| *** aswadr_ has joined #storyboard | 04:41 | |
| *** matthewbodkin has joined #storyboard | 07:35 | |
| *** yarkot1 has quit IRC | 07:40 | |
| *** fay has joined #storyboard | 07:56 | |
| *** fay is now known as faybrocklebank | 07:56 | |
| *** openstackgerrit has quit IRC | 08:48 | |
| *** openstackgerrit has joined #storyboard | 08:48 | |
| *** bethwhite_ has joined #storyboard | 08:52 | |
| Zara | hm, https://bugs.launchpad.net/trove/+bug/1294904 looks like the thing in our logs | 09:40 |
|---|---|---|
| openstack | Launchpad bug 1294904 in OpenStack DBaaS (Trove) "tox tests occasionally fail on quota tests with "Mismatch Error"" [Low,Confirmed] | 09:40 |
| Zara | is it just those patches? | 09:40 |
| SotK | yeah, I suspect there is something weird in either our code or our test code since we've never seen it before | 09:48 |
| Zara | I wonder if it's interpreting % as modulo | 09:48 |
| Zara | in https://git.openstack.org/cgit/openstack-infra/storyboard/tree/storyboard/tests/api/test_comments.py | 09:49 |
| Zara | idk why it would suddenly do that or go wrong now, though, and I feel like it'd be a more dramatic error | 09:50 |
| SotK | hooray I finally reproduced it locally | 10:29 |
| SotK | now to find out why it doesn't work | 10:29 |
| SotK | hmm, it seems to work when not authenticated, but fail when authenticated | 10:46 |
| SotK | getting somewhere | 10:46 |
| Zara | yay | 11:00 |
| Zara | that sounds hopeful | 11:01 |
| * SotK sends https://review.openstack.org/359895 again | 11:29 | |
| SotK | get it together openstackgerrit | 11:29 |
| Zara | it says 'nah, mate' | 11:29 |
| SotK | if the tests fail now I'll be extra confused | 11:29 |
| *** alexismonville has joined #storyboard | 11:33 | |
| *** alexismonville has quit IRC | 11:38 | |
| SotK | argh | 12:02 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Hide timeline events and comments of private stories https://review.openstack.org/359895 | 12:03 |
| SotK | when you remake the fix on a different machine... | 12:03 |
| SotK | good afternoon openstackgerrit | 12:03 |
| Zara | :) | 12:04 |
| *** alexismonville has joined #storyboard | 12:07 | |
| Zara | just remembered I was reading a novel the other day, and it included the sentence, "What *is* Jenkins?". made me laugh. | 12:18 |
| *** wendar_ has joined #storyboard | 12:19 | |
| persia | Wodehouse? | 12:20 |
| *** wendar has quit IRC | 12:20 | |
| Zara | Thackeray | 12:20 |
| *** persia_ has quit IRC | 12:25 | |
| *** persia_ has joined #storyboard | 12:26 | |
| Zara | oh yay, test passing | 12:50 |
| Zara | guess I can take a look again now | 12:50 |
| *** yarkot1 has joined #storyboard | 13:03 | |
| SotK | \o/ | 13:09 |
| matthewbodkin | Does anyone know if we have a .less file for the about page or anything like that to avoid me having to put a `style=""` in the html file? | 13:09 |
| persia | matthewbodkin: Does it need to be an about-specific file? Could it be an about-specific style in a more generic file? | 13:17 |
| matthewbodkin | That's what I was thinking because there doesn't seem to be an about-specific file, I'm just not sure what other generic file it would be and was hoping someone could point me in the right direction | 13:19 |
| matthewbodkin | It's no worries I can just keep looking but I didn't know if the file in question was somewhere obvious and I just couldn't see it | 13:21 |
| Zara | okay, so it's possible to find that hidden comments exist, but not what they say, as the error message for no comment isn't the same as the message for a hidden comment | 13:58 |
| Zara | 'Comment 165 not found' | 13:58 |
| Zara | is when comment is hidden | 13:58 |
| Zara | ''NoneType' object has no attribute 'event'' | 13:58 |
| Zara | is when comment is not present | 13:58 |
| Zara | also, the faultcode is 'server' for the real 404 and 'client' for the fake one | 13:59 |
| Zara | I don't know that it's a big concern but it does suggest that it's reporting that they're being treated differently, which might mean that reporting is happening elsewhere | 14:00 |
| Zara | for stories, though, the codes are the same (ie: 'story x not found' whether it's hidden or not present) | 14:00 |
| Zara | suspect this may be the case for any attribute of a hidden thing | 14:01 |
| persia | Is that just a UI failure, or are things usefully blocked in the API? | 14:01 |
| Zara | api (I'm using the browser to navigate to api/v1/ and not passing any auth) | 14:03 |
| Zara | there's no way to browse to events directly under the ui, so users will just get the story error | 14:05 |
| Zara | so that side should be fine | 14:05 |
| Zara | since that's the same for hidden/non-existent stories | 14:05 |
| Zara | you'd also have to be paying pretty close attention to notice the things were different. | 14:06 |
| persia | Right. I just feared there may have been a bug processing data, rather than sending. I do not think that is too dangerous a leak: one can get the entire population of a thing, and deduce the hidden ones by gap analysis anyway. | 14:08 |
| Zara | yeah, theoretically they *could* have been removed so the deduction would be uncertain-- but I agree that this on its own isn't a big problem | 14:09 |
| Zara | my only concern is if it suggests that something else isn't being handled in the way we want to handle 'private' things | 14:09 |
| Zara | in a way that could be more of a problem | 14:09 |
| persia | Ah, true. | 14:15 |
| SotK | hm, that is unintentional, will fix | 14:18 |
| Zara | hah, I +2d since it was still an improvement, but it doesn't have +1s for +A yet, so cool | 14:20 |
| Zara | as far as I'm aware, the about page doesn't have any specific styling; if you give more detail about what issue you're trying to fix then we might be able to point you at the right place in the code. | 14:28 |
| Zara | matthewbodkin ^ | 14:28 |
| matthewbodkin | the small print at the bottom of the page 'Webclient version: ...' runs off the edge of the page when in mobile view. This can easily be fixed by using word-wrap: break-word but I'd have to add a style="" to do that | 14:31 |
| matthewbodkin | it has a class called text-muted but if I do anything to that class in another file then I fear that could cause problems with everything else that has that class | 14:33 |
| matthewbodkin | if that makes sense haha | 14:33 |
| SotK | it does :) | 14:35 |
| SotK | you can just give it a new class as well and define it somewhere relevant | 14:36 |
| * SotK would suggest in src/theme/base/typography.less | 14:36 | |
| matthewbodkin | I was thinking of giving it an id so that I could target that id and class so it'd only affect the line I want | 14:36 |
| matthewbodkin | okay yeah | 14:37 |
| Zara | (yeah, I think the text-muted class is just meant to grey out the text so we shouldn't sneak that in there.) | 14:37 |
| Zara | storyboard has 'truncate' used in places so you might be able to use that | 14:37 |
| SotK | good point actually, we could just truncate the sha to 7 characters and have done with it | 14:38 |
| matthewbodkin | okay so if I just give it the class truncate then all will be well? | 14:38 |
| persia | So class="text-muted truncate" ? | 14:38 |
| matthewbodkin | okay great thanks | 14:38 |
| Zara | looking around, it's a filter rather than a class | 14:38 |
| matthewbodkin | simple enough | 14:38 |
| matthewbodkin | oh okay | 14:38 |
| Zara | there are examples in src/app/search/template/whatever_search_item.html | 14:39 |
| Zara | so that should let you cut the sha down to an arbitrary number of characters | 14:40 |
| Zara | (and similar for other things that stretch on into the distance) | 14:40 |
| *** openstackgerrit has quit IRC | 14:48 | |
| SotK | actually, the sha is interpolated in the build process so it may be an interesting experiment combining that with an angularjs filter somehow | 14:49 |
| SotK | or it may be simple, idk | 14:49 |
| *** openstackgerrit has joined #storyboard | 14:49 | |
| Zara | I wonder if there are any problems if we can style things so that things are wrapped globally (eg: will it mess up pasted commands?) | 14:51 |
| Zara | that'd solve the 'chaos when blah title too long' issue, too | 14:52 |
| Zara | but I assume it's akward | 14:52 |
| SotK | it will likely be awkward to not mess up things | 14:53 |
| SotK | best way to do it would be to create a class that does some wrapping and set it on everything we want wrapped | 14:53 |
| Zara | yeah, I felt like that would be 'most everything' so wondered if the inverse made more sense | 14:54 |
| SotK | even then we will probably find some interesting corner cases | 14:54 |
| Zara | I'm all for finding out the results of the experiment, anyway | 14:54 |
| SotK | well, wrapping a table row for example won't work so well (assuming you have the mobile case in mind) | 14:54 |
| *** alexismonville has quit IRC | 14:56 | |
| Zara | s/everything/every-text-thing (I'd assumed it could just identify those without it being explicitly communicated, but maybe not) | 14:58 |
| *** alexismonville has joined #storyboard | 15:00 | |
| Zara | I checked the interpolation+truncate thing; should work if it's told to treat the value as a string | 15:12 |
| openstackgerrit | Matthew Bodkin proposed openstack-infra/storyboard-webclient: Make sidebar submenu the same length as sidebar https://review.openstack.org/355554 | 15:47 |
| *** matthewbodkin has quit IRC | 16:00 | |
| Zara | sent docs for our commit messages | 16:39 |
| Zara | btw that reminds me, the other thing that makes life easier when triaging is when storyboard task title = commit message title | 16:39 |
| Zara | since things can be searched | 16:40 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Hide timeline events and comments of private stories https://review.openstack.org/359895 | 16:40 |
| Zara | yay | 16:41 |
| * Zara idly wishes 'updated' field was useful, for the millionth time | 16:43 | |
| SotK | updated_at you mean? | 16:45 |
| Zara | yeah (I was just looking at the ui at the time, but that's the data) | 16:45 |
| Zara | since it just grabs the direct attributes of the thing, so when you update a task in a story... no change | 16:46 |
| Zara | so I use recent events for that, but that will only list things I'm subscribed to (and similarly, the hacky workaround of making a miniscule change to a story description whenever updating a task will only help you find stories you've changed recently) | 16:47 |
| *** alexismonville has quit IRC | 17:04 | |
| Zara | I've been able to test the 'hide things' patch as logged in story-user and logged out, but not as 'logged in but not listed as user of that story' | 18:01 |
| Zara | ideally we can check that quickly tomorrow morning | 18:02 |
| Zara | it looks good so far and seems like an improvement so have +2'd but tests need to run anyway | 18:02 |
| Zara | and gerrit is unhappy atm anyway so I think that'll take a while | 18:02 |
| Zara | so, 'night! | 18:02 |
| *** aswadr_ has quit IRC | 18:21 | |
| -openstackstatus- NOTICE: The Gerrit service on review.openstack.org is restarting to implement some performance tuning adjustments, and should return to working order momentarily. | 19:47 | |
| *** alexismonville has joined #storyboard | 21:46 | |
| *** alexismonville has quit IRC | 22:52 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!