=== gary_poster|away is now known as gary_poster [13:55] good morning [13:55] hi hatch [14:32] so how was everyones weekend? [14:33] bac: http://pastebin.com/0M4PiXZD [14:34] hi hatch. had a lovely long weekend. hope yours was good [14:36] yeah it was really sunny here - which is a little odd for this time of year [14:37] today is looking like it's goign to be the same [14:44] the supply of Fire Emblem Awakening for 3DS is so bad people are charging a $25 premium online for it hah! [14:52] benji: howdy [14:52] hi rick_h_; so, post-merge... [14:52] yea, qa'ing fun for all [14:52] do you have a card on the kanban board? [14:53] hmm, on ours. I didn't add one to yours [14:53] after a merge the app is updated on http://uistage.jujucharms.com:8080/ you can go there to QA your change [14:54] rick_h_: that prbably makes sense; I ask because there is a UI/UX review slot on ours that tracks if there is something for the UI/UX guys to review [14:54] ok, and then once qa'd just mark the bug as fix released? [14:54] benji: ok, good to know. [15:00] rick_h_: right [15:07] bcsaller_: 'charmstore' is the default namespace correct? if no namespace is defined it's the default? [15:09] hatch: yes, thats passed in at init and set as default [15:10] alright I'm going to change that to be something more generic - I just didn't want to change it if I was understanding incorrectly [15:12] hatch, I converted your kanban card to a prototype card as we discussed [15:13] funny story I was watching a c# tutorial yesterday and the guy took a template and moved the opening curly brackets onto the new line....I cringed haha [15:13] hatch, and then I put the actual bugs as cards, to be moved through one at a time [15:13] gary_poster: sounds good! I'm fighting with edge cases right now :/ [15:13] hatch, and they are all at the top of the board now [15:14] gotcha [15:14] cool hatch. don't forget that, if you've proved the basic story in the prototype, you can just move on to the TDD approach. OTOH, if you think these edge cases might affect the underlying design, go for it. [15:14] yeah unfortunately they do - I woudln't feel right submitting it as is [15:15] sorry that's not entirely true [15:15] the one I'm finishing now does [15:15] then I can call it [15:15] cool [15:15] I have a postit note that says "split it up" [15:15] lol [15:15] :-) [15:16] Actually doing the full TDD thing would be even more fab [15:20] benji: hangout died. won't come back. let's chat after call [15:21] bac: k [15:21] benji: entityName is a string [15:21] rick_h_, thanks again for the fix for bug 1111730. We will be making a release today, hopefully. we've been experimenting with moving bugs to "fix committed" and then we actually will mark them as "fix released" when we make the tarball. That may just be too annoying, but fwiw that's where we are [15:21] <_mup_> Bug #1111730: Unit count widget misplaced

< https://launchpad.net/bugs/1111730 > [15:21] f(a,b type) means both a and b are of the same type [15:21] bac: I figured, I was just wondering why it isn't declared as such [15:21] shortcut [15:22] gary_poster: ah ok. Yea we battle the same idea of 'released' with things as well. [15:22] cool [15:22] bac: saving those 7 characters is a big win [15:26] OK fixed now to split this up [15:27] cool [15:30] jujugui call nowm oops [15:48] benji: quick hangout? [15:57] bac: I do make a mean BBQ potato [15:58] mmmmmm BBQ [15:58] mine has been covered with about 2ft of snow for too long [15:58] move someplace warmer :-P [16:04] gary_poster et al - prototype is at https://drive.google.com/a/canonical.com/#folders/0B1IM--9A1RkTV3FQLVJtbjZvUGM [16:04] thank you jovan2 looking now [16:11] Makyo: rapi is further along than suspected , it should just be an improv thing, and the key is 'env' [16:17] bcsaller_, ah, awesome. [16:20] gary_poster: haha if anything i'd move somewhere with MORE snow :) [16:20] :-) [16:29] yikes lbox just rejected me because my indentation was wrong [16:29] note to self....check indentation [16:30] hatch, make lint is what's catching that. make prep can help, too, if the beautifier can fix it. [16:30] it's also dumping on me because of an invalid jsdoc tag [16:31] I thought we used yuidoc? [16:31] hatch, we do. Wonder what that's about? [16:31] hatch, if you can turn off the option probably fine [16:31] will investimagate [16:31] (must be jshint) [16:32] The only jshint tag we're using is bitwise operations, right? [16:32] dunno [16:33] *shudder* bitwise in js [16:34] :/ [16:35] you're a bitwise fan? [16:37] I don't care too strongly either way, and the code that uses it isn't mine, but I do like tools that get the job done :) [16:39] hmm [16:39] virtualenv/bin/gjslint --strict --nojsdoc --jslint_error=all [16:39] but it's still flagging the jsdoc stuff [16:41] hatch: do you mean yuidoc? [16:43] nope it's dumping on me because my indentation is off on a couple lines and because my documentation is not 'valid' [16:44] hatch: if you pastebin the output I'll be glad to see if I can help [16:45] https://gist.github.com/anonymous/1e9e97bf336735ad288a [16:46] damnit I can't claim that gist [16:49] ohh we have the --strict tag [16:49] that's why it's failing [16:50] anyone have any objections to me removing that? [16:50] https://developers.google.com/closure/utilities/docs/linter_howto?hl=en search for --strict [16:55] hatch: I'm pretty sure we actively want --strict; we've had it enabled for quite a while without it killing us, so I'm curious why it is causing problems now [16:56] well I'm using non jsdoc tags [16:56] they are valid yuidoc tags though [16:56] if you push this to a branch that I can grab I will be glad to look at it [16:56] yup it's pushed [16:56] 1130787-subapp-app-extension [16:57] I can construct one, but a full branch URL would be more helpful. [16:57] https://code.launchpad.net/~hatch/juju-gui/1130787-subapp-app-extension [16:57] :) [16:59] if we want to be able to use all of the yuidoc syntax flag we have to drop --strict [16:59] OR, I could just delete those flags and indent the things as it likes [17:02] hatch: re. @public: just remove it; re. indentation: I think you know what to do there ;); re. Type should be immediately after @param tag: reverse the name and type ("{object} config" not "config {object}") [17:03] ♫ To the right, to the right, all of my tasks to the right to the right ♫ [17:03] re. "descriptions must end with valid punctuation such as a period.": do what it says :) [17:03] heh [17:03] I can dooo eeet [17:04] I actualy prefer it the way you have it, but it's not worth enguaging with the linter about it. [17:04] * hatch punches the linter for fun [17:04] you got it [17:05] * hatch grumbles and contemplates writing a blog post on how linters hinder innovation in hopes it gets picked up by Fox's hype machine [17:12] wow github support deleted that gist already...that was like 10m [17:13] bac: I have a passing test! [17:18] O K jshint is a little nuts.... it won't let me use ++ [17:18] lol [17:20] += 1 for jshint! [17:21] frankban: lol yup [17:22] although it's not allowing me to use fallthrough's on my switch statement [17:23] rofl it says file must use "use strict" then it says strings must use single quotes [17:24] benji: yay [17:26] ahah /* falls through */ is a workaround [17:28] gary_poster: re bug 1130793, what do you think about including the new helpers (from the charm) in the same MP? [17:28] <_mup_> Bug #1130793: For safety, Python charm-helpers juju-log command should insert a "--" before the log message. < https://launchpad.net/bugs/1130793 > [17:28] *sigh* tests are failing [17:29] has anyone seen this error before? [17:29] 1) Application basics "before all" hook: [17:29] TypeError: 'undefined' is not an object (evaluating 'c._buildCfg') [17:32] it's on test-prod not test-debug [17:34] bac: it isn't my greatest work, but this seems sufficient for the moment: http://paste.ubuntu.com/5565315/ [17:35] ohhh I didn't add the module to modules-prod [17:35] *the more you know* [17:36] hmm nope that shoudln't be it [17:43] whenever someone has a second I'd like some input on why this branch won't pass test-prod - I am almost 100% sure it's because it's not loading in the new extension https://code.launchpad.net/~hatch/juju-gui/1130787-subapp-app-extension [17:43] ^ jujugui [17:44] hatch: I can look; let me run the tests real quick to be sure I am seeing the same thing [17:44] sure I'm running `make test-prod` [17:45] I swear I'll get all of these procedures down ;) [17:45] arg, I have to reinstall mocha-phantomjs; one sec [17:47] benji: i think it is fine. it did find a a real bug. [17:47] benji: how to proceed? want me to merge it in and propose or do you want to? [17:47] bac: have at it [17:47] benji: ok, thanks [17:47] frankban, sorry don't know why I didn't see before. Probably too late but +1 [17:48] oh, I was not here that's why. :-P gradual lunching is the order of the day [17:49] it'd be really nice if the HACKING document was actually a literate shell script or something that would just run all the commands... oh! if we prefixed the commands with a command prompt, wwe could have a make target that pulled them out and ran them [17:50] I have to say I haven't used a makescript in YEARS until 2 weeks ago [17:50] :) [17:51] well at least directly [17:51] you poor, poor deprived person; I'm glad we got to you when we did [17:51] gary_poster: not too late, thanks. We also have a gui branch implementing a log_hook context manager. we did that as a first attempt to workaround the juju-log gojuju problem. it could be worth proposing it anyway [17:51] fine by me frankban, yeah [17:51] benji: lol iunnoo.....if I write a node script I can run it on windows too ;) [17:52] well unless it relies on a unix tool [17:52] heh [17:53] exactly! good hygiene requires using tools that will never work on windows, therefore you needn't fear being forced to use it [17:54] ok, hatch, I have reproduced the failure, looking quickly [17:54] lol [17:54] ok great [17:54] thanks [17:55] hatch: I am seeing lots of merge conflict markers in the code, e.g., app/assets/javascripts/app-subapp-extension.js [17:55] benji: just curious, did you deploy and hit with your ws client? [17:55] bac: nope [17:56] will do [17:56] benji: that's interesting because the files doesn't exist in trunk [18:01] hatch: I figured out where the conflicts came from, it is because I had modified my branch of your branch when looking at your lint issues; I've fixed it and am now looking at the test failure [18:01] ahh ok [18:03] hatch: if I apply this diff, all the tests pass: http://paste.ubuntu.com/5565397/ [18:04] I suspect that translates to "you did something wrong" ;P [18:04] but at the moment I don't have anything more insiteful to say [18:04] lol - so if I make it so that it doesn't extend the extension then it works [18:04] yep [18:05] i am pretty sure it has something to do with the module being loaded because test-debug passes just fine [18:05] so... maybe you need to initialize the config earlier, or defer calling some of the subclass code that needs the config until later [18:05] the file is in assets/javascripts does that mean I need to include it manually in modules-prod? [18:06] I don't know without looking at how those files are used. That sounds like something good to investigate though. [18:06] I'll check back after lunch and see if there is anything I can do to help. [18:07] yeah I think I'll go grab some lunch now then investigate [18:08] thanks for the help [18:41] and back [18:50] hmm [18:54] benji: I figured it out - I needed to add it to the 'merge-files' file [18:55] I should probably document that somewhere [19:06] hatch: hmm, that doesn't sound right; the dependency should be picked up automatically [19:06] all of the modules under the assets dir are in there [19:06] so I'm assuming that's correct [19:12] hatch: oh, I thought you were adding code that you had written, not third-party code [19:13] wait, I think you are adding code you wrote, in that case, assets is not the best place for it [19:15] oh...darn I just proposed it [19:15] I put it in there because that's where I saw bcsaller_ put the new routing stuff [19:16] I could create another folder for extensions I suppose [19:17] hatch: I suspect a location under "app" is most appropriate [19:18] alright I can fix that - can you comment that on the review? [19:18] hatch, if it is generic and potentially sharable then I don't think anyone can really complain that you put it in assets [19:19] sharable == usable by others [19:19] ok great - because yes it's very generic [19:20] wow that took way to long heh - at least now I know the processes :) [19:21] we should consider an index of assets that catalogs things like where we got them from and what version they are, etc. [19:21] hatch if you are done with the prototype--you've gotten feedback from Ben and you like it and it works and stuff--then move it to "release: daily" on the board so other people can use your spot on the board please [19:22] ok but the prototype I'm not proposing though...correct? [19:23] hatch, right you are done with it [19:23] as far as the board is concerned [19:24] you won't work on it again--it's done. yeah? [19:24] yep it's done [19:24] then it is as released as it ever will be :-) [19:26] great can you triage bug #1132943 [19:26] <_mup_> Bug #1132943: Create an app scaffold that others can use with Y.SubApp < https://launchpad.net/bugs/1132943 > [19:26] I already created a task for it [19:28] bac: what can I do to push our card forward? [19:29] benji: i have it in review. and i'm now trying to figure out how to get a single diff of the two pieces of work to annotate as a HOWTO. [19:29] hatch you have privileges to triage it, right? If not, I'll change it so you do. I am not the gate keeper, we (the team) are. The bug is high importance, as you correctly triaged it. This is demonstrable, because we have decided to work on it now, and that's what "high" means to us. Since you triaged it, and you are on the team, it is triaged. :-) [19:30] benji: so, i think getting this add-get-2 branch landed and doing the write up is all that needs to be done to finish the card. you agree? [19:30] gary_poster: OHHH ok, it said that the gatekeeper had reviewed....I assumed that was you :D [19:30] hatch I went ahead and triaged it, but (a) does that make sense? and (b) could you verify that you see the option to change the status [19:30] yes I have the ability to change it [19:30] hatch completely understandable. ok excellent [19:31] bac: yep, I agree [19:32] benji: so...i'm not sure what there is to do collaboratively, except maybe review the draft of the doc once i get it. [19:32] there's room to statr another command card if that makes sense [19:33] start [19:33] bcsaller_, merged re ls annotations, thanks [19:33] benji or you could do those design ones [19:33] bug 1125424 [19:33] <_mup_> Bug #1125424: Tweaks to loading messages and login visuals < https://launchpad.net/bugs/1125424 > [19:34] hazmat: thanks :) [19:35] jujugui: could I get some reviews on https://codereview.appspot.com/7381055/ please so I can open up my spot on the board [19:36] hatch: I'll take one [19:36] thanks sir [19:36] hatch, I'm working on it. (your tags are supposed to show you if anyone claimed it) [19:36] ahhh I see now [19:38] hatch, I'm not sure how much we should care yet, but AFAICT your new js module does not show up in the docs (make view-docs) [19:38] for some reason the diff on app/app.js is broken. wonder how that happened [19:38] hatch, if you don't think we should care then argue as such on Friday :-) we talked about it before and we were/I was wishy washy about it [19:39] bcsaller_: actually all the diffs are broken [19:39] :/ [19:39] gary_poster: nope I agree docs are important [19:39] except the new module [19:40] hatch, verified that bcsaller_ is correct. I was hoping that there are not in fact changes in those other files, but they do have changes in fact [19:40] in fact in fact [19:41] yeah single line changes [19:41] I can try re-propose? [19:44] hatch: yeah, try that [19:45] ok it's on it's way [19:45] Makyo: the new rapi-rollup should have basic annotations if you didn't see that already [19:46] bcsaller_, thanks, just saw. [19:48] looks like the same issue wrt the diffs [19:52] It's correct in the MP :-/ dunno what's going on with rietveld [19:53] hatch, re bug 1132933, you are right that it should be documented. That is only necessary for the "assets" bits [19:53] <_mup_> Bug #1132933: document the merge-files file < https://launchpad.net/bugs/1132933 > [19:53] I will comment to that effect [19:56] sounds good [19:58] ok the second branch that I did has the same diff bug [19:58] what the heck is going on there [19:58] bac: let me know if there is anything I can do to further the cause. In the meantime I will start a new card [19:59] hatch, at least as far as I am concerned, don't worry about it. The other files have trivial changes I see in the MP [19:59] ok I'm just concerned that they result didn't have the actual changes [19:59] hatch, look at https://code.launchpad.net/~hatch/juju-gui/1130787-subapp-app-extension/+merge/150414 [20:00] OH i know why [20:00] it's "related" to two branches [20:00] thats gota be it [20:01] hatch,it is? [20:01] don't see that [20:01] https://bugs.launchpad.net/juju-gui/+bug/1130790 [20:01] <_mup_> Bug #1130790: Create Y.SubApp by extending Y.App < https://launchpad.net/bugs/1130790 > [20:01] maybe....? :) [20:01] hatch, maybe that is confusing to lbox/rietveld. shouldn't be. isn't to the MP. [20:02] yeah oh well [20:02] hatch, lbox is something we can hack on [20:03] I don't think you want me anywhere near a makefile :P [20:03] not that I have [20:23] oh man I have too many active branches heh - I think I need folder color coding in sublime :D [20:25] bcsaller: sorry I forgot to add to my reply - that I agree that the parent app should be a bubble target but I haven't implemented that yet because I don't have a real world example of how that should work [20:25] rebubbling is very 'expensive' so the parent app could simply listen for those events [20:26] but it could also rebubble so its an internal event [20:26] both have advantages/disadvantages [20:26] once we have real code using this it's trivial to add one or the other [20:27] right now its 'listen to the sub event manually' option [20:27] feel free to disagre and tell me I'm stupid ;) [20:35] hatch I'm distracted but will be back to review in a sec. my biggest comment will be "where are the tests" :-) [20:36] oh boy I completely forgot about those in the rush to get the code in [20:37] * hatch creates a test that returns true [20:37] :-) [20:40] darn I'm getting that 'watch 'error again [20:43] hmm [20:44] does anyone know of a way to see how many watches are being used? [20:47] ahah! Sublime Text also adds watchers in which I have a lot of files open [21:03] benji: i've got a first draft of the doc but am now EOD. would you want to run with it for the rest of your day or are you now in the middle of something else? [21:05] bac: I'm in the middle of something, plus the fact that I'll only have an hour to work on it suggests that it would be best for me to not put my fingers in it [21:05] benji: a-ok [21:05] i forgot that you were faking EST and thought you had two more hours [21:05] i'll pick it up tomorrow [21:05] bye all [21:07] bye [21:08] cya [21:09] gary_poster: ok I spend the next couple hours learning mocha and writing tests for those two tasks that are blocking [21:09] if that's alright [21:11] hatch, blocking tasks? [21:12] well they are in the review section [21:12] but there is two of them [21:12] so they are blocking [21:12] ah! [21:12] mocha is looking pretty simple so it might not take 2h [21:12] I needed to refresh board [21:12] :) [21:12] hatch, heh, no yo ucan hack that together quickly :-) [21:20] hatch, I made a review finally of the older revision of your first branch [21:24] alright - looks like your comments are still very valid on the latest version - I'll try and get to them soon [21:27] cool thanks [21:27] hatch to reiterate I like what you are doing there generally [21:44] gary_poster: i'm going to be starting a little late tomorrow. have to take my mom to the airport. [21:45] ok bac [21:45] bac how was the weekend trip? [21:45] gary_poster: good. had a lot of fun on the east coast and then went south on sunday [21:46] cool bac, I'm glad [21:46] 100x35 miles means you can do a lot of exploring on a day trip [21:51] bcsaller, would you be willing to share your "improv landscape" work even if it's not finished? Would like to make sure I'm heading down the right path. [21:51] bcsaller, s/share/push [21:52] Makyo: the change to support get_annotation('env') is already merged, I'm working to add the env annotations to the delta stream right now [21:53] bcsaller, thanks [21:59] how do I go about running JUST one test suite? [21:59] do we have that ability? [22:00] hatch yes. See those links for each suite. open them in a new tab or click on them in the same tab and reload [22:07] sorry I don't think I follow [22:16] * Makyo walkinates dogalope. [22:17] haha [22:20] gary_poster: are you saying there is a *.html file somewhere which doesn't automatically run all the tests? [22:20] oh, hatch! I forgot you were not looking at these in the browser [22:20] hatch, start the test server, and look at them there [22:21] the section headers will be links [22:21] I suggest you use those [22:21] Maybe there is a way to do it with the phantomjs integration but I don't know it [22:21] ohhh now I see :) [22:21] thanks [22:35] Makyo: lp:~bcsaller/juju/delta-annotations has the delta streaming, its up for review now, but you can try it out, lp:~bcsaller/juju-gui/landscape-links has the initial wiring [23:33] hatch, thanks for clarifying the dispatch binding stuff. cool