| === thomi_ is now known as thomi | ||
| === Ursinha-afk is now known as Ursinha | ||
| === _salem is now known as salem_ | ||
| === salem_ is now known as _salem | ||
| === chihchun_afk is now known as chihchun | ||
| === chihchun is now known as chihchun_afk | ||
| === jibel_ is now known as jibel | ||
| === _salem is now known as salem_ | ||
| === roadmr is now known as roadmr_afk | ||
| === roadmr_afk is now known as roadmr | ||
| === Ursinha is now known as Ursinha-afk | ||
| brendand | elopio, hey! | 15:37 |
|---|---|---|
| elopio | brendand: hello. | 15:37 |
| brendand | elopio, you know those uitk failures caused by autopilot? | 15:39 |
| elopio | brendand: the ones for the date picker? | 15:40 |
| brendand | elopio, yeah | 15:40 |
| === Ursinha-afk is now known as Ursinha | ||
| balloons | elopio, brendand I had the date picker failures appear yesterday in jenkins.. magically | 15:54 |
| balloons | while I was working on this: https://code.launchpad.net/~nskaggs/ubuntu-ui-toolkit/fix-aphelper-emulator-warning/+merge/227093 | 15:54 |
| balloons | so something changed.. | 15:55 |
| balloons | if the UITK landed, I would blame that first | 15:55 |
| brendand | balloons, no - the date changed :) | 15:56 |
| brendand | balloons, because the test is adding 25 years to this year | 15:56 |
| balloons | brendand, yes, but depending on how the pickers are implemented they seem to get a large year on them, and when AP reads the list it blows | 15:57 |
| brendand | balloons, although, i tried changing the date used in the tests and it didn't seem to fix it, so there might be something else going on - or it could be i didn't understand the test properly | 15:57 |
| balloons | but I've run the UITK tests before without issue | 15:58 |
| balloons | brendand, yes, it depends on the pickers | 15:58 |
| brendand | balloons, someone needs to look closer at the tests and see if anything strange is happening | 15:59 |
| === Ursinha is now known as Ursinha-afk | ||
| balloons | pitti, perchance you are still about? | 16:35 |
| === Ursinha-afk is now known as Ursinha | ||
| balloons | so elopio are you looking at https://bugs.launchpad.net/ubuntu-calendar-app/+bug/1328600? | 16:55 |
| ubot5 | Ubuntu bug 1328600 in Autopilot "Autopilot lacks support for large timestamps" [High,Confirmed] | 16:55 |
| elopio | balloons: not yet. Looking now at the qmlscene crash | 16:56 |
| === Ursinha is now known as Ursinha-afk | ||
| balloons | elopio, just wanted to make sure I wasn't stepping on any toes. Is anyone fixing autopilot to support the large timestamps? | 16:58 |
| balloons | if not, I'll just propose my local change as an mp | 16:58 |
| elopio | balloons: not yet. Yesterday thomi and veebers were looking at the issues with QML components versions. | 16:58 |
| === rvr is now known as rvrOff | ||
| === Ursinha-afk is now known as Ursinha | ||
| === roadmr is now known as roadmr_afk | ||
| balloons | elopio, well no matter the date you set, AP sees the 2064 date in the uitk tests | 17:48 |
| elopio | balloons: that's the max date on the picker, right? | 17:49 |
| balloons | elopio, that's a question for t1mp I suppose.. It seems we should dig deeper into the UITK. I messed with AP to no avail. The fix in AP also causes some weird off by 1 hour differences. Looking at the qml for the test, modifying the year doesn't fix anything | 17:50 |
| elopio | balloons: yeah, last time we found that the exception doesn't come from selecting a specific date | 17:51 |
| elopio | it comes from autopilot trying to load all the properties on the picker. | 17:51 |
| balloons | elopio, yes I remember.. But I was trying to find a qml solution.. someway to not end up with those high dates in the picker | 17:52 |
| elopio | balloons: I don't get the hour difference with the statements that veebers pasted there. | 18:03 |
| balloons | elopio, I don't get it persay either.. I tried it locally and didn't have the issue. However, after I made the change to AP and ran the tests, the differences showed up | 18:03 |
| elopio | it could be a tz issue, as I get 11 instead of 17 or 18. | 18:04 |
| elfy | evening balloons elopio | 18:04 |
| elopio | hello elfy | 18:04 |
| balloons | elopio, I get 12 :-) so yes, it's tz based | 18:05 |
| balloons | elfy, good eve to you sir | 18:05 |
| elfy | elopio: so let's hope that letozaf has seen the hangout and can now tell the difference between us both :) | 18:05 |
| balloons | you too look slightly different.. I mean it's definitely close :-p | 18:06 |
| elopio | elfy: I am you! | 18:06 |
| balloons | oO | 18:06 |
| elfy | ha ha ha | 18:06 |
| * elfy is the young looking one :p | 18:07 | |
| balloons | elopio, so looking at trunk you see they've not touched the datepicker stuff; http://bazaar.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/trunk/files/head:/modules/Ubuntu/Components/Pickers/ | 18:08 |
| elopio | balloons: I suspect this is a lower level change | 18:11 |
| elopio | like the listview loading more values than before | 18:11 |
| balloons | ok, I'll dig deeper ;-) | 18:11 |
| balloons | elopio, that said, I see an interesting idea for a fix... | 18:12 |
| elopio | balloons: I'm not sure what's the value we are looking for. UTC, or the one on the current tz. | 18:14 |
| elopio | I actually don't know how to interpret the timestamp if 0 means something different depending on where you are. | 18:14 |
| elopio | but I think that using timedelta or time is the way to go. | 18:14 |
| balloons | elopio, so this is I think the easiest way to go forward right away: https://code.launchpad.net/~nskaggs/ubuntu-ui-toolkit/fix-1328600/+merge/227381 | 18:18 |
| elopio | balloons: well, that will unblock the toollkit landings, but it also changes the API. | 18:20 |
| balloons | yea, there's a few other things that would have to change as well.. the '50' limit exists a few other places | 18:21 |
| balloons | elopio, but what I'm really wondering is if we can set this max property in our test qml | 18:22 |
| elopio | >>> datetime.datetime.utcfromtimestamp(2047570047) | 18:22 |
| elopio | datetime.datetime(2034, 11, 19, 17, 27, 27) | 18:22 |
| elopio | >>> datetime.datetime.utcfromtimestamp(0) + datetime.timedelta(seconds=2047570047) | 18:22 |
| elopio | datetime.datetime(2034, 11, 19, 17, 27, 27) | 18:22 |
| balloons | that fixes everything.. I think we can, just need to try | 18:22 |
| elopio | I think that explains to veebers the weird behaviour he was seeing. | 18:22 |
| elopio | what I don't understand is why he has one call utc, and the other on his timezone. | 18:22 |
| elopio | balloons: oh, yes, on the tests we can change the max date. | 18:23 |
| balloons | elopio, yes, syntatically I haven't figured it out tho | 18:24 |
| elopio | balloons: you need it to be a javascript date object. | 18:26 |
| elopio | http://developer.ubuntu.com/api/qml/sdk-14.04/Ubuntu.Components.Pickers.DatePicker | 18:26 |
| elopio | let me find an example. | 18:26 |
| balloons | elopio, I see setting maximum.. it's just not working | 18:27 |
| elopio | balloons: this might also fix it | 18:29 |
| elopio | maximum: Date.prototype.getInvalidDate.call() | 18:29 |
| elopio | that means infinite, so I guess that the property won't get a big integer. | 18:30 |
| elopio | maybe -1, or empty. | 18:30 |
| balloons | mmm.. so I'm still getting that large 50 year date | 18:33 |
| balloons | I'll try your idea | 18:35 |
| balloons | mm.. that makes the date max be 2106 :-) | 18:38 |
| balloons | elopio, so setting the max value I can see the max value, but there are several objects, and the 2064 year object still exists | 18:40 |
| elopio | hum, I think that we just need to solve it in autopilot | 18:41 |
| elopio | it's clear that it's autopilot's fault. | 18:41 |
| balloons | well, I guess that does swing the pendulum back into AP's territory | 18:42 |
| balloons | elopio, so I wonder if we can look closer at what's really on the dbus tree, vs what AP is creating | 18:43 |
| elopio | balloons: the vis shows the values before convertion, I think. | 18:50 |
| elopio | or you can put a breakpoint just before autopilot makes the convertion. I guess you could also inspect dbus, but I'm not sure what to look for there. | 18:50 |
| balloons | ohh, silly me. I'm doing this all on the phone. | 18:51 |
| balloons | I can use the desktop, even if it doesn't blow up, to look at the objects | 18:51 |
| * balloons facepalms | 18:51 | |
| balloons | elopio, sadly I believe vis shows translated values.. but maybe I'm wrong | 18:52 |
| elopio | balloons: and it makes it harder to reproduce that our machines are 64bits | 18:52 |
| elopio | jenkins is 32, so that's why it is failing. | 18:52 |
| balloons | elopio, i figured out the limit for my box this morning, heh | 18:53 |
| balloons | elopio, try 91234567891123456789 | 18:53 |
| balloons | it'll blow up | 18:53 |
| elopio | here I get an exception when trying the year 10000, but it's not the same exception. | 18:53 |
| === roadmr_afk is now known as roadmr | ||
| balloons | elopio, success! I isolated the value from the dbus session | 19:54 |
| balloons | so one object has the maximum date I set, while the other still has the default maximum | 19:57 |
| elopio | balloons: so there are two pickers? I'm not sure what you meant | 19:58 |
| balloons | elopio, I'm playing around looking at the dbus session. It's not an autopilot thing. It's definitely in the toolkit | 19:59 |
| balloons | I'm looking at the raw values AP is reading, and it's in there | 19:59 |
| balloons | elopio, here's one of the objects: http://paste.ubuntu.com/7816047/ | 19:59 |
| balloons | so let me try the fix now :-) | 20:00 |
| balloons | woot | 20:00 |
| === Ursinha is now known as Ursinha-afk | ||
| balloons | elopio, TLDR, setting the max works. But you also have to set the max for the timepicker | 20:02 |
| balloons | because it too has m/d/yyyy on it | 20:03 |
| === Ursinha-afk is now known as Ursinha | ||
| Aki-Thinkpad | what is this channel for exactly? I see unicorns... but that does not exactly tell me much... | 21:53 |
| balloons | Aki-Thinkpad, this is for the quality team | 21:53 |
| balloons | testing, bugs, fun stuff | 21:53 |
| Aki-Thinkpad | balloons, ahhh; okay. | 21:54 |
| === salem_ is now known as _salem | ||
| === _salem is now known as salem_ | ||
| === salem_ is now known as _salem | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!