NHibernate join 2 tables and project to DTO with collection - nhibernate

I am trying to join 2 tables, and project directly to DTOs (NHibernate 5).
I have following entities:
public class Person {
public Guid Id {get;set;}
public string Name {get;set;}
}
public class Car {
public Guid Id {get;set;}
public string Brand {get;set;}
public Person Owner {get;set;}
}
as we see, there is just reference from Car to Person (car knows its owner), and this is ok in my whole project.
However, there is one place, where I need to query all Persons, and make each person with collection of owned cars.
I created such DTOs:
public class PersonDto {
public Guid Id {get;set;}
public string Name {get;set;}
public IList<CarDto> {get;set;}
}
public class CarDto {
public Guid Id {get;set;}
public string Brand {get;set;}
}
it is kind of present the data linked together upside-down.
This task seems trivial using SQL or LINQ (GroupJoin) however I found it extremly hard to do in NH, since GroupJoin is not implemented in NH.
Can you please help me how to solve above issue?

Just add a Car collection to the existing Person entity and mark the collection inverse. Collections in NHibernate are lazy by default, so when you query Person, it won't read the cars from the DB, until you start iterating them. In other words adding the Car collection won't affect the way your code works.
When you want to efficiently query persons together with cars, force NH to do a join
.QueryOver<Person>.Fetch(person => person.Cars).Eager

I'd like to answer my own question. Thanks Rafal Rutkowski for his input.
To get data that has association in only 1 direction, we need to introduce a new data model, and then, manually convert result to our entities. I hope following example is best possible answer:
1) create such data model to store NH response:
private class PersonCarModelDto
{
public Guid PersonId { get; set; }
public string PersonName { get; set; }
public Guid CarId { get; set; }
public string CarBrand { get; set; }
}
2) create such a model to store hierarchical data as output:
private class PersonModel
{
public Guid Id { get; set; }
public string Name { get; set; }
public IList<CarModel> Cars { get; set; }
}
private class CarModel
{
public Guid Id { get; set; }
public string Brand { get; set; }
}
3) now the NH query:
Person personAlias = null;
Car carAlias = null;
PersonCarModelDto resultAlias = null;
var response = GetCurrentSession().QueryOver<Car>(() => carAlias) // notice we are going from from 'downside' entity to 'up'
.Right.JoinAlias(c => c.Owner, () => personAlias) // notice Right join here, to have also Persons without any car
.SelectList(list => list
.Select(() => personAlias.Id).WithAlias(() => resultAlias.PersonId)
.Select(() => personAlias.Name).WithAlias(() => resultAlias.PersonName)
.Select(() => carAlias.Id).WithAlias(() => resultAlias.CarId)
.Select(() => carAlias.Brand).WithAlias(() => resultAlias.CarBrand)
.TransformUsing(Transformers.AliasToBean<PersonCarModelDto>())
.List<PersonCarModelDto>()
;
4) now we have flat data as a list of PersonCarModelDto, but we want to make output model:
var modelResult = response.GroupBy(p => p.PersonId)
.Select(x => new PersonModel
{
Id = x.Key,
Name = x.Select(y => y.PersonName).First(), // First() because each PersonName in that group is the same
Cars = x.Select(y => new CarModel
{
Id = y.CarId,
Name = y.CarBrand
})
.ToList()
})
.ToList()
;
Conclusions:
the problem is much easier to solve having bidirectional associations
if for some reason, you don't want bidirectional associations, use this technique
this approach is also useful if your entities has a lot of other properties, but for some reason you need just a small part of them
(optimiziation of data returned for DB)

Related

One-to-Many relationship with ORMLite

The only examples I can find addressing this sort of scenario are pretty old, and I'm wondering what the best way is to do this with the latest version of ORMLite...
Say I have two tables (simplified):
public class Patient
{
[Alias("PatientId")]
[Autoincrement]
public int Id { get; set; }
public string Name { get; set; }
}
public class Insurance
{
[Alias("InsuranceId")]
[Autoincrement]
public int Id { get; set; }
[ForeignKey(typeof("Patient"))]
public int PatientId { get; set; }
public string Policy { get; set; }
public string Level { get; set; }
}
Patients can have multiple Insurance policies at different "levels" (primary, secondary, etc). I understand the concept of blobbing the insurance information as a Dictionary type object and adding it directly to the [Patient] POCO like this:
public class Patient
{
public Patient() {
this.Insurances = new Dictionary<string, Insurance>(); // "string" would be the Level, could be set as an Enum...
}
[Alias("PatientId")]
[Autoincrement]
public int Id { get; set; }
public string Name { get; set; }
public Dictionary<string, Insurance> Insurances { get; set; }
}
public class Insurance
{
public string Policy { get; set; }
}
...but I need the insurance information to exist in the database as a separate table for use in reporting later.
I know I can join those tables in ORMLite, or create a joined View/Stored Proc in SQL to return the data, but it will obviously return multiple rows for the same Patient.
SELECT Pat.Name, Ins.Policy, Ins.Level
FROM Patient AS Pat JOIN
Insurance AS Ins ON Pat.PatientId = Ins.PatientId
(Result)
"Johnny","ABC123","Primary"
"Johnny","987CBA","Secondary"
How can I map that into a single JSON response object?
I'd like to be able to map a GET request to "/patients/1234" to return a JSON object like:
[{
"PatientId":"1234",
"Name":"Johnny",
"Insurances":[
{"Policy":"ABC123","Level":"Primary"},
{"Policy":"987CBA","Level":"Secondary"}
]
}]
I don't have a lot of hope in this being do-able in a single query. Can it be done in two (one on the Patient table, and a second on the Insurance table)? How would the results of each query be added to the same response object in this nested fashion?
Thanks a ton for any help on this!
Update - 4/29/14
Here's where I'm at...In the "Patient" POCO, I have added the following:
public class Patient
{
[Alias("PatientId")]
[Autoincrement]
public int Id { get; set; }
public string Name { get; set; }
[Ignore]
public List<Insurance> Insurances { get; set; } // ADDED
}
Then, when I want to return a patient with multiple Insurances, I do two queries:
var patientResult = dbConn.Select<Patient>("PatientId = " + request.PatientId);
List<Insurance> insurances = new List<Insurance>();
var insuranceResults = dbConn.Select<Insurance>("PatientId = " + patientResult[0].PatientId);
foreach (patientInsurance pi in insuranceResults)
{
insurances.Add(pi);
}
patientResult[0].Insurances = insurances;
patientResult[0].Message = "Success";
return patientResult;
This works! I get nice JSON with nested items for Insurances while maintaining separate related tables in the db.
What I don't like is that this object cannot be passed back and forth to the database. That is, I can't use the same nested object to automatically insert/update both the Patient and InsurancePolicy tables at the same time. If I remove the "[Ignore]" decorator, I get a field in the Patient table called "Insurances" of type varchar(max). No good, right?
I guess I'm going to need to write some additional code for my PUT/POST methods to extract the "Insurances" node from the JSON, iterate over it, and use each Insurance object to update the database? I'm just hoping I'm not re-inventing the wheel here or doing a ton more work than is necessary.
Comments would still be appreciated! Is Mythz on? :-) Thanks...
An alternate more succinct example:
public void Put(CreatePatient request)
{
var patient = new Patient
{
Name = request.Name,
Insurances = request.Insurances.Map(x =>
new Insurance { Policy = i.Policy, Level = i.Level })
};
db.Save<Patient>(patient, references:true);
}
References are here to save the day!
public class Patient
{
[Alias("PatientId")]
[Autoincrement]
public int Id { get; set; }
public string Name { get; set; }
[Reference]
public List<Insurance> Insurances { get; set; }
}
public class Insurance
{
[Alias("InsuranceId")]
[Autoincrement]
public int Id { get; set; }
[ForeignKey(typeof("Patient"))]
public int PatientId { get; set; }
public string Policy { get; set; }
public string Level { get; set; }
}
I can then take a JSON request with a nested "Insurance" array like this:
{
"Name":"Johnny",
"Insurances":[
{"Policy":"ABC123","Level":"Primary"},
{"Policy":"987CBA","Level":"Secondary"}
]
}
...to create a new record and save it like this:
public bool Put(CreatePatient request)
{
List<Insurance> insurances = new List<Insurance>();
foreach (Insurance i in request.Insurances)
{
insurances.Add(new Insurance
{
Policy = i.Policy,
Level = i.Level
});
}
var patient = new Patient
{
Name = request.Name,
Insurances = insurances
};
db.Save<Patient>(patient, references:true);
return true;
}
Bingo! I get the new Patient record, plus 2 new records in the Insurance table with correct foreign key references back to the PatientId that was just created. This is amazing!
First you should define a foreign collection in Patient class. (with get and set methods)
#ForeignCollectionField
private Collection<Insurance> insurances;
When you query for a patient, you can get its insurances by calling getInsurances method.
To convert all into a single json object with arrays inside you can use a json processor. I use Jackson (https://github.com/FasterXML/jackson) and it works very well. Below will give you json object as a string.
new ObjectMapper().writeValueAsString(patientObject);
To correctly map foreign fields you should define jackson references. In your patient class add a managed reference.
#ForeignCollectionField
#JsonManagedReference("InsurancePatient")
private Collection<Insurance> insurances;
In your insurance class add a back reference.
#JsonBackReference("InsurancePatient")
private Patient patient;
Update:
You can use Jackson to generate objects from json string then iterate and update/create database rows.
objectMapper.readValue(jsonString, Patient.class);

Not sure how to join tables using Fluent NHibernate

I'm using NHibernate on legacy tables and after looking through our codebase, I seem to be the only person with this need. I need to join two tables so I can run a query, but I haven't made any progress today. I'll try to abbreviate where it makes sense in my code snippets. Care to help?
Tables--
Order
OrderID (primary key)
OrderName
OrderType
OrderLocation
OrderAppendix
ID (composite key)
Key (composite key)
Value (composite key)
The ID portion of the OrderAppendix composite key is associated with OrderID in the Order table; therefore, and Order can have several entries in the OrderAppendix table.
Domain Objects--
[Serializable]
public class Order
{
public virtual string OrderID { get; set; }
...
public virtual string OrderLocation { get; set; }
}
[Serializable]
public class OrderAppendix
{
public virtual string ID { get; set; }
public virtual string Key { get; set; }
public virtual string Value { get; set; }
public override bool Equals(object obj)
{
...
}
public override int GetHashCode()
{
...
}
}
Mapping
internal sealed class OrderMap : ClassMap<Order>
{
Table("Order");
Id(x => x.OrderID).Column("OrderID").Length(20).GeneratedBy.Assigned();
Map( x => x.OrderName).Column("OrderName")
....
}
internal sealed class OrderAppendixMap : ClassMap<OrderAppendix>
{
Table("OrderAppendix");
CompositeId()
.KeyProperty(x => x.ID, "ID")
....
Map( x => x.ID).Column("ID);
...
}
I won't muddy this up with my futile attempts at joining these tables, but what I would like to do is query by things like OrderType or OrderLocation given that the results all have the same Value from the OrderAppendix table.
Example desired SQL
SELECT * FROM ORDER
INNER JOIN
ON Order.OrderID = OrderAppendix.ID
WHERE OrderAppendix.Key = "Stuff"
Edit
Here's where I've gotten by reading the documentation on "QueryOver"
http://nhibernate.info/doc/nh/en/index.html#queryqueryover
Order Order = null;
OrderAppendix OrderAppendix = null;
resultList = session.QueryOver<Order>(() => Order)
.JoinAlias(() => Order.OrderAppendix, () => OrderAppendix)
.Where(() => OrderAppendix.Key == "MatchThis")
.List();
I think I'm on the right track using aliases to join the table, but obviously I haven't found a way to inform NHibernate of the many-to-one mapping that's needed. Also, you can see that I've added a property of type OrderAppendix to Order in order to the use the alias functionality.
Well, I didn't get why you did not used standard parent-child approach using HasMany on Order side and Reference of Appendix. This is because of composite key on appendix side?
If you would do that, you will be able to provide following HQL:
SELECT o FROM Order o
LEFT JOIN FETCH o.apxs a
WHERE a.Key="Stuff"

Trying to Query RavenDB Entity with a Collection via Index

I have a RavenDB mvc applicaton that has a document entity called Member. Each Member document has a list of users that are considered administrators. Only they can view and manage that Member document. On one of the pages I have a member search and have created an index to assist in the search.
public class Members_ByName : AbstractIndexCreationTask<Member>
{
public Members_ByName()
{
Map = members => from member in members select new {member.Title};
Indexes.Add(x => x.Title, FieldIndexing.Analyzed);
Sort(x => x.Title, SortOptions.String);
}
}
public class UserReference
{
public string Id { get; set; }
public string Name { get; set; }
}
public class Member
{
public string Id { get; set; }
public string Title { get; set; }
public ICollection<UserReference> Administrators { get; set; }
}
Since the user can only view/manage Member documents where they are an Administrator when I do the following to get the members
RavenQueryStatistics stats;
var query = RavenSession.Query<Member, Members_ByName>().Statistics(out stats);
query = query.Where(x => x.Title.StartsWith("anything"));
query = query.Where(x => x.Administrators.Any(y => y.Id == CurrentUser.Id));
var list = query.OrderBy(x => x.Title).Paging(CurrentPage, Configuration.DefaultPage, CurrentPageSize).ToList();
When the above code runs I get "The field 'Administrators_Id' is not indexed, cannot query on fields that are not indexed" which I understand but every thing I have attempted to get Administrator's Id in the index has not worked and not sure how to make it work at this point.
Try this:
Map = members => from member in members
select new {member.Title, Administrators_Id = members.Administrators.Select(x=>x.Id)};

Entity Framework Code First Class with parent and children of same type as it's own class

I have a class of Content which should be able to have a parentId for inheritance but also I want it to have a list of child content which is nothing to do with this inheritance tree.
I basically wanted a link table as ChildContentRelationship with Id's for parentContent and childContent in it and the Content class would have a list of ChildContentRelationship.
This has caused a lot of errors.
Here's waht I sort of want to do
public class Content
{
public int Id { get; set; }
public int? ParentContentId { get; set; }
public virtual Content ParentContent { get; set; }
public string Name { get; set; }
public int ContentTypeId { get; set; }
public virtual ContentType ContentType { get; set; }
public virtual ICollection<Property> Properties { get; set; }
public virtual ICollection<ChildContentRelationship> ChildContent { get; set; }
}
How would I set this up in EF?
I am not sure if I understand your model correctly. Let's discuss the options.
For a moment I omit this additional entity ChildContentRelationship and I assume the ChildContent collection is of type ICollection<Content>.
Option 1:
I assume that ParentContent is the inverse property of ChildContent. It would mean that if you have a Content with Id = x and this Content has a ChildContent with Id = y then the ChildContents ParentContentId must always be x. This would only be a single association and ParentContent and ChildContent are the endpoints of this same association.
The mapping for this relationship can be created either with data annotations ...
[InverseProperty("ParentContent")]
public virtual ICollection<Content> ChildContent { get; set; }
... or with Fluent API:
modelBuilder.Entity<Content>()
.HasOptional(c => c.ParentContent)
.WithMany(c => c.ChildContent)
.HasForeignKey(c => c.ParentContentId);
I think this is not what you want ("...has nothing to do with..."). Consider renaming your navigation properties though. If someone reads Parent... and Child... he will very likely assume they build a pair of navigation properties for the same relationship.
Option 2:
ParentContent is not the inverse property of ChildContent which would mean that you actually have two independent relationships and the second endpoint of both relationships is not exposed in your model class.
The mapping for ParentContent would look like this:
modelBuilder.Entity<Content>()
.HasOptional(c => c.ParentContent)
.WithMany()
.HasForeignKey(c => c.ParentContentId);
WithMany() without parameters indicates that the second endpoint is not a property in your model class, especially it is not ChildContent.
Now, the question remains: What kind of relationship does ChildContent belong to? Is it a one-to-many or is it a many-to-many relationship?
Option 2a
If a Content refers to other ChildContents and there can't be a second Content which would refer to the same ChildContents (the children of a Content are unique, so to speak) then you have a one-to-many relationship. (This is similar to a relationship between an order and order items: An order item can only belong to one specific order.)
The mapping for ChildContent would look like this:
modelBuilder.Entity<Content>()
.HasMany(c => c.ChildContent)
.WithOptional(); // or WithRequired()
You will have an additional foreign key column in the Content table in your database which belongs to this association but doesn't have a corresponding FK property in the entity class.
Option 2b
If many Contents can refer to the same ChildContents then you have a many-to-many relationship. (This is similar to a relationship between a user and roles: There can be many users within the same role and a user can have many roles.)
The mapping for ChildContent would look like this:
modelBuilder.Entity<Content>()
.HasMany(c => c.ChildContent)
.WithMany()
.Map(x =>
{
x.MapLeftKey("ParentId");
x.MapRightKey("ChildId");
x.ToTable("ChildContentRelationships");
});
This mapping will create a join table ChildContentRelationships in the database but you don't need a corresponding entity for this table.
Option 2c
Only in the case that the many-to-many relationship has more properties in addition to the two keys (ParentId and ChildId) (for example something like CreationDate or RelationshipType or...) you would have to introduce a new entity ChildContentRelationship into your model:
public class ChildContentRelationship
{
[Key, Column(Order = 0)]
public int ParentId { get; set; }
[Key, Column(Order = 1)]
public int ChildId { get; set; }
public Content Parent { get; set; }
public Content Child { get; set; }
public DateTime CreationDate { get; set; }
public string RelationshipType { get; set; }
}
Now your Content class would have a collection of ChildContentRelationships:
public virtual ICollection<ChildContentRelationship> ChildContent
{ get; set; }
And you have two one-to-many relationships:
modelBuilder.Entity<ChildContentRelationship>()
.HasRequired(ccr => ccr.Parent)
.WithMany(c => c.ChildContent)
.HasForeignKey(ccr => ccr.ParentId);
modelBuilder.Entity<ChildContentRelationship>()
.HasRequired(ccr => ccr.Child)
.WithMany()
.HasForeignKey(ccr => ccr.ChildId);
I believe that you want either option 2a or 2b, but I am not sure.

Recursively get all children from a self referencing many to many object with ef-core

My first post on Stackoverflow. So Apologies in advance if I do not understand something here :)
I have a abstract class called ElementEntity.
Inside this class i have 2 many to many relations, ChildRelations and ParentRelations.
I need a list with id's from all the Children and grandchildren etc.. form a certain ElementEntity.
We don't know how deep this tree is so somthing like:
Context.Hospitals().Include(b => b.ChildRelations).ThenInclude(c => c.child).ThenInclude etc.. is not an option. ( Hospital is a derived ElementEntity )
What would be the best (most inexpensive) approach?
As indication a certain element could have hundreds of thousands of offspring objects.
public abstract class ElementEntity
{
public long Id {get; set;}
public virtual ICollection<BaseGroupParentChildRelation> ParentRelations { get; } = new List<BaseGroupParentChildRelation>();
public virtual ICollection<BaseGroupParentChildRelation> ChildRelations { get; } = new List<BaseGroupParentChildRelation>();
}
public class BaseGroupParentChildRelation
{
public long ParentId { get; set; }
public virtual ElementEntity Parent { get; set; }
public long ChildId { get; set; }
public virtual ElementEntity Child { get; set; }
}
Possible fix:
await projectDbContext.Set<BaseGroupParentChildRelation>().ToListAsync();
var Hotels = await projectDbContext.Hotels.Include(b =>
b.ChildRelations).ThenInclude(b => b.Child).ToListAsync();
Hotels now include children, grandchildren, great-grandchildren etc.. etc..