=== chriadam|away is now known as chriadam
[07:00] <dholbach> good morning
[07:01] <kivi> hey all
=== chihchun_afk is now known as chihchun
[07:20] <didrocks> zsombi: hey, so, here we go: https://github.com/didrocks/splitthebill. I still have the issue when changing section (which just toggles the visibility of a page on and off). I put some questions in it based on yesterday's discussion. If you can grep on "ZSOMBI:" you should find them all :)
=== chriadam is now known as chriadam|away
[07:21] <didrocks> zsombi: but from what I understand, any RowLayout for example can get an implicitHeight automatically based on children's height (contrary to for instance, Item or Rectangle…)
[07:21] <zsombi> didrocks: no, you did not understand it right :)
[07:21] <didrocks> zsombi: tell me once you get a chance to have a look, thanks a bunch in advance!
[07:22] <zsombi> didrocks: I did not say yesterday that the RowLayout does that, I said the Row does that
[07:22] <zsombi> didrocks: I said RowLayout works same way as Item or Rectangle
[07:23] <didrocks> zsombi: that's weird, because from my tests, it does seem to me that RowLayout does it (you will see in my // ZSOMBI: that some examples have no height)
[07:24] <zsombi> didrocks: heh?
[07:24] <didrocks> zsombi: for instance (when you get some time to look), the "id: priceRow" seems to have that behavior to me
[07:24] <didrocks> it's a RowLayout and its height seems to be set by its content
[07:25] <didrocks> zsombi: feel free to have a look when you have the time (you can maybe even commit your answers if you prefer this way)
[07:25] <zsombi> didrocks: which one is that file_
[07:25] <zsombi> ?
[07:26] <zsombi> ah, DetailsPage.qml
[07:26] <didrocks> yep :)
[07:28] <zsombi> didrocks: that component makes no sense... why do you use RowLayout there, if none of the component sare using the layouting? Then, each UITK component has an implicit size
[07:29] <zsombi> didrocks: I see TextArea uses the maximumWidth, doe sthat work at all?
[07:30] <zsombi> didrocks: so in this case the layout works as positioner, as the size is not decided by the layout, but by the components themselves, and the layout cannot affect its sizes
[07:30] <didrocks> zsombi: it's indeed for the maximumWidth in the TextArea, and it seems to work?
[07:31] <zsombi> didrocks: what happens if you size the view to be smaller in width_
[07:31] <zsombi> ?
[07:31] <didrocks> zsombi: setting it to 3 gu and it works as expected
[07:31] <zsombi> didrocks: no, resize the view with the mouse
[07:31] <zsombi> didrocks: I did not ask to change the value
[07:32] <zsombi> of the component, the view
[07:32] <didrocks> zsombi: ah, so it's basically fallbacking to width: maxWidth
[07:32] <zsombi> didrocks: will the text area resize? no
[07:32] <didrocks> and it's set
[07:32] <didrocks> so no resizing
[07:32] <zsombi> didrocks: no, it doesnát
[07:32] <zsombi> did if you set the minimumWidth to ex 10GU?
[07:34] <didrocks> yeah, I see, basically the width is set, without resizing capability
[07:34] <zsombi> didrocks: as the first and the last label layouting is not set, nothing will happen
[07:34] <didrocks> so RowLayout is transformed as a positionner
[07:34] <didrocks> and so, the height is implicit from children
[07:34] <didrocks> that's right?
[07:36] <zsombi> didrocks: have you printed its height out onHeightChanged?
[07:38] <didrocks> zsombi: just did that, get one event (I guess once all children are initialized)
[07:38] <zsombi> didrocks: and the height is set?
[07:38] <didrocks> yeah
[07:39] <zsombi> didrocks: hmm... interesting then! so then somewhat acts as the positioner... I did remember it differently, seems they fixed that
[07:40] <didrocks> zsombi: that's a behavior which makes sense to me (at least ;)), but yeah, maybe they changed it if too many people were puzzled about it
[07:40] <zsombi> didrocks: but then woudl be interesting to see how it works with the proper layouting used... try to set minimum and maximum for both blabels and text area, then resize the window runtime
[07:41] <didrocks> zsombi: I guess that's the case of id: tipRow (a little bit lower)
[07:41] <didrocks> it has one element with Layout.fillWidth: true
[07:41] <didrocks> no height set
[07:42] <didrocks> and indeed, the slider changes its width based on parent's size, and the height is set (the event is fired once with a value)
[07:42] <zsombi> didrocks: no, there you have a width set for the RowLayout
[07:43] <zsombi> didrocks: I meant to set the max and min inside this RowLayout, where you don't have width or height set
[07:43] <zsombi> didrocks: in the priceRow
[07:45] <didrocks> zsombi: well, without any set width on RowLayout, as you told, minimum/maximumWidth doesn't work
[07:45] <didrocks> I just have a set width on my TextField, when resizing the window (the whole RowLayout is shifting)
[07:46] <zsombi> didrocks: yes, but I asked you to set all min and max for the two labels and the text area
[07:46] <zsombi> didrocks: or you say you've set it and it doesn't work?
[07:48] <didrocks> zsombi: I have set min/max to all of the children (so the 2 labels and text area), the RowLayout don't have any width and it doesn't work (it has the behavior I explained above ^)
[07:49] <zsombi> didrocks: as expected
[07:49] <zsombi> didrocks: so the auto-height seem to be fixed in the layouter, so I'd expect the same in the ColumnLayout
[07:50] <didrocks> zsombi: yeah, that removes some of my questions.
[07:50] <zsombi> didrocks: so then make sure you set the width
[07:50] <didrocks> zsombi: yeah, or I'm using a Row, which makes more sense in that one
[07:50] <zsombi> didrocks: but instead of width: parent.width binding, I'd suggest you to switch to anchoring
[07:50] <didrocks> ah?
[07:50] <didrocks> why?
[07:51] <zsombi> didrocks: well, depends, if you want to control the width of the text area to fill the remainder space, you should use RowLayout
[07:51] <zsombi> didrocks: it is more performant to use anchors than width bindings
[07:51] <didrocks> zsombi: good to know, changing that and pushing a new rev then
[07:51] <zsombi> didrocks: width binding is evaluated each time the width is changed, anchors do not do that
[07:52] <didrocks> of course! didn't think about it :)
[07:52] <didrocks> zsombi: I guess TextField and Label's height are implicitely set based on Font's size?
[07:52] <zsombi> didrocks: I mean they do, but that is solved under the hood, and does not go through the binding system, which width: parent.width doies
[07:52] <zsombi> didrocks: nopez
[07:52] <zsombi> Label height is, TextField not
[07:53] <didrocks> zsombi: hum, so if you look at billName, I didn't set an height for it?
[07:53] <zsombi> didrocks: Label paintedHeight is, I mean. the size is not
[07:56] <didrocks> zsombi: any idea why I didn't need to set the height on billName thus?
[07:57] <zsombi> didrocks: implicitHeight = clientRect.height, that's why :)
[07:57] <zsombi> childrenRect not clientRect, sorry... childrenRect.height takes into account the paintedHeight
[07:57] <didrocks> so, it does set an implicitHeight based on its content?
[07:57] <didrocks> ok :)
[07:57] <zsombi> and this is done internally
[07:59] <didrocks> zsombi: ok, changed this and switched to anchors (pushed a new rev). So there are only 2 remaining ZSOMBI: question in the code (one about the icon size and one about the element one) + there is then this bug when clicking on sections
[08:00] <zsombi> didrocks: could you point me to the files having those Qs?
[08:01] <didrocks> zsombi: sure, one is in Main.qml, and the other in pages/DetailsPage.qml (at the very end)
[08:01] <zsombi> didrocks: so Page?
[08:01] <didrocks> zsombi: sorry, the other one is in components/AddRemoveInt.qml
[08:01] <didrocks> zsombi: yeah, Page
[08:02] <didrocks> I guess it's its internal implementation doing that
[08:02] <zsombi> didrocks: Page fills its parent automatically... that was a mistake we have to live with... and header is always overlayd the Page content, and MainView is responsible on scrolling in/out the header
[08:02] <zsombi> overlaid
[08:02] <didrocks> zsombi: ok, that's what I reckoned
[08:03] <zsombi> didrocks: if you want to disable this, set anchors.fill: undefined, then you have to do the anchoring youtself!
[08:03] <zsombi> didrocks: but beware that header management may get broken afterwards!!!
[08:03] <zsombi> next
[08:03] <didrocks> zsombi: yeah, that makes sense ;)
[08:03] <didrocks> so the Button + icon?
[08:04] <zsombi> didrocks: hmm....
[08:05] <zsombi> didrocks: the Button reisizes its width automatically... unless you set a width, or a maximumWidth
[08:05] <zsombi> didrocks: is that what you are asking there?
[08:06] <didrocks> zsombi: well, the resizing is weird for the last element in DetailsPage.qml
[08:06] <didrocks> zsombi: if you uncomment //iconName: "add", then the size is way bigger than just the icon's additional width space
[08:08] <zsombi> didrocks: uh, I checked the AddRemoveInt.qml... not at the DetailsPage yet
[08:09] <zsombi> didrocks: DetailsPage, when you set an icon, it also adds some margins, perhaps that is what you see... screenshot?
[08:12] <didrocks> zsombi: the margins are a little bit large (;)) as you can see here: http://people.canonical.com/~didrocks/qmltest/withicons.png
[08:12] <didrocks> in comparison of: http://people.canonical.com/~didrocks/qmltest/withouticons.png
[08:12] <zsombi> didrocks: wow!!! that's bad!
[08:13] <zsombi> didrocks: smells of a bug!
[08:14] <didrocks> zsombi: isn't it? ;) I wonder if that's due to the RowLayout around the Buttons…
[08:16] <zsombi> didrocks: as long as you don't have any layouting property attached, it shouldn't, but it would be good to check that
[08:16] <zsombi> didrocks: byt the button is not in the RowLayout, it is below it, right?
[08:17] <didrocks> zsombi: right, sorry, it's in an Item
[08:17] <zsombi> didrocks: and all you do is you anchor it to th eright of the item...
[08:17] <didrocks> right, the first to the left, the second to the right
[08:17] <zsombi> exactly
[08:28] <didrocks> zsombi: ok, I have a minimal reproducer, let me open a bug
[08:28] <didrocks> zsombi: http://paste.ubuntu.com/11952349/ is enough
[08:33]  * zsombi break, bbl
[08:34] <didrocks> bug #1478839
[08:34] <ubot5> bug 1478839 in ubuntu-ui-toolkit (Ubuntu) "Adding iconName to Button screw its automatic sizing" [Undecided,New] https://launchpad.net/bugs/1478839
[08:35] <mcphail> popey: have been away for the weekend so just got your message. I haven't documented anything about a non-minimal SDL as it is basically just the minimal SDL with most of the config options switched on! I see bschaefer has opened an interesting new branch which I'm keen to test
[08:35] <didrocks> zsombi: thanks a lot for your answer on the layout system! Now that it seems that the RowLayout changed its behavior compared to what we discussed yesterday, things make sense to me! :)
[08:35] <zsombi> didrocks: yeah... I was outdated on that front, sorry :)
[08:36] <didrocks> zsombi: no worry, I prefer that that me being totally out of any rationale understanding on why that was different from my observations :)
[10:10] <popey> Kaleo: is https://bugs.launchpad.net/camera-app/+bug/1468341 being worked on?
[10:10] <ubot5> Ubuntu bug 1468341 in camera-app "Camera pictures have the wrong resolution" [High,Confirmed]
[10:10] <popey> (Got bitten by this at the weekend, annoyed that video taken on mx4 was 720p and not 1080p 🙁 )
=== chihchun is now known as chihchun_afk
[12:08] <dpm> hi JamesTait, I've been pointed out to bug 1478653 - do you happen to know where the "Highlights of the week" message can be translated? I can't find it on the templates here: https://translations.launchpad.net/software-center-agent/trunk/
[12:08] <ubot5> bug 1478653 in Ubuntu Translations "Ubuntu Store: number of strings are in English instead of being in Russian" [Undecided,New] https://launchpad.net/bugs/1478653
[12:17]  * JamesTait looks
[12:31] <JamesTait> dpm, https://translations.launchpad.net/software-center-agent/trunk/+pots/devportal/ru/+translate?direction=backwards&memo=1030&start=1020 is where it appears.
[12:32] <JamesTait> dpm, in fact, https://translations.launchpad.net/software-center-agent/trunk/+pots/devportal/ru/+translate?batch=10&show=all&search=highlight is probably better. ☺
[12:32] <dpm> JamesTait, ah, thanks. It seems the bug report mistakenly said highlight*s*, which is why I could not find the string!
[12:34] <JamesTait> dpm, it took me a couple of attempts as well. ☺
=== _salem is now known as salem_
[14:13] <didrocks> kalikiana: hey, are you sure that bug #1271973 was fixed? I can confirmed that I don't get the whole document
[14:13] <ubot5> bug 1271973 in U1DB Qt/ QML "Results from U1db.Query contain only indexed fields, not the whole document" [Medium,Confirmed] https://launchpad.net/bugs/1271973
[14:32] <kalikiana> didrocks: lemme check
[14:34] <kalikiana> didrocks: that was not fixed. the bug status is correctly set to Confirm. there was a change meant to address it but it broke ABI
[14:35] <kalikiana> didrocks: the better approach is here https://code.launchpad.net/~kalikiana/u1db-qt/indexRoles/+merge/211771
[14:36] <kalikiana> not sure what the state of it is, though, I haven't had time for a while to actively work on things
[14:37] <didrocks> zbenjamin: maybe you can give a look (I saw that you reviewed other u1db branches) on that one? ^
[14:37] <didrocks> kalikiana: any workaround for now than listing in Indexes all fields you need?
[14:38] <kalikiana> didrocks: I don't think so
[14:41] <didrocks> kalikiana: ok, do you think you have time for pushing this MP a little bit? I think that's quite a bad one for developer experience in particular
[19:34] <cwayne_> hiya, is there a way to programatically get my app's writable directory path?
=== salem_ is now known as _salem
=== karni-away is now known as karni
=== chriadam|away is now known as chriadam