Let's suppose I have the following code:
TEModule teModule = Context.TEModules.Where(module => module.EnumValue.Equals(text.ModuleName)).FirstOrDefault();
if (teModule == null)
{
teModule = new TEModule();
teModule.EnumValue = text.ModuleName;
Context.TEModules.AddObject(teModule);
//Context.SaveChanges();
TEModule aux = Context.TEModules.Where(module => module.EnumValue.Equals(teModule.ModuleName)).FirstOrDefault();
}
My problem is that if I keep the "SaveChanges" commented, then on the next query the aux object is always null, because Context.TEModules is empty, even when I call the "AddObject" method. However, if I call SaveChanges after AddObject, then on the next query the aux object is not null. The problem is that I don't want to call SaveChanges so often, because this is not the only piece of code in which I add objects, and the performance goes down if I do so.
So question is: Do I have to call SaveChanges after every AddObject call, if later I need to know if the object already exists?
The purpose of linq-to-entities query is to be executed and the execution is performed in the database so if you didn't saved the entity its database representation doesn't exist.
If you need to find locally stored entities (not persisted yet) you must query ObjectStateManager
instead.
var entity = Context.ObjectStateManager.GetObjectStateEntries(EntitiState.Added)
.Where(e => !e.IsRelationship)
.Select(e => e.Entity)
.OfType<TEModule>()
.FirstOrDefault(m => m.EnumValue.Equals(teModule.ModuleName));