitializeDistroSeriesJobSource' | 07:11 |
| cjwatson | But on a local instance you're mostly on your own, or at least will have to dig through log files yourself and generally provide more useful problem reports than "it doesn't work" :-) | 07:11 |
| cjwatson | I suspect derived distributions will be pretty complex to get going on a local instance, as you'd have had to import some other distribution to derive from. | 07:12 |
| cjwatson | You could in principle import Ubuntu using scripts/gina.py or something, but there'd be a lot of details for you to work out for yourself. | 07:13 |
| maozhou_ | cjwatson: my workmate said that he once init the "DerivativeDistributions" succeed by run some script, but he can't remember which script he run. may be i need try to run 'cronscripts/process-job-source.py IInitializeDistroSeriesJobSource' | 07:24 |
| maozhou_ | cjwatson: thanks for your help, Let me try by your way. | 07:27 |
| wgrant | maozhou_: Oh, are you working on the kylinos.com.cn Launchpad instance that was spamming people a few months back? | 07:27 |
| cjwatson | ... and have you deleted sampledata yet? :-) | 07:30 |
| maozhou_ | wgrant: yes, may be. I am sorry about that. but how to stop it spamming people? | 07:33 |
| wgrant | maozhou_: As far as we could tell, you took a normal Launchpad development instance, including all the development sample data, enabled outbound email, and started using it. The sample data includes some real email addresses, and is generally a terrible basis for a production deployment. | 07:36 |
| wgrant | I don't know what your goal for your Launchpad instance is, but starting with sampledata is almost certainly the wrong way to go. | 07:36 |
| cjwatson | Also may well have security holes of various kinds since the sample data includes people with e-mail addresses you don't control in privileged teams. | 07:38 |
| cjwatson | The development sample data is meant to make it easier to experiment with changes to Launchpad itself locally, by giving you a bit of test data so that important pages will render something at least a little bit useful. But it should never be included in production deployments. | 07:41 |
| cjwatson | I don't know if there's a sensible way to excise it from an already-running instance; it would probably involve non-trivial database surgery, and might well be easier to start again from scratch. | 07:42 |
| maozhou_ | ok, i will tell my workmate to check it . may be we need to modification datebase. | 07:54 |
| maozhou | if i alter datebase about mailaddress,is it effectivity? | 07:59 |
| cjwatson | You really ought to delete the predefined non-team accounts, and there will be various other junk in e.g. your list of registered projects, your bug/branch/specification tables, etc. that you probably want to get rid of. | 08:04 |
| wgrant | A fresh bootstrapped database would be a much better idea. | 08:07 |
| maozhou_ | there are many data already in datebase ,if do a fresh bootstrapped database, all the data will be losed , may be we can't do that. | 08:14 |
| maozhou_ | we will check it to stop it spamming people. | 08:26 |
| maozhou_ | cjwatson:when i run 'cronscripts/process-job-source.py IInitializeDistroSeriesJobSource' , it's error . the error log is “Running <InitializeDistroSeriesJob for distribution: kylin, distroseries: dota, parent[overlay?/pockets/components]: utopic[False/Release/main], architectures: (), archindep_archtag: None, packagesets: None, rebuild: True> (ID 2) in status Waiting” | 09:14 |
| maozhou_ | Job resulted in OOPS: OOPS-ea5bde41e13487cb6437263a6b26b44d | 09:14 |
| maozhou_ | 1 InitializeDistroSeriesJob jobs did not complete. | 09:14 |
| maozhou_ | what is “OOPS: OOPS-ea5bde41e13487cb6437263a6b26b44d” ？ | 09:15 |
| cjwatson | You'll need to find the actual text of that OOPS. | 09:15 |
| cjwatson | It should have been dumped out somewhere, but exactly where depends on your configuration. | 09:15 |
| cjwatson | The error_reports lazr config section should have an error_dir key, which might be a place to start. | 09:16 |
| maozhou_ | ok，i know .thank you. | 09:17 |
| === anthonyf is now known as Guest89967 | ||
| cjwatson | wgrant: Would you mind having another quick look over https://code.launchpad.net/~cjwatson/launchpad/git-repository-ui-edit-target/+merge/261232 ? I know you already gave an Approve vote, but I made some fairly extensive changes to cope with the changes related to target_default. | 14:33 |
| === anthonyf is now known as Guest41199 | ||
| blr | morning | 20:55 |
| blr | hmm would be nice to have the option to delete a comment within x minutes of creation for people that haven't had enough coffee. | 21:38 |
| mwhudson | i think there is a bug with a very small number about that :-) | 21:56 |
| blr | mwhudson: made a start on your request last night, have the import meta rendering on Product+index, will try to sort out the bzr meta and ProductSeries tag before eow | 21:57 |
| mwhudson | blr: i just noticed, cool :-) | 21:57 |
| mwhudson | i'll get set up to be able to test it | 21:58 |
| mwhudson | is launchpad still running on precise? | 21:58 |
| blr | mwhudson: yep | 21:59 |
| mwhudson | okidoke | 22:00 |
| === anthonyf is now known as Guest21341 | ||
| blr | mwhudson: have another branch (which allows a project to have an associated 'default' vcs, amoungst other things) which needs to land before this branch. Not far away however | 23:09 |
| blr | will govern whether a bzr or git meta tag is rendered | 23:09 |
| mwhudson | blr: ah yes, that seems relevant :-) | 23:20 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!