| *** mrmartin has joined #storyboard | 05:44 | |
| *** mrmartin has quit IRC | 05:48 | |
| *** mrmartin has joined #storyboard | 06:38 | |
| *** jtomasek has joined #storyboard | 07:46 | |
| *** mrmartin has quit IRC | 08:55 | |
| *** mrmartin has joined #storyboard | 09:44 | |
| * SotK apologises for yesterday's rush patch bomb :) | 10:15 | |
| * SotK goes to review krotscheck's access tokens stuff | 10:15 | |
| Zara | krotscheck: the 'undefined' error is my work, I fear. It's looking for subscriptions, but there is no subscription for a logged out user. (I changed subscriptions so it wouldn't do one GET per subscription, but also didn't really know what I was doing, so that code isn't great. | 10:32 |
|---|---|---|
| Zara | I thought the getter setter error was something that was there before we got there, but maybe not. in which case perhaps the pagination stuff is responsible, since that refers to offset. | 10:33 |
| *** mrmartin has quit IRC | 10:34 | |
| SotK | I think the getter setter thing is caused by the autoresizeWidth directive | 10:34 |
| *** mrmartin has joined #storyboard | 10:35 | |
| *** mrmartin has quit IRC | 10:35 | |
| Zara | ahhh | 10:36 |
| *** faybrocklebank has joined #storyboard | 10:40 | |
| *** mrmartin has joined #storyboard | 10:47 | |
| *** mrmartin has quit IRC | 11:12 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Update `PUT /v1/worklists/:id/items/:id` to allow assigning a due date https://review.openstack.org/284316 | 11:36 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Add Due Dates to the WSME models for boards, worklists, tasks, and stories https://review.openstack.org/284315 | 11:36 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Add Due Dates to the API https://review.openstack.org/284314 | 11:36 |
| Zara | :0 | 11:36 |
| pedroalvarez | krotscheck: it does make sense yes | 11:45 |
| pedroalvarez | I have to admit I don't fully understand our authentication code, though | 11:45 |
| pedroalvarez | I'll have to read more about it if I want to understand it I imagine | 11:46 |
| Zara | tested behaviour, agree that the refresh token behaviour is a bit confusing for users, but this is so much less annoying than before, so have +1'd. | 11:48 |
| * SotK reads the js error log for his patches more closely and wonders how he missed that we already use momentjs | 11:48 | |
| * SotK facepalms | 11:48 | |
| Zara | I missed that too when reviewing :/ | 11:48 |
| *** mrmartin has joined #storyboard | 12:11 | |
| Zara | any objections to me merging both parts of king in the north | 12:14 |
| Zara | ? | 12:14 |
| pedroalvarez | _o_ | 12:14 |
| * Zara workflow +1s | 12:15 | |
| SotK | \o/ | 12:15 |
| Zara | so much nicer than before :) | 12:15 |
| Zara | thanks so much, krotscheck | 12:15 |
| pedroalvarez | we can always revert if something goes really really wrong | 12:15 |
| pedroalvarez | but I doubt it | 12:15 |
| Zara | yeah, the only problem I could find was with the refresh token behaviour, and that's a separate thing. | 12:16 |
| Zara | (my reviews always jump straight to the negative things! but I'm very pleased about this really :D) | 12:17 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Display a modal when a card is clicked on https://review.openstack.org/284280 | 12:36 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Allow a custom format string to be passed to time-moment https://review.openstack.org/284275 | 12:36 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Add a calendar directive https://review.openstack.org/278508 | 12:36 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Add a $resource wrapper for Due Dates https://review.openstack.org/284278 | 12:36 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Add Due Dates to boards https://review.openstack.org/284279 | 12:36 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Add onBlur and onFocus callbacks to user-typeahead https://review.openstack.org/284276 | 12:36 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Improve the board CSS a little https://review.openstack.org/284277 | 12:36 |
| SotK | argh, forgot to edit a commit message | 12:37 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Display a modal when a card is clicked on https://review.openstack.org/284280 | 12:38 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Allow a custom format string to be passed to time-moment https://review.openstack.org/284275 | 12:38 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Add a calendar directive https://review.openstack.org/278508 | 12:38 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Add a $resource wrapper for Due Dates https://review.openstack.org/284278 | 12:38 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Add Due Dates to boards https://review.openstack.org/284279 | 12:38 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Add onBlur and onFocus callbacks to user-typeahead https://review.openstack.org/284276 | 12:38 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Improve the board CSS a little https://review.openstack.org/284277 | 12:38 |
| Zara | xD | 12:39 |
| pedroalvarez | /ignore SotK | 12:40 |
| krotscheck | np everyone | 12:41 |
| *** mrmartin has quit IRC | 12:43 | |
| *** mrmartin has joined #storyboard | 12:44 | |
| *** jtomasek_ has joined #storyboard | 12:47 | |
| *** mrmartin has quit IRC | 12:48 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Update `PUT /v1/worklists/:id/items/:id` to allow assigning a due date https://review.openstack.org/284316 | 12:48 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Add Due Dates to relevant WSME models https://review.openstack.org/284315 | 12:48 |
| pedroalvarez | should w recheck king_of_the_north patch set? | 12:57 |
| Zara | ah, thanks for the tip | 13:01 |
| Zara | the very last test timed out, ugh | 13:01 |
| Zara | testing the due dates api patches is going to be exciting, since my db is currently a different version 52 (because of task links patch) | 13:08 |
| *** jtomasek_ has quit IRC | 13:14 | |
| openstackgerrit | Merged openstack-infra/storyboard-webclient: Logging in no longer causes page refresh. https://review.openstack.org/270882 | 13:19 |
| openstackgerrit | Merged openstack-infra/storyboard-webclient: Access tokens are now refreshed in preflight promises. https://review.openstack.org/284356 | 13:42 |
| SotK | \o/ | 13:43 |
| Zara | \\\\\\o////// | 13:45 |
| *** faybrocklebank has quit IRC | 14:04 | |
| *** faybrocklebank has joined #storyboard | 14:19 | |
| * Zara checks out 'fix permissions', calls branch 'permissionspermissions' | 14:33 | |
| *** b3rnard0_away is now known as b3rnard0 | 14:34 | |
| *** jmccrory has quit IRC | 15:12 | |
| *** jmccrory has joined #storyboard | 15:13 | |
| Zara | heh, I think some of the odd tag choices in 'move search icon' were to do with expected tags displaying in strange ways | 15:17 |
| Zara | (not that it doesn't still look strange, but it looked even more strange :D) | 15:18 |
| anteaya | so jeblair commented that searching for people to add as owners to a board is easier in storyboard than searching for people in gerrit | 15:26 |
| anteaya | so well done | 15:26 |
| *** Zara_ has joined #storyboard | 15:26 | |
| *** Zara has quit IRC | 15:26 | |
| pedroalvarez | as long as there's only one with that name | 15:27 |
| anteaya | well that is interesting you say that | 15:28 |
| anteaya | there are two yolanda's in storyboard | 15:28 |
| anteaya | how do we know which is the real yolanda? | 15:29 |
| SotK | yeah, the name thing is a problem | 15:29 |
| anteaya | what is your understanding of the problem? | 15:29 |
| SotK | the only identifying information for users we have is their name, email, and openid URL | 15:30 |
| *** Zara_ is now known as Zara | 15:30 | |
| SotK | we keep the email private, and launchpad openid URLs aren't useful for humans trying to determine identity | 15:31 |
| SotK | (mine is https://login.launchpad.net/+id/dQxpCYE for example) | 15:31 |
| SotK | so we only have the name as useful information to display, but name alone isn't enough to uniquely identify a person | 15:32 |
| SotK | I think StoryBoard used to ask for a username from the openid provider too, but that meant people couldn't log in with launchpad until they'd logged in to ubuntu one with their launchpad id IIRC | 15:33 |
| pedroalvarez | avatars will help, but is not enough | 15:34 |
| SotK | indeed | 15:34 |
| SotK | (my gravatar is a cloud) | 15:34 |
| pedroalvarez | I'm starting to think that emails is the only way | 15:35 |
| SotK | (and that isn't even for the right email address...) | 15:35 |
| SotK | pedroalvarez: I suspect that may be the case too | 15:35 |
| anteaya | hmmmm | 15:36 |
| anteaya | sounds like it might be a discussion item for a larger audience | 15:36 |
| SotK | I think so too | 15:36 |
| anteaya | perhaps start an email thread on the infra mailing list? | 15:36 |
| * Zara loses some backscroll, checks logs | 15:40 | |
| Zara | (yay, well done SotK!) | 15:41 |
| * SotK wonders what he did well :) | 15:42 | |
| pedroalvarez | having a cloud as avatar | 15:42 |
| persia | So, playing with https://storyboard.openstack.org/#!/board/ , I discovered it is possible to distinguish between board IDs to which I don't have access vs. board IDs which don't exist. Is this important? | 15:42 |
| persia | e.g. board 4 is not visible by me, and board 5 has been deleted. | 15:43 |
| persia | The behaviour difference is that I get a 404 for a nonexistent board, and a redirect to the dashboard for a board to which I don't have permission. | 15:43 |
| SotK | erm, that is weird | 15:44 |
| SotK | I can see board 4 but it appears to have no lanes | 15:44 |
| persia | As long as we're exposing this difference, I'd rather get a 403 when I don't have permission, as it means I don't have to go start from scratch if I mistype the URI. | 15:44 |
| persia | Interesting: now I can see board 4, board 5 redirects to dashboard, and board 6 gets 404. | 15:45 |
| persia | (on storyboard.openstack.org, for clarity) | 15:45 |
| persia | (as a side note, that https://storyboard.openstack.org/#!/board/6 produces 404: GET /api/v1/boards/6: Board 6 not found is vaguely amusing: note the difference between "board" and "boards") | 15:46 |
| SotK | that plural/singular difference is there for all the resources I think | 15:48 |
| SotK | its weird that the webclient is showing different behaviour, we should be only ever cause 404 on boards you can't see for whatever reason | 15:49 |
| SotK | any hint in your browser console as to what the error which triggered the redirect was? | 15:49 |
| Zara | mine just seems to give 404s for both 5 and 6 | 15:50 |
| Zara | what's the board 5 url that is redirecting you to the dashboard? | 15:51 |
| persia | https://storyboard.openstack.org/#!/boards/5 | 16:02 |
| persia | Err | 16:03 |
| persia | https://storyboard.openstack.org/#!/board/5 | 16:03 |
| pedroalvarez | there you have it | 16:04 |
| persia | Note that if I'm on the dashboard, I can't get the 404s: those only appear if I'm viewing a board currently | 16:04 |
| persia | pedroalvarez: The first URL was a typo (I lose the URL on redirect), not that I was getting odd behaviour from the wrong one. | 16:05 |
| pedroalvarez | hah, just wondering, because when the url is boards/* (with s) I get the redirect too | 16:05 |
| SotK | persia: can you use your browsers inspector to figure out which API request is failing? | 16:05 |
| persia | http://paste.openstack.org/show/488213/ is some console log from my browser | 16:08 |
| persia | While "5" was the latest request (causing the redirect), I was able to get a 404 from "5" earlier up. | 16:09 |
| * SotK is suddenly enlightened on how to fix that stupid marked problem, on an unrelated note | 16:09 | |
| persia | heh | 16:10 |
| * SotK remains clueless as to why you're getting a redirect though | 16:13 | |
| Zara | I can't reproduce it. :/ | 16:13 |
| * SotK neither | 16:13 | |
| persia | Might be a caching thing: my browser caches more than it should. | 16:14 |
| persia | I don't know precisely why. | 16:14 |
| persia | I found it by starting from board/1 and increasing the digit until I had different behaviour | 16:14 |
| persia | But as I test more, sometimes I get the dashboard on a given board, and sometimes I get the board, making me think it is me, and not the code. | 16:15 |
| persia | Apologies for the confusion. | 16:15 |
| Zara | heh. :) oh well, I've now found it *is* possible to differentiate between 'never a board' and 'board that was archived or you don't have permissions for' | 16:15 |
| Zara | 500: GET /api/v1/boards/98/permissions: 'NoneType' object has no attribute 'permissions' | 16:15 |
| Zara | will come up if you go to a board that doesn't exist and never existed | 16:16 |
| Zara | well, I think archived boards behave differently, maybe not | 16:16 |
| persia | Hrmm. I went up to 12 before, and now I can replicate the 500 on 11+12, and the 404 on 10 or 6. | 16:17 |
| persia | But I wasn't getting a 500 before on 11 or 12. | 16:18 |
| persia | I think my browser isn't behaving right. | 16:18 |
| Zara | I think your browser did weird things a while ago | 16:18 |
| Zara | this seems familiar | 16:18 |
| Zara | but I can't remember what they were and can't help debug it at all =D | 16:18 |
| * SotK can't see the 500 on anything | 16:19 | |
| Zara | hm, that's strange. | 16:19 |
| Zara | https://storyboard.openstack.org/#!/board/555 | 16:20 |
| Zara | gives me two errors | 16:20 |
| Zara | a 404 and a 500 | 16:20 |
| SotK | oh, I wasn't logged in | 16:20 |
| Zara | haha | 16:20 |
| SotK | https://review.openstack.org/#/c/284313/ might fix that I think | 16:21 |
| Zara | heh | 16:22 |
| SotK | hm, it makes it a 404 rather than a 500 | 16:22 |
| SotK | does the same on boards that exist but aren't visible though | 16:23 |
| Zara | an extra one? | 16:23 |
| SotK | yeah | 16:23 |
| Zara | hah, didn't catch that | 16:24 |
| * Zara can't procrastinate any more from testing due dates api patch | 16:53 | |
| Zara | and all the migrating that involves | 16:53 |
| SotK | change the 52 to a 53 in the migration | 16:54 |
| Zara | ahahaha | 16:54 |
| Zara | that would work xD | 16:55 |
| SotK | (and the 51 to a 52) | 16:55 |
| SotK | (the 51 in the migration script that is) | 16:55 |
| Zara | I think my brain just hasn't turned on today | 16:55 |
| Zara | yeah | 16:55 |
| Zara | does it ever turn on? sources differ | 16:55 |
| Zara | I will do that, thanks | 16:55 |
| * SotK looks at the tasks links stuff | 16:59 | |
| Zara | \o/ | 16:59 |
| Zara | I suspect more -1s await | 16:59 |
| Zara | but the boolean search patch should be SOLID | 16:59 |
| SotK | The code looks good, but I wonder if we want to be able to assign more than one link to a task? | 17:00 |
| Zara | yeah, I wondered about that. I figured 'start off with one, change it to more later if we have to' | 17:04 |
| Zara | though I had assumed that would be fairly simple, when it might not be | 17:05 |
| persia | More than one link to a task? | 17:07 |
| * SotK thinks its probably worth just going straight to more than one links | 17:07 | |
| * SotK thinks | 17:08 | |
| SotK | maybe its worth not insisting its a link and just allowing some text metadata with a key and value that clients can use to mean whatever they want | 17:09 |
| SotK | that might be pointless complexity | 17:09 |
| Zara | I think it's doing that practically-speaking, it's just called 'link' (and displays in a way that implies it can be clicked) | 17:10 |
| SotK | yeah, being called link is what I meant, insisting was too strong a word | 17:10 |
| Zara | heh | 17:11 |
| Zara | yeah, I wondered about that, too; I couldn't see how that'd be used in any way that I was gonna get round to coding the UI for any time soon | 17:11 |
| Zara | I could see people using it for comments on tasks, etc | 17:12 |
| Zara | (and I think the field lets you do that in practice, since I used the same thing as story description iirc) | 17:12 |
| Zara | buuuut the layout side of it would need fixing | 17:12 |
| Zara | I'm not opposed to calling it something other than 'link', anyway, I just couldn't think of a good alternative and that's all it's useful for atm xD | 17:13 |
| * SotK would like it to be called link if we only allow one, but if we allow many I think it should be a broader name | 17:14 | |
| persia | Now that I understand "link", I agree that multiple makes more sense. | 17:21 |
| persia | For example, for a given task in my project, there may be a couple of downstream trackers that contain useful information (e.g. distro trackers), so I'd want to reference more than one. | 17:22 |
| pedroalvarez | I agree that various links might be better | 17:24 |
| pedroalvarez | although that will complicate things for whenever I implement the thing that auto-updates tasks | 17:25 |
| SotK | it'll also be more complex to show on the UI, but I suspect its worth doing | 17:26 |
| Zara | :) well from me it's not really a 'it's not worth it' and more 'I am not going to get to that for some time', so it's whether it's better to have one link in the interim, or none. | 17:28 |
| * SotK wonders how urgently we want task links | 17:29 | |
| pedroalvarez | oh, question: why multiple links? | 17:30 |
| pedroalvarez | and what are the links going to be for | 17:31 |
| Zara | I could see them being for a commit, a spec, some relevant docs, that kind of thing. The most important immediate one was the commit | 17:31 |
| * SotK had the same kind of things in mind | 17:31 | |
| Zara | since I figured other things tend to apply to several tasks and can generally go in the description in the interim | 17:32 |
| pedroalvarez | I only had "a patch" in mind | 17:32 |
| pedroalvarez | but I can see how allowing multiple sould be useful | 17:32 |
| pedroalvarez | although maybe they should be separate things... for example, a field for the patch link, and a description box in where you can put links, and whatever else | 17:33 |
| pedroalvarez | (just a thought) | 17:33 |
| Zara | yeah, for me, it's a case of: I had time to work further on this earlier this fortnight, now I don't (and won't 'til I work through due_dates review), and it requires a fair bit of time on the UI side | 17:33 |
| * SotK apologises for not looking at it until now | 17:34 | |
| Zara | nah, you were busy coding the due_dates! :D | 17:34 |
| persia | One link is better than none. | 17:34 |
| pedroalvarez | nod | 17:35 |
| persia | Although I start to worry about the fragility. | 17:35 |
| persia | While I *really* want to be able to click a task in Storyboard and open the gerrit review in a new window (for tasks both in review and merged), having to specify this in the commit message hoping a bot will parse this and update Storyboard so I can link seems poorly integrated somehow. | 17:36 |
| persia | I'm not sure of the best way to solve that, but I thought I'd share my worry in the hopes someone else can think of a way for it to be robust. | 17:36 |
| * SotK doesn't see what more we could do | 17:37 | |
| SotK | Short of putting the issue tracker in Gerrit, I don't see any better way of synchronisation than a bot or plugin based on something in the commit message | 17:39 |
| persia | It's more the flexibility of the things. | 17:40 |
| persia | If we assert SB is for gerrit, we can do cross-integration things afterwards to maintain data consistency. | 17:40 |
| persia | If we want SB to work for arbitrary patch trackers (and I think we do), I think we end up fragile. | 17:41 |
| persia | I don't know the solution today. | 17:41 |
| persia | And fragile-but-useful is better than not-useful | 17:41 |
| * SotK sees the point, and agrees that fragile-but-useful is better than nothing | 17:44 | |
| *** b3rnard0 is now known as b3rnard0_away | 18:01 | |
| * SotK wonders how archiving cards instead of deleting them should work when it comes to creating a new card for a task that has an archived card | 18:04 | |
| persia | Does card identity depend on the worklist and/or board housing it? | 18:04 |
| SotK | yes | 18:05 |
| persia | Then it makes sense that if I archive foo from a worklist, and then later want to add foo, to restore old foo, so I have context, in case I set a date, etc. | 18:05 |
| persia | And nobody needs fear that if they add foo to their worklist they will end up with my recycled foo | 18:05 |
| persia | Note that if I delete foo from my worklist, and add it to a *different* worklist, it should only be unarchived if those worklists participate in the same board. | 18:06 |
| persia | (at least, that's how I imagine using it) | 18:06 |
| SotK | that makes sense to me | 18:07 |
| openstackgerrit | Merged openstack-infra/storyboard: Add column for task links https://review.openstack.org/277481 | 18:08 |
| openstackgerrit | Merged openstack-infra/storyboard: Adjust API Endpoints for Task Links https://review.openstack.org/277831 | 18:09 |
| Zara | \o/ | 18:09 |
| Zara | I didn't get that much reviewed today, should get more done tomorrow. :/ | 18:13 |
| Zara | night, storyboard! :D | 18:31 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Update `PUT /v1/worklists/:id/items/:id` to allow assigning a due date https://review.openstack.org/284316 | 18:37 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Fix permissions being able to be set by any authenticated user https://review.openstack.org/284313 | 18:37 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Move permissions checks into storyboard.db.api https://review.openstack.org/284312 | 18:37 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Add Due Dates to relevant WSME models https://review.openstack.org/284315 | 18:37 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Add Due Dates to the API https://review.openstack.org/284314 | 18:37 |
| *** jtomasek has quit IRC | 18:49 | |
| *** b3rnard0_away is now known as b3rnard0 | 18:49 | |
| *** krotscheck is now known as krotscheck_dcm | 19:50 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!