I am using Robolectric for my Unit testing and updated my Robolectric jar from 1.2 to 2.2 and trying to figure out how to bind shadow classes in this new version. This is what I was doing before:
Robolectric.bindShadowClass(ShadowLog.class);
#Implements(Log.class)
public static class ShadowLog {
public static int i(java.lang.String tag, java.lang.String msg) {
System.out.println("[" + tag + "] " + msg);
return 0;
}
}
But I think now there is no bindShadowClass API available. I tried using addShadowClass but I am not sure if this is the right way to add a shadow class. Can I just use
ShadowMap a = new ShadowMap.Builder().addShadowClass(ShadowLog.class).build();
Do I need to create a classHandler or something using this shadowMap and if yes, how do I create and use that classHandler to get access to my Log class methods?
#Implements(Log.class)
public static class ShadowLog {
public static int i(java.lang.String tag, java.lang.String msg) {
System.out.println("[" + tag + "] " + msg);
return 0;
}
}
And then Log.i("LogTest", "log message ");
Thanks
Abhi
The binding of shadow classes is now replaced with #Config annotations.
Example:
#Config(shadows = {ShadowLog.class})
See also my answer to this other question and the Robolectric blog.
Related
I have an Xamarin Forms app with MvvmCross for Android and IOS and I would like to add a dark theme. My idea was to have to dictionaries with the ressources for either the dark or the light theme and load the one I need on startup.
I added this after I registered the dependencies in my MvxApplication:
if (Mvx.IoCProvider.Resolve<ISettingsManager>().Theme == AppTheme.Dark)
{
Application.Current.Resources.Add(new ColorsDark());
}
else
{
Application.Current.Resources.Add(new ColorsLight());
}
ColorsDark and ColorsLight are my ResourceDictionary. After that i can see the new Dictionary under Application.Current.Resources.MergedDictionaries but the controls can't find the resources as it seems. However it does work when I add it to the App.xaml
<ResourceDictionary Source="Style/ColorsDark.xaml" />
Do I have to put move that another part in the code or is that a wrong approach at all?
Personally don't like this approach at all. What i do: have a static class with all the colors, sizes etc. defined in static fields. At app startup or at app reload after changing skin just call ex: UiSettings.Init() for this ui definitions static class, like follows:
public static class UiSettings
{
public static Init()
{
if (Settings.AppSettings.Skin=="butterfly")
{
ColorButton = Color.Blue;
TitleSize= 12.0;
}
else
if (Settings.AppSettings.Skin=="yammy")
{
ColorButton = Color.Red;
if (Core.IsAndroid)
ButtonsMargin = new Thickness(0.5,0.6,0.7,0.8);
}
// else just use default unmodified field default values
}
public static Color ColorButton = Color.Green;
public static Thickness ButtonsMargin = new Thickness(0.3,0.3,0.2,0.2);
public static double TitleSize= 14.0;
}
in XAML use example:
Color= "{x:Static xam:UiSettings.ColorProgressBack}"
in code use example:
Color = UiSettings.ColorProgressBack;
UPDATE:
Remember that if you access a static class from different assemblies it is possible that you will access a fresh copy of it with default values where Init() didn't happen, if you face such case call Init() from this assembly too.
If you want something to load up when your app loads then you have to code it in App.xaml.cs
protected override void OnStart ()
{
if (Mvx.IoCProvider.Resolve<ISettingsManager>().Theme == AppTheme.Dark)
{
Application.Current.Resources.Add(new Xamarin.Forms.Style(typeof(ContentPage))
{
ApplyToDerivedTypes = true,
Setters = {
new Xamarin.Forms.Setter { Property = ContentPage.BackgroundImageProperty, Value = "bkg7.png"},
}
});
}
else
{
Application.Current.Resources.Add(new Xamarin.Forms.Style(typeof(ContentPage))
{
ApplyToDerivedTypes = true,
Setters = {
new Xamarin.Forms.Setter { Property = ContentPage.BackgroundImageProperty, Value = "bkg7.png"},
}
});
}
}
In this code I'm setting the BackgroungImage of all of my pages. Hope you'll get the idea from this code.
Any idea how to do what the title says? Only thing I found was on the original Velocity site, and I don't think
ve.setProperty( RuntimeConstants.RUNTIME_LOG_LOGSYSTEM_CLASS,
"org.apache.velocity.runtime.log.Log4JLogChute" );
ve.setProperty("runtime.log.logsystem.log4j.logger",
LOGGER_NAME);
will work wonderfully well on .NET. I am using log4net, which should make it quite easy, but the documentation on NVelocity is really a mess.
Implement NVelocity.Runtime.Log.ILogSystem (you could write a simple implementation that bridges to log4net) and set this impl type in the property RuntimeConstants.RUNTIME_LOG_LOGSYSTEM_CLASS
How I got this information:
Get the code.
Search for "log" in the codebase
Discover the classes in NVelocity.Runtime.Log.
Read those classes' source, they're very simple and thoroughly documented.
Update:
Currently, NVelocity does not support logging. The initializeLogger() and Log() methods in RuntimeInstance Class are commented out.
If you need to log, uncomment the two methods, add a private ILogSystem logSystem; property
Here's our on-the-fly implementation:
public class RuntimeInstance : IRuntimeServices
{
private ILogSystem logSystem;
...
...
private void initializeLogger()
{
logSystem = LogManager.CreateLogSystem(this);
}
...
...
private void Log(LogLevel level, Object message)
{
String output = message.ToString();
logSystem.LogVelocityMessage(level, output);
}
...
}
Then, we implemented ILogSystem for log4net
using log4net;
using NVelocity.Runtime;
using NVelocity.Runtime.Log;
namespace Services.Templates
{
public class Log4NetILogSystem : ILogSystem
{
private readonly ILog _log;
public Log4NetILogSystem(ILog log )
{
_log = log;
}
public void Init(IRuntimeServices rs)
{
}
public void LogVelocityMessage(LogLevel level, string message)
{
switch (level)
{
case LogLevel.Debug:
_log.Debug(message);
break;
case LogLevel.Info:
_log.Info(message);
break;
case LogLevel.Warn:
_log.Warn(message);
break;
case LogLevel.Error:
_log.Error(message);
break;
}
}
}
}
Then, when creating the engine:
var engine = new VelocityEngine();
var props = new ExtendedProperties();
props.SetProperty(RuntimeConstants.RUNTIME_LOG_LOGSYSTEM,
new Log4NetILogSystem(LogManager.GetLogger(typeof(NVelocityEngine))));
engine.Init(props);
The Goal is to have a list of options (that a user can chose through radio buttons) in one place(for eg: a yaml config file). No other place should have this list hard-coded
I've done something similar to create select elements, and I think enums worked just fine. Doing radio buttons should be very similar. I've set it up so that the labels can be defined in the messages file. I'm going to try to excerpt the relevant portions from my larger auto-form-generation code (using FastTags) the best I can. It's a bit heavy for this one case but it makes sense in the larger system.
I use the tag like #{form.selector 'order.status' /}, which looks find the variable named order in the template, sees that status is declared as public Status status, and then goes to find all the values of the Status enum and generate options for them in the select element.
First, I use a FieldContext object which just contains a bunch of info that's used by the other code to determine what to generate along with some utility methods:
public class FieldContext {
public final Map<?,?> args;
public final ExecutableTemplate template;
public final int fromLine;
public Class clazz = null;
public Field field = null;
public Object object = null;
public Object value = null;
private Map<String,String> attrs = new HashMap<String,String>();
private Map<String,Boolean> printed = new HashMap<String,Boolean>();
private List<Option> options;
...
Then I have this in another helper class (its info gets added to the FieldContext):
public List<Option> determineOptions(FieldContext context) {
List<Option> options = new ArrayList<Option>();
if (context.field.getType().isEnum()) {
for (Object option : context.field.getType().getEnumConstants()) {
options.add(new Option(option.toString(), Message.get(option.toString())));
}
}
return options;
}
then the tag declaration is
public static void _selector(Map<?,?> args, Closure body, PrintWriter out, ExecutableTemplate template, int fromLine) {
String field_name = args.get("arg").toString();
TagContext.current().data.put("name", field_name);
SelectHelper helper = HelperFactory.getHelper(SelectHelper.class);
try {
FieldContext context = new FieldContext(field_name, args, template, fromLine);
helper.autoconfigure(context);
TagContext.current().data.put("selected", helper.determineValue(context));
out.print("<div class=\"formutil-field formutil-selector\">");
out.print("<label for=\"" + context.getAttr("id") + "\">");
out.print(helper.findOrCreateLabel(context));
out.print("</label>");
out.print("<select");
context.printAttribute(out, "id", "name");
out.print(">");
if (context.hasOptions()) {
for (Option option : context.getOptions()) {
out.print("<option value=\"" + option.value + "\">" + option.label + "</option>");
}
}
out.print("</select>");
context.printErrorIfPresent(out);
context.printValidationHints(out);
out.println("</div>");
}
...
}
I'm sort of jumping in headfirst to some Flex/AIR stuff. I have a pretty solid background with AS3, but given the inherent hierarchal complexity of Flex (compared to regular Flash), I'm running into an issue.
Let's assume that you have an app where pretty much everything is event driven (common). Accessing elements in the near vicinity of the event target, or the event target itself, is trivial. I'm trying to find, however, the most practical (read: best, most efficient) way to find children that are far removed from the current context.
I know there are functions like getChildAt() and getChildByName(), but that assumes a parent context; what if the element (Flex) you're looking for is several parents up, in a sibling, and then several children down? We take for granted things like jQuery that do this easily, but obviously we don't have that luxury in AS3.
Are any of the following valid? Is there a better way?
Iterate through parents and parents' parents until you find a stop point, find the sibling, and iterate through children and their children until you find your target;
Keep key objects in a global object store (sic) and reference them as necessary (yech)
Use specific dot notation to reach the target, including elements (like skins and their containers - yech again)
Any thoughts would be appreciated.
Edit:
To clarify, let's take an empty Flex 4 AIR app. We have WindowedApplication as the root, obviously, and let's add two SkinnableContainer children with IDs navContainer and mainContainer, respectively. Both have custom skins. Within mainContainer, we have another SkinnableContainer with a vertical layout and ID mainContent, and as one of its children, it has an object (any will do - a spark BorderContainer, maybe) with the ID animatedBox, for example. Within the navContainer, we have a spark Button, which has a listener bound for MouseEvent.CLICK. Within that function, we are going to want to access animatedBox (nativeWindow.mainContainer.mainContent.animatedBox) and animate it to change, say, it's width.
The goal is to access that distant DisplayObject (animatedBox) in a way that is as unobtrusive and efficient as possible, while still conforming to Flex standards that I clearly have yet to possess. :)
in my implementation it is easy to do (however it's in pure AS3):
in display object which handles the click:
private function onClick(e:MouseEvent):void{
Radio.broadcast(new CustomEvent(id, ..params));
}
in animatedBox:
Radio.addListener(id, new Reciever(uid, animate));
private function animate(e:CustomEvent) {
//needed code and access of CustomEvent props you pass
}
upd:
package lazylib.broadcast
{
/**
* ...
* #author www0z0k
*/
public class Reciever
{
private var id: String;
private var toRun: Function;
/*#param nm - unique listener id - required
*#param fn - event handler function - required*/
public function Reciever(nm:String, fn:Function)
{
id = nm;
toRun = fn;
}
public function onEvent(e:* = null):String {
if (e == null) { return id; }
toRun(e);
return id;
}
public function get ID():String { return id; }
}
}
and
package lazylib.broadcast
{
import flash.events.Event;
import flash.events.EventDispatcher;
/**
* ...
* #author www0z0k
*/
public final class Radio extends EventDispatcher
{
private static var listeners: Object = new Object();
private static var archive: Array = new Array();
private static var forSlowpokes: Object = new Object();
public static function get ForSlowpokes():Object { return forSlowpokes; }
public static function addListener(type: String , listener: Reciever):Boolean {
listeners['anchor'] = null;
if (!listeners[type]) {
var o: Object = new Object();
listeners[type] = o;
}
if (!listeners[type][listener.ID]) {
listeners[type][listener.ID] = listener;
return true;
}else {
return false;
}
}
public static function broadcast(evt: * , singleUse:Boolean = false):void {
var type:String = (evt as Event).type;
if (listeners[type]) {
var returned: Array = new Array();
for (var i: String in listeners[type]) {
if(listeners[type][i]){
var fnRetVal: String = listeners[type][i].onEvent(evt);
returned.push(fnRetVal);
}else{
//trace("no listener for id = " + i + ' , type = ' + type);
}
}
}else {
//trace("nobody's interested in : \"" + type + "\"");
}
if (singleUse) {
forSlowpokes[type] = 'you missed it realtime';
delete listeners[type];
}
}
public static function clearDeadFuncs(namez:Object):void {
for (var a:String in namez) {
if (a != 'anchor') {
killListener(a, namez[a]);
}
}
}
public static function killListener(type: String , id: String):Boolean {
if (!listeners[type]) {
//trace("there are no listeners for event : " + "\"" + type + "\"");
return false;
}else {
if (!listeners[type][id]) {
//trace("there is no \"" + id + "\" listener for event : " + "\"" + type + "\"");
return false;
}else {
listeners[type][id] = null;
//trace("removed listener \"" + id + "\" for event : " + "\"" + type + "\"");
var evt2kill: Number = 0;
for (var str: String in listeners[type]) {
if (listeners[type][str]) {
evt2kill++;
}
}
if (evt2kill == 0) {
delete listeners[type];
//trace("no more listeners for event : " + "\"" + type + "\"");
return true;
}
return true;
}
}
}
}
}
delivered as is ;)
We take for granted things like jQuery that do this easily, but obviously we don't have that luxury in AS3.
well there is this: http://tech.nitoyon.com/blog/2008/01/as3query_alpha.html
I asked myself this question also a lot of times. Still haven't figured out an ultimate solution to the problem. Iterating through parents and parents is definately a way but has to be taken with caution, cause relations might change in your application during runtime. I wrote a simple method a few days ago that lets you iterate through all parents of a given object. Definitely not an elegant solution but it works so far. the SWIZ framework also offers good methods to facilitate the communication between objects via code injection and Event mediation. Maybe worth a look...
I'm trying to unit test some code that calls into VirtualPathUtility.ToAbsolute.
Is this possible with the unit testing tools provided with VS 2008? If not, is it possible with a later version of Visual Studio?
We're well past VS 2008 but for anyone who is still grappling with this issue, I've found a solution on: http://forums.asp.net/t/995143.aspx?Mocking+HTTPContext+object.
Use the following code in your test init to override the default AppDomain values. (The VirutalPathUtility static methods will use your new values.)
[TestInitialize]
public void Initialize()
{
// Fake out env for VirtualPathUtility.ToAbsolute(..)
string path = AppDomain.CurrentDomain.BaseDirectory;
const string virtualDir = "/";
AppDomain.CurrentDomain.SetData(".appDomain", "*");
AppDomain.CurrentDomain.SetData(".appPath", path);
AppDomain.CurrentDomain.SetData(".appVPath", virtualDir);
AppDomain.CurrentDomain.SetData(".hostingVirtualPath", virtualDir);
AppDomain.CurrentDomain.SetData(".hostingInstallDir", HttpRuntime.AspInstallDirectory);
TextWriter tw = new StringWriter();
HttpWorkerRequest wr = new SimpleWorkerRequest("default.aspx", "", tw);
HttpContext.Current = new HttpContext(wr);
}
Static classes and methods are really hard to work with in unit tests (which is one reason why i try to avoid them). In this case, I would probably develop a wrapper around the static class, containing just those methods that I use. I would then use my wrapper class in place of the real class. The wrapper class would be constructed so that it is easy to mock out.
Example (sort of) using RhinoMocks. Note that it uses dependency injection to give the class under test a copy of the wrapper. If the supplied wrapper is null, it creates one.
public class MyClass
{
private VPU_Wrapper VPU { get; set; }
public MyClass() : this(null) {}
public MyClass( VPU_Wrapper vpu )
{
this.VPU = vpu ?? new VPU_Wrapper();
}
public string SomeMethod( string path )
{
return this.VPU.ToAbsolute( path );
}
}
public class VPU_Wrapper
{
public virtual string ToAbsolute( string path )
{
return VirtualPathUtility.ToAbsolute( path );
}
}
[TestMethod]
public void SomeTest()
{
string path = "~/path";
string expected = "/app/path";
var vpu = MockRepository.GenerateMock<VPU_Wrapper>();
vpu.Expect( v => v.ToAbsolute( path) ).Return( expected );
MyClass class = new MyClass( vpu );
string actual = class.SomeMethod( path );
Assert.AreEqual( expected, actual );
vpu.VerifyAllExpectations();
}
Using Microsoft Fakes we can fake VirtualPathUtility ToAbsolute Method easily.
Browse System.Web in References > Right Click > Add Fakes Assembly.
Use Following Code
using Microsoft.QualityTools.Testing.Fakes;
using Microsoft.VisualStudio.TestTools.UnitTesting;
using System.Web.Fakes;
public class TestCode
{
[TestMethod]
public void TestSummaryTabLinks()
{
using (ShimsContext.Create())
{
//Fake VirtualPathUtility ToAbsolute method to Work properly in case of Unit Test Project
//For Finding Relative url.
ShimVirtualPathUtility.ToAbsoluteString = (string s) => { return s; };
MyClass class = new MyClass( vpu );
string actual = class.SomeMethod( path );
Assert.AreEqual( expected, actual );
}
}
}