I want to create a Windows 8 Metro application that has a fancy "hub" page similar to the following:
http://blogs.msdn.com/cfs-file.ashx/__key/communityserver-blogs-components-weblogfiles/00-00-01-51-31-metablogapi/3482.landing_2D00_branded_5F00_thumb_5F00_287109A4.jpg
The key elements are: Horizontal Panorama style scrolling page but with "sections" that are data bound. I've had a look around for implementations and have come up short.
Seems to me it would be possible to do with a ScrollViewer that contains a Horizontal Stackpanel that has the elements that make up each section. But seeing as this is such a common Metro design pattern, I figure there is probably an easier / recommended way of implementing.
Just wondering if someone can give me some tips. How would you do it?
When you start using VariableSizedWrapGrid all things get their own relative sizes. http://msdn.microsoft.com/en-us/library/windows/apps/br227651.aspx
Set the ColumnSpan and RowSpan properties to make things double size, tripple size etc...
For me I make the first item in the collection double width, double height by setting ColumnSpan and RowSpan to 2. All other items get this ColumnSpan and Rowspan to 1, and it's getting a nice effect.
In addition to what Hans mentioned, I would highly recommend going through the existing Metro Style App Samples. Learning from working examples works better than any documentation.
Related
Good morning!
I'm currently trying to create a view similar to this mockup.
I have put down 3 different screen sizes so you can see the issue.
I have a header background image (grey box) with an angled bottom. On the right I want to display an image, which obviously needs to be positioned.
Positioning it horizontally is no issue but how can I position the image vertically? I have it positioned fixed for one screen size but obviously need to make it flexible.
Any ideas? Help would be much appreciated!
David
You can definitely use measure as #rajesh pointed out, or you can use Dimensions. As far as getting the layout consistent across devices, using position absolute and measuring the device height should allow you to get consistency across these devices.
Check out this example I set up, it should be a good starting point at least.
https://rnplay.org/apps/pSzCKg
Now I could be wrong about this but after testing it all day, I have discovered...
When adding a widget and setting the z-index, the value "0" seems to be the magic depth.
If a widget's Z is at 0, it will be drawn on top of everything that's not at 0, Z wise.
It doesn't matter if a widget has a z-index of 99, -999, 10, -2 or what ever... It will not appear on top of a widget who's z-index is set to 0.
It gets more strange though...
Any index less than -2 or greater than 2 seems to create an "index out of range" error. Funny thing is...when I was working with a background and sprite widget, the background's Z was set to 999 and no errors. When I added another sprite widget, that's when the -2 to 2 z-index limitation appeared.
Yeah I know...sounds whacked!
My question is, am I right about "0" being the magic Z value?
If so, creating a simple 23D effect like making a sprite move being a big rock will take some unwanted code.
Since you can only set Z when adding,a widget, one must remove and immediately add back, with the new Z value...a widget.
You'll have to do this with the moving sprite and the overlapping object in question. Hell, I already have that code practically written but I want to find out from Kivy pros, is there a way to set z-index without removing and adding a widget.
If not, I'll have to settle for the painful way.
My version of Kivy is 1.9.0
What do you mean by z-order? Drawing order is determined entirely by order of widgets being added to the parent, and the index argument to add_widget is just a list index at which the widget will be inserted. The correct way to change drawing order amongs widgets is to remove and add them (actually you can mess with the canvases manually but this is the same thing just lower level, and not a better idea).
I found a working solution using basic logic based on the fact widgets have to be removed and added again in order to control depth/draw order.
I knew the Main Character widget had to be removed along with the object in question...so I created a Main Character Parent widget, which defines and control the Main Character, apart from its Graphic widget.
My test involves the Main Character walking in front of a large rock, then behind it...creating a 23D effect.
I simply used the "y-" theory along with widget attach and detach code to create the desired effect.
The only thing that caught me off guard was the fact my Graphic widget for my Actor was loading textures. That was a big no no because the fps died.
Simple fix, moved the texture loading to the Main Character Parent widget and the loading is done once for all-time.
PS, if anyone knows how to hide the scrollbars and wish to share that knowledge, it'll be much appreciated. I haven't looked for an API solution for it yet but I will soon.
Right now I'm just trying to make sure I can do the basic operations necessary for creating a commercial 23D game (handhelds).
I'm a graphic artist and web developer so coming up with lovely visuals won't be an issue. I'm more concerned with what'll be "under the hood" so to say. Hopefully enough, lol.
I am designing a simple music app where the user gets to play instruments i.e. Drums, and the problem that I am facing is with resolutions.
The drums are images, which I have converted them into buttons. Everything looks great at the state that I have designed it.
However, when I switch to other resolution states, the button(image) are distorted, e.g. skewed, scaled, and looks nasty.
I have tried designing or arranging them via selecting 'Enable state Recording', but the specific designs for that state are not being saved.
Have you tried the approaches discussed here? http://msdn.microsoft.com/en-us/library/windows/apps/hh465362.aspx For the actual button sizes, make sure you are not fixing the width/height with pixel values. Use * weighted rows and columns to layout your grids and have the buttons autosize to fill a given cell in the grid. Then match with the appropriate image resource per the article.
Grids are great for dividing up available space but they can't account for changes in aspect ratios. If your items are still set to Stretch (or Fill) then they can end up out of aspect ratio. Another option is to design the entire layout at a fixed size (let's say 1024 x 768 or 1366 x 768) and wrap the entire thing in a ViewBox. ViewBox will scale all elements equally and maintain the aspect ratio, adding letterboxing (or empty space) on the sides / top & bottom if necessary. This might be a better approach for a drum kit.
Hope that helps.
Redid the whole project of designing again.
This time, I put the image inside a specific grid and that made things lot better. :)
I'm developing a Windows 8 Metro-style application and I want to use vector images. As there seems to be no direct support for svg images, I am trying to use a xaml fragment consisting of multiple shapes (a path and some lines) as an image. I would like to have a resource dictionary entry with the composite shape and be able to include it in different pages. Ideally, I would also like to be able to resolve a specific composite shape from a data bound property.
From what I've read, the WPF approach was to have a VisualBrush or DrawingBrush consisting of the shapes, but there are no such classes in Windows 8 (and it seems like it's not even possible to derive from Brush).
How am I supposed to do this using WinRT UI?
No, you cannot use a DrawingBrush as the value of a background property in WinRT XAML. It's too bad, huh? Seems like a very powerful features to setup the fill of an object with vector layouts. In fact DrawingBrush is not even part of Windows 8 yet. It is what it is. For now, images are a fine solution. But we feel your pain.
I might as well toss in that VisualBrush is not part of WinRT-XAML either.
Each XAML fragment is really a UI element. I think the simplest approach would be to put the XAML into it's own user control and then add the user control to each page you wanted to display the "drawing" on. If you want the user control to display different shapes, you can expose a property on the control, data bind to that property, and inside the setter of the property read the value and toggle the visibility of the various XAML shapes to show or hide whatever parts of the composite image you want. It is a bit brute force, but will do what you want.
Windows8 way to have vector graphics in it is just to use ViewBox+Canvas and Path elements. It works well in my opinion, though I do miss VisualBrush.
I was looking at http://www.msnbc.msn.com/id/26295161/ns/weather/ and noticed that the flash map is provided by bing and have a custom color scheme. I have a similar need to show maps with a black & white colortheme, is it possible using their api or have they made custom tiles ? (doesnt seem feasible to provide tiles for alls maps on all levels :)
If you are not wedded to Bing maps, you could look into Mapbox, which provide maps based on OpenStreetMap in any style that you desire.
This question is still not answered and even if it's kind of old, I think it matters.
In fact, they're not using different tiles or a custom tile scheme. You can see those in the network inspection tool.
Since they're inside a Flash application, they can easily change the color of the image using filter on image (ColorMatrixFilter) and changing dynamically on the client side the rendered colors.
Well now some years later you can: https://msdn.microsoft.com/en-us/library/mt823636.aspx :-)
(I got here by looking for some sample styles and thought I add the link for anybody else finding this question)
Perhaps a custom tile skinner is that you need: http://rbrundritt.wordpress.com/2009/11/27/bing-maps-custom-tile-skinner/
In this example, the author changes the color of roads using ColorMatrix and ColorMap:
The two key methods used to transform the colors
consist of using ColorMatrix or a ColorMap. By
using a color matrix all the colors on the tile
can be changed within a couple lines of code. Using
a ColorMap allows you to change one color at a
time. This is useful if you only want to change a
few colors on the map.