The layout of an ExtensionSDK

Extension SDKs are a very powerful way distribute your control libraries for use in Windows Store and Windows Phone apps.
This article will go through the layout of the extension sdk, and later take that knowledge to build an extension sdk from an already released app.

An ExtensionSDK essentially consists of 3 parts:

  • Files to use during design time
  • Files to deploy as content
  • Assemblies to use for reference

In addition there's a metadata file 'SDKManifest.xml' that describes the content.

The root layout then looks like the following:
…where 'EXTENSIONAME' is the name of your extension, and VERSION is version number in the format "".
For each of these groups you can control what gets deployed in debug and release or both. If you don't want to control whether you use debug or release, you will below these folders use the folder 'CommonConfiguration'. For debug specific configuration use 'Debug', and for release configuration use 'Retail'. In most case you will be using 'CommonConfiguration' though.
This means our folder structure now looks like this:

Next level down in the folders describe if files are related to AnyCPU, x86, x64 or ARM builds (the latter is very useful for C++ projects). For AnyCPU use 'neutral', meaning it doesn't matter. So use this for .NET Assemblies compiled for AnyCPU, image resources, winmd files etc. You will want to use the architecture specific folder if you deploy binaries that are architecture specific.

So what goes in what folders:

  • DesignTime: This is where you will put .Design assemblies if you have specific design time binaries for your assemblies, as well as Generic.xaml. You only need to deploy 'neutral' and/or 'x86' architectures, since VS runs in a 32bit process.
  • Redist: Images, shaders, Generic.xbf, videos etc, AND C++ binaries.
  • References: .NET DLLs, C++ WinMDs, xml doc.

Here's an example layout of an extension sdk that consists of two libraries: One C++ WinRT component (NativeLib) and a Managed library:

The SDKManifest could look like the following:

    <?xml version="1.0" encoding="utf-8" ?>
      DisplayName="My Super Duper Control Library"
      Description="My Control Library"
      Identity="MyControlLib, Version="
      AppliesTo="WindowsAppContainer + ( Managed )"
      DependsOn="Microsoft.VCLibs, version=12.0"
        <File Reference="NativeLib.winmd" Implementation="NativeLib.dll" />
        <File Reference="ManagedLib.dll"/>

Note that if you don't have native dependencies, this would change quite a lot. The full set of properties are pretty poorly documented today, so generally I download and install a wealth of extension sdks and look at them and see if they do similar things to me and then copy from that.

Building an Extension SDK from an installed app

So now that we know the layout of an extension sdk, let us apply that to 'reverse-engineering' an already deployed app into an extension sdk and use that to build our own app on top. Because Windows Store apps aren't fully encrypted, this means you can often take parts of an app that's separated out into libraries and build a new app from these libraries. This is something to consider when you build your app - if you are really good are separating your stuff into sub-libraries, you also make it easier for others to reuse your stuff. As an example let's download the Bing Maps Preview app and reverse it into an SDK and build our own 3D Map App.

When you installed the app, you will be able to access a folder with a name similar to the following with administrator rights:
"c:\Program Files\WindowsApps\Microsoft.Maps3DPreview_2.1.2326.2333_x64__8wekyb3d8bbwe\"

In here we'll find a lot of logic for the app, but the main one we are interested in is the "Bing.Maps" folder and the Bing.Maps dll+winmd. The folder is essentially the content and is image resources and shaders. The Bing Maps.dll and Winmd are C++ WinRT components. Since the dll is C++, the architecture will either be ARM, x86 or x64 depending on what PC you downloaded it on. In my case it's x64 so I should be able to build an extension sdk that will support 64 bit PCs from this alone. If I want to support more, I will have to install the app on a x86 or ARM PC and copy the dll from there as well (the other files are neutral and will be the same).

So let's first create the following folder : "Bing.Maps\\".

Next, let's copy the "Bing.Maps" folder that has all the images and shaders into
Next copy the Bing.Maps.dll into (x64 if that's what you have, change/add ARM/x86 if your binary isn't x64)
Lastly, copy the Bing.Maps.winmd into:

Lastly we need to create a new file in \Bing.Maps\\SDKManifest.xml to describe the SDK:

    <?xml version="1.0" encoding="utf-8" ?>
      DisplayName="Bing Maps"
      Identity="Bing.Maps, Version="
      DependsOn="Microsoft.VCLibs, version=12.0"
        <File Reference="Bing.Maps.winmd" Implementation="Bing.Maps.dll" />

Voila! We now have an ExtensionSDK. There's several ways you can 'install' this into Visual Studio. The simplest way is to copy the folder into your user folder under %USERPROFILE%\AppData\Local\Microsoft SDKs\<target platform>\v<platform version number>\ExtensionSDKs.
In this case %USERPROFILE%\AppData\Local\Microsoft SDKs\Windows\v8.1\ExtensionSDKs\

If you're building an installer you can also install it into
    %Program Files%\Microsoft SDKs\Windows\v8.1\ExtensionSDKs
Or specify a link to the location of the folder in a registry key:
    HKLM\Software\Microsoft\Microsoft SDKs\Windows\v8.1\ExtensionSDKs\Bing.Maps\\

Lastly you can do this in your project file    by using the 'SDKReferenceDirectoryRoot' tag. Add the following right before the <Target Name="BeforeBuild"/> tag at the very bottom.


Note that for the latter, the folder should point to a root extension sdk folder, meaning the SDK above must be located in a certain tree under this folder. In this case:

When you've done any of these install options, you can now get started building an app. Go to "Add References" and the Bing Maps entry should show up.


Now add the following XAML to your page:

        <bing:Map x:Name="map" xmlns:bing="using:Bing.Maps">
                <bing:ThreeDimensionalMapProjection />

And in code-behind after "InitializeComponents":

            map.BaseLayers = Bing.Maps.BaseLayers.CreatePhotoRealisticOverlay();

Run the app and you should see a 3D globe!


Note: This Bing Maps SDK is not based on anything officially released but on a un-finished app. This is by all means a giant hack and only meant as an exercise to build an Extension SDK. Use this at your own risk and don’t attempt to publish any apps using it.


Notes from Mix08

Virtual Earth announced during their session that they will release a Silverlight Control for embedding maps in your silverlight app. It will be interesting to see if the silverlight adds any benefit over the HTML/JS control.

Deep Zoom: This is an unmanaged part of SilverLight and not very customizable. The tilingscheme is hardcoded. You cannot create images larger than 4x4billion pixels. That's still enough for having a sub-millimeter precision image of the entire earth, so I think that should be sufficient!

The Steve Ballmer Keynote was very entertaining. It was more of a live interview than a keynote. Rumour has it that the questions he got really upset him, but he hides it very well. WATCH IT!

Other sessions worth watching:
Developing ASP.NET applications using the Model View Controller Pattern
Scott Hanselmans' presentation on WPF user controls was the best session that I attended, and also very entertaining. I talked to him today and he admitted to be nervous which was partly why he kept shooting off jokes. It turned out to be be the best presentation, so it worked out :-)

Developing Cutting Edge Web Applications With Internet Explorer 8
Lots of good tips on changes from IE7 to IE8. Espcially the part about the stuff they fixed that will break existing apps is a must read. Also a bit about new features, XDomainRequest objects, new accessibility features etc.

Creating Rich, Dynamic interfaces with Silverlight 2


Huge wind turbines in Virtual Earth

Since today is Blog Action Day and the topic is the environment, I thought I would share this amazing birds eye view of one of the 2mW windmills in the offshore windfarm "Middelgrunden" just off the coast of Copenhagen that was published last week. I've knew these windmills were BIG, but looking from the peer it's hard to tell how big they really are. The boat next to it really gives a sense of its size. I can't even begin to imagine how big the 5mW mills they make these days are... 

There are 20 of these mills, all together producing 3% of Copenhagens total electricity consumption. Around 20% of Denmarks electricity comes from windmills.

Warning... bad joke coming up: Now you know why I moved to Sunny California - The weather really "blows" in Denmark :-)


New Virtual Earth release

I just visited and it looks like they finally put the new release out there. A revamped design and Birds Eye View imagery inside the 3D viewer. You can now also create animations using your collections.

Nyhavn, Copenhagen - A place I missed a lot this summer

Another new feature is the new "Create model" button in the collection editor when you are in 3D mode. Using it will prompt you to download "3dvia" enabling you to model your own buildings and insert them into VE. It looks somewhat similar to SketchUp used for the same thing in Google Earth. A neat feature is that you get the map as a background to assist you in your modelling, and comes with a bunch of default textures as well:

The Microsoft Live Maps and Google Maps projection

I have lately seen several blogposts confused about which datum and projection Microsoft’s Live Maps (Virtual Earth) and Google Maps use. As most people already know by now, they render the round earth onto a flat screen using a Mercator projection.

I think the confusion comes from that they actually use two spatial reference systems at the same time:

  1. Geographic  Longitude/Latitude coordinatesystem based on the standard WGS84 datum.
  2. Mercator projection using a datum based on WGS84, BUT modified to be spheric.

So when is what used?

The Javascript API’s use (1) as input when you want to add points, lines and polygons. That is, they expect you to input any geometry in geographical coordinates, and click events etc. will also return geometry in this spatial reference. This is the coordinate system most javascript developers will use. The API will automatically project it to the spheric mercator projection.

If you want to create image overlays, or roll your own tile server on top of the map, you will need to project your images into a spheric mercator projection. The JavaScript APIs are not able to do this for you.

Here’s a bit of facts about the two projections:

The valid range of (1) is: [-180,-85.05112877980659] to [180, 85.05112877980659].

The valid range of (2) is: [-20037508.3427892, -20037508.3427892] to [20037508.3427892, 20037508.3427892]

Well-known Text for (1):

Well-known Text for (2):
PROJCS["Mercator Spheric", GEOGCS["WGS84basedSpheric_GCS", DATUM["WGS84basedSpheric_Datum", SPHEROID["WGS84based_Sphere", 6378137, 0], TOWGS84[0, 0, 0, 0, 0, 0, 0]], PRIMEM["Greenwich", 0, AUTHORITY["EPSG", "8901"]], UNIT["degree", 0.0174532925199433, AUTHORITY["EPSG", "9102"]], AXIS["E", EAST], AXIS["N", NORTH]], PROJECTION["Mercator"], PARAMETER["False_Easting", 0], PARAMETER["False_Northing", 0], PARAMETER["Central_Meridian", 0], PARAMETER["Latitude_of_origin", 0], UNIT["metre", 1, AUTHORITY["EPSG", "9001"]], AXIS["East", EAST], AXIS["North", NORTH]]

Proj.4 definition for (1):
+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs

Proj.4 definition for (2) (see here for an explanation of the weird ’nadgrids’ parameter):
+proj=merc +a=6378137 +b=6378137 +lat_ts=0.0 +lon_0=0.0 +x_0=0.0 +y_0=0 +k=1.0 +units=m +nadgrids=@null +no_defs

So why this weird extent for the latitude? First of all, the poles in the Mercator projection extends towards infinity, so at some point they have to cut them off (and who cares about ice anyway - apart from that its melting). If you look at (2) these latitude/longitude values projected into the spheric mercator results in a perfect square, fitting perfectly with squared image tiles, that are simple to sub-divide over and over again, as you zoom in. I expect the reason for the spheric datum is for simplicity and perfomance when reprojecting points from longitude/latitude to screen coordinates. Charlie Savage also has a more mathematical approach to deriving these values.

For a quick introduction to projections, coordinate systems and datums see here.

If you want to know more about how these mapping api's work, keep an eye on Jayant's blog.

Update: We now have an offical EPSG code for the projection. See details here.