First I want to say I saw a site:
https://cwiki.apache.org/WICKET/autocomplete-using-a-wicket-model.html
Still I have some problem with implementation.
Context
I have form where want to edit Order from my database.
I want to autocomplete client's name in form and save it to database.
Problem
I can see suggestions of client names and I can choose which client's name I'll use.
Just when I submit form new client won't be saved in order, in database.
I don't know how to relate form Model with AutoCompleteTextField input.
Any ideas ?
Pseudo code of my classes:
Order{
Long id;
String date;
Client client;
Status status;
...
}
Client{
Long id;
String name;
String nip;
String address;
String postcode;
String city;
String phone;
String mail;
...
}
Status{
Long id;
String name;
String value;
}
Edited:
Yes you are right.
My implementation of AbstractAutoCompleteTextField from site:
AbstractAutoCompleteRenderer autoCompleteRenderer = new AbstractAutoCompleteRenderer() {
protected final String getTextValue(final Object object) {
Client client = (Client) object;
return client.getName();
}
protected final void renderChoice(final Object object, final Response response, final String criteria) {
response.write(getTextValue(object));
}
};
// textfield
AbstractAutoCompleteTextField<Client> name = new AbstractAutoCompleteTextField<Client>("name", new PropertyModel(order, "client"), autoCompleteRenderer) {
protected final List<Client> getChoiceList(final String input) {
return clientService.findByNames(10, 0, input);
}
protected final String getChoiceValue(final Client choice) throws Throwable {
return choice.getId().toString();
}
};
form.add(name);
My form implementation you asked for:
form = new Form("orderForm", new CompoundPropertyModel(order)) {
#Override
public void onSubmit() {
orderService.update((Order) getDefaultModelObject());
setResponsePage(OrdersPage.class);
// Form validation successful. Display message showing edited
// model.
}
};
With this code I got: "'Hurtownia Techniczna "ADA"' is not proper Client." (translated from Polish) in feedback panel.
I think, that's where things go wrong:
AbstractAutoCompleteTextField<Client> name = new AbstractAutoCompleteTextField<Client>("name", new PropertyModel(order, "client"), autoCompleteRenderer)
Rest of the answer edited to reflect the correct use case
To be specific: You're creating a PropertyModel of your orders client value, which is a Client-Object and tie it to a TextField. To create a Client-object from the TextField, wicket needs a converter. There are lots of build-in converters but none of them is capable of converting to a custom object. Thus you'll need to provide the converter by implementing IConverter. Since I don't know how you store and retrieve your Client-objects I can only show you a generic example, a Locale-Converter used in one of my projects:
public class LocaleConverter implements IConverter {
private static final long serialVersionUID = 3251433094703013493L;
/* (non-Javadoc)
* #see org.apache.wicket.util.convert.IConverter#convertToObject(java.lang.String, java.util.Locale)
*/
#Override
public Object convertToObject(String value, Locale locale) {
Locale retValue = null;
try {
retValue = LocaleUtils.toLocale(value);
} catch (IllegalArgumentException e) {
throw (new ConversionException("" + value + " is not a valid locale.", e));
}
return retValue;
}
/* (non-Javadoc)
* #see org.apache.wicket.util.convert.IConverter#convertToString(java.lang.Object, java.util.Locale)
*/
#Override
public String convertToString(Object value, Locale locale) {
return value.toString();
}
}
Then you'll need to register your new converter to your application. Again, this will be slightly different for you since you've got a different converter...
#Override
protected IConverterLocator newConverterLocator() {
ConverterLocator newConverterLocator = (ConverterLocator) super.newConverterLocator();
newConverterLocator.set(Locale.class, new LocaleConverter());
return newConverterLocator;
}
This method needs to be included in your application class.
Related
The documentation for graphql-java-annotations doesn't do such a great job of telling me how to add a Custom Scalar to my schema: https://github.com/Enigmatis/graphql-java-annotations/tree/v8.0.1#annotations-schema-creator
What I need is to create some 'scalar Date' in the Schema. It is unclear how to do this with the AnnotationsSchemaCreator builder thing.
GraphQLSchema schema = AnnotationsSchemaCreator.newAnnotationsSchema()
.query(Query.class) // to create you query object
.mutation(Mutation.class) // to create your mutation object
.subscription(Subscription.class) // to create your subscription object
.directive(UpperDirective.class) // to create a directive
.additionalType(AdditionalType.class) // to create some additional type and add it to the schema
.typeFunction(CustomType.class) // to add a typefunction
.setAlwaysPrettify(true) // to set the global prettifier of field names (removes get/set/is prefixes from names)
.setRelay(customRelay) // to add a custom relay object
.build();
The docs give me just that. Is a typeFunction what I need here? Do I have to first get the graphql-java "Custom Scalar" stuff set up and put that into the typeFunction?
What's happening right now is that my graphql-java-annotations Types which need the Date type...
public abstract class BasePart {
#GraphQLField
#GraphQLNonNull
#JsonIgnore
public String id;
...
#GraphQLField
public Date createdOn;
...
}
Get into the Schema without the Date scalar defined so the GraphiQL UI is rejecting it with errors like...
Error: Date fields must be an object with field names as keys or a function which returns such an object.
at invariant (http://localhost.blueorigin.com:8080/webjars/graphiql/0.10.1/graphiql.min.js:13:12678)
at defineFieldMap (http://localhost.blueorigin.com:8080/webjars/graphiql/0.10.1/graphiql.min.js:14:16395)
at e.getFields (http://localhost.blueorigin.com:8080/webjars/graphiql/0.10.1/graphiql.min.js:14:22028)
at http://localhost.blueorigin.com:8080/webjars/graphiql/0.10.1/graphiql.min.js:15:22055
at typeMapReducer (http://localhost.blueorigin.com:8080/webjars/graphiql/0.10.1/graphiql.min.js:15:22227)
at http://localhost.blueorigin.com:8080/webjars/graphiql/0.10.1/graphiql.min.js:15:22200
at Array.forEach (<anonymous>)
at http://localhost.blueorigin.com:8080/webjars/graphiql/0.10.1/graphiql.min.js:15:22082
at typeMapReducer (http://localhost.blueorigin.com:8080/webjars/graphiql/0.10.1/graphiql.min.js:15:22227)
at typeMapReducer (http://localhost.blueorigin.com:8080/webjars/graphiql/0.10.1/graphiql.min.js:15:21564)
I'm trying to figure out how to get that information into the, what, AnnotationsSchemaCreator.newAnnotationsSchema() builder?
How do you add a Custom Scalar to a graphql-java-annotations project?
The TypeFunction is the key. You pass the TypeFunction when you are building the Schema with the AnnotationsSchemaCreator. The following code effectively got scalar Date into the service's GraphQL Schema
graphQLSchema = AnnotationsSchemaCreator.newAnnotationsSchema()
.query(QuerySchema.class)
.setAlwaysPrettify(true)
.typeFunction(new MyDateTypeFunction()) // <-- This got scalar Date onto the schema
.build();
The TypeFunction itself realizes the support for the scalar Date.
public class MyDateTypeFunction implements TypeFunction {
#Override
public boolean canBuildType(Class<?> clazz, AnnotatedType annotatedType) {
return clazz == java.util.Date.class;
}
#Override
public GraphQLType buildType(
boolean b,
Class<?> clazz,
AnnotatedType annotatedType,
ProcessingElementsContainer processingElementsContainer) {
return MY_DATE;
}
public static final GraphQLScalarType MY_DATE = GraphQLScalarType
.newScalar()
.name("Date")
.description("Coerce java.util.Date to/from a String representation of the long value of getTime().")
.coercing(
new Coercing() {
#Override
public Object serialize(Object dataFetcherResult) throws CoercingSerializeException {
if (dataFetcherResult instanceof Date) {
final String result = String.format("%d", ((Date) dataFetcherResult).getTime());
return result;
}
final String message =
String.format("Expected type java.util.Date but found %s", typeName(dataFetcherResult));
throw new CoercingSerializeException(message);
}
#Override
public Object parseValue(Object input) throws CoercingParseValueException {
if (input instanceof String) {
try {
return stringToDate((String) input);
} catch (NumberFormatException nfe) {
final String message = String.format("NumberFormatException %s", nfe.getMessage());
throw new CoercingParseValueException(message);
}
}
final String message = String.format("Unable to parseValue %s to a java.util.Date", input);
throw new CoercingParseValueException(message);
}
#Override
public Object parseLiteral(Object input) throws CoercingParseLiteralException {
if (input instanceof StringValue) {
try {
final String inputStringValue = ((StringValue) input).getValue();
return stringToDate(inputStringValue);
} catch (NumberFormatException nfe) {
final String message = String.format("NumberFormatException %s", nfe.getMessage());
throw new CoercingParseLiteralException(message);
}
}
final String message = String.format("Unable to parseLiteral %s to a java.util.Date", input);
throw new CoercingParseLiteralException(message);
}
}
)
.build();
public static Date stringToDate(String input) throws NumberFormatException {
final long inputAsLong = Long.parseLong(input);
return new Date(inputAsLong);
}
public static String typeName(Object input) {
return input == null ? "null" : input.getClass().getName();
}
}
Note that I'm not recommending that you represent java.util.Date as the String value of the long getTime(), java.time.Instant's ISO-8601 is so much more readable, but my service needed this string value and this is how I got it into a graphql-java-annotation's project schema.
I have spring application which expose REST endpoint, lets name it "doAction". As the request it consumes object:
class Person{
private String name;
private String email;
}
Some clients can call this endpoint by passing data with different practice of writing words, like:
Peter_1
name = Peter
email = peter#gmail.com (lower case)
Mark_2
name = mark
email = MARK#gmail.com (upper case)
Julia_3
name = julia
email = JuliaToward#gmail.com (camel case)
Is there some approach to force all income data be parsed to lowercase(lets assume all fields are Strings)?
So as a result I desire to have:
Peter_1
name = peter
email = peter#gmail.com
Mark_2
name = mark
email = mark#gmail.com
Julia_3
name = julia
email = juliatoward#gmail.com
Solution for Jackson is appreciated.
Short answer Call toLower in the setter
Here is an example:
class Animal
{
private String name;
public void setName(final String newValue)
{
StringUtils.trimToNull(StringUtils.lowerCase(newValue));
}
}
I also recommend either trimToNUll or trimToEmpty.
If you are using Spring Data Rest with spring mvc and you want all incoming string data to be in lower case then define following
public class StringSerializer extends StdDeserializer<String>{
public StringSerializer() {
this(null);
}
public StringSerializer(Class<String> vc) {
super(vc);
}
#Override
public String deserialize(JsonParser p, DeserializationContext ctxt) throws IOException, JsonProcessingException {
JsonToken t = p.getCurrentToken();
if (t==JsonToken.VALUE_STRING){
String receivedValue = p.getText();
if (receivedValue == null)
return null;
else
return receivedValue.toLowerCase();
}else{
return null;
}
}
}
And following:
#Configuration
public class RestDataConfig extends RepositoryRestMvcConfiguration {
#Override
#Bean
public ObjectMapper halObjectMapper() {
ObjectMapper mapper = super.halObjectMapper();
SimpleModule module = new SimpleModule();
module.addDeserializer(String.class, new StringSerializer());
mapper.registerModule(module);
return mapper;
}
}
Basically I have a following form class.
public static class BasicForm {
#Required
public String name;
#Required
#Email
public String email;
#Required
public String password;
#Required
public String confirmPassword;
public List<ValidationError> validate() {
if (User.findByEmail(email) != null) {
List<ValidationError> validationErrorList = new ArrayList<ValidationError>();
validationErrorList.add(new ValidationError("email", "error.alreadyexists", new ArrayList<Object>()));
return validationErrorList;
}
return null;
}
}
As you can see, I am trying to validate the uniqueness of the email address.
If email is not unique, I would like to display the error message on the email field, NOT as a global error message
What is the correct way of implementing validate() method to achieve this?
You should use a validate method with the following signature:
public Map<String, List<ValidationError>> validate()
This way you can add errors to single fields like this:
Map<String, List<ValidationError>> errors = null;
if (emailIsBad) {
errors = new HashMap<String, List<ValidationError>>();
List<ValidationError> list = new ArrayList<ValidationError>();
list.add(new ValidationError("email", "email is bad"));
errors.put("email", list);
}
return errors;
please note that if you should return an empty map, it still will render the form as erroneous. If you want the validate() method to succeed, you need to return null.
Not enough rep to comment (yet), but don't forget to include the proper import statement:
import.play.data.validation.*;
I wasted a few minutes due to an incorrect statement.
This is what I came up with based on Daniel's
code:
My code checks for password confirmation, tho.
public Map<String, List<ValidationError>> validate() {
Map<String, List<ValidationError>> errors = null;
if (!password.equals(confirmPassword)) {
errors = new HashMap<>();
List<ValidationError> list = new ArrayList<>();
list.add(new ValidationError("password", "Passwords do not match"));
errors.put("password",list);
errors.put("confirmPassword", list);
}
return errors;
}
Having my ValueObject
UserVO {
long id;
String username;
}
I created custom editor for parsing this object from string id#username
public class UserVOEditor extends PropertyEditorSupport {
#Override
public void setAsText(String text) throws IllegalArgumentException {
Preconditions.checkArgument(text != null,"Null argument supplied when parsing UserVO");
String[] txtArray = text.split("\\#");
Preconditions.checkArgument(txtArray.length == 2, "Error parsing UserVO. Expected: id#username");
long parsedId = Long.valueOf(txtArray[0]);
String username = txtArray[1];
UserVO uvo = new UserVO();
uvo.setUsername(username);
uvo.setId(parsedId);
this.setValue(uvo);
}
#Override
public String getAsText() {
UserVO uvo = (UserVO) getValue();
return uvo.getId()+'#'+uvo.getUsername();
}
in my controller i register
#InitBinder
public void initBinder(ServletRequestDataBinder binder) {
binder.registerCustomEditor(UserVO.class, new UserVOEditor());
}
having in my model object ModelVO
ModelVO {
Set<UserVO> users = new HashSet<UserVO>();
}
after custom editor is invoked all you can see after form submission is
ModelVO {
Set<String> users (linkedHashSet)
}
so when trying to iterate
for(UserVO uvo : myModel.getUser()){ .. }
Im having classCastException .. cannot cast 1234#username (String) to UserVO ..
HOW THIS MAGIC IS POSSIBLE ?
It is not magic, it is because of Generics will be only proved at compile time. So you can put every thing in a Set at runtime, no one will check if you put the correct type in the Set.
What you can try, to make spring a bit more clever, is to put the ModelVO in your command object.
<form:form action="whatEver" method="GET" modelAttribute="modelVO">
#RequestMapping(method = RequestMethod.GET)
public ModelAndView whatEver(#Valid ModelVO modelVO){
...
}
How do you handle the case where you want user input from a form to be htmlEscape'd when
you are binding to a command object?
I want this to sanitize input data automatically in order to avoid running through all fields in command object.
thanks.
If you are using a FormController you can register a new property editor by overriding the initBinder(HttpServletReques, ServletRequestDataBinder) method. This property editor can escape the html, javascript and sql injection.
If you are using a property editor the values from the request object will be processed by the editor before assigning to the command object.
When we register a editor we have to specify the type of the item whose values has to be processed by the editor.
Sorry, now I don't the syntax of the method. But I'm sure this is how we have achieved this.
EDITED
I think the following syntax can work
In your controller override the following method as shown
#Override
protected void initBinder(HttpServletRequest request,
ServletRequestDataBinder binder) throws Exception {
super.initBinder(request, binder);
binder.registerCustomEditor(String.class,
new StringEscapeEditor(true, true, false));
}
Then create the following property editor
public class StringEscapeEditor extends PropertyEditorSupport {
private boolean escapeHTML;
private boolean escapeJavaScript;
private boolean escapeSQL;
public StringEscapeEditor() {
super();
}
public StringEscapeEditor(boolean escapeHTML, boolean escapeJavaScript,
boolean escapeSQL) {
super();
this.escapeHTML = escapeHTML;
this.escapeJavaScript = escapeJavaScript;
this.escapeSQL = escapeSQL;
}
public void setAsText(String text) {
if (text == null) {
setValue(null);
} else {
String value = text;
if (escapeHTML) {
value = StringEscapeUtils.escapeHtml(value);
}
if (escapeJavaScript) {
value = StringEscapeUtils.escapeJavaScript(value);
}
if (escapeSQL) {
value = StringEscapeUtils.escapeSql(value);
}
setValue(value);
}
}
public String getAsText() {
Object value = getValue();
return (value != null ? value.toString() : "");
}
}
Hopes this helps you
You can use #Valid and #SafeHtml from hibernate validator. See details at https://stackoverflow.com/a/40644276/548473