I want to define a SuggestBox, which behaves like the search bar in Google Maps: When you begin to type, real addresses, starting with the typed letters, appear.
I think, that I need to use the Geocoder.getLocations(String address, LocationCallback callback) method, but I have no idea how to connect this with the oracle, which is needed by the suggest box to produce its suggestions.
Can you please give me ideas how do I connect the getLocations Method with the SuggestOracle?
I solved this by implementing a subclass of SuggestBox, which has it's own SuggestOracle. The AddressOracle deals as a Wrapper for the Google Maps Service, for which the class Geocoderin the Google Maps API for GWT offers abstractions.
So here is my solution:
First we implement the Widget for a SuggestBox with Google Maps suggestions
public class GoogleMapsSuggestBox extends SuggestBox {
public GoogleMapsSuggestBox() {
super(new AddressOracle());
}
}
Then we implement the SuggestOracle, which wraps the Geocoder async method abstractions:
class AddressOracle extends SuggestOracle {
// this instance is needed, to call the getLocations-Service
private final Geocoder geocoder;
public AddressOracle() {
geocoder = new Geocoder();
}
#Override
public void requestSuggestions(final Request request,
final Callback callback) {
// this is the string, the user has typed so far
String addressQuery = request.getQuery();
// look up for suggestions, only if at least 2 letters have been typed
if (addressQuery.length() > 2) {
geocoder.getLocations(addressQuery, new LocationCallback() {
#Override
public void onFailure(int statusCode) {
// do nothing
}
#Override
public void onSuccess(JsArray<Placemark> places) {
// create an oracle response from the places, found by the
// getLocations-Service
Collection<Suggestion> result = new LinkedList<Suggestion>();
for (int i = 0; i < places.length(); i++) {
String address = places.get(i).getAddress();
AddressSuggestion newSuggestion = new AddressSuggestion(
address);
result.add(newSuggestion);
}
Response response = new Response(result);
callback.onSuggestionsReady(request, response);
}
});
} else {
Response response = new Response(
Collections.<Suggestion> emptyList());
callback.onSuggestionsReady(request, response);
}
}
}
And this is a special class for the oracle suggestions, which just represent a String with the delivered address.
class AddressSuggestion implements SuggestOracle.Suggestion, Serializable {
private static final long serialVersionUID = 1L;
String address;
public AddressSuggestion(String address) {
this.address = address;
}
#Override
public String getDisplayString() {
return this.address;
}
#Override
public String getReplacementString() {
return this.address;
}
}
Now you can bind the new widget into your web page by writing the following line in the onModuleLoad()-method of your EntryPoint-class:
RootPanel.get("hm-map").add(new GoogleMapsSuggestBox());
Related
I am creating a small test. In Code behind I have two classes. Pages, LoginPage.
The first part is running. I dont know how to integrate with second part. Currently I am able to open the browser. Also I am trying to use the Page obect model pattern .
Fitnesse code
!|import|
|TestFramework|
!|script|Pages|
|Goto||https://gmail.com|
|LoginPage|CheckRequiredElementsPresent|Pass|
Fixtures
public class Pages
{
string url;
private LoginPage loginPage;
public static void Goto(string url)
{
Browser.Goto(url);
}
}
public class LoginPage
{
static string PageTitle;
[FindsBy(How = How.Id, Using = "TextUsername")]
private static IWebElement username;
[FindsBy(How = How.Id, Using = "TextPassword")]
private static IWebElement password;
[FindsBy(How = How.Id, Using = "_ButtonLogin")]
private static IWebElement submit;
public string IsAtLoginPage()
{
return "";
}
public string CheckRequiredElementsPresent()
{
if (username != null && password != null && submit != null)
{
return "Pass";
}
return "Fail";
}
}
}
You need to do something like below:
Fitnesse Code
!|import|
|TestFramework|
!|script|Pages|
|Goto||https://gmail.com|
|check Required Element|Pass|
You need to call your second class from your Pages class, please see the code changes & fitnesse fixture changes that I've made.
Fixtures
public class Pages
{
string url;
private LoginPage loginPage;
public static void Goto(string url)
{
Browser.Goto(url);
}
// This is what you need to do to refer method of second class.
// This method will be called after Goto method in sequence.
public boolean checkRequiredElement(){
return loginPage.CheckRequiredElementsPresent()
}
}
I'm using (and new to) RequestFactory in GWT 2.5, with JDO entities with a one-to-many relationship, on AppEngine datastore. I've just started using the GWT RequestFactoryEditorDriver to display/edit my objects.
The Driver traverses my objects fine, and displays them correctly. However, when I try to edit a value on the "related" objects, the change doesn't get persisted to the datastore.
When I change b.name on my UI and click "save", I notice only A's persist() call is called. B's persist() is never called. How do I make the editorDriver fire on both ARequest as well as BRequest request contexts? (since what I want is for B's InstanceRequest<AProxy,Void> persist() to be called when my edits are to B objects only.)
Also, AFAICT, if I have an editor on BProxy, any object b that is being shown by the editor (and following the Editor Contract) should automatically be "context.edit(b)"ed by the Driver to make it mutable. However, in my case "context" is an ARequest, not a BRequest.
Do I have to make a ValueAwareEditor like mentioned here: GWT Editor framework
and create a fresh BRequest inside the flush() call and fire it, so that changes to B separately persist in a BRequest before the ARequest is fired?
editorDriver.getPaths() gives me:
"bs"
Also, the driver definitely sees the change to B's property, as editorDriver.isChanged() returns true before I fire() the context.
There are no errors on my client-side or server-side logs, and the Annotation Processor runs with no warnings.
Here's how I setup my driver:
editorDriver = GWT.create(Driver.class);
editorDriver.initialize(rf, view.getAEditor());
final ARequest aRequest = rf.ARequest();
final Request<List<AProxy>> aRequest = aRequest.findAByUser(loginInfo.getUserId());
String[] paths = editorDriver.getPaths();
aRequest.with(paths).fire(new Receiver<List<AProxy>>() {
#Override
public void onSuccess(List<AProxy> response) {
AProxy a = response.get(0);
ARequest aRequest2 = rf.aRequest();
editorDriver.edit(a, aRequest2);
aRequest2.persist().using(a);
}
});
This is how my entities look:
public abstract class PersistentEntity {
public Void persist() {
PersistenceManager pm = getPersistenceManager();
try {
pm.makePersistent(this);
} finally {
pm.close();
}
return null;
}
public Void remove() {
PersistenceManager pm = getPersistenceManager();
try {
pm.deletePersistent(this);
} finally {
pm.close();
}
return null;
}
}
#PersistenceCapable(identityType = IdentityType.APPLICATION)
#Version(strategy=VersionStrategy.VERSION_NUMBER, column="VERSION",
extensions={#Extension(vendorName="datanucleus", key="field-name", value="version")})
public class A extends PersistentEntity {
... (Id, version omitted for brevity)
#Persistent
private String name;
#Persistent
private List<B> bs;
public String getName() {
return name;
}
...
public void setName(String name) {
this.name = name;
}
public List<B> getBs() {
return bs;
}
public void setBs(List<B> bs) {
this.bs = bs;
}
}
... (same annotations as above omitted for brevity)
public class B extends PersistentEntity {
... (Id, version omitted for brevity)
#Persistent
private String name;
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
}
Here are the proxies:
#ProxyFor(A.class)
public interface AProxy extends EntityProxy {
String getName();
List<BProxy> getBs();
void setName(String name);
void setBs(List<BProxy> bs);
}
#ProxyFor(B.class)
public interface BProxy extends EntityProxy {
String getName();
void setName(String name);
}
Here are my service stubs:
#Service(A.class)
public interface ARequest extends RequestContext {
Request<List<A>> findAByUser(String userId);
InstanceRequest<AProxy, Void> persist();
InstanceRequest<AProxy, Void> remove();
}
#Service(B.class)
public interface BRequest extends RequestContext {
Request<List<A>> findB(String key);
InstanceRequest<BProxy, Void> persist();
InstanceRequest<BProxy, Void> remove();
}
Edit:
I've now changed my ARequest interface and service implementation to support a "saveAndReturn" method, so that I can recursively "persist" "a" on the server side:
Request<UserSandboxProxy> saveAndReturn(AProxy aProxy);
I find now that when I "flush" my RequestFactoryEditorDriver, the client-side context object has my new "b.name" value. However, if I call "context.fire()" and inspect my "saveAndReturn" method on the server side, the resulting server-side object "a", just before I "persist" it, doesn't contain the change to "b.name" on any item of the List.
Why could this be happening? How do I debug why this client-information doesn't go across the wire, to the server?
Options I've considered, tried and ruled out:
1) Ensuring the APT has been run, and there are no warnings/errors on Proxy or Service interfaces
2) Ensuring that my proxies does have a valid setter in AProxy for the List
You have to use a session-per-request pattern for RequestFactory to work properly. More details here: https://code.google.com/p/google-web-toolkit/issues/detail?id=7827
I'm trying to consume a RESTful JSON web service using WCF on the client side. The service is 3rd party, so I cannot make any changes to the server response.
The server is sending back a response that looks something like this when there's only one data point...
Single Data Point
{
"Data":
{
"MyPropertyA":"Value1",
"MyPropertyB":"Value2"
},
}
and something like this when there's more than one data point...
Multiple Data Points
{
"Data":
[
{
"MyPropertyA":"Value1",
"MyPropertyB":"Value2"
},
{
"MyPropertyA":"Value3",
"MyPropertyB":"Value4"
},
{
"MyPropertyA":"Value5",
"MyPropertyB":"Value6"
}
],
}
I have my service contract set up like this...
[ServiceContract]
public interface IRewardStreamService
{
[OperationContract]
[WebInvoke]
MyResponse GetMyStuff();
}
and a data point's data contract like this...
[DataContract]
public class MyData
{
[DataMember]
public string MyPropertyA { get; set; }
[DataMember]
public string MyPropertyB { get; set; }
}
and the only way I can get the single data point response to work is if I have a single instance property like this, but this does not parse the multiple data point response...
Response for Single Instance
[DataContract]
public class MyResponse
{
[DataMember]
public MyData Data { get; set; }
}
and the only way I can get the multiple data point response to work is if I have an array / list instance property like this, but this does not parse the single data point response...
Response for Multiple Instance
[DataContract]
public class MyResponse
{
[DataMember]
public IList<MyData> Data { get; set; }
}
I understand the issue is that the response is omitting the brackets when there's only one data point returned, but it seems that WCF doesn't play well with deserializing that syntax. Is there some way I can tell the DataContractJsonSerializer to allow single element arrays to not include brackets and then tell my service to use that serializer? Maybe a service behavior or something?
Any direction would be helpful.
You can use a custom message formatter to change the deserialization of the JSON into the data contract you want. In the code below, the data contract is defined to have a List<MyData>; if the response contains only one data point, it will "wrap" that into an array prior to passing to the deserializer, so it will work for all cases.
Notice that I used the JSON.NET library to do the JSON modification, but that's not a requirement (it just has a nice JSON DOM to work with the JSON document).
public class StackOverflow_12825062
{
[ServiceContract]
public class Service
{
[WebGet]
public Stream GetData(bool singleDataPoint)
{
string result;
if (singleDataPoint)
{
result = #"{
""Data"":
{
""MyPropertyA"":""Value1"",
""MyPropertyB"":""Value2""
},
}";
}
else
{
result = #"{
""Data"":
[
{
""MyPropertyA"":""Value1"",
""MyPropertyB"":""Value2""
},
{
""MyPropertyA"":""Value3"",
""MyPropertyB"":""Value4""
},
{
""MyPropertyA"":""Value5"",
""MyPropertyB"":""Value6""
}
],
} ";
}
WebOperationContext.Current.OutgoingResponse.ContentType = "application/json";
return new MemoryStream(Encoding.UTF8.GetBytes(result));
}
}
[DataContract]
public class MyData
{
[DataMember]
public string MyPropertyA { get; set; }
[DataMember]
public string MyPropertyB { get; set; }
}
[DataContract]
public class MyResponse
{
[DataMember]
public List<MyData> Data { get; set; }
public override string ToString()
{
return string.Format("MyResponse, Data.Length={0}", Data.Count);
}
}
[ServiceContract]
public interface ITest
{
[WebGet]
MyResponse GetData(bool singleDataPoint);
}
public class MyResponseSingleOrMultipleClientReplyFormatter : IClientMessageFormatter
{
IClientMessageFormatter original;
public MyResponseSingleOrMultipleClientReplyFormatter(IClientMessageFormatter original)
{
this.original = original;
}
public object DeserializeReply(Message message, object[] parameters)
{
WebBodyFormatMessageProperty messageFormat = (WebBodyFormatMessageProperty)message.Properties[WebBodyFormatMessageProperty.Name];
if (messageFormat.Format == WebContentFormat.Json)
{
MemoryStream ms = new MemoryStream();
XmlDictionaryWriter jsonWriter = JsonReaderWriterFactory.CreateJsonWriter(ms);
message.WriteMessage(jsonWriter);
jsonWriter.Flush();
string json = Encoding.UTF8.GetString(ms.ToArray());
JObject root = JObject.Parse(json);
JToken data = root["Data"];
if (data != null)
{
if (data.Type == JTokenType.Object)
{
// single case, let's wrap it in an array
root["Data"] = new JArray(data);
}
}
// Now we need to recreate the message
ms = new MemoryStream(Encoding.UTF8.GetBytes(root.ToString(Newtonsoft.Json.Formatting.None)));
XmlDictionaryReader jsonReader = JsonReaderWriterFactory.CreateJsonReader(ms, XmlDictionaryReaderQuotas.Max);
Message newMessage = Message.CreateMessage(MessageVersion.None, null, jsonReader);
newMessage.Headers.CopyHeadersFrom(message);
newMessage.Properties.CopyProperties(message.Properties);
message = newMessage;
}
return this.original.DeserializeReply(message, parameters);
}
public Message SerializeRequest(MessageVersion messageVersion, object[] parameters)
{
throw new NotSupportedException("This formatter only supports deserializing reply messages");
}
}
public class MyWebHttpBehavior : WebHttpBehavior
{
protected override IClientMessageFormatter GetReplyClientFormatter(OperationDescription operationDescription, ServiceEndpoint endpoint)
{
IClientMessageFormatter result = base.GetReplyClientFormatter(operationDescription, endpoint);
if (operationDescription.Messages[1].Body.ReturnValue.Type == typeof(MyResponse))
{
return new MyResponseSingleOrMultipleClientReplyFormatter(result);
}
else
{
return result;
}
}
}
public static void Test()
{
string baseAddress = "http://" + Environment.MachineName + ":8000/Service";
WebServiceHost host = new WebServiceHost(typeof(Service), new Uri(baseAddress));
host.Open();
Console.WriteLine("Host opened");
ChannelFactory<ITest> factory = new ChannelFactory<ITest>(new WebHttpBinding(), new EndpointAddress(baseAddress));
factory.Endpoint.Behaviors.Add(new MyWebHttpBehavior());
ITest proxy = factory.CreateChannel();
Console.WriteLine(proxy.GetData(false));
Console.WriteLine(proxy.GetData(true));
Console.Write("Press ENTER to close the host");
((IClientChannel)proxy).Close();
factory.Close();
Console.ReadLine();
host.Close();
}
}
I don't know about using WCF so I'll change to Asp.Net WCF. Here is an article that will get you one the way
http://www.west-wind.com/weblog/posts/2012/Aug/30/Using-JSONNET-for-dynamic-JSON-parsing
I just can't figure out how to determine if it's an array or a single object. Here is a little code.
[TestMethod]
public void SingleObject()
{
using (var client = new HttpClient())
{
var result = client.GetStringAsync("http://localhost:8080/api/JSONTestOne");
string content = result.Result;
JObject jsonVal = JObject.Parse(content);
dynamic aFooObj = jsonVal;
Console.WriteLine(aFooObj.afoo.A);
}
}
[TestMethod]
public void ArrayWithObject()
{
using (var client = new HttpClient())
{
var result = client.GetStringAsync("http://localhost:8080/api/JSONTest");
string content = result.Result;
JObject jsonVal = JObject.Parse(content);
dynamic foos = jsonVal;
Console.WriteLine(foos[0].A);
}
}
im trying to build a google app engine projekt with JPA, JAX-RS and JAX-B. My POST and GET Methods work, but my DELETE method doesn't delete the data.
Resource
#DELETE
#Path("card/{id}")
public void deleteCardById (#PathParam ("id") Long id) {
Service.removeCard(id);
}
Service
public static void removeCard(Long id) {
EntityManager em = EMFService.get().createEntityManager();
Card emp = findCard(id);
if (emp != null) {
em.remove(emp);
}
em.close();
}
public static Card findCard(Long id) {
EntityManager em = EMFService.get().createEntityManager();
Card card = em.find(Card.class, id);
em.close();
return card;
}
Entity
#XmlRootElement
#Entity
public class Card {
#Id
#GeneratedValue(strategy = GenerationType.IDENTITY)
Long id;
String begriff;
String tabu1;
String tabu2;
String tabu3;
public Card(String begriff, String tabu1, String tabu2, String tabu3) {
super();
Begriff = begriff;
Tabu1 = tabu1;
Tabu2 = tabu2;
Tabu3 = tabu3;
}
public Card() {
}
public Long getId() {
return id;
}
public void setId(Long id) {
this.id = id;
}
public String getBegriff() {
return Begriff;
}
public void setBegriff(String begriff) {
Begriff = begriff;
}
public String getTabu1() {
return Tabu1;
}
public void setTabu1(String tabu1) {
Tabu1 = tabu1;
}
public String getTabu2() {
return Tabu2;
}
public void setTabu2(String tabu2) {
Tabu2 = tabu2;
}
public String getTabu3() {
return Tabu3;
}
public void setTabu3(String tabu3) {
Tabu3 = tabu3;
}
#Override
public String toString() {
return "Card [Begriff=" + Begriff + ", Tabu1=" + Tabu1 + ", Tabu2="
+ Tabu2 + ", Tabu3=" + Tabu3 + "]";
}
When i Debug the app it gives the correct Object to the remove function. But it just don't remove the data ...
You mean you're using v1 of the GAE JPA plugin, and you don't bother putting a transaction around your remove (so the remove is delayed until the next transaction ... which never happens)?
Obviously you could either put a transaction around the remove, or better still you use v2 of the GAE JPA plugin
I was facing similar issue too. the JPA delete actually deletes the entity in the datastore,but it doesn't delete the entity from the JPA Cache.. You page is actually using the JPA Cached result list to display..
The way I used to resolve the issue is to have the JPA Cache cleared every time after a delete.
Sample Code would be something like this:
EM.getTransaction().begin();
EM.remove(current_record);
EM.getTransaction().commit();
EM.getEntityManagerFactory().getCache().evictAll();
ok i think i should write it like this
*edit the problem was the findCard function, i think because of the secone instance of the EntityManager. I chnaged it without using this method to this and now it works.
public static void removeCard(Long id) {
EntityManager em = EMFService.get().createEntityManager();
EntityTransaction tx = em.getTransaction();
try {
tx.begin();
Card card = em.find(Card.class, id);
if (card != null) {
em.remove(card);
}
tx.commit();
} finally {
if (tx.isActive()) {
tx.rollback();
}
em.close();
}
}
I've seen this kind of thing described in various examples showing how to create a REST service which takes arrays or a list of objects as part of the URL.
My question is, how to implement this using RESTeasy?
Something like the following would be how i would assume this to work.
#GET
#Path("/stuff/")
#Produces("application/json")
public StuffResponse getStuffByThings(
#QueryParam("things") List<Thing> things);
Create a StringConverter and a use a wrapper object. Here is a quick and dirty example:
public class QueryParamAsListTest {
public static class Thing {
String value;
Thing(String value){ this.value = value; }
}
public static class ManyThings {
List<Thing> things = new ArrayList<Thing>();
ManyThings(String values){
for(String value : values.split(",")){
things.add(new Thing(value));
}
}
}
static class Converter implements StringConverter<ManyThings> {
public ManyThings fromString(String str) {
return new ManyThings(str);
}
public String toString(ManyThings value) {
//TODO: implement
return value.toString();
}
}
#Path("/")
public static class Service {
#GET
#Path("/stuff/")
public int getStuffByThings(
#QueryParam("things") ManyThings things){
return things.things.size();
}
}
#Test
public void test() throws Exception {
Dispatcher dispatcher = MockDispatcherFactory.createDispatcher();
dispatcher.getProviderFactory().addStringConverter(new Converter());
dispatcher.getRegistry().addSingletonResource(new Service());
MockHttpRequest request = MockHttpRequest.get("/stuff?things=a,b,c");
MockHttpResponse response = new MockHttpResponse();
dispatcher.invoke(request, response);
Assert.assertEquals("3", response.getContentAsString());
}
}
I think you can also use a StringParamUnmarshaller
I had some luck with this, using Collection rather than List. I was unable to make a StringConverter for List work.
#Provider
public class CollectionConverter implements StringConverter<Collection<String>> {
public Collection<String> fromString(String string) {
if (string == null) {
return Collections.emptyList();
}
return Arrays.asList(string.split(","));
}
public String toString(Collection<String> values) {
final StringBuilder sb = new StringBuilder();
boolean first = true;
for (String value : values) {
if (first) {
first = false;
} else {
sb.append(",");
}
sb.append(value);
}
return sb.toString();
}
}
I did the toString from my head. Be sure to write unit tests for it to verify. But of course, everything is easier and clearer when you use Guava. Can use Joiner and Splitter. Really handy.
Just use a wrapper on its own, no need for anything else.
In your endpoint
#Produces({MediaType.APPLICATION_JSON, MediaType.APPLICATION_XML})
#Path("/find")
#GET
MyResponse find(#QueryParam("ids") Wrapper ids);
And you wrapper looks like this :
public class Wrapper implements Serializable {
private List<BigInteger> ids = Collections.emptyList();
public String toString() {
return Joiner.on(",")
.join(ids);
}
public List<BigInteger> get() {
return ids;
}
public Wrapper(String s) {
if (s == null) {
ids = Collections.emptyList();
}
Iterable<String> splitted = Splitter.on(',')
.split(s);
Iterable<BigInteger> ids = Iterables.transform(splitted, Functionz.stringToBigInteger);
this.ids = Lists.newArrayList(ids);
}
public Wrapper(List<BigInteger> ids) {
this.ids = ids;
}
}