I am using RIA Services Beta 2 with Linq2Sql and SL3. In my SL3, I have a datagrid where I can do some mappings of data (Updates, Inserts and Deletes). I override the Submit method when SubmitChanges() is called. In the submit method in the domain service, I do some validation. If a validation fails for a particular ChangeSetEntry in the ChangeSet, a ValidationErrors is added. Then I call the base.Submit(changeSet).
So, if the changeset has 3 entities and one of the entities results in validation error, the other 2 entities are also rolled back. It looks like, RIA Services does an implicit transaction and hence it either submits all 3 or none even if 2 out of 3 does not have any validation error. Is there a way for the RIA service, to prevent rollback of the valid entities and only invalidate the ones that has validation failed.
Inputs will be appreciated.
Thanks
Mohit
First I am a bit confused as to how or why you are using SL3 because it was only supported with Beta 1.
Is there a reason for not using the built in validation? If you did most validation would be caught client side and could be corrected before the server submit.
You might be able to achieve what you want by overriding the PersistChangeSet method but I would worry about keeping your client in sync with the server if only some of the entities are persisted.
http://weblogs.asp.net/fredriknormen/archive/2009/12/29/wcf-ria-services-domainservice-life-cycle-and-adding-transactions.aspx
Related
We are building a RESTful web service where there are sometimes different required fields for a POST than a PUT. For example, a field like CustomerSinceDate is allowed to be set on an insert, but not on an update. Is there is a way to set that up with Data Annotations?
EntityFramework does not (and should not) know anything about your web service. It deals only with what rules exist in the persistence layer.
What you are looking for is validation.
So in your REST service, you should check whether CustomerSinceData has been changed, and the entity is being updated. If so, you should throw an Exception with an appropriate message to the consumer.
Here is an article on writing your own DataAnnotations, if you prefer using those:
http://msdn.microsoft.com/en-us/data/jj819164#attributes
Otherwise, take a look at this article on how to write your own custom validation: http://msdn.microsoft.com/en-us/data/gg193959.aspx
(in particular, the section on IValidatableObject).
Your rule could be formulated as (pseudo code)
//if object exists in db AND CustomerSinceData has changed
DataAnnotations will get you a long way, but can be tedious to write if you are writing business logic that will never be reused anywhere else.
I my client application I am calling entityManager.saveChanges to send all currently changed entities from the client up to the server. Then in the BeforeSaveEntity event on the server I am performing some server side validation on each entity to see if it should be excluded from the save map. So for example, my entity may have a value for description that is too long. So I return false from BeforeSaveEntity, and generate a new EntityError which will then be added to the saveResult.EntityErrors collection. All of the valid records that haven't been excluded from the save map then save off successfully, and my saveResult is returned to the client. But because of this single entity error, the auto-patchup of the entities returned does not occur back on the client. I looked at the source and basically there seems to be a check that says if there is anything in the saveResult.EntityErrors collection, don't bother with the patch-up. But There was only 1 entity that purposefully wasn't saved, so I still want to be able to patch up the others. Is this behavior by design? I want to be able to exclude certain entities from the save (which I can do using the BeforeSaveEntity event), but there doesn't seem to be any way of then getting the entity errors back to the client using the inbuilt mechanism, without the full patch-up being abandoned.
Saves in breeze are transactional if at all possible ( some backend providers, like MongoDb are not because they don't support it.). This means that if any failures are experienced with any entities within a save bundle the entire save is reverted and an error is returned to the client. This is by design.
I am having trouble getting related entities to be loaded on the client using RIA Services and EF 4.1 with Silverlight.
I'm currently using the Include() method on my DbDomainService with an Expression parameter and am finding that when stepping through my service the related entities are loaded just fine. However, when the Queryable results are returned to the client NO related entities are loaded - they are null. All of my entities are marked with the [DataMember] attribute so I have assumed that it isn't a serialization issue. Moreover, my DbDomainService query method is marked with the [Query] attribute.
I was wondering if there is anything specific that has to be set up on the client when using RIA Services with EF 4.1 code first? I must be missing something, but I'm not sure what.
Any help would be appreciated.
Thanks,
sfx
While you may have used the .Include() in your service call, you also have to add the [Include] attribute in the metadata class that is also created.
The .Include() statement tells EF to generate the SQL necessary to retrieve the data, while the Include attribute tells WCF RIA Services to make sure the Entity Class is also created on the client.
Once the data arrives at the client, it needs to know what type of structure to put it in as well.
HTH
I have a problem using Entity Framework and COM+. The architecture is this:
A windows service calls four methods of a COM, every n minutes.
COM+ then calls the corresponding methods on a Business Layer.
Business Layer calls DAL. DAL then returns a list to Business Layer. This is done by calling a .ToList()
When I try running the service, the DAL methods return a timeout inner exception. When I try to view the table from Enterprise Manager, it returns a timeout as well! From what I 've seen, the SELECT statements block the other connection instances.
Has anyone else experienced similar problems?
P.S. I cannot post any code yet because I am not at my work... Will do so tomorrow.
Well, as it seems, Entity Framework didn't have to do with any of the above.
As it turns out, the problem was within COM+. I should have ended each method of COM+ with ContextUtil.SetComplete(). Apparently, I didn't do that so the transaction stayed active and after the first few calls, it locked my db.
e.g.
Using MyEntity As New EntityObject
MyEntity.Connection.Open()
Dim rows = From tbl In MyEntity.MyTable _
Select tbl
list = rows.ToList
End Using
ContextUtil.SetComplete()
Please note that if an exception occurs, you should place a ContextUtil.SetAbort(). I should also like to note that the above code is mixed. It's part of my DAL layer and part of my COM+. I just put it like that to make the example more clear...
I have a Product object with a property that is a collection of type Workflows. In my "GetProducts" method on the domaincontext object I have set a breakpoint at the return statement to see if the workflows collection is filled.
It is.
On the client side I check Context.Products[0].Workflows in another breakpoint and I see 0 results. Is there a way to persist this nested data for consumption on the client side or is RIA Services inhibited from doing this?
If you have or can download the RiaServicesOverviewPreview.pdf document section 4.8 details how to do this. The basic summary it.
Make sure your L2S query specifies the .LoadWith<>() parameter. Lazy loading doesn't work with RIA services so you have to use implicit loading.
You need to apply the "IncludeAttribute" to the associated member. For example add the [Include] attribute on your Workflows field in the Product metadata class.
Ensure that your Workflow (child) type is exposed as a client type so it gets genned to the client side.
You can get the document here: http://www.microsoft.com/downloads/details.aspx?FamilyID=76bb3a07-3846-4564-b0c3-27972bcaabce&displaylang=en
I should kick myself. I realized that I needed to add "[Include]" to the property in Product within the DataService.metadata.cs file and now it gets sent to the client.