When sorting on a column in the GridView bound to an Enum, it sorts by the order of the Enum. I need it to sort by the string representation of the Enum. Here are the options I have seen, of which I like none of them.
Reorder the Enum in alphabetical order - Bad because now the presentation is relying on the Business and Data Access Layer to "pre-sort" the data.
Create a new object (datatable, new list, whatever) with myEnum.ToString() and bind this to the GridView - This one is not bad, but I would rather not.
In my search, check to see if the column sorted is an Enum, then sort by the string representation of the column - Do I have to say why this is bad?
Number 2 would be my favorite so far, but like I said, I don't like it.
More info just in case - I am binding a List of IWhatever to the grid, and 2 columns are enums that need to be sorted by strings. There are also guid-type, string, and decimal columns in the grid that need to be sorted.
Try to use ViewModels. Basically you create ViewModel Objects that contains your Model Object (your IWhatever). This ViewModel exposes then new Properties and Methods which are used in your View. In Your case you would expose a property with the string representation of your Enum. The advantage is, that you could do any transformation logic you want.
See MVVM Pattern. http://en.wikipedia.org/wiki/Model_View_ViewModel
EDIT: Little Example:
public class WhateverViewModel
{
public WhateverViewModel(IWhatever model)
{
this.model = model;
}
...
public string MyEnumView
{
get
{
return model.MyEnum.ToString();
}
}
public string MyEnumView2
{
get
{
switch(model.MyEnum)
{
case MyEnumType.A: return "Hello";
case MyEnumType.B: return "World";
}
}
}
}
Related
I'm using Entity Framework (DB First) on a new project and wanted to add some customisation to the classes generated. However, my changes are obviously lost every time that the edmx is refreshed. I was just wondering if there is a design pattern for handling this sort of thing?
As an example, suppose I have a class with a integer property; StatusID - and I'd like to extend the entity class so that the status value can also be accessed/set via the related enum and finally a property that gets a text representation of that Enum from the description attribute. This all works, but those customisations are lost when the model is refreshed. I appreciate that the property can be converted to an enum, so the latter property that gets the description of the enum is perhaps a better example for this question.
I think I know the answer but I just wanted to put this out there in case there were some magic tricks that would allow this to work and prevent those customisations from being lost.
public int StatusID { get; set; }
public Enumerations.ValidationStatus StatusEnum
{
get
{
return (Enumerations.ValidationStatus)StatusID;
}
set
{
StatusID = (int)value;
}
}
public string StatusText
{
get
{
return MyMethodThatGetsTheEnumDescription(StatusEnum);
}
}
Two Solutions to work around the problem:
User Data Transfer Object(DTO) nd put the enum there. then use Automapper or manually map between the DB Model and the DTO Model (best practice)
Instead of enum you can use extension functions on the model and define your getter, setters and any extra properties you want as extension functions to the class
(will add some complexity to your models)
I have a simple JavaFX Table with a column like this:
#FXML
private TableColumn<PropertyModel, String> columnPropertyProdValue;
Within the initialize method I used this
columnPropertyProdValue.setCellFactory(TextFieldTableCell.forTableColumn());
in order to get an textarea at a double click event on this column. This is working fine.
The column should store different property values but the datatype depends on a different column in the table. The first column "datatype" defines the datatype like boolean, string or integer and the property column should store its value. But in the moment it is always a String.
Its fine for me if I store the property value as a String in the database but the application should check for correctness of the datatype at runtime.
Does anyone has a good idea how to realize that?
Thanks a lot
Hauke
For TextInputControl and subclasses you can apply TextFormatter. This class allows you to control input. For example (for float number):
setTextFormatter(new TextFormatter<String>(
s -> {
if (s.getControlNewText().isEmpty())
return s;
try{
Float.parseFloat(s.getControlNewText());
return s;
} catch (NumberFormatException e) {
return null;
}
}
I have a plain textfield in Tridion that can have multiple values. The itemtype is a SingleLineTextField.
In the TBB code I have the following (removed the non-essential parts):
ItemFields itemFields = new ItemFields(folder.Metadata, folder.MetadataSchema);
foreach (ItemField itemField in itemFields)
{
string itemFieldValue = string.Empty;
switch (Utilities.GetFieldType(itemField))
{
case FieldType.SingleLineTextField:
itemFieldValue = itemField.ToString();
break;
}
}
Now the result in case of two entries is just two strings with a character line break in it.
String A
String B
The method used is a generic one, which also works on other fields, so I was looking for some way to find out if a SingleLineTextField has more values in it.
You can cast the field to a SingleLineTextField type, then iterate through the Values collection, something along these lines:
SingleLineTextField field = (SingleLineTextField)itemField;
foreach(string value in field.Values)
{
// do something with value
}
// or if all you want is the count of values
int i = field.Values.Count;
Firstly, I would advise against relying on the ToString() method on objects unless it is specifically documented. In this case it works with the abstract class ItemField, but this may not always be the case.
The TOM.Net API only defines Definition and Name properties for ItemField, so you need to cast your ItemField object to something more specific.
the TextField abstract class, which SingleLineTextField inherits from, defines a ToString() method, but also Value and Values properties, which are much better suited to what you're trying to do. Looking at the documentation, we can see that Values will give us an IList<String> of the values, even if your field is not multi-valued. Perfect!
So, to answer your question, "I was looking for some way to find out if a SingleLineTextField has more values in it", you need to cast your ItemField as a TextField and check the number of Values it provides, thus:
TextField textField = (TextField)itemField;
// If you need to deal with multi-valued fields separately
if (textField.Values.Count > 1)
{
//Logic to deal with multiple values goes here
}
else
{
//Logic to deal with single valued goes here
}
// Much better... If you can deal with any number of values in a generic fashion
foreach (string value in textField.Values)
{
// Generic code goes here
}
Possibly not specific to webservices, but...
I have a webmethod that returns:
List<Tadpole> myList = getList();
return new { data = myList , count = 5 };
It returns this as JSON.
my code checks myList[x].fishsticks which isn't actually part of the Tadpole class (so it errors). I am wondering, can I add a fishsticks attribute to myList somehow to avoid the error, so it gets included when I return the data?
Is there perhaps another elegant solution for doing this?
In your example, you'll have to add a fishsticks property to Tadpole.
public class Tadpole
{
//....
public int Fishsticks { get; set; }
}
Also, why are you adding a .Count property to your JSON type? Wouldn't it make more sense to just .data.Count, or just return the list and skip the wrapper entirely?
I haven't checked what properties of List<> get serialized lately, so it's possible that it's not included, but even if that's the case it would make more sense to do this:
List<Tadpole> myList = getList();
return new { data = myList , count = myList.Count };
Or, create a descendant class that overrides .Count and adds a serialization attribute.
Edit
If I remember correctly, anonymous/dynamic types are internally implemented as dictionaries, while classes are, well, not. (BTW, anonymous types and dynamic objects bring a host of performance and maintenance issues along with them.)
If you don't want to modify Tadpole for some reason, you could always create a descendant class:
public class HungryTadpole : TadPole
{
public int FishSticks { get; set; }
}
Strong typing is your friend and will save you many headaches down the road.
I have a dynamic class that I have created
public dynamic class SiteZoneFileUploadVO
{
public var destination:String = "sitezone";
public var siteZoneId:uint;
public var fileType:String;
public var fileContents:String;
public function SiteZoneFileUploadVO()
{
}
}
when I try to iterate over this object's property names it only iterates the dynamically added properties.
parameters.dynVar= "value";
for(var name:String in parameters)
{
trace(name);
}
Even though the object has all the properties equal to a value (ive checked this in the debugger) the only property name that will be traced is dynVar.
How can I iterate over all the property names and not just the dynamically added ones?
You can use describeType() to get an XML with all methods and variables of your class and then filter out the properties you want to iterate over (e.g. all variables) and store them in an XMLList.
As the next step you would then iterate over the XMLList and use square bracket notation on your object to access the filtered properties by their names. However, you can only access public properties this way because describeType() won't look at private properties.
If you're running flex:
Looked at a few posts, ObjectUtil.toString was the most promising, then looked at the flex source code for it, it uses another method ObjectUtil.getClassInfo which is exactly what you need. If you just want property names:
ObjectUtil.getClassInfo(myClass).properties
returns an Array of QName objects, each has a localName property which will give you a string for each property name
Just use trace(ObjectUtil.toString(parameters)); That should give you your entire object.