16:03:44 #startmeeting Horizon 16:03:45 Meeting started Tue Jul 1 16:03:44 2014 UTC and is due to finish in 60 minutes. The chair is david-lyle. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:03:46 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:03:48 The meeting name has been set to 'horizon' 16:03:50 o/ 16:03:56 hallo 16:03:56 hello everyone 16:03:58 hello 16:04:00 hi 16:04:03 hello o/ 16:04:07 hello 16:04:10 hey 16:04:16 hello 16:04:16 hi 16:04:23 HI! 16:04:27 hiya! 16:05:19 First general announcement: Welcome Ana and Zhenguo to the Horizon core team 16:05:27 Welcome \o/ 16:05:39 welcome! 16:05:44 Welcome! 16:05:48 welcome 16:05:48 w00t, good job you two 16:05:51 congrats 16:05:52 congratulations! 16:05:55 Congratulations 16:05:56 congrats : ) 16:06:03 I think ana is on pto, though 16:06:05 congrats, well deserved! :) 16:06:12 congrats! 16:06:20 no she is not tzumainn 16:06:26 she is at the mid-sprint for Horizon in Paris 16:06:29 congrats! 16:06:46 second item is https://launchpad.net/horizon/+milestone/juno-2 16:07:07 o/ 16:07:09 this has over 50 blueprints targeting it 16:07:21 that is a huge review load 16:07:45 I would be happy if those slated as Needs Code Review made it 16:08:10 considering J-2 closes on July 24 16:08:36 most items that haven't been started are just started will likely end up in J-3 16:09:33 yikes! 16:10:11 There was only one item on the agenda today 16:10:25 #topic SCSS Patches 16:10:44 doug-fish or jtomasek? 16:10:48 this came about during review of https://review.openstack.org/#/c/90371/ 16:10:52 * jrist votes merge 16:10:52 discussion regarding this is here: http://lists.openstack.org/pipermail/openstack-dev/2014-June/038971.html 16:11:18 basically, the patch is for converting to SCSS, but in the level of bootstrap we have there are defects .... 16:11:36 the defects will get fixed when we upgrade bootstrap 16:11:36 that are particular to the SCSS version? 16:11:40 * jtomasek votes for merging too as that will definitely speed up the process 16:11:49 david-lyle: yes 16:12:08 I just don't think it's worth sitting on this patch for weeks/months, especially since we're near the beginning of Juno 16:12:10 it is the bootstrap-scss pipy package that has the error 16:12:23 Question is should we merge the SCSS conversion now, with its low impact but obvious bugs, or should we merge it now and fix later 16:12:41 hi all! 16:12:42 IMO we should be using patch dependencies to handle this 16:12:44 what is the bug that surfaces? 16:12:57 some of the buttons are styled strangely 16:13:01 buttons on hover have wrong ackground 16:13:08 how long would it take to get a patch for bootstrap going? 16:13:38 tqtran: it depends, but I have started looking into it and I'd like to have it finished next week 16:14:13 it requires to get new bootstrap-scss version to requirements - I have created patch for that already 16:14:45 and then go through the templates and code and update whatever is needed to adjust to new bootstrap 16:15:06 I have already done this patch once, so I can base on that 16:15:26 the bugs are merely cosmetic then? as they are SCSS related. we can probably live with that until we can get the upgrade in 16:15:36 +1 16:15:40 yes, they are cosmetic 16:15:43 I would be nice if the lag isn't too great 16:15:53 agree 16:16:03 lag should be smaller with follow-on patches 16:16:08 but a week or two with cosmetic issues is acceptable in my eyes 16:16:13 much higher if we wait for dependent hell 16:16:13 david-lyle: I'll try to makit as fast as possible 16:16:34 So then its just a matter of getting core reviewers on https://review.openstack.org/#/c/90371/ 16:16:35 fwiw we've been bugging him for months to finish it :) 16:16:37 jk 16:16:37 haha 16:17:00 jrist :) 16:17:01 It sound like there is agreement merging is the right thing to do, despite cosmetic defects 16:17:18 they are quick to fix and easier to uncover with more eyes 16:17:18 looks like that patch needs a rebase 16:17:37 better motivation to upgrade bootstrap 16:18:22 #info merge less to scss conversion as soon as ready, bootstrap upgrade to follow shortly 16:18:45 #topic Open Discussion 16:19:09 Sahara merge stuff.... 16:19:34 We've had some more reviews this week. A few from cores. Changes continue to be small. Since those reviews appear to be settling down, I have started porting a handful of changes that were made after the merging effort got underway (way back in April). The velocity of reviews needs to increase in order to hit j2. 16:20:02 yeah lets get cores on that! :) 16:20:14 I've reviewed your massive patches like 5 times now 16:20:18 it's a pain but lets get it done 16:20:31 Cores are working but there are a lot of nits and a lot of locs 16:20:43 crobertsrh: I think the only blocker for the first patch at this point is the strange namespace workaround 16:21:14 Right. I'm not so sure that the code wasn't intended to work that way. 16:21:21 But you guys would know for sure. 16:21:47 I have it loaded in my devstack and I'm playing around with it now, to figure out the issue 16:22:29 Ok. I went ahead and made that change after nobody responded to my email about the issue. I took that as "go fly a kite, our code is just fine....deal with it" 16:23:27 And I still found issues when testing sahara stuff... 16:23:39 crobertsrh: I think the issue is we're not sure 16:23:46 :) 16:23:58 crobertsrh, you mean "Other" naming? 16:24:05 would like to figure it out before we approve a workaround 16:24:22 tmazur: Yeah, the "workaround" or "work-within" is for the "Other" naming 16:25:36 crobertsrh, I already wrote you at the bug report page my thoughts about this 16:26:38 so, we need to work this out, but the plus side is it's getting lots of eyes 16:27:07 Great. 16:29:12 Any other items? 16:29:58 any update on the client-side meetup? 16:29:58 i'd like to ask about best practices for testing changes in client code 16:29:58 https://etherpad.openstack.org/p/juno-horizon-meetup indicates there is limited interest in a late mid-cycle meetup 16:30:22 yeah.... 16:30:36 I'd like to give it another week to see if there is any more interest 16:31:01 namely, I've started fixing the bug https://bugs.launchpad.net/horizon/+bug/1191006, but don't whether or not should I write tests for it 16:31:11 I think it's difficult to express too strong an interest without a location in mind, so people know what travel arrangements they would have to get approved 16:32:11 thats true 16:33:07 party at my house? jk =P 16:33:33 tsufiev: it would certainly be nice to have tests around it, but the reward/effort ratio has to be favorable 16:34:49 Fixing the bug would be valuable even if we can't get a test to back it 16:35:43 do these test cases that we write help towards the integration test effort that jpich is leading or is that a seperate set of test cases 16:35:55 david-lyle, are there some general guidelines on writing js-code unit-tests in horizon? i'm a bit new to this kind of tests... 16:36:04 we have a ton of javascript test debt that you can't be expected to take on to fix the bug. New clientside functionality however should include tests 16:36:49 tsufiev: there are two (yes two) javascript test frameworks in Horizon now, qunit and Jasmine, I would base a test of such examples 16:37:25 there are not many, although for interaction testing the integration tests using Selenium are probably a better example to follow 16:37:48 johnma: we have both 16:37:56 david-lyle, should I write tests using both frameworks :) or one of them will suffice? 16:38:16 tsufiev: one is plenty 16:38:22 legacy vs future 16:38:38 Jasmine will work better for angular 16:38:49 ok, got it 16:40:44 johnma: to follow on, I'm not sure testing this particular case in integration tests is necessary or beneficial as integration is not required 16:40:55 for the tested functionality 16:41:35 david-lyle, the other thing I wanted to ask is how the overall process of assigning bugs in Horizon is organized (as I'm starting to contribute more systematically to Horizon)? So, if bug is assigned to no one, I'm free to take it, otherwise (say, I have ready fix) I should discuss it on irc meeting, am I right? 16:41:47 right, that makes sense David. Thank you 16:42:43 I'm looking at the j-2 content list - there is a bug regarding quotas not being editable and I don't see that on the j2 list. I think the fix has been out for review for a while 16:42:44 https://bugs.launchpad.net/nova/+bug/1299517 16:42:52 is that something that could be targetted for j-2? 16:43:07 tsufiev: if there is not an owner, feel free to assign yourself to it, if there is an owner and you have a fix or interest, feel free to contact the assignee and ask if they are still interested/working on the bug 16:43:37 if they have a strong desire to keep working on it, feel free to provide a suggestion in the bug on LP 16:43:57 and offer your help reviewing the fix once it's ready :-) 16:44:22 david-lyle, ok, will do :) 16:44:30 if it sits stagnant for an extended period of time, a core can reassign it, but that doesn't need to happen very often 16:45:51 I think if you've contacted the owner / left a comment on a bug that's been assigned for a long time and there's no answer it's ok to just take it on - but it's best etiquette to ask first 16:46:02 doug-fish: if I recall that issue correctly, the support is in Horizon already, the problem is they removed the functionality from nova and need to restore it 16:46:08 that's true 16:46:15 but the support in nova has been readded 16:46:37 and backported to icehouse 16:47:04 I set the milestone to increase visibility 16:47:15 I wonder if we should aim to backport it too 16:47:35 Backporting makes sense to me, but it was unclear how translation might play in to that 16:47:59 jpich, yep, will keep that in mind 16:47:59 I assume at one time that panel was translation 16:48:39 s/translation/translated 16:49:39 First step is to get it into master anyway, we can ask the translation team for input later on. Previously they weren't opposed to additional small additions in translations in stable releases as long as they had some notice to work on them 16:50:09 doug-fish: somehow I had missed that patch altogether, thanks for raising the issue 16:50:14 sure, np 16:53:27 I'm going to end early just so that I can say that I've done that at least once. And we seem to run out of things to say to each other ;) Thanks everyone for the reviews and have a great week. 16:53:34 #endmeeting