[06:31] Good Morning Aall :) [06:54] for those ubuntu developers out there am in college and ave decided to give it a go [06:55] how do i go about it ?? [06:57] Hargard, about what [06:57] developing for ubuntu? [07:02] Nimble: Hi, Thank you for interest [07:02] Nimble: Are you interested in developing? [07:03] mihir_, yes, a bit [07:03] originally I came here because the instructions on using the messaging menu in the docs didn't seem to work [07:03] Nimble: Okay, you can start doing resolving bugs.. [07:04] well, why not [07:04] I take it I should probably sign up for launchpad [07:04] you can look visit this link http://developer.ubuntu.com/2013/07/join-the-ubuntu-touch-core-apps-development-team/ [07:04] Yup, you should do that first [07:04] alright, will do. [07:07] Nimble: :) [07:10] good morning [07:19] oSoMoN: ping [07:21] mihir_, will this work with my ubuntu single sign on? [07:21] and if I have one, will it not allow me to use my email address registered to that? [07:24] Nimble: yes , you should use one email across the ubuntu :) [07:24] alright, my single sign on worked [07:24] so that's good [07:27] Nimble: That's great it worked :) [07:35] zsombi: pojng [07:35] pong [07:36] oSoMoN: jenkins did the job now, you have the deb packages there :) [07:36] zsombi: awesome, I’ll test right away [07:37] dpm: Good Morning :) [07:44] morning mihir_ and all [07:46] dpm: Hi , I have proposed two MRs could you please review if you have time [08:06] zsombi: do you know if the uitk’s autopilot tests can be run on devices? [08:07] bzoltan: ^ [08:07] oSoMoN: It should be OK [08:09] bzoltan, zsombi: I’m getting the following error when trying to run them on maguro: http://pastebin.ubuntu.com/5954209/ [08:10] oSoMoN: bzoltan: no idea what can this be... [08:10] also getting an error on stdout about gcc not found… [08:11] installed gcc and this error went away, but not the runtime error === mpt_ is now known as mpt [09:20] gusch: hi, i think i asked you already, but forgot the answer: can we find out if the currently active camera has light (torch/flash) support ? [09:20] gusch: also, how to find out the number of cameras on the current device [09:21] nerochiaro: for the number of camera, you need to use the same class, that you use to switch cameras [09:21] nerochiaro: http://qt-project.org/doc/qt-5.0/qtmultimedia/qvideodeviceselectorcontrol.html#deviceCount [09:22] dpm: Ping !! [09:24] nerochiaro: need to check the best way to get the flash information [09:25] gusch: ok. do you only have a galaxy nexus, right ? [09:25] nerochiaro: and the N10 [09:25] gusch: ok, but no n7, which apparently has only one camera [09:27] oSoMoN: timp: to store configuration info, are we currently using dconf or something else ? [09:27] nerochiaro: nope [09:28] nerochiaro: dunno [09:36] nerochiaro: we have an AP in the SDK to provide API for gsettings/dconf, so that will be the one you should use... when it will be ready [09:37] nerochiaro: for the focus you can use this one http://qt-project.org/doc/qt-5.0/qtmultimedia/qml-qtmultimedia5-camerafocus.html#isFocusModeSupported-method [09:37] zsombi: so what should i do now ? [09:37] zsombi: i mean, in the meantime before it's ready ? [09:37] nerochiaro: yes, I got it :) [09:38] nerochiaro: well, do something :) [09:38] gusch: don't need it for focus ATM, only num of cameras and flash modes [09:38] gusch: but thanks [09:38] zsombi: ok, use whatever for now then when your stuff is ready use that. fair enough [09:38] nerochiaro: for flash you'll need to use C++, and I'll have to update the plugin to do this: "Some camera devices may not have flash hardware, or may not be configurable. In that case, there will be no QCameraFlashControl available." [09:39] gusch: that will work fine for me [09:40] nerochiaro: :) yep... We were trying to find out whether we could put under QSettings and provide QML binding to QSettings, but it ended up that Qt guys are planning something new and don't really want to see any backend added to QSettings anymore :/ [09:43] nerochiaro: does the notes patch work? were you able to test it? [09:46] zsombi: what patch ? missing something here [09:46] (12:32:26 PM) zsombi: nerochiaro: set propagateComposedEvents: true to the MA which activates the note card, also mouse.accepted = false in onClicked [09:47] zsombi: ah, let me try that [09:48] zsombi: in addition also propagateComposedEvents: true on the IMA, right ? [09:49] nerochiaro: yes, this is needed as the "inactive" card MA may be higher in the paint order, so that also needs to propagate the composed events [09:49] zsombi: ok, giving it a shot [09:52] zsombi: ok, seems to work. i'll submit an MR [09:52] nerochiaro: awesome!!! :) [09:52] zsombi: :9 [09:53] zsombi: btw, is it normal that on the latest image when i drag the launcher all the way across to reveal the apps lens, the dash is completely transparent and shows the running app underneath ? [09:53] zsombi: it's new in today's image [09:54] nerochiaro: no idea, don't have my device woken up yet... [09:56] zsombi: zzzzz [09:57] :) [09:57] nerochiaro: I've set it on charger, but it is still dead [09:59] zsombi: shut it down before going to bed ;) [09:59] nerochiaro: I thought that it was, as when I pressed the power button it aint woke up... :) [10:02] zsombi: happens to me all the time [10:02] zsombi: i think we need a more reliable shutdown mechanism [10:03] nerochiaro: and a safety power-off, which turns off the device when the battery dries down to 3%, so a possible wake-up c an be done afterwards safely [10:06] zsombi: yeah [10:09] I want to create .dmg file from my source file on ubuntu so that I can distribute it on mac? | http://askubuntu.com/q/329249 [10:13] * zsombi lunch [10:43] bzoltan: re: http://pad.lv/1202207 dpm e-mailed about... I remember some past hud discussion with Wellark, what was the end results? I thought it would be compiled in SDK PPA for older Ubuntus or something like that. [10:43] Launchpad bug 1202207 in Dropping Letters "12.04 install problems" [Undecided,New] [10:46] zsombi: https://code.launchpad.net/~amanzi-team/notes-app/notes-app-event-propagation/+merge/178714 [10:46] nerochiaro: checking... [10:46] zsombi: haven't run the tests yet, i'll try that now on the device [10:47] nerochiaro: ok === MacSlow is now known as MacSlow|lunch === gusch is now known as gusch|lunch [11:05] mhall119: ping [11:06] seb128: ping === nik90|Office is now known as nik90|Lunch [11:15] nik90|Lunch, context less ping = no pong usually (just saying) [11:16] seb128: oh [11:16] :) [11:17] seb128: so I saw that in the system settings app, you guys show the timezone info "Europe/Amsterdam" etc..Is that just a mockup string? Or do you guys actually read the system timezone info? [11:17] nik90|Lunch, having "ping" without context is a trap, sometime you don't have time for a long discussion so you just default to ignore the ping to not be trapped [11:17] nik90|Lunch, we read the tz [11:17] seb128: How do you do that? Did you create a custom C++ plugin to read the tzdata file? [11:20] seb128: I need that same info for the clock app. However since the clock app is all javascript and qml, I am unable to get the timezone ID info. So before I start writing my own custom C++ plugin to do that, I wanted to know if other apps like the system app did that already to prevent code duplication unnecessarily. [11:20] nik90|Lunch, yes, http://bazaar.launchpad.net/~system-settings-touch/ubuntu-system-settings/trunk/view/head:/plugins/time-date/timedate.cpp [11:20] dpm: ping !! [11:21] nik90|Lunch, that seems like something useful enough, that we should have a sdk api for it imho [11:22] seb128: I agree. However I am not really sure when that would be implemented since all sdk devs are quite busy. And I need this info asap :( [11:22] nik90|Lunch, right, you probably better copy what we are doing meanwhile [11:23] seb128: yeah. thnx for getting the info for me [11:23] yw! [11:23] mhall119: Ping !! [11:24] nik90|Lunch, hum, maybe https://github.com/nemomobile/nemo-qml-plugin-time/tree/master/src would be an option as well [11:27] nik90|Lunch, their current backend is for meego though [11:32] seb128: be sure to file a bug with the sdk that points at your branch, it might speed up development once it happens ;-) [11:33] kalikiana, what project is the right one for those sort of apis? [11:33] zsombi: tests are also successful on device (on top of jenkins being happy about them). maybe i should add another ap test for this specific case [11:34] nerochiaro: excellent! an extra test is always welcome [11:34] nerochiaro alos, the code looks right, so I'll approve from my side, but let's wait till timp approves it [11:35] zsombi: ok [11:39] hey gusch|lunch, when you get back from lunch, could you please review https://code.launchpad.net/~osomon/webbrowser-app/simplify-historydomainlistmodel/+merge/178687 ? I already ran the autopilot tests on maguro, they all pass === gusch|lunch is now known as gusch [11:41] oSoMoN: what a timing ;) - I review ... [11:41] thanks :) [11:50] seb128: just file it in the sdk, it's easy to move, and it'd depend on what other api it might relate to. just timezone data won't be its own component [11:51] it might also coincide with zsombi's alarm stuff [11:51] s/coincide/fit/ [11:57] oSoMoN: approved [11:58] gusch: awesome, thanks, I’ve got a couple more MRs in the pipe, waiting to run autopilot tests on device before requesting a review [11:59] oSoMoN: keep them going ;) [12:00] kalikiana: seb128: hehe, the github branch you've pointed actually connects to timed, the one we were suggesting to port to Ubuntu, as that handles all the time related stuff (including proper NITZ support with all its exceptions) + alarms... [12:00] but seems we will do a different adaptation and have a different approach for that... === nik90|Lunch is now known as nik90|Office [12:02] zsombi: Should I go about writing my own C++ plugin to get system timezone ID or will this come as part of the alarm package? [12:02] zsombi: I need to make a decision depending on when a suitable API might land in the sdk for it === dholbach_ is now known as dholbach [12:04] nik90|Office this may not come as part of the alarm API as this is more related to time management, but we may create a separate API for Time management. I say may, as currently we don't have any API planned for that. But you could always contribute to the SDK by offering to do such an API ;) [12:04] gusch: next one: https://code.launchpad.net/~osomon/webbrowser-app/blank-thumbnails/+merge/178716 [12:04] oSoMoN: ok [12:06] zsombi: I would love to help but I am learning qml/c++ as I go. So what I implement may not have sdk standards and would rather be suited for an app instead. [12:06] That said I will give it a shot and see where it leads me [12:06] nik90|Office don't be afraid, we can help on that! :) [12:07] zsombi: I will let you know when I have a working branch [12:08] nik90|Office cool!! thx! === MacSlow|lunch is now known as MacSlow [12:18] om26er: is there any way in autopilot tests to wait for an animation to finish ? [12:18] oSoMoN: maybe you know that too ? ^ [12:19] nerochiaro, most of the times, object.animating is what you need to assert [12:19] om26er: oSoMoN: the reason i ask is because i need the animation to finish to get the new geometry of the object so that i can click in the right place for my test [12:19] om26er: I reviewed your autopilot branch fix for clock app. [12:20] there are also cases where object.moving is used as well, but that's mostly when the page stack is being moved etc. [12:20] om26er: object.animating does not exist [12:20] nerochiaro: you may need to add an animating property to the object, and wait for it to become false [12:21] nerochiaro: look for example at the SDK’s Panel implementation, it has an 'animating' property [12:21] om26er: I see that we continue to use our own custom toolbar emulator function. Should we convert that to use elopio's emulator function now or in a later commit? [12:21] oSoMoN: adding properties only for the sake of tests does not seem very clean [12:21] nik90|Office, I think doing that in an other branch will be better, also I will fix the problems you highlighted today [12:22] nerochiaro: you asked for a solution, I’m sharing the only solution I know [12:22] om26er: okay [12:22] nerochiaro, which type of object is that ? [12:22] oSoMoN: sure, i wasn't complaining for the sake of complaining, i was trying to get a discussion running. it's an ok workaround if nothing else is available. [12:23] gusch: next MR: https://code.launchpad.net/~osomon/webbrowser-app/cosmetics/+merge/178725 [12:23] In cases I have used .animating or .moving and also 'swipeState' depending on the type of object I am interacting with [12:23] nerochiaro, ^ [12:23] om26er: it's one of the notes in notes-app. i need to wait for the expanding animation to finish before it settles on a stable geometry [12:23] om26er: so i can continue with the rest of the test [12:24] om26er: oh, but wait, i think i have a property that represents the final height, and i animate towards that. so i can just check if height eventually equals final_height and that should work [12:24] nerochiaro: you could probably write a custom matcher that waits for the height to be stable, i.e. not changing during a certain interval of time, but that sounds way more hackish than the other solution [12:24] nerochiaro: yep, height == final_height looks like the cleanest solution so far [12:24] oSoMoN: ok let's try that [12:26] nerochiaro, you could tests the state of the NoteItem, it has collapsed and expanded [12:27] *test [12:27] om26er: was my first idea, but it doesn't work because the state changes immediately and then the animation is triggered, as far as I can see [12:28] nerochiaro, there is also 'isExpanded' [12:28] om26er: same problem with that [12:28] oSoMoN: om26er: the height test works, i'll use that. thanks for the input [12:29] cool [12:34] oSoMoN: cosmetics approved [12:40] oSoMoN: thumbnails approved as well === dholbach_ is now known as dholbach [12:47] gusch: thanks! [12:48] zsombi: added more autopilot tests to https://code.launchpad.net/~amanzi-team/notes-app/notes-app-event-propagation/+merge/178714 [12:53] nerochiaro, gusch if I turn on autopilot tests for camera-app for each merge proposal do you mind ? [12:53] I'll keep fixing if we face any blockers along the way === _salem is now known as salem_ [12:54] *On Maguro and Mako I meant [12:56] om26er: well, what's the current status ? will it fail if you do turn it on ? [12:56] nerochiaro, currently they are passing as seen on daily image testing [12:56] om26er: is it possible to enable the camera-app autopilot tests for the qtubuntu-camera as well? [12:57] gusch, it should be, I can look into that. never tried qtubuntu-camera suite [12:58] om26er: would be cool if you could try to do that at one point (no hurry) [12:59] gusch, sure [13:03] zsombi, great, is there a bug/blueprint/somewhere I can see what's going on about this timed? [13:12] om26er: then go ahead with that ,i have no problem [13:12] nerochiaro, ok, thanks [13:48] boiko: could help me regarding this bug ?https://bugs.launchpad.net/ubuntu-calculator-app/+bug/1205020 [13:48] Ubuntu bug 1205020 in Ubuntu UI Toolkit "functionality for copying result in clipboard" [Undecided,Confirmed] [13:48] Still design team needs to approve it [13:49] mihir_: I think for this one what you need to do right now is to wait for something to be provided by the UI Toolkit team [13:50] boiko: Okay thank you :) === _salem is now known as salem_ [14:36] zsombi: can we top approve https://code.launchpad.net/~amanzi-team/notes-app/notes-app-event-propagation/+merge/178714 as jenkins likes my new tests and they pass on the device too ? [14:56] mhall119, ping [15:00] iBelieve: pong [15:00] mhall119, I've got a question about merge requests for File Manager [15:02] mhall119, I was reading the backlog for the last meeting (I wasn't their), and Arto said he doesn't have much time for development except for responding to emails. I've had a merge request that's been waiting for a long time. Should I have somebody else, like you or popey, review it? Or should I just approve it myself? [15:05] iBelieve: I'll take a look, can you link me to the MP? [15:05] mhall119, I'm going to make a few changes to the old merge request (https://code.launchpad.net/~mdspencer/ubuntu-filemanager-app/advanced-options), but you could take a look at the new one (https://code.launchpad.net/~mdspencer/ubuntu-filemanager-app/sidebar/+merge/178774) [15:06] iBelieve: I'm in a meeting atm, but I'll look at them after [15:07] mhall119, thanks. Also, I just checked on the first merge request that I linked (advanced-options), and that's ready to be reviewed, I'm not going to make any changes. [15:09] nerochiaro: ping [15:09] gusch: [15:10] gusch: hi [15:10] nerochiaro: about detecting flash support [15:10] nerochiaro: you should check it a bit different [15:11] nerochiaro: it's not possible to dynamically add/remove the C++ flashControl object (so that QML detects it) [15:12] nerochiaro: so get in C++ the QCameraFlashControl [15:12] nerochiaro: and there use bool QCameraFlashControl::isFlashModeSupported(QCameraExposure::FlashModes mode) [15:12] nerochiaro: as mode use QCameraExposure::FlashOn - if that is supported, than a flash exists ... [15:13] nerochiaro: is that ok? [15:14] gusch: ok, i guess that doesn't tell me if a torch mode exists, i have to check that separately. other than that it works ok for me [15:14] nerochiaro: yep - that should be checked separately [15:17] nerochiaro: you need to have a running camera to get the correct values btw. [15:20] gusch: running as in initialized properly [15:20] nerochiaro: yep [15:26] seb128: we don't have blueprint for timed as we didn't do anything from SDK side. All we have is a BP for Alarm API, but that's not really related to the time functionality you have pasted from github [15:27] zsombi, I looked at timed, it doesn't seem to do stuff like system tz handling (or I looked at the wrong place for it) [15:27] nerochiaro: I'm affraid to top-approve that as may introduce regressions without my MR approved. [15:28] seb128 it does, it handles alarms and also has NITZ support. [15:28] timp, ping [15:29] nerochiaro: can you review this? https://code.launchpad.net/~schwann/qtubuntu-camera/qtcamera-vp-flash/+merge/178784 [15:29] zsombi, NITZ is not system timezone though, it wouldn't work on a tablet or desktop [15:32] seb128 it does if the tablet has GSM module in it. NITZ however helps you keep the time zone in sync when traveling or when the daylight saving starts/ends. So that one we also need to check, perhaps not in our time backend in SDK, but somewhere in the system. So far I know there is some work done around NITZ somewhere else in the stack... [15:32] seb128: so whatever is needed right now for the clock, we don't have any BP/API planned. But we definitely need one [15:32] zsombi, right, I'm not saying NITZ it's not useful, it's only a piece of the puzzle though [15:32] iBelieve: pong [15:32] seb128: +1 [15:33] timp, could you explain the reasoning for marking LP #1202469 as Invalid? [15:33] Launchpad bug 1202469 in Ubuntu UI Toolkit "Actions should be hidable " [Undecided,Invalid] https://launchpad.net/bugs/1202469 [15:34] timp, I'm working on advanced options for File Manager and need a way to make actions hidden. [15:36] iBelieve: hold on, let me see [15:36] seb128: timed is mostly there to handle and schedule alarms, keep those in sync with the system time changes + NITZ changes, and also handles when to play alarm sound (there are quite many exceptions, like when in a call, when multimedia plays, etc) [15:37] zsombi, ok, it seems a bit orthogonal to the set manually time/date/timezone, use ntp features (which are more desktopish) [15:44] zsombi: ping [15:44] nerochiaro: pong [15:44] zsombi: can we top approve https://code.launchpad.net/~amanzi-team/notes-app/notes-app-event-propagation/+merge/178714 as jenkins likes my new tests and they pass on the device too ? [15:45] nerochiaro: so it doesn't bother it with the MA event propagation? [15:46] zsombi: what do you mean ? [15:47] iBelieve: where do you use the Action? In the toolbar? [15:47] nerochiaro: these changes you did were because the IMA changes. Do these changes work without any problems with the SDK from archive? [15:48] timp, I was planning on having the Actions in an ActionSelectionPopover. For example, I was going to have an "Open in Terminal" action that is only shown when the user has set a setting that shows advanced features. [15:48] zsombi: ah, i don't know. i tested them with the sdk from your branch [15:48] zsombi: but it seems jenkins thinks they are ok with the sdk from archive as well [15:49] nerochiaro: ok, so let's approve this, then timp can approve IMA MR afterwards [15:50] iBelieve: for the toolbar we replaced the Actions by a visual representation of actions (ToolbarButton) that has the visible property now [15:51] iBelieve: I don't think we have a visual representation for the actions in the ActionSelectionPopover yet. [15:51] zsombi: you wrote the ActionSelectionPopover? Is that still a recommended component to use? [15:52] iBelieve: I'm reading the docs for ActionSelectionPopover now, I'm not familiar with it [15:52] timp: why not? it gets a list of actions and displays them. However can be modified to work better with the new Action stuff [15:53] zsombi: ok cool [15:53] zsombi: thanks [15:53] timp, why are you saying that the visible property should be part of the visual representation of an Action (such as a ToolBarButton)? To me, it seems like it should be part of the Action itself. [15:56] iBelieve: we don't have our "own" Actions in the SDK any more, but we use them from Unity.Actions. Those actions are integrated with HUD and will be with the launcher [15:57] iBelieve: we had some discussions about the "visible" property, and I used to share your opinion. [15:58] iBelieve: probably we'll need a visual representation for actions in an actionselectionpopover [15:59] timp, so why isn't there a visible property anymore? I'm just curious :) [16:00] iBelieve: well, for the HUD, the idea is that it can show the user all the possible available actions [16:01] iBelieve: and for other uses (toolbar), we thought that the ActionItem/ToolbarButton would take care of it [16:02] iBelieve: your case still needs some thoughts [16:02] Wellark: ^ do you have any suggestions for the visibility of actions in an ActionSelectionPopover? [16:02] timp, that makes sense, somewhat, though it seems like Actions could still have a visible property and be shown no matter what in the HUD. [16:03] timp, anyway, what should I do for an ActionSelectionPopover? Is there a way to do what I want now, or do I need to wait for support in it? [16:05] iBelieve: you can specify the delegate of the ActionSelectionPopover, of that you can set the visible property. [16:05] iBelieve: if you use action.enabled, you can say visible: action.enabled in the delegate [16:06] zsombi: in the docs for ActionSelectionPopover, it says: text: action.text [16:06] zsombi: shouldn't that be text: item.text ? [16:06] timp, that might work, I'll try it. Thanks for your explanation of the reasoning behind the bug and for help with a solution [16:07] iBelieve: thanks for making me doubt our previous decision to remove visible ;) [16:07] timp, :) [16:08] timp: no, because that delegate is the component which is used to show the action, and the "action" itself is the property from the Standard [16:09] zsombi: ah yes, ListItem.Standard has an action property. Then setting the text here does nothing (it is action.text by default) [16:09] timp: huhh?? [16:10] timp, iBelieve: so, the rationale here is that Action is separated from it's visualization [16:10] zsombi: what huh? [16:10] timp: I was looking after this action property :) [16:10] timp: but that has been moved to AbstractButton [16:11] zsombi: no [16:11] zsombi: to ActionItem, the parent of AbstractButton :) [16:11] timp: actually AbstractButton is derived from ActionItem :) [16:11] timp: right [16:11] zsombi: many of our components are now derived from ActionItem [16:11] timp: and that's right! [16:12] Action is an abstract entity describing the functionality the application can do and when it's visualized you always assosiate the action with some form of delegate [16:15] this allows a single action to be shared or contributed between different UI and shell elements and each delegate has a separate visibility property [16:15] or optional visibility propertyy if it makes sense [16:19] Wellark, timp, so is there an easy way to control the visibility of an Action in an ActionSelectionPopover without a custom delegate? timp suggested visible: action.enabled, but I think that would require a custom delegate? [16:20] iBelieve: I might be using the term delegate ambiguously [16:20] I haven't looked in to ActionSelectionPopover yet [16:20] iBelieve: does it work if you add "delegate.visible: action.enabled" to set the visible property of the standard delegate? [16:20] iBelieve: that ^ sounds reasonable [16:21] Wellark: ActionSelectionPopover has a property Component delegate which is used in the Repeater that visualizes the Actions [16:21] if your action is not enabled then it's invisible [16:21] Wellark, timp: QML won't let me do sub-properties like that :( [16:23] Wellark, timp: maybe it would, but it doesn't show that property since it is a Component, not an Empty === greyback_ is now known as greyback|food [16:28] iBelieve: ah yes, I think you're right [16:28] timp, I tried using another property like property Empty delegate2: delegate, and then used delegate2.visible, but then QML said Invalid property nesting [16:30] iBelieve: actually the delegate is very simple: http://pastebin.ubuntu.com/5955564/ [16:31] timp, I suppose I'll just copy it if there is no better way, but I never like to duplicate code, especially if the original is in the SDK and subject to change. [16:32] iBelieve: I agree, it is better not to copy code. But like this you can continue working on it now [16:33] timp, right, that's what I'll do. Thanks for all the help! === greyback|food is now known as greyback [20:32] Hi [20:32] I cannot enter payment information into profile while trying to submit app [21:45] balloons, ping [21:46] iBelieve, pong.. I wanted to chat with you :-) [21:46] balloons, about LP #1208999? [21:46] Launchpad bug 1208999 in Ubuntu File Manager App "Cannot paste a file on desktop or device" [Undecided,New] https://launchpad.net/bugs/1208999 [21:46] iBelieve, indeed. that and https://bugs.launchpad.net/autopilot-qt/+bug/1209004 are the only failing tests on the device [21:46] Ubuntu bug 1209004 in Autopilot Qt Support "Process Manager tries to call Xlib on phablet devices" [Undecided,New] [21:47] balloons, so does copying and pasting work on the desktop? [21:47] iBelieve, no.. I can't get it to work at all [21:47] no matter what device I use :-) [21:48] balloons, does the Actions popover popup in the center of the app, or does it popup from the Actions button? [21:48] center of app on desktop, causing issues [21:48] from the popover on the device I think.. [21:48] it's odd.. it's not the same failure depending on how you do it [21:49] balloons, the reason it centers is because of how the popover gets opened in code. I fixed that, but I'm getting errors about the copy and cut functions not existing. [21:49] balloons, can you try lp:~mdspencer/ubuntu-filemanager-app/sidebar and see if that works? [21:50] balloons, though wait a sec until I push more changes [21:50] iBelieve, no rush, as I'm working on the clock app now and probably won't be back to file manager until tomorrow [21:51] I just wanted to make you aware and see if you knew anything about it. Sounds like you do which is good :-) [21:53] balloons, I was already working on fixing the action selection popovers among other things, so thanks for noticing the copying problem. I uploaded the branch if you wanted to take a look at it, but it still isn't working for me. [22:08] balloons, I fixed the copying/cutting of files in the UI, but now those tests and others are broken, so I need to fix those. [22:11] iBelieve, heh, always something eh? Good though, I'm glad to hear [22:11] thoughts on fixing the open_file test to not depend on processmanager? [22:11] iBelieve, ^^ [22:13] balloons, I haven't seen the new tests yet, I'll take a look though [22:14] balloons, that's cool how that test works. Is there a reason why it should not depend on ProcessManager? [22:15] iBelieve, https://bugs.launchpad.net/autopilot-qt/+bug/1209004 [22:15] Ubuntu bug 1209004 in Autopilot Qt Support "Process Manager tries to call Xlib on phablet devices" [Undecided,New] [22:15] it blows up on the device, but it's not your fault.. however, I think it can be done without processmanager.. [22:17] balloons, ah, that was the other bug you mentioned before. I wasn't the one who wrote that test, elopio did, so I'm not familiar with how it works, but I'll try to think of some way to do it. [22:18] balloons, I've got a question on how I should write the autopilot tests for my new changes, if you have time. [22:19] sure, go for it.. and yea, I know you didn't write it. I would leave it for now till we get feedback on whether or not we should be using processmanager for it I'd suspect [22:22] balloons, I've added a sidebar (like in Nautilus) that is visible when the app is wide enough (here is the screenshot: http://i.imgur.com/GZOVaLf.jpg). This basically replaces the Places menu in the toolbar. I've also changed the default size to 100 gu, so it looks good on the desktop, assuming the phone will correctly size it to 50 gu. [22:22] * balloons wonders what it will look like on my tablet [22:23] balloons, now my question is, how should I write the tests for the new layout? There needs to be tests for both the tablet/desktop mode and phone mode. Should these be in one test, or in separate tests? [22:24] balloons, and is it possible to change QML properties from autopilot so I can put the app into phone mode to test it? [22:24] iBelieve, ahh yes.. Well you could put it into the same test and detect the proper size then act accordinly [22:24] and yea, forcing a particular size is interesting too [22:25] balloons, so should only do one test depending on what device it is running on, or should I force both sizes to test? [22:25] I'm concerned with how a device will respond if you force the size. On the desktop, sure should be fine. On the phone, well it can't fit a tablet size :-) [22:26] balloons, good point, I hadn't thought about that. So I guess I'll just write the tests to test for the particular configuration the app is run in? [22:27] yes, I would do that. As to wanting to hack on the different size and test them, I suppose the desktop test could run both.. [22:29] balloons, for now, I'll just run the tests based on the initial size. Thanks for the help! [23:04] balloons, ping [23:05] iBelieve, pong [23:07] balloons, so one of the reasons the copy autopilot test fails is that there is no wait for the destiniation directory to be opened. I'd like to be able to incorporate that into the FolderListDelegate emulator, but it seems that assertThat isn't available outside of the test case. Any suggestions? [23:08] balloons, here is a snippet of what I'm trying to do: http://paste.ubuntu.com/5956870/. First I tried calling assertthat on self, then on AutopilotTestCase, but AutopilotTestCase requires an instance. [23:32] balloons, I need to go, I'll try and see if I can find some way to do it, otherwise I'll ask again later. [23:33] iBelieve, yes, in short, don't use asserts outside of testcases. That said, I'm not sure I get the hangup here [23:33] balloons, what do you mean "I'm not sure I get the hangup here" [23:38] I mean I'm knee deep in ubuntu clock, and losing my mind as it's getting late here :-) [23:38] in other words, I can't be of much more help atm ;-p