I am using Xamarin Forms. I would like to download jpg file (it is done) and then open that jpg in default application on Android/iOS (opening photo browser with this photo). Of course photo is single example, I would like to open any file in default application.
I found several solutions native-only but my application is designed to be cross-platform. I though that I can use Launcher from Xamarin.Essentials package but apparently I can't.
How to achieve this?
You can have a try with Xamarin.Essentials: Launcher:
var fn = "File.txt";
var file = Path.Combine(FileSystem.CacheDirectory, fn);
File.WriteAllText(file, "Hello World");
await Launcher.OpenAsync(new OpenFileRequest
{
File = new ReadOnlyFile(file)
});
I found several solutions native-only
Opening something in another app is quite close to the system for a mobile application and there are some things to consider, which dependend on the platform. Usually, mobile apps run in a sandbox with very limited access to the surrounding system. Particularly this means that, if you downloaded a file to the sandbox of your app, other apps (which native viewers are), aren't allowed to access the file.
On Android, you can copy the file to a shared space (see Application.Context.GetExternalFilesDir(null)) and then open it. This might be possible with Essentials, but I'm not quite sure, but since we're on the Android platform anyway now, you could create an intent now anyway.
On iOS you create controllers from within your app (for example the QLPreviewController to preview the file) that may access items in your sandbox. Depending on the type of controller (e.g. UIActivityViewController) they may open other apps.
How to use this platform-independently?
Since you are programming a platform independent app, you'll have to take care that the correct class is called to the platform dependent work. There are several options how you can achieve this
Use the DependencyService
Use a real dependency injection framework
Use an abstract base class with initialization in the platform dependent projects
DependencyService
To use the Xamarin.Forms DependencyService you need two things
An interface for the functionality you'd like to implement
One implementation per platform
Assuming you hvae a simple interface to share a file
public IShareFile
{
void ShareFile(string fileName);
}
you can implement an implementation of this interface on each platform and add the DependencyAttribute to the assembly. e.g. for iOS:
[assembly: Dependency(typeof(MyApp.iOS.DeviceOrientationService))]
namespace MyApp.iOS
{
public class ShareFile : IShareFile
{
public void Share(string fileName)
{
// implementation goes here
}
}
}
The general scaffold is the same for Android, albeit the implementation differs.
Using a real dependency injection framework
Basically it's pretty much the same. You can skip the DependencyAttribute, though. In order to make the implementation available you'll have to get hold of the DI container from your platform specific code, which might be tricky. This might be an overshoot for a single dependency, but if you're using a DI container anyway and there are X dependencies, it might be worth the effort.
Using an abstract base class
Add an abstract base class to your project
public abstract class ShareFile
{
public static ShareFile Instance { get; protected set; }
public abstract void Share(string fileName);
}
and in your implementation in the platform specific project, you add an Init() method
internal class ShareFileImpl : ShareFile
{
public static void Init()
{
ShareFile.Instance = new ShareFileImpl();
}
public void Share(string fileName)
{
// implementation goes here
}
}
This init method must be called from your platform specific code. Most likely during initialization. The implementation can then be accessed via its abstraction from your platform independent code (of course you'll see only the abstraction, public methods added to ShareFileImpl won't be visible from your platform independent code).
ShareFile.Instance.Share(fileName);
A combination of the abstract class approach and dependency injection is also conceivable. When registering your classes in the DI framework, you could register the platform instance like
container.RegisterInstance<ShareFile>(ShareFile.Instance);
This way you can make use of the DI container features (e.g. constructor injection), while keeping the hassles of using the DI container from your platform specific project away from you. The drawback is, that you'll still have to call ShareFileImpl.Init() from your platform specfic code.
Related
I have a store app that uses the mvvmcross sqlite plugin (community edition). This app has a periodic background task that accesses the database to get data to be shown in a live tile. I can't see how I can get access to this database from the background task. I would like to use the mvvmcross sqlite plugin in the background task, but I don't see how to initialize the mvvmcross environment properly.
If you want to initialize the full MvvmCross framework including all of your app, then you'll need to run your Setup class.
In WinRT, this could be as simple as calling:
var setup = new Setup(null /*rootFrame*/);
setup.Initialize();
although it may require you to do a little work to:
Make sure your presenter does not use the null rootFrame
Provide some other means to create a UI thread dispatcher - currently MvxStoreViewDispatcher relies on .Dispatcher access - see https://github.com/MvvmCross/MvvmCross/blob/v3.1/Cirrious/Cirrious.MvvmCross.WindowsStore/Views/MvxStoreViewDispatcher.cs - to do this, you could override InitializeViewDispatcher with something like:
protected override void InitializeViewDispatcher()
{
if (_rootFrame != null)
{
base.InitializeViewDispatcher(); return;
}
var dispatcher = new NonMainThreadDispatcher();
Mvx.RegisterSingleton<IMvxMainThreadDispatcher>(dispatcher);
}
public class NonMainThreadDispatcher : MvxMainThreadDispatcher
{
public bool RequestMainThreadAction(Action action)
{
action();
}
}
If you want to initialize less functionality than the entire framework (e.g. for memory reasons) then you can also consider creating special Setup and App classes just for your background task.
Aside> This is similar to questions like these in Android - Using MvvmCross from content providers and activities and MvvmCross initialization
I was able to solve the problem in a straightforward way. Since the background task only needed the SQLite data service from the PCL core project, I did the following:
Included a reference to the Core project.
Added the nuget packages for MvvmCross and the SQLite community plugin.
Deleted all of the files and folders added when doing the mvvmcross install: Bootstrap/, Todo-Mvvmcross/, Views/, DebugTrace.cs, and Setup.cs.
There is a current limitation in the nuget installer that requires some additional edits to the project file to handle multiple store platforms (x86, ARM, and x64), see 'Cirrius.Mvvmcross.Community.Plugins.SQLite.WindowsStore needs platform-specific dlls for X86 and ARM' on Stack Overflow for details. Make sure you put the Choose statement after the default SQLite.WindowsStore reference and you need to leave the default reference in the project file. You will also need to adjust the HintPath based on the location/names of your references.
Initialized the SQLite data service by explicitly calling the factory and creating a new instance of the data service:
var factory = new MvxStoreSQLiteConnectionFactory();
IMyDataService repository = new MyDataService(factory);
I then have access to the data service with no other overhead associated with mvvmcross.
I am modifying the N-10-KittensDb sample solution.
I See how to create a SQLite database, but I wish to use an existing database. I am guessing that I need to copy the database to the proper UI data folder. Maybe it is done within the Core project? And if so how is the correct path injected into the running Exec? As the Core can be used across many UI's? What method is called to see if the database exists or needs to be copied?
Sample from DataService:
public DataService(ISQLiteConnectionFactory factory)
{
const string file = "Cats.sldb";
var connection = factory.Create(file);
connection.CreateTable<Kitten>();
}
I believe the paths are different for Android vs Phone vs Touch vs Wpf?
Please direct me to a sample piece of code that uses the Cirrious.MvvmCross.Plugins.Sqlite for Phone or Wpf.
Thank you
Dan
Each platform by default creates a database in a folder location appropriate for the platform - e.g. Touch uses:
public ISQLiteConnection Create(string address)
{
var path = Environment.GetFolderPath(Environment.SpecialFolder.Personal);
return new SQLiteConnection(Path.Combine(path, address));
}
from https://github.com/slodge/MvvmCross/blob/v3/Plugins/Cirrious/Sqlite/Cirrious.MvvmCross.Plugins.Sqlite.Touch/MvxTouchSQLiteConnectionFactory.cs#L18
To read/write files, MvvmCross does bundle a File plugin - this also operates by default in platform specific locations - but the two may not match perfectly - e.g. see:
protected override string FullPath(string path)
{
if (path.StartsWith(ResScheme))
return path.Substring(ResScheme.Length);
return Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.MyDocuments), path);
}
from https://github.com/slodge/MvvmCross/blob/v3/Plugins/Cirrious/File/Cirrious.MvvmCross.Plugins.File.Touch/MvxTouchFileStore.cs#L22
Because of this mismatch, in order to share the same database-specific copy code across platforms you may find it easier to just inject your own platform specific copy on each platform - for more on injecting platform specific services, see http://slodge.blogspot.co.uk/2013/06/n31-injection-platform-specific.html
Hi I have written a script using HTMLUnit that fetches a web page given a url and performs certain operations on them. For instance searching for a string on a page or clicking on a link and so on. I am creating a runnable jar using eclipse IDE.
HTML-Unit consists of about 21 different library jars that i am extracting in my single final runnable jar. These dependencies cause the single jar to occupy a space of about 9.3MB. I have been trying to reduce the individual jar size using obfuscation. I am using a tool called proguard for it.
Here is a sample proguard configuration i am using to obfuscate a single library jar by the name of "commons-logging-1.1.1.jar":
`-libraryjars <java.home>/lib/rt.jar
-injars C:/Users/Desktop/Jars/commons-logging-1.1.1.jar
-outjar C:/Users/Desktop/SmallJars/commons-logging-1.1.1.jar
-printmapping C:/Users/Desktop/SmallJars/out_commons-logging-1.1.1.map
-renamesourcefileattribute SourceFile
-dontnote
-keepattributes InnerClasses,SourceFile,LineNumberTable,Deprecated
-keep public class * {
public protected *;
}
-keepclassmembernames class * {
java.lang.Class class$(java.lang.String);
java.lang.Class class$(java.lang.String, boolean);
}
-keepclasseswithmembernames class * {
native <methods>;
}
-keepclassmembers class * implements java.io.Serializable
{
static final long serialVersionUID;
private void writeObject(java.io.ObjectOutputStream);
private void readObject(java.io.ObjectInputStream);
java.lang.Object writeReplace();
java.lang.Object readResolve();
}`
The config is pretty much the same as one given on the proguard website in usage -> typical libraries. On rebuilding the project in eclipse using these 21 reduced jars and running it, the script fails at runtime with the exception:
"java.lang.NoSuchMethodException: org.apache.http.conn.ssl.SSLSocketFactory.createDefaultSSLContext()"
Looks like i have obfuscated the individual jars in a manner that certain methods will now not be found. Could you guide me as to what may be causing these exceptions. and is there something wrong with the config file above.If
so what would be the best proguard configuration for this scenario.
I am aware another member was chasing a similar problem. The question is posted at link:
[a link] (Determine used libraries to reduce JAR file size)
Thank you!!
The method is accessed by reflection, which ProGuard can't know from its static analysis. You have to preserve it in your configuration. Cfr. ProGuard manual > Troubleshooting > NoSuchMethodException.
Processing the application as a whole will be much more effective than processing the libraries individually, because ProGuard may then remove entirely unused libraries or at least larger parts of them.
I'm having trouble figuring out what the best approach is these days for Ninject and ASP.NET MVC 3.
I have used Install-Package Ninject.MVC3 on my application and have my bootstrap class with the following methods:
public static void Start()
public static void Stop()
private static IKernel CreateKernel()
private static void RegisterServices(IKernel kernel)
It's all great and it loads my modules as expected. But historically what I have done is something like this:
MyApp.dll
Kernel.Bind<ISomething>().To<Something>();
Kernel.Bind<IBlah>().To<Blah>();
Kernel.Bind<IFoo>().To<Foo>();
MyApp.Tests.dll
Here I want to override ONLY ISomething's binding, so I used to just unbind the thing I needed to mock/whatever and rebind:
Kernel.Unbind<ISomething>();
Kernel.Bind<ISomethig>().To<TestSomething>();
But there isn't a method in the Nuget package that implies a thought through way to achieve this with the App_Start class from the original library. If I put another Ninject bootstrap class in my test app it only seems geared up to build a new kernel:
[assembly: WebActivator.PreApplicationStartMethod(typeof(TestNinjectBootstrapper), "Configure")]
I could store the kernel in the original bootstrapper statically and call from the tests project, but this feels wrong. Am I thinking too much and missing something? Or thinking too little :)
Argh. What is a good approach?
To reuse interface/class mapping registration in different project there is ability to create NInject modules. Modules just need to implement the INinjectModule interface, but most should extend the NinjectModule class for simplicity.
So you can place interface/class mapping inside module like in the following example:
public class WarriorModule : NinjectModule
{
public override void Load()
{
Bind<IWeapon>().To<Sword>();
Bind<Samurai>().ToSelf().InSingletonScope();
}
}
After you define such module you can instantiate Kernel with mapping defined in this module.
All that you need is to specify this module as argument during creating Kernel object:
IKernel kernel = new StandardKernel(new WarriorModule());
Note that you can create and instantiate kernel with multiple modules.
So, modules will help you to reuse default mapping configuration. Mapping configuration will be defined in one place which will simplify maintance especially if there are several projects which uses the same interface/class mapping configuration.
There are also some other features like 'Dynamic Module Loading' and etc. More information about modules can be found here.
I have the requirement that the end-user can change localized resources and the changes should be visible in the application without the need to restart the application.
Update to clarify the scenario:
I am talking about changing the localized resources at runtime. Lets say I have a typo in the german translation of a page. Then some admin-user should have the possibility to change that typo at runtime. There should be no need for a redeployment or restart in order for this change to be reflected in the UI.
I am using ASP.NET MVC3.
What options do I have?
I have been looking into writing a custom ResourceProvider that loads resources from the database.
This seems not too much effort, however so far I pointed out two drawbacks:
It is not working with the DataAnnotations that are used for convenient validation in MVC3 (DataAnnotations work with a ErrorMessageResourceType parameter, which only works with compiled resources)
We basically have to provide our own tooling around managing resources (like translating etc.) which is a pity, since there are a lot of tools for this that work with resx-files.
What are the other options? Would manipulation of the deployed resx-files at runtime be an option?
But I suspect that the application is automatically "restarted" when it detects those changes: I suspect ASP.NET realizes that the resx-files have changed, it then recycles the application-pool and compiles the new resx-files on the fly.
Is this correct? Is there any way around this?
I have not yet looked into compiling the resources into satellite assemblies before deployment. Is this even a recommended scenario for web applications?
But even with compiled satellite assemblies I suspect that ASP.NET restarts the application, when those assemblies are changed on the fly. Is this correct?
I would be interested in any experience in how the original requirement can be satisfied?
And I would be interested in any comments about the options I have mentioned above.
DataAnnotations accept a ErrorMessageResourceType which tells the ValidationAttrributes where to access resources. You can pass this as follows:
[Required(
ErrorMessageResourceType = typeof(DynamicResources),
ErrorMessageResourceName = "ResourceKey")]
public string Username { get; set; }
By creating a type for this parameter with static properties for each key you can create an implementation that loads resources from a database or other implementation. You could then combine this with a dynamic object for DRY and move the implementation into TryGetMember. Potentially then use T4 templates to generate the statics from your database at compile time, ending up with this:
public class DynamicResources : DynamicObject
{
// move these into partial and generate using T4
public static string MyResource
{
get { return Singleton.MyResource; }
}
public static string MyOtherResource
{
get { return Singleton.MyOtherResource; }
}
// base implementation to retrieve resources
private static dynamic singleton;
private static dynamic Singleton
{
get { return singleton ?? (singleton = new DynamicResources()); }
}
public override bool TryGetMember(GetMemberBinder binder, out object result)
{
// some logic here to look up resources
result = GetResourceKeyFromDatabase(binder.Name);
return true;
}
}
Of course it would be perfect if resources weren't static properties.