=== rogpeppe1 is now known as rogpeppe [12:21] interesting, it looks like the Ubuntu One team is working on a statsd client and server: https://launchpad.net/pystatsd (project registered 11 hours ago) [12:22] benji: fork/copy of https://github.com/sivy/py-statsd it looks like [12:22] peeked at that before [12:23] thanks for the link, rick_h__ [12:23] Is there an Ubuntu One team at this point? Or is just Sidnei? [12:26] Sidnei is a Team of One, like the US Army. [12:26] frankban: i'm available to QA your branch if it is still needed [12:26] :-) [12:27] i need two reviews of https://codereview.appspot.com/10024046 if anyone has some time. [12:31] bac: done [12:37] thx [12:38] bac: I'll do a review [12:38] * benji tags the card. [12:38] hey luca__, you have a good visit? [12:38] gary_poster: I did thanks, how are things? [12:40] luca__, I'm glad. had a good weekend, but would be nice if it were longer. also somewhat worried for my wife's sanity starting tomorrow, because our boys will then be out of school for the summer. :-P [12:40] luca__, call about scenarios looks good. I assume you mean right after the standup? [12:40] gary_poster: hehe that doesn't sound too good :P [12:40] gary_poster: yeah [12:41] cool [12:41] gary_poster: I'm not sure how long we need to discuss them but we can talk more tomorrow if needed, just wanted to start working through them as soon as. [12:42] luca__, I started on doc on Thursday to brainstorm stories. I never got it to where I wanted it [12:42] luca__, https://docs.google.com/a/canonical.com/document/d/1YTOmCCwSGMpZSXCQ_aRZz5UHXx6Tm4hr8X32Zz9iogw/edit#heading=h.y144flj0cwx4 [12:42] rick_h__: making the GA id configurable is interesting but seems counter to our goals. [12:42] gary_poster: fantastic, I'll give it a read over and we can talk about it later :) [12:43] rick_h__: the point, as i understand it, is for us to be able to gather usage patterns for anyone deploying the gui. giving an easy way to subvert that seems counterproductive. [12:44] but it is an interesting thing for us to discuss [12:45] luca__, cool. I forget where I was going to take the last "Building an environment" discussion right now, but I know part of it was going to be that I like what you had in your story. I think I was going to brainstorm different users/scenarios for when you might build from charms (experimenting, small use case, or building a bundle for the future) and when you might build from a deployer file/bundle (duplicating d [12:45] ev or prod environment, deploying known good config from charm store and then tweaking, ...) [12:46] gary_poster: I see, sound good :) [12:49] bac: I am back. It would be great if we can take a look at the error you reported. It's weird. The usual chat seems open. [12:51] rick_h__, hey. When does sinzui usually start his day? [12:57] bac benji frankban teknico, any of you interested in attending a call about grid units/making the gui friendly to mobile/responsive web design (I used to know this as "flexible web design" but now it seems to also refer to different form factors)? It is in 4 minutes. I'll attend. It's not a good time for Ben Matt or Jeff or I would have invited them. [12:57] 3 minutes :-P [12:58] not especially, but I'll be glad to drop in if no one else will take the bullet [12:58] gary_poster: yep, pass the link [12:58] teknico, priv msg [12:58] gary_poster: usually between now and our 9:30 standup [12:59] ok rick_h__. I would have asked him to ask you the question above :-) [12:59] jcsackett, too [12:59] gary_poster: heh, responsive means using media queries in the css, so just an extra layer on flexible [13:00] gary_poster: sure, I'll attend [13:00] rick_h__, gotcha. cool, will privmsg [13:01] luca__, responsive call is now, yeah? Rick and I are here. [13:01] gary_poster: coming! [13:01] bac: yea, but will most production deploys agree to share that info? and it solves the jujucharms.cmo case for my needs :) [13:01] :-) [13:02] rick_h__: on a call so i can't think. :) [13:02] bac: I think a cool default would help but allow people to use it as they might find handy :) [13:12] benji: i like this "The looks good." i think i'll adopt it as a reviewer catch phrase. [13:12] The looks, they are good, yes? [13:14] Time to fail for the "As seen on TV" self-shrinking garden hose I got for my birthday: 10 days. [13:33] frankban: tests finished. http://paste.ubuntu.com/5751661/ === BradCrittenden is now known as bac_ [13:34] bac: intersting, does that file exist? [13:35] frankban: you mean /dev/null? [13:36] bac: /home/bac/charms/precise/juju-gui/tests/.venv/local/lib/python2.7/site-packages/xvfbwrapper/xvfbwrapper.py [14:02] Anyone up for another review/QA? https://codereview.appspot.com/10048043/ [14:02] bac_: my call is done if you want to continue discussion of the GA id as config? [14:02] Note that that includes a comment WRT a deferred task. Card is already created and assigned. [14:04] * gary_poster had to actually google to discover what bac's hose could possibly be. http://windo.en.alibaba.com/product/636728871-214175883/2012_Shrinking_Garden_Hose_As_Seen_On_TV.html for those with similar interest [14:05] gary_poster: it wasn't really an endorsement [14:05] although those were some of the most fun 10 days of watering [14:06] hi rick_h__ [14:06] sure [14:06] Man, I love watering. [14:07] morning all [14:07] bac_: so I'm not up on the back story of the feature. I assume that it's defaulted to on, allow users to turn it off, and used mainly for helpful reporting of usage to make life better for all users [14:07] bac_: my argument is that, allowing users to do their own reporting is a nice cool feature that our team would love to have in the charm, and doesn't adversely effect the original goal since most users won't change the default config [14:08] the users that go into the config to chage the GA id are likely to be the same that would just turn it off anyway. [14:08] rick_h__: which deployment are you specifically talking about? [14:08] uistage? [14:09] rick_h__: wanna chat? [14:09] bac_: sure [14:12] luca__, thank you for organizing that mtg. was great to have. would be thrilled to share more across teams [14:13] gary_poster: thanks Gary :) I was really hoping it would be worth everyones time. [14:14] orangesquad: I created an mp for a mid-implementation review: My branch is 2 days old, now and there are a few issues I am not sure about: https://code.launchpad.net/~sinzui/charmworld/charm-model-fixes-views/+merge/168175 [14:17] sinzui: Oh, good grief. I assumed "optional" mean "may be none", but it appears to mean "may not be present": http://docs.mongodb.org/manual/reference/gridfs/#gridfs-files-collection [14:20] abentley, If you mean "aliases" then I too think the value is None. [14:21] abentley, This is the dict problem we have with charms isn't it? [14:21] sinzui: I meant files.contentType [14:22] yep, it is listed about aliases as optional. We both think the value can be None, but the apparent use is the key may not be present [14:23] sinzui: That's okay with a dict, but optional attributes on an instance are a PITA. [14:27] luca__: yea, thanks. I'm bouncing off walls at the idea of a shared CSS lib. [14:28] orangesquad: could you please review https://code.launchpad.net/~abentley/charmworld/same-name-related/+merge/168190 [14:28] * sinzui looks [14:28] rick_h__: haha, glad it was helpful :) [14:30] * rick_h__ looks forward to the day when he doesn't have to look up the #-code for "ubuntu orange" [14:30] rick_h__, sinzui in prep for meeting, https://bugs.launchpad.net/juju-gui/+bug/1186299 is done right? [14:30] <_mup_> Bug #1186299: The charm browser quality tab does not have data

[14:30] gary_poster: yes, I'll close up and link to the branch. Sorry I missed it [14:31] gary_poster: reminds me, luca__ did you see my email about the quality tab and get a chance to peek at it? [14:31] luca__: kind of a 'proposed' solution and would love feedback. [14:31] rick_h__, great thank you. when will charm browser default land? that was another goal for today [14:31] gary_poster: jcsackett is working on it now I believe [14:31] cool rick_h__ thanks [14:31] sinzui: Is anyone chasing down the Tarmac issue today? [14:32] abentley, r=me, matsubara fell out of the channel the moment I tab completed him [14:32] fastest review evar! on that branch [14:32] sinzui: ! [14:32] abentley, we are thinking the same I think :) [14:32] I will follow up with matsubara [14:47] rick_h__: can I set up some sort of config to open a tmux window with a certain window split every time? [14:48] hatch: see pm [14:51] thx [14:56] gary_poster: rick_h__ makes the good point that until the GA opt in/out is easily available via the charm we should have the default for useAnalytics set to false. i agree and would like to land this initial branch with it turned off. a-ok? [14:57] bac +1 [14:57] gary_poster: he also suggests a chat with jcastro to brief from a community perspective, which i'm also +1 and will be happy to do [14:58] thank you bac_ , cool [14:58] we have a sync call in 2 minutes anyway! [14:58] :-) [14:58] y [15:13] sinzui: can we ask for an update to mjc to get v2 api available on there? [15:13] I will after this meeting, you you can ask [15:14] sinzui: k, I'll bug you post-meeting. I thuoght we needed to file an RT and wanted to check on the process. [15:16] rick_h__, we do, and then we prod webops. We know the deploy will fail so hand holding is needed [15:16] man I can never remember the syntax for linking with rst [15:17] .. _something: google.com [15:17] ...something something something....dark side [15:17] lol [15:20] rick_h__: https://twitter.com/FromAnEgg/status/344111810431049728 :) [15:21] hatch: :P http://www.webupd8.org/2012/03/retext-30-released-text-editor-for.html [15:21] sure but that's another app :) [15:21] maybe there is a sublime plugin [15:21] I think there's a few editors that do live viewing [15:22] hatch: yea, but it's live so you don't have to type/save/type/save/etc [15:22] oh kewl [15:33] rick_h__: why did you say // THIS IS BAD!!!!!!!!! on ln 1000 of app.js? [15:33] it's not bad, the router is just broken :) [15:34] although Ben fixed it this weekend I think [15:34] hatch: oh crap was that in there? [15:34] lol [15:34] hatch: I was walking through the issue with jcsackett in a call and just marked out the line as we talked [15:34] completely missed that in the diff, thought I reverted when I switched colo-branch [15:35] hatch: lmao, thanks for catching. Mechanical whatever I guess... [15:35] store/charm.js ln 164 "charmworldj" [15:36] wtf, it's not in my local branch [15:36] man it's a good thing I'm going through this :P [15:36] * rick_h__ is so confused [15:36] bad merge? [15:37] must be, both of those aren't in my local code [15:37] * rick_h__ does a re-propose [15:40] doesn't look like codereview lets you do a self review - it doesn't turn your comment green [15:40] this made me laugh this morning http://xkcd.com/742/ [15:42] hatch: cool, thanks for peeking at it. Yea, looks like a mess up from when I did a bzr switch between branches [15:43] is that a bzr issue when switching branches like that? Or did you mess something up? (curious) [15:43] so I don't know exactly what I typed/did but I had uncommitted changes in a branch I didn't want. I created a new one and went to work and it carried over the uncommitted changes I think [15:44] ahhh yeah I noticed it does that [15:44] found that odd [15:44] but I suppose git does something similar [15:48] orangesquad. We want to deploy, but we know migrations will fail. I propose landing a branch that makes migration 7 a no-op so that the deploy works without intervention. I am preparing a script that walks through the charms and generate a report to discover which charms have circular references. [15:49] sinzui: Makes sense. [15:50] jujugui call in 10. kanban now [15:50] abentley, matsubara is working on the ssh-to-Lp problem. I don't expect tarmac to work today [15:51] sinzui: Okay. [16:00] jujugui call now [16:02] Welp. [16:05] abentley, do you have a moment to bless or curse this merge: https://code.launchpad.net/~sinzui/charmworld/disable-migration/+merge/168482 [16:07] sinzui: I am confused about whether we still follow Launchpad policies. If so, https://dev.launchpad.net/PolicyandProcess/XXXPolicy would dictate a bug number. [16:08] abentley, yeah, I sould have one since I have a card on the board for it [16:13] abentley, I updated the comment and reported bug 1189531 [16:14] sinzui: r=me [16:14] thank you [16:15] luca__, alejandraobregon shall we meet in guichat or somewhere else? [16:15] guichat is fine gary_poster === deryck is now known as deryck[lunch] [17:08] * hatch needs a break from docs [17:11] is there some reason errors are thrown with this format? "INVALID_STATE_ERR : Connection is open to another client." [17:11] specifically, the space before the colon; that seems like a weird thing to do in English [17:12] English has consistent grammar rules? [17:12] since when? [17:12] :P [17:45] gary_poster: I'm at a place with the documentation where I'd like to get some input from Ben before moving forward so should I move on to the ghost config with the view controller or did you have something else in mind? [17:47] hatch, ...could you (a) land bcsaller's nsrouter branch and (b) work on replacing the alerts dropdown bootstrap stuff? [17:48] hatch: on the ns-router lines I've changed it in my local copy, just have getHash do return window.location.hash; [17:48] hatch: please take a peek before landing it as it is still broken for our needs in getting the active tab support landed [17:50] gary_poster: can do [17:50] thank you hatch [17:52] no problemo === deryck[lunch] is now known as deryck [17:54] rick_h__: guichat? [17:54] hatch: sure thing [18:04] orangesquad: could you please review https://code.launchpad.net/~abentley/charmworld/file-etags/+merge/168497 [18:05] * sinzui looks [18:07] * hatch lunching [18:11] first they steal 'launchpad' and now 'mavericks' (sic) [18:12] pourquoi? [18:12] * hatch used the wrong word [18:12] hatch: you're not up on your apple-fu [18:13] sorry I've been WORKING today instead of reading blogs :P [18:13] hatch: you need more screens :P [18:13] I know! [18:13] maybe now you can upgrade that laptop and get 3 thunderbolt displays on it so you can work with news on the side and know what bac knows :) [18:13] damn shared video ram on a mac mini [18:13] lol [18:13] damn rights! [18:14] man i wish i had an external monitor [18:14] xbox one $500? small price to pay for an invasion of privacy! [18:14] hatch: come on, you get that for free with your gmail, twitter account, and ISP [18:15] what is lacking with that is a camera and microphone in my livingroom [18:17] glad MS is planning on filling that void [18:17] abentley, r=me [18:17] sinzui: Thanks! [18:17] jujugui Another call for reviews: https://codereview.appspot.com/10048043/ Just need one more :) [18:19] rick_h__: sorry I just don't trust MS enough to place an always-on video camera and microphone into my livingroom [18:21] s/MS/anyone/ [18:21] jujugui: i too must ask for reviews https://codereview.appspot.com/10156043 [18:21] jcsackett: I'll peek in a second [18:22] jcsackett: hah truth! Although if it was OSS then at least people could look into the source to find the inevitable bugs [18:22] jcsackett, on it. [18:26] jcsackett: couple of notes while I look into the test regex part [18:33] rick_h__: the sad thing is that I actually want the new xbox but not with the kinect [18:42] jcsackett: I see what's up with the test issue, I'm working on some code to help it out. Will be a few min. [18:45] How long until people figure out you can just put the xbox in a well-lit diorama of your livingroom with the proper number of people painted on the couch? [18:48] jcsackett: check out lp:~rharding/juju-gui/sanity-check-default please and let me know how that looks [19:04] abentley, This is the script to look for errors in charm data. Dev and staging are fine as I expect. Do you have any additions to the script? http://pastebin.ubuntu.com/5752622/ [19:07] sinzui: No, it looks good to me. [19:07] fab [19:07] Makyo: lol! [19:13] hey hazmat. May I schedule 30 minutes with you tomorrow to talk about bundles/deployer stuff? Almost certainly will need more later, and with other people, but this can get things rolling again [19:15] arosales, hey. In https://docs.google.com/a/canonical.com/document/d/1FU29PCI2I2a-PUKCMHvt9jK6Jr1TcfyvUFDypHOQ_ic/edit# does ecosystems need to do anything to have test results ready on time or are they already ready to go? [19:15] on the ecosystems side I mean [19:16] Makyo: do I have to re-review your branch? [19:17] hatch, Don't think so, though you're welcome to. [19:21] gary_poster, they should be ready to go [19:21] Makyo: didn't we decide that we weren't going to fire events on other classes? re topology/panzoom.js ln 181 [19:22] if orange run into any issues on the parsing of test results though please let me and marco know [19:22] great thanks arosales, so that deliverable is all on UX and GUI (Orange). I'll call that out explicitly on list [19:23] hatch, I don't remember any such decision, but I do recall us deciding that Topology was it's on special sort of madness. [19:24] yeah I remember that part haha [19:24] gary_poster, sure [19:24] Maybe we just discussed it but didn't make any decisions [19:24] anyways, my LGTM stands :) [19:25] gary_poster, aiui that testing data from jenkins was in a consumable form. So yes next actions for testing are on orange and UI. [19:25] s/ui/design [19:26] thanks hazmat . created event. gave you edit power if you need to move it, but it worked with your calendar as it is now [19:27] arosales, cool, thanks. sinzui, do you all already have a plan for slurping up the jenkins data? does it go in m.j.c and then the GUI gets it from there? [19:28] We have been slurping data since before orange took control [19:28] cool sinzui, so this is entirely simply a matter of exposing something in the GUI? The data already is moving to the GUI? [19:28] we also know know then data format changes because it breaks staging and production 15 minutes later [19:28] that's right [19:29] gary_poster, np. [19:29] gary_poster, thanks, done. [19:29] sinzui, let us (me and marco) know if you run into any issues slupring the testing jenkins data [19:30] sinzui, data format changes: ok. what's the status of that for long term m.j.c happiness? [19:30] is that call-worthy? [19:30] arosales, we will :). All has been good for that last 6 weeks. [19:30] sinzui, thanks, and good to know :-) [19:31] gary_poster, Its in the long-term charmworld stability blueprint [19:31] ok sinzui thanks will see that when I get there then [19:40] sinzui: I would like to discuss bug #1176452 with you. [19:40] <_mup_> Bug #1176452: ElasticHttpError:NumberFormatException[For input string: "yes" [19:41] in a few minutes abentley, otp with dan [19:41] sinzui: ack [19:41] I always thought otp = on the potty [19:42] but it makes a lot more sense thaat it means on the phone [19:42] lol [19:45] benji: do you need to re-review the test i added? 'lbox propose' is running now and 'lbox submit' is on hit heels. [19:45] * BradCrittenden hates his droppy internet === BradCrittenden is now known as bac [19:46] bac: nah, if you're happy with it then I'm fine [19:46] benji: i'm ecstatic [19:46] * benji throws confetti [20:07] orangesquad, tarmac is back. We were blacklisted for hammering bazaar.lp.net [20:08] sinzui: Ah. [20:16] hatch: did you qa teknico's branch? [20:17] bac: I did - I didn't LGTM it though [20:18] hatch: i changed useAnalytics to true and ran it but don't see the banner. am i missing something obvious? [20:18] I think you need to be on sandbox: false with rapi running [20:19] really? huh. [20:20] oh wait - no that was just for prod [20:20] nm [20:21] bac: I think app.js line 1217 is the issue [20:21] abentley, sorry for the delay. I an talk now [20:21] I am not sure if that is being set [20:22] sinzui: Cool. [20:22] abentley, I can talk now [20:24] sinzui: Your microwave may not be wi-fi compatible :-) [20:24] abentley, i just got kicked out [20:25] sinzui: I have reinvited you, but it's just ringing and ringing and finally said "It looks like Curtis Hovey isn't available." [20:25] gary_poster: thoughts on GA on uistage? leave it in or rip it out for now? [20:25] bac: oh nm it appears to be being set [20:26] hatch: well i can't get it to work [20:26] hmm sorry I'm not sure - I can look into it further if you would like but I have a few other things in front of it atm [20:26] hatch: nope, don't bother [20:26] bac we have to have it as the cornerstone of everywhere else for GA to work, I thought. wrong? [20:26] s/of/for/ [20:26] bac: alright - imho it needs a bit more work before it's landable anyways [20:26] or something... [20:27] gary_poster: on uistage i manually inserted the snippet into index.html. my branch has now landed that has it for real, but with the config turned off. [20:27] gary_poster: i propose reverting uistage to not be divergent [20:27] oic. +1 bac [20:27] gary_poster: but that will disable GA for now [20:27] bac will that screw up GA for other sites? [20:28] gary_poster: other sites? other deployments will not have it turned on by default and i assume no one is going to hack it in [20:28] bac, if so, I suggest reverting index.html but changing config to turn it on [20:28] er.... [20:28] bac, :-) [20:28] gary_poster: what other sites are you referring to? [20:29] bac, I mean for when we turn this on in the charm [20:29] oh [20:29] if we turn this on in the charm, GA needs to already be happy [20:29] the jujucharms must be on stuff [20:29] yeah, well, i don't know how that'll work since it has been seen once. [20:30] bac, I vote to leave it turned on in uistage, assuming that the cookies warning is working [20:30] wdyt [20:30] gary_poster: so for uistage, revert index.html to the branch version and then set config-prod: useAnalytics true. seems like a good plan [20:30] exactly [20:30] cool, thank you [20:30] the cookie warning will not land until tomorrow at earliest [20:30] oh [20:30] ok [20:30] lots of work from review [20:30] mm [20:31] then useAnalytics false bac, till that lands [20:31] ok [20:32] no new MBP yet :( guess that's good for my credit card balance :) [20:39] gary_poster: done [20:39] thanks bac. [20:39] yeah hatch, was disappointed too :-) [20:40] other sites are claming September which is quite a ways away for those who are in the market.. [20:40] * hatch <---- this guy [20:41] huh. it's 364 days since last update, which is most in their history (http://buyersguide.macrumors.com/#MacBook_Pro) [20:41] oh wait [20:42] never mind [20:42] http://buyersguide.macrumors.com/#Retina_MacBook_Pro [20:42] yeah it's definitely not been a long time....but haswell! [20:42] I know [20:42] completely with you [20:45] to make up for my negative review this morning, i heartily endorse this product: http://www.amazon.com/dp/B00008GS96/ref=tsm_1_fb_lk [20:45] bac: lol nice comment in the review :P [20:45] bac: here the bugs aren't big enough for those :( but what they lack in size they make up for in quantity [20:46] we have one upstairs and down [20:46] they make up by transmitting deadly disease [20:46] i mean, ours are tiny but awful [20:50] that's unfortunate - ours are mostly disease free (as far as I know) but the ticks apparently aren't the cleanest [20:51] * bac walks dogs, jojo + 2 little white things who spend 90% of the time tangled up in the leash [20:52] lol [20:58] Dogwalk. Eyes need a break;. [21:12] jcsackett: still around? [21:12] rick_h__: yup. [21:13] jcsackett: so looking at the changes, was there something to not moving the browser subapp up into the original subApplications def vs a push in init? [21:13] rick_h__: i don't follow. [21:13] jcsackett: and the problem is that when the browser isn't FF'd it always tries to run and at least show the sidebar. There's no store though. [21:14] rick_h__: i know that part; but the phantom parsing of "test" is still not sorted. [21:14] jcsackett: line 55 can have the subapp vs line 355 [21:14] as to the push, i have no preference. i whacked the flag and then hunted down test errors. [21:14] jcsackett: ok, right, my first pass I mentioned that there's no need to have the subapp in the init and the extra call to push it [21:14] rick_h__: i don't think i saw that. [21:14] * jcsackett looks [21:15] jcsackett: so the 'test' is that someone is testing urls that are just /something/test and it caused the code in the subapp browser to check it as an id [21:15] jcsackett: so adding it to the id ignore was just a lucky 'ignore what someone decided to use in a test' vs a real fix to not try to run browser code when it's not intended [21:17] jcsackett: so LGTM after moving the browser subapp def back to the top of the class per https://codereview.appspot.com/10156043/diff/1/app/app.js#newcode355 [21:17] rick_h__: huh; i grepped for that in the test file but didn't see anything to that effect. [21:17] jcsackett: it was in other tests [21:17] ah, and we have nothing for test isolation. right. [21:17] * rick_h__ goes to look if he has the failures [21:17] jcsackett: right, all tests around the App() will hit browser the subapp since it's part of it now that it's not FF'd [21:18] jcsackett: so the 'hack' is that other tests won't set a charmworld.url as part of their setup. [21:18] jcsackett: so the browser subapp just ignores things if there's not a valid store. Browsing makes no sense without it. [21:18] right; i like the store check as a fix. [21:18] it was the phantom url that was irking me. [21:18] might as well be a "isTesting" flag, but it's a bit more specific [21:19] test/test_charm_running.py: self.load('/test/') [21:19] might have been teh cause [21:19] just a quick grep, but anyway. Changes look good to me [21:20] jcsackett: thanks for the update and un-FF'ing the browser. Look forward to not needed :flags:/ :) [21:29] rick_h__: I think I have it #.[^?/]*[$]? [21:29] hatch: why the [$]? [21:29] hatch: you should be able to remove that part [21:29] ahh yes I can [21:29] thx [21:29] hatch: np [21:30] turns out to be exactly what you said all along [21:30] I was thrown by the wako results in regexpal [21:30] once in a while I get lucky :) [21:30] needed to revert to the console [21:31] yea, tests or bust! [21:31] regex pal for some reason highlights http: as well [21:31] oh hmm, strange. [21:32] the console results are correct though [21:34] ohh the regex is slightly off [21:34] it should drop the first # [21:35] hatch: ah, well the regex is only matching. In the match you can split/slice/trim to get rid of it. [21:35] hatch: I think there's some rtrim/etc functions available in there /me isn't at his desktop atm [21:35] yep but the regex should only return the proper data [21:35] why take another step when you don't have to [21:36] hatch: well, you're asking it to match. It's going to match the character. I tried once to do a #(xxx) type of things but didn't see a good way to get at the () part [21:36] ahh ok [21:36] hatch: well, if you find a good way let me know, but I've manually trimmed/etc in my use. You know it'll be there because it's part of the match so it'll always be there [21:41] rick_h__: ([#])(.[^?/]*)? [21:41] use this url: http://gui:8888/sidebar/precise/ceph-9/:flags:/browser_enabled/#bws-configuration=foobar&stuff=awesome?query=charms [21:41] the 3rd match.... [21:41] pretty sure this isn't ideal [21:41] haha [21:42] can drop the [] in the first parens I guess [21:44] hatch: this is nicer than just url.match(/#[^?/]/).replace('#', ''); ? [21:44] sorry [21:44] hatch: this is nicer than just url.match(/#.[^?/]*/).replace('#', ''); ? [21:44] it's a lot harder to understand so yes [21:44] :P [21:44] lol [21:48] rick_h__: did you want to pull down my branch to test? [21:48] hatch: not atm, I'm EOD and on the tablet. [21:48] ahh ok jcsackett still around? [21:48] hatch: my branch if up if you want to test it out 100%, but if that urls passes I'm good [21:49] hatch: go ahead with it and I'll pull the update down into my branch tomorrow [21:49] here is the diff, look good? http://bazaar.launchpad.net/~hatch/juju-gui/routing-hash-query/revision/714 [21:50] https://code.launchpad.net/~rharding/juju-gui/direct-2-tab is my branch [21:51] hatch: looks ok from here [21:51] I figured you could just see the tests to see that it appears to cover all of the use cases we need [21:51] urls are the epitome of edge cases [21:52] hatch: yea, don't need to get 100% atm, but having tests and getting the - we know we need will help :) [22:05] rick_h__: so I can't propose this because lint is giving me an odd error [22:05]