POST EDITED - see edit below
I have a query about the FLuent Automapping which is used as part of the SHarp Architecture. Running one of the tests cases will generate a schema which I can use to create tables in my DB.
I'm developing a site with Posts, and Tags associated with these posts. I want a tag to be able to be associated with more than one post, and for each post to have 0 or more tags.
I wanting to achieve a DB schema of:
Post {Id, Title, SubmitTime, Content}
Tag {Id, Name}
PostTag {PostId, TagId}
Instead, I'm getting:
Post {Id, Title, SubmitTime, Content}
Tag {Id, Name, PostID (FK)}
I'm using sharp architecture, and may classes look as follows (more or less):
public class Post : Entity
{
[DomainSignature]
private DateTime _submittime;
[DomainSignature]
private String _posttitle;
private IList<Tag> _taglist;
private String _content;
public Post() { }
public Post(String postTitle)
{
_submittime = DateTime.Now;
_posttitle = postTitle;
this._taglist = new List<Tag>();
}
public virtual DateTime SubmitTime { get { return _submittime; } private set { _submittime = value; } }
public virtual string PostTitle { get { return _posttitle; } private set { _posttitle = value; } }
public virtual string Content { get { return _content; } set { _content = value; } }
public virtual IList<Tag> TagList { get { return _taglist; } set { _taglist = value; } }
public class Tag : Entity
{
[DomainSignature]
private String _name;
public Tag() { }
public Tag(String name)
{
this._name = name;
}
public virtual String Name
{
get { return _name; }
private set { _name = value; }
}
public virtual void EditTagName(String name)
{
this.Name = name;
}
}
I can see why it's gone for the DB schema set up that it has, as there will be times when an object can only exist as part of another. But a Tag can exist separately.
How would I go about achieving this? I'm quite new to MVC, Nhibernate, and SHarp architecture, etc, so any help would be much appreciated!
EDIT*
OK, I have now adjusted my classes slightly. My issue was that I was expecting the intermediate table to be inferred. Instead, I realise that I have to create it.
So I now have (I've simplified the classes a bit for readability's sake.:
class Post : Entity
{
[DomainSignature]
String Title
[DomainSignature]
DateTime SubmitTime
IList<PostTag> tagList
}
class Tag : Entity
{
[DomainSignature]
string name
}
class PostTag : Entity
{
[DomainSignature]
Post post
[DomainSignature]
Tag tag
}
This gives me the schema for the intermediate entity along with the usual Post and Tag tables:
PostTag{id, name, PostId(FK)}
The problem with the above is that it still does not include The foreign key for Tag. Also, should it really have an ID column, as it is a relational table? I would think that it should really be a composite key consisting of the PK from both Post and Tag tables.
I'm sure that by adding to the Tag class
IList<PostTag> postList
I will get another FK added to the PostTag schema, but I don't want to add the above, as the postList could be huge. I don't need it every time I bring a post into the system. I would have a separate query to calculate that sort of info.
Can anyone help me solve this last part? Thanks for your time.
Ok, I'd been led to believe that modelling the composite class in the domain was the way forward, but I finally come across a bit of automapper override code which creates the composite table without me needing to create the class for it, which was what I was expecting in the first place:
public class PostMappingOverride
: IAutoMappingOverride
{
public void Override(AutoMapping map)
{
map.HasManyToMany(e => e.TagList)
.Inverse()
.Cascade.SaveUpdate();
}
}
This will now give me my schema (following schema non simplified):
create table Posts (
Id INT not null,
PublishTime DATETIME null,
SubmitTime DATETIME null,
PostTitle NVARCHAR(255) null,
Content NVARCHAR(255) null,
primary key (Id)
)
create table Posts_Tags (
PostFk INT not null,
TagFk INT not null
)
create table Tags (
Id INT not null,
Name NVARCHAR(255) null,
primary key (Id)
)
alter table Posts_Tags
add constraint FK864F92C27E2C4FCD
foreign key (TagFk)
references Tags
alter table Posts_Tags
add constraint FK864F92C2EC575AE6
foreign key (PostFk)
references Posts
I think the thrower is that I've been looking for a one-to-many relationship, which it is, but it is called HasManytoMAny here...
Related
I'm using Dapper Extensions and have defined my own custom mapper to deal with entities with composite keys.
public class MyClassMapper<T> : ClassMapper<T> where T : class
{
public MyClassMapper()
{
// Manage unmappable attributes
IList<PropertyInfo> toIgnore = typeof(T).GetProperties().Where(x => !x.CanWrite).ToList();
foreach (PropertyInfo propertyInfo in toIgnore.ToList())
{
Map(propertyInfo).Ignore();
}
// Manage keys
IList<PropertyInfo> propsWithId = typeof(T).GetProperties().Where(x => x.Name.EndsWith("Id") || x.Name.EndsWith("ID")).ToList();
PropertyInfo primaryKey = propsWithId.FirstOrDefault(x => string.Equals(x.Name, $"{nameof(T)}Id", StringComparison.CurrentCultureIgnoreCase));
if (primaryKey != null && primaryKey.PropertyType == typeof(int))
{
Map(primaryKey).Key(KeyType.Identity);
}
else if (propsWithId.Any())
{
foreach (PropertyInfo prop in propsWithId)
{
Map(prop).Key(KeyType.Assigned);
}
}
AutoMap();
}
}
I also have this test case to test my mapper:
[Test]
public void TestMyAutoMapper()
{
DapperExtensions.DapperExtensions.DefaultMapper = typeof(MyClassMapper<>);
MySubscribtionEntityWithCompositeKey entity = new MySubscribtionEntityWithCompositeKey
{
SubscriptionID = 145,
CustomerPackageID = 32
};
using (var connection = new SqlConnection(CONNECTION_STRING))
{
connection.Open();
var result = connection.Insert(entity);
var key1 = result.SubscriptionID;
var key2 = result.CustomerPackageID;
}
}
Note that I set the default mapper in the test case.
The insert fails and I notive that my customer mapper is never called. I have no documentation on the github page on the topic, so I'm not sure if there's anything else I need to do to make dapper extensions use my mapper.
Thanks in advance!
Looking at your question, you are attempting to write your own defalut class mapper derived from the existing one. I never used this approach; so I do not know why it is not working or whether it should work.
I explicitly map the classes as below:
public class Customer
{
public int CustomerID { get; set; }
public string Name { get; set; }
}
public sealed class CustomerMapper : ClassMapper<Customer>
{
public CustomerMapper()
{
Schema("dbo");
Table("Customer");
Map(x => x.CustomerID).Key(KeyType.Identity);
AutoMap();
}
}
The AutoMap() will map rest of the properties based on conventions. Please refer to these two resources for more information about mapping.
Then I call SetMappingAssemblies at the startup of the project as below:
DapperExtensions.DapperExtensions.SetMappingAssemblies(new[] { Assembly.GetExecutingAssembly() });
The GetExecutingAssembly() is used in above code because mapping classes (CustomerMapper and other) are in same assembly which is executing. If those classes are placed in other assembly, provide that assembly instead.
And that's it, it works.
To set the dialect, I call following line just below the SetMappingAssemblies:
DapperExtensions.DapperExtensions.SqlDialect = new DapperExtensions.Sql.SqlServerDialect();
Use your preferred dialect instead of SqlServerDialect.
Apparently, the solution mentioned here may help you achieve what you are actually trying to. But, I cannot be sure, as I said above, I never used it.
I'm trying to create a simple schema using ReflectiveSchema and then trying to project an Employee "table" using Groovy as my programming language. Code below.
class CalciteDemo {
String doDemo() {
RelNode node = new CalciteAlgebraBuilder().build()
return RelOptUtil.toString(node)
}
class DummySchema {
public final Employee[] emp = [new Employee(1, "Ting"), new Employee(2, "Tong")]
#Override
String toString() {
return "DummySchema"
}
class Employee {
Employee(int id, String name) {
this.id = id
this.name = name
}
public final int id
public final String name
}
}
class CalciteAlgebraBuilder {
FrameworkConfig config
CalciteAlgebraBuilder() {
SchemaPlus rootSchema = Frameworks.createRootSchema(true)
Schema schema = new ReflectiveSchema(new DummySchema())
SchemaPlus rootPlusDummy = rootSchema.add("dummySchema", schema)
this.config = Frameworks.newConfigBuilder().parserConfig(SqlParser.Config.DEFAULT).defaultSchema(rootPlusDummy).traitDefs((List<RelTraitDef>)null).build()
}
RelNode build() {
RelBuilder.create(config).scan("emp").build()
}
}
}
I seem to be correctly passing in the "schema" object to the constructor of the ReflectiveSchema class, but I think its failing while trying to get the fields of the Employee class.
Here's the error
java.lang.StackOverflowError
at java.lang.Class.copyFields(Class.java:3115)
at java.lang.Class.getFields(Class.java:1557)
at org.apache.calcite.jdbc.JavaTypeFactoryImpl.createStructType(JavaTypeFactoryImpl.java:76)
at org.apache.calcite.jdbc.JavaTypeFactoryImpl.createType(JavaTypeFactoryImpl.java:160)
at org.apache.calcite.jdbc.JavaTypeFactoryImpl.createType(JavaTypeFactoryImpl.java:151)
at org.apache.calcite.jdbc.JavaTypeFactoryImpl.createStructType(JavaTypeFactoryImpl.java:84)
at org.apache.calcite.jdbc.JavaTypeFactoryImpl.createType(JavaTypeFactoryImpl.java:160)
at org.apache.calcite.jdbc.JavaTypeFactoryImpl.createStructType(JavaTypeFactoryImpl.java:84)
What is wrong with this example?
Seems that by just moving the Employee class a level above, ie. making it a sibling of the DummySchema class, makes the problem go away.
I think the way the org.apache.calcite.jdbc.JavaTypeFactoryImpl of Calcite is written doesn't handle Groovy's internal fields well.
This code works fine.
using (ContextDB db = new ContextDB())
{
var custAcct = (from c in db.CustAccts
select new
{
c.AcctNo,
c.Company,
c.UserName
}).ToList();
But this one doesn't
public class CustAcct
{
public int AcctNo { get; set; }
public string Company { get; set; }
public string UserName { get; set; }
}
....
....
....
using (ContextDB db = new ContextDB())
{
CustAcct custAcct = (from c in db.CustAccts
select new
{
c.AcctNo,
c.Company,
c.UserName
}).ToList();
It returns this error:
Cannot implicitly convert type 'System.Collections.Generic.IEnumerable' to 'EMailReader.Models.CustAcct'. An explicit conversion exists (are you missing a cast?)
I used Google, found many related topics but still couldn't put it to work using the available solutions
I just need to return data to a strong typed model.
EDITED:
After more research I found this solution bellow, but I wonder why I cannot retrieve directly in the list from LinqToSql.
List<CustAcct> temp = new List<CustAcct>();
IEnumerable<dynamic> items = custAcct;
foreach (var item in items)
{
temp.Add(new CustAcct()
{
AcctNo = item.AcctNo,
Company = item.Company,
UserName = item.UserName,
});
}
You are re defining those properties by creating new Class. And this will override LINQ2SQL generated class.
Just change "public class CustAcct" to "public partial class CustAcct".
This will solve your problem, and you do not need to define those properties again. Remove those from your class. Those will be automatically create for you.
If you can just post your class, and I will change it for you.
//Shyam
I have an aggregate named Campaigns every with a root entity named campaign, this root entity has a list of attempts (entity)
public class Attempts: IEntity<Attempts>
{
private int id;
public AttempNumber AttemptNumber {get;}
//other fields
}
public class Campaign: IEntity<Campaign> //root
{
private int id;
public IList<Attempt> {get;}
//other fields
}
Im using a method to add a campaign attempt
public virtual void AssignAttempts(Attempts att)
{
Validate.NotNull(att, "attemps are required for assignment");
this.attempts.add(att);
}
Problem comes when i try to edit a specific item in attempts list. I get Attempt by AttempNumber and pass it to editAttempt method but i dont know how to set the attempt without deleting whole list and recreate it again
public virtual void EditAttempts(Attempts att)
{
Validate.NotNull(att, "attemps are required for assignment");
}
Any help will be appreciated!
Thanks,
Pedro de la Cruz
First, I think there may be a slight problem with your domain model. It seems to me like 'Campaign' should be an aggregate root entity having a collection of 'Attempt' value objects (or entities). There is no 'Campaigns' aggregate unless you have a parent concept to a campaign which would contain a collection of campaigns. Also, there is no 'Attempts' entity. Instead a collection of 'Attempt' entities or values on the 'Campaign' entity. 'Attempt' may be an entity if it has identity outside of a 'Campaign', otherwise it is a value object. The code could be something like this:
class Campaign {
public string Id { get; set; }
public ICollection<Attempt> Attempts { get; private set; }
public Attempt GetAttempt(string id) {
return this.Attempts.FirstOrDefault(x => x.Number == id);
}
}
class Attempt {
public string Number { get; set; }
public string Attribute1 { get; set; }
}
If you retrieve an Attempt from the Campaign entity and then change some of the properties, you should not have to insert it back into the campaign entity, it is already there. This is how the code would look if you were using NHibernate (similar for other ORMs):
var campaign = this.Session.Get<Campaign>("some-id");
var attempt = campaign.GetAttempt("some-attempt-id");
attempt.Attribute1 = "some new value";
this.Session.Flush(); // will commit changes made to Attempt
You don't need an Edit method. Your code can modify the Attempts in-place, like so:
Attempt toModify = MyRepository.GetAttemptById(id);
toModify.Counter++;
toModify.Location = "Paris";
MyRepository.SaveChanges(); // to actually persist to the DB
Of course how you name the SaveChanges() is up to you, this is the way Entity Framework names its general Save method.
If I'm trying to serialize a normal CLR object, and I do not want a particular member variable to be serialized, I can tag it with the
[NonSerialized]
attribute. If I am creating a table services entity, is there an equivalent attribute I can use to tell Azure table services to ignore this property?
For Version 2.1 there is a new Microsoft.WindowsAzure.Storage.Table.IgnoreProperty attribute. See the 2.1 release notes for more information: http://blogs.msdn.com/b/windowsazurestorage/archive/2013/09/07/announcing-storage-client-library-2-1-rtm.aspx.
There's no equivalent I know of.
This post says how you can achieve the desired effect - http://blogs.msdn.com/b/phaniraj/archive/2008/12/11/customizing-serialization-of-entities-in-the-ado-net-data-services-client-library.aspx
Alternatively, if you can get away with using "internal" rather than "public" on your property then it will not get persisted with the current SDK (but this might change in the future).
For version 2.0 of the Table Storage SDK there is a new way to achieve this.
You can now override the WriteEntity method on TableEntity and remove any entity properties that have an attribute on them. I derive from a class that does this for all my entities, like:
public class CustomSerializationTableEntity : TableEntity
{
public CustomSerializationTableEntity()
{
}
public CustomSerializationTableEntity(string partitionKey, string rowKey)
: base(partitionKey, rowKey)
{
}
public override IDictionary<string, EntityProperty> WriteEntity(Microsoft.WindowsAzure.Storage.OperationContext operationContext)
{
var entityProperties = base.WriteEntity(operationContext);
var objectProperties = this.GetType().GetProperties();
foreach (PropertyInfo property in objectProperties)
{
// see if the property has the attribute to not serialization, and if it does remove it from the entities to send to write
object[] notSerializedAttributes = property.GetCustomAttributes(typeof(NotSerializedAttribute), false);
if (notSerializedAttributes.Length > 0)
{
entityProperties.Remove(property.Name);
}
}
return entityProperties;
}
}
[AttributeUsage(AttributeTargets.Property)]
public class NotSerializedAttribute : Attribute
{
}
Then you can make use of this class for your entities like
public class MyEntity : CustomSerializationTableEntity
{
public MyEntity()
{
}
public string MySerializedProperty { get; set; }
[NotSerialized]
public List<string> MyNotSerializedProperty { get; set; }
}