Develop Reference

When should I update a database to reflect property changes?

I am new to WPF and I am building a small app with Linq To Entities (and SQLite database).
I just would like to know, where do I have to call my methods in order to update the database, when a property has changed ?
I would say in the property in ViewModel like this :
public string FirstName
{
get
{
return this.person.FirstName;
}
set
{
this.person.FirstName = value;
OnPropertyChanged("FirstName");
this.person.updateFirstname(value);
}
}
I am not sure if this is the best solution...

The problem of when to save to the database gives rise to the Unit of Work pattern. Linq-to-Entities has a reasonable implementation of this with the ObjectContext, where data is queued up in the context and then saved to the database when the logical unit of work is complete.
In your example, you are already setting the property on the L2E entity, Person, which is likely connected to the context. When you call ObjectContext.SaveChanges, this will be saved without the need for the updateFirstname method.
The thing you have to decide is when to call ObjectContext.SaveChanges (and thus end the unit of work), and doing this when the user explicitly saves or when the form is closed (optionally propmting for the user to commit or discard changes) is a reasonable approach here. To implement this, your viewmodels reference the ObjectContext and can call the SaveChanges method when the user action (usually modeled with a WPF ICommand published by the viewmodel and bound to the view) is executed.

You should concentrate your updates around unit-of-work rather than around individual fields. If your database is properly normalized each row will represent an entity and should be treated as such, updates to an entity should keep an entity in "valid" state. In your scenario if you update person's first name with intention of also updating last name if the app or server blows up your person record will be invalid.
In terms of MVVM, I usually either piggyback on grid's "update entire row at once" strategy and route that event into viewmodel or I just give them a save button :)

It is best to inject a service Interface into your ViewModel constructor and use some type of service to update the database.
This way you end up with loosely coupled system and your ViewModel stays agnostic of your Data Access Layer as it should...

Entity Framework inserts row without me calling AddToObject()?

Okay, so it's late and I'm tired, but this shouldn't be happening... or so I believe. I have the following code for an event handler for a button click where I create a new customer from a web form.
int customerId = <from_somewhere_doesnt_matter>;
Customer cust;
if (int.TryParse(Request.QueryString["cid"], out customerId)) {
// update existing customer
cus = db.Customers.Single(c => c.CustomerId == customerId);
cus.UpdatedByUser = user;
cus.Updated = DateTime.Now;
}
else {
// create new customer
cus = new Customer();
cus.InsertedByUser = user;
cus.Inserted = DateTime.Now;
}
SetFields(cus);
db.SaveChanges();
The SetFields() method just populates the different properties for the customer from the corresponding web form fields.
I have been running this code in production for quite a while and it's been working fine. However, recently a user told me it doesn't work to add a new user (doesn't happen very often). I checked it, and sure enough, I filled in the form and tried to add the user but was just redirected back to the user list without any error message and without a new user.
I checked the code, and realised I had forgotten to add the db.Users.AddObject(usr) when adding a new user. I added the method call, and the user was added correctly. I then went to the customer code, just to check how and when I call the AddObject-method there, and it turns out I don't!
I might be blind, but I have searched the source code and I do not call the method anywhere and it still works to add a customer! The only thing I can think of is that the customer is added because it refers to another object (the current user), and that somehow triggers an add. The user does not depend on any other fields.
What is happening!?
Thanks in advance!

The reason might be automatic association fixup. The T4 Template POCO Generator of EF 4.0 for instance creates such methods which are called in the property setters. Possibly it happens in this line in your code:
cus.InsertedByUser = user;
If your User entity has a collection of customers (as the inverse navigation property to InsertedByUser) then the POCO Generator would create a fixup method which does something like ...
InsertedByUser.Customers.Add(this);
... where this is cus in your example. This is called in the setter for the InsertedByUser property. This means that your new created customer cus is automatically added to the Customers collection of the user which you assign. EF change detection will recognize this and put the new customer in Added state into the context. When you call SaveChanges a new customer record in the database is created.
Just a hypothesis. You could check in detail what's going on by debugging into the property setter.

When you set
cus.InsertedByUser = user;
Your customer will be added to the user.Customers collection that represents the other end of the many-to-one. This gives EF a handle on the object which will then be added to the context and saved away since it is new.

Can entities be attached to an ISession that weren't previously attached?

I'm playing around with NHibernate 3.0. So far things are pretty cool. I'm trying to attach an entity that wasn't detached previously:
var post = new Post(){ Id = 2 };
session.Update(post); // Thought this would work but it doesn't.
post.Title = "New Title After Update";
session.Flush();
What I'm trying to write is the following:
var post = new Post(){ Id = 2 };
session.Attach(post);
post.Title = "New Title After Update";
session.Flush(); // Sql should be something like: UPDATE Post SET Title='New Title After Update' WHERE Id=2
Is this possible so that only Title gets updated? This is currently possible in EntityFramework. I'd like to not have to load Post from the database when I just need to update a few properties. Also, I'm trying to avoid a method call that would create the object... since it's moving away from an object oriented approach in my opinion.
EDIT: I know about using transactions, I just used Flush() to make the code simple. Ok so I think we're sort of getting on the right track for what I'm trying to achieve. I'd like to be able to create an entity with a known Id using the constructor, like I have in the 2nd code block above. I don't want to have to make a call to Get<T> or Load<T> since it feels rather wrong constructing objects like this that already exist in the database. For example, in Entity Framework I can write the 2nd code example and it will "just work". It only updates the Title property.

You can session.Save() or session.SaveOrUpdate()
update
Okay, I think I see now what you are trying to do. You are trying to update a single property on a Post that was previously persisted, not a new Post, and to do that you're instantiating a new Post and giving it the Id of one in the database.
I'm not sure what you mean when you say you're trying to avoid a method call that would create the object, but the way to do this with NHibernate is this:
var post = session.Load<Post>(2);
post.Title = "New Title";
session.SaveOrUpdate(post);
In general, you should not be calling Flush() on your sessions.
The important thing to note here is the use of session.Load. Calling Load with an id in and of itself does not load the entity from the database. The entity's property values will only be loaded when/if you access them.
Of course, in this scenario, I believe that NHibernate will load the properties for the Post, (but not collections unless you've specified an eager fetch mode), and that makes sense (frankly, I don't understand why EF would not load the entity). What if the setter for your Title property does something important, like check it against the existing title, validate the title's length, check your credentials, or update another property? Simply sending an UPDATE to the database isn't sufficient.

It's possible to only update changed properties by setting dynamic-update in the mapping. However, as far as I know, it is not possible (without reverting to SQL) to perform an update without retrieving the object from the database at some point.

Use the Merge method. You have to create a new instance variable to accept the attached entity = nhibernate will not do anything else with your detached instance.
var post = new Post(){ Id = 2 };
post.Title = "New Title After Update";
// Must create a new instance to hold final attached entity
var attachedPost = session.Merge(post);
session.Update(attachedPost);
session.Flush();
// Use attachedPost after this if still needed as in session entity
That covers the "attach" functionality you are looking for, but I don't see how you are going to be able to only update the one property. if the object instance has not been populated from the database, the properties will be different. Dynamic mapping will not solve this - NHibernate sees the properties as "updated" to a bunch of nulls, empty strings.
Gotta say, you are creating a new instance but what you are actually doing is updating an existing instance. You are working directly with IDs not objects. And you are setting a single property and now have an instance potentially hanging around and doing more things but it has not enforced any invariants and may in fact bear no resemblence to the real deal other than the id property...
It all feels pretty anti-object oriented to me personally.

Using Attach with Linq To Sql and Stored Procs

I am trying to use the attach method to update an entity that was retrieve via a stored proc.
The stored proc is set up to return a specific instance, which is present in my dbml. The retrieval works as expected and returns a fully populated object. The reason I need to use a stored proc is that I need to update a property on that entity at the same time that it is retrieved.
After I have retrieved this entity, I am mapping it using AutoMapper to another model which is used in another tier of the app. This tier performs a few operations, and makes a change to the entity, and passes it back to the repository for updating.
The repository converts this business model back into a database model, and attempts to attach it to the datacontext in order to take advantage of the automagic updating.
No matter what combination of Attach(entity, true) Attach(entity) etc, and it gives me messages like "Row not found or changed" or "Unable to add an entity with the same primary key".
Does anyone have any experience with the Attach method and how it can be used to update entities that did not necessarily come from the data context using query syntax (ie in this case a stored proc)?
Thanks alot

First, if you are creating a copy of the object, making changes and then trying to attach the copied object to the same DataContext as the one with the original object in it, then this would probably result in the "Unable to add an entity with the same primary key" message. One way to handle this is:
1. Get object from DataContext
2. Make changes and map object (or vice versa - whatever order)
3. Update the original object with the new values made in the other tier
4. SubmitChanges on the DataContext containing the original object
or
Get the object from a DataContext and close the DataContext
Make your changes and do your mapping
Retrieve the object from the DataContext to which you want to save
Update that object with the values from your mapped object
SubmitChanges
Alternately, when you say you are using the proc because you need to update a property at the same time that you retrieve it, I'd need to see the proc, but if you are somehow committing this update after retrieving the information, then indeed the message "row not found or changed" is correct. This would be hard to do, but you could do it if you're loading the data into a temp table, doing the update, and then using a select from the temp table to populate the object. One thing you could try is setting that property, in the L2S designer, to AutoUpdate = Never and see if that makes the problem go away. If so, this is your problem.

1: is it the same data-context, and
2: is it the same entity instance (or one that looks like it)
This would only happen for the same data-context, I suspect. If it is the same entity, then it is already there; just call SumbitChanges. Otherwise, either use a second data-context or detach the original entity.

So if I retrieved the entity via a stored proc, is it being tracked by the datacontext?
The thing is.. I'm going from the data model, to a another model that is used by another component, and then back. Its not.. really the same instance, but it does have all the same properties.
IE
public Models.Tag GetEntity()
{
var dbTag = db.PROJ_GetEntity((int)EntityStatuses.Created, (int)EntityStatuses.CreatingInApi).SingleOrDefault();
return FromDb Entity(dbEntity);
}
var appModel = GetEntity(); // gets an Entity from a stored proc (NOT GetEntity_RESULT)
appModel.MakeSomeChanges();
_Repo.Persist(appModel);
public void Persist(Models.AppModel model)
{
var dbEntity = Mapper.Map(model);
db.Attach(dbEntity);
db.SubmitChanges();
}
This is somewhat pseudo code like.. but it demostrates pretty much exactly what I am doing.
Thanks

I'm upvoting weenet's answer because he's right - you can't use Attach to apply the changes.
Unlike Entity Framework, you can only attach an L2S object to a datacontext if it has never been attached before - i.e. it's a newed entity that you want to Insert into a table.
This does cause numerous problems in multi-layered environments - however I've been able to get around many of the issues by creating a generic entity synchronisation system, which uses reflection and expression trees.
After an object has been modified, I run the dynamic delegate against a new object from the DC and the modified object, so that only the differences are tracked in the DC before generating the Update statement. Does get a bit tricky with related entities, though.

How do you save a Linq object if you don't have its data context?

I have a Linq object, and I want to make changes to it and save it, like so:
public void DoSomething(MyClass obj) {
obj.MyProperty = "Changed!";
MyDataContext dc = new MyDataContext();
dc.GetTable<MyClass>().Attach(dc, true); // throws exception
dc.SubmitChanges();
}
The exception is:
System.InvalidOperationException: An entity can only be attached as modified without original state if it declares a version member or does not have an update check policy.
It looks like I have a few choices:
put a version member on every one of my Linq classes & tables (100+) that I need to use in this way.
find the data context that originally created the object and use that to submit changes.
implement OnLoaded in every class and save a copy of this object that I can pass to Attach() as the baseline object.
To hell with concurrency checking; load the DB version just before attaching and use that as the baseline object (NOT!!!)
Option (2) seems the most elegant method, particularly if I can find a way of storing a reference to the data context when the object is created. But - how?
Any other ideas?
EDIT
I tried to follow Jason Punyon's advice and create a concurrency field on on table as a test case. I set all the right properties (Time Stamp = true etc.) on the field in the dbml file, and I now have a concurrency field... and a different error:
System.NotSupportedException: An attempt has been made to Attach or Add an entity that is not new, perhaps having been loaded from another DataContext. This is not supported.
So what the heck am I supposed to attach, then, if not an existing entity? If I wanted a new record, I would do an InsertOnSubmit()! So how are you supposed to use Attach()?
Edit - FULL DISCLOSURE
OK, I can see it's time for full disclosure of why all the standard patterns aren't working for me.
I have been trying to be clever and make my interfaces much cleaner by hiding the DataContext from the "consumer" developers. This I have done by creating a base class
public class LinqedTable<T> where T : LinqedTable<T> {
...
}
... and every single one of my tables has the "other half" of its generated version declared like so:
public partial class MyClass : LinqedTable<MyClass> {
}
Now LinqedTable has a bunch of utility methods, most particularly things like:
public static T Get(long ID) {
// code to load the record with the given ID
// so you can write things like:
// MyClass obj = MyClass.Get(myID);
// instead of:
// MyClass obj = myDataContext.GetTable<MyClass>().Where(o => o.ID == myID).SingleOrDefault();
}
public static Table<T> GetTable() {
// so you can write queries like:
// var q = MyClass.GetTable();
// instead of:
// var q = myDataContext.GetTable<MyClass>();
}
Of course, as you can imagine, this means that LinqedTable must somehow be able to have access to a DataContext. Up until recently I was achieving this by caching the DataContext in a static context. Yes, "up until recently", because that "recently" is when I discovered that you're not really supposed to hang on to a DataContext for longer than a unit of work, otherwise all sorts of gremlins start coming out of the woodwork. Lesson learned.
So now I know that I can't hang on to that data context for too long... which is why I started experimenting with creating a DataContext on demand, cached only on the current LinqedTable instance. This then led to the problem where the newly created DataContext wants nothing to do with my object, because it "knows" that it's being unfaithful to the DataContext that created it.
Is there any way of pushing the DataContext info onto the LinqedTable at the time of creation or loading?
This really is a poser. I definitely do not want to compromise on all these convenience functions I've put into the LinqedTable base class, and I need to be able to let go of the DataContext when necessary and hang on to it while it's still needed.
Any other ideas?

Updating with LINQ to SQL is, um, interesting.
If the data context is gone (which in most situations, it should be), then you will need to get a new data context, and run a query to retrieve the object you want to update. It's an absolute rule in LINQ to SQL that you must retrieve an object to delete it, and it's just about as iron-clad that you should retrieve an object to update it as well. There are workarounds, but they are ugly and generally have lots more ways to get you in trouble. So just go get the record again and be done with it.
Once you have the re-fetched object, then update it with the content of your existing object that has the changes. Then do a SubmitChanges() on the new data context. That's it! LINQ to SQL will generate a fairly heavy-handed version of optimistic concurrency by comparing every value in the record to the original (in the re-fetched) record. If any value changed while you had the data, LINQ to SQL will throw a concurrency exception. (So you don't need to go altering all your tables for versioning or timestamps.)
If you have any questions about the generated update statements, you'll have to break out SQL Profiler and watch the updates go to the database. Which is actually a good idea, until you get confidence in the generated SQL.
One last note on transactions - the data context will generate a transaction for each SubmitChanges() call, if there is no ambient transaction. If you have several items to update and want to run them as one transaction, make sure you use the same data context for all of them, and wait to call SubmitChanges() until you've updated all the object contents.
If that approach to transactions isn't feasible, then look up the TransactionScope object. It will be your friend.

I think 2 is not the best option. It's sounding like you're going to create a single DataContext and keep it alive for the entire lifetime of your program which is a bad idea. DataContexts are lightweight objects meant to be spun up when you need them. Trying to keep the references around is also probably going to tightly couple areas of your program you'd rather keep separate.
Running a hundred ALTER TABLE statements one time, regenerating the context and keeping the architecture simple and decoupled is the elegant answer...

find the data context that originally created the object and use that to submit changes
Where did your datacontext go? Why is it so hard to find? You're only using one at any given time right?
So what the heck am I supposed to attach, then, if not an existing entity? If I wanted a new record, I would do an InsertOnSubmit()! So how are you supposed to use Attach()?
You're supposed to attach an instance that represents an existing record... but was not loaded by another datacontext - can't have two contexts tracking record state on the same instance. If you produce a new instance (ie. clone) you'll be good to go.
You might want to check out this article and its concurrency patterns for update and delete section.

The "An entity can only be attached as modified without original state if it declares a version member" error when attaching an entitity that has a timestamp member will (should) only occur if the entity has not travelled 'over the wire' (read: been serialized and deserialized again). If you're testing with a local test app that is not using WCF or something else that will result in the entities being serialized and deserialized then they will still keep references to the original datacontext through entitysets/entityrefs (associations/nav. properties).
If this is the case, you can work around it by serializing and deserializing it locally before calling the datacontext's .Attach method. E.g.:
internal static T CloneEntity<T>(T originalEntity)
{
Type entityType = typeof(T);
DataContractSerializer ser =
new DataContractSerializer(entityType);
using (MemoryStream ms = new MemoryStream())
{
ser.WriteObject(ms, originalEntity);
ms.Position = 0;
return (T)ser.ReadObject(ms);
}
}
Alternatively you can detach it by setting all entitysets/entityrefs to null, but that is more error prone so although a bit more expensive I just use the DataContractSerializer method above whenever I want to simulate n-tier behavior locally...
(related thread: http://social.msdn.microsoft.com/Forums/en-US/linqtosql/thread/eeeee9ae-fafb-4627-aa2e-e30570f637ba )

You can reattach to a new DataContext. The only thing that prevents you from doing so under normal circumstances is the property changed event registrations that occur within the EntitySet<T> and EntityRef<T> classes. To allow the entity to be transferred between contexts, you first have to detach the entity from the DataContext, by removing these event registrations, and then later on reattach to the new context by using the DataContext.Attach() method.
Here's a good example.

When you retrieve the data in the first place, turn off object tracking on the context that does the retrieval. This will prevent the object state from being tracked on the original context. Then, when it's time to save the values, attach to the new context, refresh to set the original values on the object from the database, and then submit changes. The following worked for me when I tested it.
MyClass obj = null;
using (DataContext context = new DataContext())
{
context.ObjectTrackingEnabled = false;
obj = (from p in context.MyClasses
where p.ID == someId
select p).FirstOrDefault();
}
obj.Name += "test";
using (DataContext context2 = new ())
{
context2.MyClasses.Attach(obj);
context2.Refresh(System.Data.Linq.RefreshMode.KeepCurrentValues, obj);
context2.SubmitChanges();
}