Thursday, 2018-03-29

-openstackstatus- NOTICE: Zuul has been restarted to update to the latest code; existing changes have been re-enqueued, you may need to recheck changes uploaded in the past 10 minutes00:00
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Fixes incorrect usage of page size preference key  https://review.openstack.org/55755300:41
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Fixes incorrect usage of page size preference key  https://review.openstack.org/55755300:48
*** jamesmcarthur has joined #storyboard00:49
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Fix incorrect usage of page size preference key  https://review.openstack.org/55755300:51
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Fix incorrect usage of page size preference key  https://review.openstack.org/55755300:52
*** jamesmcarthur has quit IRC00:57
*** jamesmcarthur has joined #storyboard00:58
*** jamesmcarthur has quit IRC01:08
*** jamesmcarthur has joined #storyboard01:09
*** jamesmcarthur has quit IRC01:13
*** jamesmcarthur has joined #storyboard01:18
*** jamesmcarthur has quit IRC01:27
*** jamesmcarthur has joined #storyboard01:35
*** jamesmcarthur has quit IRC01:38
*** jamesmcarthur has joined #storyboard01:43
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: The task list of a story should be sorted by ID, instead of status so that a user can quickly navigate to a specific task  https://review.openstack.org/55756901:43
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Sort task list by ID instead of status  https://review.openstack.org/55756901:44
*** jamesmcarthur has quit IRC01:54
*** jamesmcarthur has joined #storyboard01:56
*** jamesmcarthur has quit IRC02:06
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Set default project for new story dialog from header  https://review.openstack.org/55757502:22
*** vaishnavip[m] has joined #storyboard02:55
*** udesale has joined #storyboard03:29
vaishnavip[m]Hi! I'm Vaishnavi, an Outreachy Round 16 applicant to StoryBoard.03:35
vaishnavip[m]I realize I'm late, but I've been fully engaged with this for the past day and will continue to keep working now.03:35
vaishnavip[m]I've pushed some code reviews on Gerrit, left comments on a few stories and I'm going through past meeting logs right now.03:35
vaishnavip[m]I would like to work on all three components of the application that are listed on the portal, since I'm comfortable with Python, and want to learn frontend web development too.03:36
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Set default project for new story dialog from header  https://review.openstack.org/55757503:44
vaishnavip[m]Oh.. there's a bot. I'm sorry for repeated messages.03:45
vaishnavip[m]I was just amending my commits to make them more self explanatory.03:45
vaishnavip[m]Also need to test locally before committing :s03:48
*** udesale has quit IRC03:51
vaishnavip[m]I just need a little bit of help in deciding the scale of tests, like if I go deep enough I can start writing tests for variable assignments too... It's something that has always confused me about tests04:04
*** jamesmcarthur has joined #storyboard04:07
*** jamesmcarthur has quit IRC04:11
*** udesale has joined #storyboard05:04
*** udesale_ has joined #storyboard06:13
*** udesale has quit IRC06:16
*** udesale_ has quit IRC06:38
*** udesale has joined #storyboard06:39
*** udesale_ has joined #storyboard07:03
*** udesale__ has joined #storyboard07:05
*** udesale has quit IRC07:07
*** udesale_ has quit IRC07:09
*** udesale has joined #storyboard07:45
*** udesale__ has quit IRC07:47
*** tosky has joined #storyboard07:53
SotKvaishnavip[m]: hey, thanks for the patches!09:48
SotKregarding tests, I tend to prefer testing that the actual behaviour of the code matches the expectations than going so deep as to test things like variable assignments09:49
SotKat that point the tests become entwined with the actual implementation details, which makes refactoring difficult and doesn't add a whole lot of value09:50
*** dtantsur|afk is now known as dtantsur10:55
*** sambetts has joined #storyboard11:30
sambettsSotK: ping, I'm having issues creating stories on storyboard.openstack.org, the save button remains permenently greyed out even after adding tasks and everything, any ideas?11:31
*** robcresswell has joined #storyboard11:43
*** udesale has quit IRC11:46
*** udesale has joined #storyboard11:46
SotKsambetts: did you type the project name in completely before selecting it from the dropdown?12:01
SotKthere is a bug currently where doing that causes things to not update properly12:02
SotKif you just type part of the name and then select it from the dropdown it should work (assuming this is the bug you hit)12:02
sambettsyup, looks like thats what I'm hitting12:04
toskyI've seen that, and I usually ended up playing with the selection of the project, until it is really "accepted" and the button is enabled12:05
sambettsshame there is no feedback to suggest that it hasn't registered the project selection correctly, like a "You need to select a project" message when I go to click save instead of just greying it out and providing no feedback mechanism at all12:09
*** udesale_ has joined #storyboard12:14
*** udesale has quit IRC12:16
*** udesale_ has quit IRC12:19
SotKyep, that is something we should sort out too (there was a patch for it a long time ago that never got finished properly, https://review.openstack.org/#/c/264193/)12:21
vaishnavip[m]SotK: For per-user theming, is it appropriate to generate a fixed set of stylesheets on build time, then present the user with an option under preferences and fetch the relevant stylesheet using a scope variable12:26
vaishnavip[m](And then dumping it in the user preferences table of course)12:26
vaishnavip[m]I can't think of any decent hack-free way of compiling the less stylesheets dynamically (upon request from the frontend)12:30
SotKthat is likely the simplest approach to solving the underlying issue, yeah12:32
SotKwe definitely shouldn't compile the less (or hopefully sass soon) dynamically, that will be painful12:33
* zara_the_lemur__ waves at her patch from january 201612:35
SotKthe other option is to convert our stylesheets to use css custom properties instead of less/sass variables for things like the colorscheme, which will allow more fine-grained choice for users12:35
persiaNote that some effort was spent early on to support per-deployment customisation of themes, and some early deployments had sigificantly different themes.  Any implementation in this area should probably provide support for folk who actively wish to enforce certain elements in the theme for their deployment (e.g. everything is red or everything is green)12:35
SotKthe way I see it its the same for both options, have a configuration option to disable changing the deployment-default theme12:36
persiaI was thinking about having user-selectable themes be controlled by the deployment, so any given deployment could have multiple permitted themes, if they so wished.12:38
persiaUnfortunately, I'm not closely in touch with many deployers at the moment.  I think most current active deployments are SoftwareFactory users, but don't currently have a path to reach them for input.12:38
zara_the_lemur__why would a deployer want to restrict it?12:40
SotKsure, I think that just pushes the underlying issue down the road a bit though12:40
SotKas in, instead of bugs about how the styling is bad, there will just be complaints about how all the available themes are bad12:40
vaishnavip[m]We could probably allow user input of hex codes?12:41
vaishnavip[m]In addition to a predefined deployment specific set of themes12:42
vaishnavip[m]It can become a part of the user preferences12:44
SotKyeah, that's the kind of UI I'd expect for an implementation using custom properties12:45
SotKpersia: https://storyboard.openstack.org/#!/story/2000870 being a symptom of the underlying issue I'm referring to12:45
persiazara_the_lemur__: Some folk feel that once they have agreed to paint the bikeshed blue, other people shouldn't be able to use AR glasses to overly personal colours on it.12:48
zara_the_lemur__I'd argue their need for control over other people isn't our problem :)12:49
persiaSotK: That's precisely the story I thought we were discussing :)  I would expect a deployer to want to support folk who need that sort of variation.  I also remember discussions with deployers who insisted on particular colours, even when those colours caused complaints from users.12:49
persiazara_the_lemur__: While I appreciate that argument, I think we need to determine whether current (or future) deployers consider the ability to declare "this is blue" or "this is red" as a critical feature for adoption, and if they do, determine whether adoption will cause donation of resources to our project, and then balance our interest in supporting variability based on expected benefits to the project.12:51
persiaBut, as I said before, I'm remembering a past discussion, where a deployer with such control issues was willing to support development, but only with certain colours.  I don't know if that applies today.12:51
zara_the_lemur__I think accepting resources from that kinda source will lead to other headaches12:53
SotKpersia: in that case I think the tool should support the best level of customisation we can, and also allow turning it off12:53
persiaSotK: That would likely have been sufficient for the case of which I am thinking.12:54
SotKwe already have features which work similarly (editing of comments for example)12:54
persiaExcellent choice of precedent.  Yes.12:58
SotK(note that there is a limit to my preference for this type of configuration, I don't want to have any core parts of the workflow configurable)12:59
zara_the_lemur__I remember reviewing those patches and I think you're making a lot more work for yourself for a strange edge case.13:01
persiaThat's a good point: I'm defending people who aren't present (and aren't incented to be present), which suggests that they don't care enough.13:03
persiaIf we had a good way to reach deployers, I still think it would be good to ask, but in the absence, developing an easier-to-maintain solution is probably better.13:04
*** jamesmcarthur has joined #storyboard13:08
vaishnavip[m]A little off topic:13:08
vaishnavip[m]Are we considering migrating to Bootstrap 4? I ask because of the mention of SASS above. If so, I'll incorporate that into my project timeline since angular-bootstrap's last release was for 3.3.7 and it is no longer being maintained.13:08
vaishnavip[m]Also, I wish I could make more contributions and present a well rounded application that covers all aspects (increasing coverage, working on the python client, I also found that meetbot idea quite interesting), but due to time constraints, if there's something under high consideration, I'll try to get that done first since the deadline is in about 3 hours from now.13:08
*** jamesmcarthur has quit IRC13:13
zara_the_lemur__in general we want to keep components up to date though I don't know the specifics of bootstrap right now13:14
persiavaishnavip[m]: There are some patches for less->sass that depend on a few things, although I don't remember which.13:15
vaishnavip[m]persia_:  existing patches to the web client?13:19
vaishnavip[m]Oh, looks like I found one. https://review.openstack.org/#/c/379595/13:21
SotKthat's the one, I intend to go through the no doubt miserable experience of untangling it from the fontawesome change at some point soon so that it isn't depending on a thing with a -1 (and also fix the bug I left a workflow-1 for)13:24
vaishnavip[m]I'm up for doing the fontawesome patch13:28
persiavaishnavip[m]: The problem with the fontawesome patch is disagreement over the aesthetics of the icon changes: it isn't something that is easily amenable to a fix.13:29
persiarebasing less->sass to not depend on the fontawesome changes is worthwhile, but tedious.13:29
vaishnavip[m]Okay13:49
vaishnavip[m]Btw, is boartty within the scope of the Outreachy project?13:53
vaishnavip[m]Since python-storyboardclient is mentioned, we can probably have the CLI use that too13:53
*** udesale has joined #storyboard13:53
*** udesale has quit IRC13:59
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Set default project for new story dialog from header  https://review.openstack.org/55757515:54
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Set default project for new story dialog from header  https://review.openstack.org/55757516:00
openstackgerritVaishnavi Pamulapati proposed openstack-infra/storyboard-webclient master: Fix incorrect usage of page size preference key  https://review.openstack.org/55755316:04
*** sambetts is now known as sambetts|afk16:09
*** jamesmcarthur has joined #storyboard16:35
*** dtantsur is now known as dtantsur|afk17:15
*** jamesmcarthur has quit IRC17:50
diablo_rojofungi, so it looks like we need to have our tests in tree it seems. I can try to figure it out if you have those examples we talked about yesterday.18:30
fungidiablo_rojo: i started looking into it yesterday... we _could_ add tox-py27 and tox-py35 variants in project-config, but ultimately moving our jobs into the storyboard repo itself will give us increased control over them. there's some documentation on moving jobs into consuming repositories at https://docs.openstack.org/infra/manual/zuulv3.html#moving-legacy-jobs-to-projects but it's a more complex18:33
fungiendeavor than i anticipated18:33
diablo_rojofungi, okay...thats not great.18:34
fungii was going to take a stab at it today anyway18:34
diablo_rojofungi, oh okay cool.18:37
fungioh, conveniently some of our jobs are already defined in-repo18:42
diablo_rojoNice! Also, unrelated but another thing related to an item you were working on- I talked to Tacker about migrating again and they wanted to see the gerrit posting on storyboard stories again before they migrate.18:43
*** jamesmcarthur has joined #storyboard18:49
openstackgerritJeremy Stanley proposed openstack-infra/storyboard master: Increase unit test job timeout to an hour  https://review.openstack.org/55780118:50
fungidiablo_rojo: ^ is the first part18:51
fungihttps://review.openstack.org/557802 is the corresponding removal from project-config18:51
* diablo_rojo opens tabs and starts reviewing18:52
diablo_rojoI think you can thank dhellmann for the sqlite things18:53
diablo_rojoAnd getting that set up.18:54
*** jamesmca_ has joined #storyboard19:11
*** jamesmcarthur has quit IRC19:14
*** jamesmca_ has quit IRC19:52
*** jamesmcarthur has joined #storyboard19:54
*** jamesmcarthur has quit IRC19:59
*** jamesmcarthur has joined #storyboard20:00
*** jamesmcarthur has quit IRC20:04
*** tosky__ has joined #storyboard20:40
*** tosky has quit IRC20:40
diablo_rojofungi, happy to single core approve the SB  timeout increase patch after the recheck comes back in an effort to move along the utfmb4 stuff if you want.21:13
persiaAs a reminder, the general policy for Storyboard has been 1 +1 and 1 +2 for the past couple years.21:24
persiaAlso, I don't think that change is safe to merge, but happy to be educated differently.21:28
*** jamesmcarthur has joined #storyboard21:32
*** tosky__ has quit IRC21:49
*** tosky__ has joined #storyboard21:52
*** jamesmcarthur has quit IRC21:58
*** tosky__ has quit IRC22:51
*** jamesmcarthur has joined #storyboard23:00
*** jamesmcarthur has quit IRC23:04
*** jamesmcarthur has joined #storyboard23:16
*** jamesmcarthur has quit IRC23:23
*** jamesmca_ has joined #storyboard23:24
fungipersia: see my replies, including a link to another repo doing the same23:28
persiaThanks for helping me understand: have +1'd.23:30
fungiawesome23:32
fungithanks for reviewing it!23:33

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!