File size: 10,925 Bytes
4aa5fce |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 |
[02:01] <blr> huh, `ag` can find zope templates with --plone [02:07] <blr> wgrant: thank you for the review (imagine that took some time!) [02:07] <wgrant> blr: Yup, mostly tiny things. [02:10] <blr> wgrant: do we want the target ref to default to heads/master? [02:10] <blr> context +register-merge ref picker [02:11] <wgrant> blr: It should default to the target repo's HEAD ref, which is GitRepository.default_branch as of yesterday. [02:12] <blr> ah cool, handy [02:13] <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. [02:14] <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 [02:15] <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 [02:32] <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? [02:35] <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... [02:36] <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 [02:39] <blr> welp, apparently there's a carshare service in melbourne called GoGet. Thanks google. [02:40] <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) [02:41] <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 [02:42] <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 :-) [02:45] <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 [02:46] <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 [02:49] <blr> thanks mwhudson, at a glance it seems easy enough [02:49] <mwhudson> blr: yeah, seems like it should be fairly easy [02:50] <mwhudson> blr: go 1.5 will be released in august, would be nice to be able to get the relevant change into that [02:51] <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? [02:54] <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 [02:55] <mwhudson> blr: depends on whether i can con one of you guys into doing it first [03:00] <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 :-) [06:54] <maozhou_> Can someone help me? how to init series? [06:55] <cjwatson> maozhou_: Um, what are you trying to do exactly? [06:56] <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. [07:02] <maozhou_> i set up launchpad server on local, then i add a series.but when i init it ,it dosen't work [07:04] <cjwatson> maozhou_: This is not even slightly easy to get going, but https://wiki.ubuntu.com/NewReleaseCycleProcess may provide some clues. [07:09] <maozhou> cjwatson: i do it by this way.(https://dev.launchpad.net/LEP/DerivativeDistributions),but when i init it ,it dosen't work. [07:11] <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" :-) [07:12] <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. [07:13] <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. [07:24] <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' [07:27] <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? [07:30] <cjwatson> ... and have you deleted sampledata yet? :-) [07:33] <maozhou_> wgrant: yes, may be. I am sorry about that. but how to stop it spamming people? [07:36] <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. [07:38] <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. [07:41] <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. [07:42] <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. [07:54] <maozhou_> ok, i will tell my workmate to check it . may be we need to modification datebase. [07:59] <maozhou> if i alter datebase about mailaddress,is it effectivity? [08:04] <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. [08:07] <wgrant> A fresh bootstrapped database would be a much better idea. [08:14] <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. [08:26] <maozhou_> we will check it to stop it spamming people. [09:14] <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. [09:15] <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. [09:16] <cjwatson> The error_reports lazr config section should have an error_dir key, which might be a place to start. [09:17] <maozhou_> ok,i know .thank you. === anthonyf is now known as Guest89967 [14:33] <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. === anthonyf is now known as Guest41199 [20:55] <blr> morning [21:38] <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. [21:56] <mwhudson> i think there is a bug with a very small number about that :-) [21:57] <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 :-) [21:58] <mwhudson> i'll get set up to be able to test it [21:58] <mwhudson> is launchpad still running on precise? [21:59] <blr> mwhudson: yep [22:00] <mwhudson> okidoke === anthonyf is now known as Guest21341 [23:09] <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 [23:20] <mwhudson> blr: ah yes, that seems relevant :-) |