What I'm trying to accomplish is to make a base class that returns a general purpose IQueryable<T> and use that IQueryable from another class instance to filter further.
Es.
The base class return all records, and the calling class filters from there on.
This is the base class and getDealer_query is the method that returns the main IQueryable
public class Dealer
{
public DAL.DBConn_Nav db = null;
public G4.Elements.Identity oIdentity = null;
public Dealer()
{
oIdentity = new G4.Elements.Identity();
oIdentity.read();
}
public IQueryable<outerViewModel.NavisionDAL.vmDealer> getDealer_query()
{
IQueryable<vmDealer> dealer = (from d in db.DENTALICA_S_P_A__Customer
select new vmDealer()
{
idDealer = d.No_,
name = d.Name + (d.Name_2.Length > 0 ? " " + d.Name_2 : ""),
address = d.Address + (d.Address_2.Length > 0 ? " " + d.Address_2 : ""),
city = d.City,
zip = d.Post_Code,
county = d.County,
email = d.E_Mail,
phone = d.Phone_No_,
fax = d.Fax_No_,
vat = d.VAT_Registration_No_,
ssn = d.Fiscal_Code,
businessAreaCode = d.Business_Area_Code,
businessZoneCode = d.Business_Zone_Code,
clienteDiretto = (d.Cliente_Diretto == (byte)1 ? true : false),
aggancioOrdiniWeb = (d.Deposito_Agg__Ordini_Web == (byte)1 ? true : false),
aggancioOrdiniG4 = (d.Deposito_Agg__Ordini_G4 == (byte)1 ? true : false),
notUse = (d.Not_Use == (byte)1 ? true : false),
blocked = d.Blocked,
shipmentMethodCode = d.Shipment_Method_Code,
paymentTermsCode = d.Payment_Terms_Code,
paymentMethodCode = d.Payment_Method_Code,
shippingAgentCode = d.Shipping_Agent_Code,
customerPriceGroup = d.Customer_Price_Group,
customerDiscountGroup = d.Customer_Disc__Group,
identificaCliente = d.Identifica_Cliente
});
return dealer;
}
Here is the other Class Customer that calls the method from an instance of Dealer and finally Lists data.
public class Customer
{
G4.Elements.Identity oIdentity = null;
DAL.DBConn_Nav navDB = null;
public Customer()
{
//oIdentity = new G4.Elements.Identity();
//oIdentity.read();
}
///// <summary>
///// Recupera l'elenco dei clienti Navision (i Depositi pr G4) visibili ad un particolare utente
///// </summary>
///// <returns></returns>
public IQueryable<outerViewModel.NavisionDAL.vmDealer> getCustomerForOperator_query()
{
string idOperator = !string.IsNullOrEmpty(oIdentity.selectedId) ? oIdentity.selectedId : oIdentity.id;
using (navDB = new DAL.DBConn_Nav()) {
G4.Models.NavisionDAL.Dealer oD = new G4.Models.NavisionDAL.Dealer();
// Tutti i Clienti
IQueryable<outerViewModel.NavisionDAL.vmDealer> allCustomerList_query = oD.getDealer_query();
IQueryable<outerViewModel.NavisionDAL.vmDealer> customerForOperator_query = (from c in allCustomerList_query join cForOp in navDB.DENTALICA_S_P_A__SalesPerson_Customer on c.idDealer equals cForOp.Customer_No_
where cForOp.Salesperson_Code.Equals(idOperator)
&& cForOp.Starting_Date < DateTime.Now
&& (cForOp.Ending_Date.Equals(new DateTime(1753, 1, 1)) || cForOp.Ending_Date > DateTime.Now)
&& c.notUse.Equals(false)
select c).Distinct();
return customerForOperator_query;
}
}
public List<G4.outerViewModel.NavisionDAL.vmDealer> getCustomerForOperator()
{
using (navDB = new DAL.DBConn_Nav())
{
G4.Models.NavisionDAL.Dealer oD = new G4.Models.NavisionDAL.Dealer();
return getCustomerForOperator_query().ToList();
}
}
But it doesn't work! I get a generic "Object not set to an instance...".
Even if I try to debug I get nowhere cause the cursor doesn't stop.
To make it work I have to inherit the Dealer class from the Customer, but it is not what I want.
I think I'm missing something basic here.
Any Idea ?
EDIT - Simplified sample
I'm adding a simplified sample of the same code.
public class Dealer1
{
public DAL.DBConn_Nav db = null;
public Dealer1() {
}
public IQueryable<outerViewModel.NavisionDAL.vmDealer> getDealer_query()
{
IQueryable<vmDealer> dealer = (from d in db.DENTALICA_S_P_A__Customer
select new vmDealer()
{
idDealer = d.No_,
name = d.Name + (d.Name_2.Length > 0 ? " " + d.Name_2 : "")
});
return dealer;
}
}
public class Customer1
{
public DAL.DBConn_Nav db = null;
public Customer1()
{
}
public IQueryable<outerViewModel.NavisionDAL.vmDealer> getCustomerForOperator_query()
{
// All customers
G4.Models.NavisionDAL.Dealer1 oD = new G4.Models.NavisionDAL.Dealer1();
IQueryable<outerViewModel.NavisionDAL.vmDealer> allCustomerList_query = oD.getDealer_query().Where(i => i.name.StartsWith("AAA"));
return allCustomerList_query;
}
public List<G4.outerViewModel.NavisionDAL.vmDealer> getCustomerForOperator()
{
using (db = new DAL.DBConn_Nav())
{
return getCustomerForOperator_query().ToList();
}
}
}
There's nothing in your class Dealer that is setting the instance of db.
public DAL.DBConn_Nav db = null;
So even though in Customer you call using (navDB = new DAL.DBConn_Nav()) you are not passing navDB to Dealer so when you call getDealer_query() you get a null reference exception.
Based on your simplified sample I would recommend this code:
public class Dealer1
{
internal IQueryable<vmDealer> getDealer_query(DAL.DBConn_Nav db)
{
return
from d in db.DENTALICA_S_P_A__Customer
select new vmDealer()
{
idDealer = d.No_,
name = String.Join(" ", d.Name, d.Name_2)
};
}
}
public class Customer1
{
internal IQueryable<vmDealer> getCustomerForOperator_query(DAL.DBConn_Nav db)
{
return new Dealer1().getDealer_query(db).Where(i => i.name.StartsWith("AAA"));
}
public List<vmDealer> getCustomerForOperator()
{
using (var db = new DAL.DBConn_Nav())
{
return getCustomerForOperator_query(db).ToList();
}
}
}
This cleanly allows you to segregate responsibility, but it keeps you totally in control of the db instance - which you need to dispose of when done.
There shouldn't be a reason why allCustomerList_query is null.
To confirm that the linq statement in the getDealer_query method is working, I would change your code to look this:
List<outerViewModel.NavisionDAL.vmDealer> something = new List<outerViewModel.NavisionDAL.vmDealer>();
IQueryable<outerViewModel.NavisionDAL.vmDealer> test = something.AsQueryable();
return test;
In any case, to prevent the object ref error, you need to guarantee that an instance is returned, which, from the code example, I can't actually tell.
It's rather strange that you can't get a break point to stop inside the getDealer_query method. Are you running this in a test with multiple threads running? What could happen is that the other thread is having an unhandled exception raised, and thus the whole test stops. Try putting break points before return dealer in getDealer_query and on the line G4.Models.NavisionDAL.Dealer oD = new G4.Models.NavisionDAL.Dealer();. Assuming there are no other threads, the code has to stop on one of those.
Last note, looking at the code, I don't see where db is instantiated. If it isn't instantiated, you'll get an object ref error on this line:
`IQueryable<vmDealer> dealer = (from d in db.DENTALICA_S_P_A__Customer`
Related
I have a below method which simply unnest the list within list and and create merged record all sub lists
e.g.
testResultDTO similar to
{
dept_name :'',
dept_id:'',
personList :'<List>'
}
personList is list similat to
{
first_name :'',
Last_name:'',
workHistList:'<List>'
}
workHistList is similar to
{
task_name:'',
task_description:''}
`
And the method is below
public async Task<FileDTO> GetWorkList(InputParameter inputParameter)
{
TestResultDTO testResultDTO = await GetPersonWorkList(inputParameter);
var testDto = new List<TestDto>();
if (testResultDTO != null)
{
testDto = personWorkFile(testResultDTO);
}
return testDto;
}
The problem is in method personWorkFile
private List<testDTO> personWorkFile(testResultDTO testResultDTO)
{
List<testDTO> testDTO = new List<testDTO>();
if (testResultDTO.workList is null) return new();
foreach (var testItem in testResultDTO.workList)
{
if (testItem.personList != null && testItem.personList.Any())
{
for (var j = 0; j < testItem.personList.Count; j++)
{
List<testDTO> testDTOTemp = new List<testDTO>();
for (var i = 0; i < testItem.personList[j].workHistList.Count; i++)
{
testDTOTemp.Insert(i, mapper.Map(testItem.personList[j].workHistList[i], mapper.Map<testDTO>(testItem.personList[j])));
}
testDTO.AddRange(testDTOTemp);
}
}
}
return testDTO;
}
`
what it does is it create a finale list with record similar to
{dept_name,dept_id,first_name, last_name, task_name, task_description}
Now when i create a unit test for it, it always fail because, it always insert null in list testDTOTemp.insert
my content of test case code is below
mapperMock.Setup(m => m.Map<List<TestResultDTO>, List<TestDTO>>(testResultDTO)).Returns(testDTO);
//call actual service
var result = await workService.GetWorkList(inputParameter);
Can someone please help me understand why it is not able to insert any data testDTOTemp list
The mock replays the setup by matching the method arguments. In this particular case, it will try to match the method call with the specified instance of TestResultDTO.
Override the Equals() method in TestResultDTO to let the mock compare the argument in order to return as per the setup.
Ex:
public class TestResultDTO
{
public int Id {get; set;}
}
var dto01 = new TestResultDTO(){Id = 1};
var dto02 = new TestResultDTO(){Id = 1};
dto01.Equals(dto02); // This will result in false in spite of having same Ids
If the above class is modified to
public class TestResultDTO
{
public int Id {get; set;}
public override bool Equals(object other)
{
var right = other as TestResultDTO;
if(right == null) return false;
var left = this;
var result = left.Id == right.Id;
return result;
}
}
var dto01 = new TestResultDTO(){Id = 1};
var dto02 = new TestResultDTO(){Id = 1};
dto01.Equals(dto02); // This will result in true
I'm trying to make a master-detail Web Form working with Entity Framework and performing insert and update on the same page. I'm new at EF, so I must be making a lot of mistakes here. Can you help me pointing me what's the best practices to perform insert/update on EF? What am I doing wrong here?
In this code, the "New" mode works well, but the "Edit" mode gets this error: "An entity object cannot be referenced by multiple instances of IEntityChangeTracker".
OrdersEntities ordersEntities = new OrdersEntities();
private Order myOrder
{
get { return (Order)Session["myOrder"]; }
set { Session["myOrder"] = value; }
}
public DataTable dtOrderDetails
{
get { return (DataTable)ViewState["dtOrderDetails"]; }
set { ViewState["dtOrderDetails"] = value; }
}
private string Mode
{
get { return (string)ViewState["mode"]; }
set { ViewState["_modo"] = value; }
}
private void btnSaveOrder_Click(object sender, EventArgs e)
{
if (dtOrderDetails.Rows.Count > 0)
{
using (ordersEntities)
{
using (var contextTransaction = ordersEntities.Database.BeginTransaction())
{
try
{
if (Mode == "New")
{
Order newOrder = new Order();
OrderDetails newOrderDetails;
int maxOrderNumber = ordersEntities.Order.Select(o => o.OrderNumber).DefaultIfEmpty(0).Max();
maxOrderNumber++;
newOrder.OrderNumber = maxOrderNumber;
newOrder.Date = DateTime.ParseExact(txtOrderDate.Text, "dd/MM/yyyy", CultureInfo.InvariantCulture);
newOrder.CustomerID = Convert.ToInt32(ddlCustomer.SelectedValue);
newOrder.Status = 1;
ordersEntities.Orders.Add(newOrder);
foreach (DataRow dt in dtOrderDetails.Rows)
{
newOrderDetails = new OrderDetails();
newOrderDetails.OrderNumer = maxOrderNumber;
newOrderDetails.ProductId = Convert.ToInt32(dt["ProductId"]);
newOrderDetails.Quantity = Convert.ToInt32(dt["Quantity"]);
ordersEntities.OrderDetails.Add(newOrderDetails);
}
ordersEntities.SaveChanges();
contextTransaction.Commit();
myOrder = newOrder;
}
if (Mode == "Edit")
{
Order editedOrder = myOrder;
OrderDetails editedOrderDetails;
editedOrder.Date = DateTime.ParseExact(txtOrderDate.Text, "dd/MM/yyyy", CultureInfo.InvariantCulture);
editedOrder.CustomerID = Convert.ToInt32(ddlCustomer.SelectedValue);
ordersEntities.Order.Attach(editedOrder);
ordersEntities.Entry(editedOrder).State = System.Data.Entity.EntityState.Modified;
editedOrder.OrderDetails.Clear();
foreach (DataRow dt in dtOrderDetails.Rows)
{
editedOrderDetails = new OrderDetails();
editedOrderDetails.OrderNumer = editedOrder.OrderNumber;
editedOrderDetails.ProductId = Convert.ToInt32(dt["ProductId"]);
editedOrderDetails.Quantity = Convert.ToInt32(dt["Quantity"]);
ordersEntities.OrderDetails.Add(editedOrderDetails);
}
ordersEntities.SaveChanges();
contextTransaction.Commit();
}
}
catch (Exception ex)
{
contextTransaction.Rollback();
}
}
}
}
}
Here is how you should approach it.
It would be best if you abstract the DbContext away, with this simple interface:
public interface IDataRepository : IDisposable
{
IDbSet<Order> Orders { get; set; }
void Save();
}
Of course, your implementation of IDataRepository is based on EntityFramework. Note that you will need to have a dataRepositoryConnection connection string in your web.config file:
public class EfDataRepository : DbContext, IDataRepository
{
public EfDataRepository() : base("dataRepositoryConnection")
{
}
public IDbSet<Order> Orders { get; set; }
public void Save()
{
this.SaveChanges();
}
}
In my experience, you also need a 'factory', which gives you a new instance of the data repository. This allows you to be the 'owner' of the instance, and you can safely dispose it. Note that the interaction with the DataContext should be minimal - you do your Unity of Work and get rid of it. Don't reuse! You will see it as an example below.
public class DataRepositoryFactory<T> where T : IDataRepository
{
private Type dataRepositoryImplementationType;
public DataRepositoryFactory(T dataRepositoryImplementation)
{
if (dataRepositoryImplementation == null)
{
throw new ArgumentException("dataRepositoryImplementation");
}
this.dataRepositoryImplementationType = dataRepositoryImplementation.GetType();
}
public T Create()
{
return (T)Activator.CreateInstance(this.dataRepositoryImplementationType);
}
}
In your controller (if it were MVC app), or Page backend (forms), it would be best if you use Microsoft Unity to get an instance of DataRepositoryFactory. For now, a manual construction would suffice too.
IDataRepository dataRepository = new EfDataRepository();
var dataRepositoryFactory = new DataRepositoryFactory<IDataRepository>(dataRepository);
Also, you don't need all this Transaction/Commit stuff you have put. It should be transparent for you. EF supports it implicitly, you don't have to be explicit about it.
// See, now you are the 'owner' of the dataRepository
using (var dataRepository = this.dataRepositoryFactory.Create())
{
if (Mode == "New")
{
Order newOrder = new Order();
// This doesn't make sense. Either generate a random order number (e.g. a Guid), or just use the Order.Id as an order number, although I don't recommend it.
int maxOrderNumber = dataRepository.Orders.Select(o => o.OrderNumber).DefaultIfEmpty(0).Max();
maxOrderNumber++;
newOrder.OrderNumber = maxOrderNumber;
newOrder.Date = DateTime.ParseExact(txtOrderDate.Text, "dd/MM/yyyy", CultureInfo.InvariantCulture);
newOrder.CustomerID = Convert.ToInt32(ddlCustomer.SelectedValue);
newOrder.Status = 1;
dataRepository.Orders.Add(newOrder);
foreach (DataRow dt in dtOrderDetails.Rows)
{
OrderDetails newOrderDetails = new OrderDetails();
newOrderDetails.OrderNumer = maxOrderNumber;
newOrderDetails.ProductId = Convert.ToInt32(dt["ProductId"]);
newOrderDetails.Quantity = Convert.ToInt32(dt["Quantity"]);
newOrder.OrderDetails.Add(newOrderDetails);
}
myOrder = newOrder;
}
if (Mode == "Edit")
{
Order editedOrder = dataRepository.Orders.FirstOrDefault(o => o.Id == myOrder.Id);
editedOrder.Date = DateTime.ParseExact(txtOrderDate.Text, "dd/MM/yyyy", CultureInfo.InvariantCulture);
editedOrder.CustomerID = Convert.ToInt32(ddlCustomer.SelectedValue);
editedOrder.OrderDetails.Clear();
foreach (DataRow dt in dtOrderDetails.Rows)
{
OrderDetails editedOrderDetails = new OrderDetails();
editedOrderDetails.OrderNumer = editedOrder.OrderNumber;
editedOrderDetails.ProductId = Convert.ToInt32(dt["ProductId"]);
editedOrderDetails.Quantity = Convert.ToInt32(dt["Quantity"]);
editedOrder.OrderDetails.Add(editedOrderDetails);
}
}
dataRepository.Save();
}
Also, I am pretty sure you have setup the relation between Order and OrderDetails classes incorrectly, in your EF code-first approach.
This is just wrong:
OrderDetails newOrderDetails = new OrderDetails();
newOrderDetails.OrderNumer = maxOrderNumber;
If you post them here, I can fix them for you.
I tried to compare two lists by using the Except method. But when I did, I got an error saying:
Cannot convert from 'Systems.Collections.Generic.List<>' to 'System.Linq.IQueryable<>'
'System.Collections.Generic.List<> does not contain a definition for 'Except' and the best extension method overload 'System.Linq.Queryable.Except(System.Linq.IQueryable, System.Collections.GEneric.IEnumerable)' has some invalid arguments
I also experienced this when I tried Intersect. I'm trying to compare Sent list and Result list (code and list shown below) and return items that does not have any match. So when I googled for how to do so, I came across the Except method as well as the Intersect.
public class Sent
{
public string Address;
public string Data;
}
public class Result
{
public string AddressOK;
public string DataOK;
}
var sent = new List<Sent>();
sent.Add(new Sent() { Address = linaddr1, Data = lindat1 });
var res = new List<Result>();
res.Add( new Result() { AddressOK = linaddr2, DataOK = lindat2 } );
//linaddr1 and 2, lindat1 and 2 contains the address and data shown in the list below
//taken from another part of the entire program
The lists look like such:
sent res
Address Data Address Data
04004C 55AA55 04004C 55AA55
040004 0720 040004 0720
040037 30
04004A FFFF 04004A FFFF
I only tried using this code:
var diff = sent.Except(res).ToList()
but as I've mentioned, it results with the aforementioned errors above.
EDIT: I edited the list. Sorry for that. It's just only a matter of the res list missing one or two or more items from the original list and then comparing both lists to see which item/s is/are missing from the res list.
Use Any:
using System;
using System.Collections.Generic;
using System.Linq;
public class Program
{
public static void Main()
{
var sent = new List<Sent>()
{
new Sent { Address = "04004C", Data = "55AA55" },
new Sent { Address = "040004", Data = "0720" },
new Sent { Address = "040037", Data = "31" },
new Sent { Address = "04004A", Data = "FFFF" }
};
var res = new List<Result> () {
new Result { AddressOK = "04004C", DataOK = "55AA55" },
new Result { AddressOK = "040004", DataOK = "0721" },
new Result { AddressOK = "040038 ", DataOK = "31" },
new Result { AddressOK = "04004A", DataOK = "FFFF" }
};
var diff =
sent.Where (s => !res.Any (r => s.Address == r.AddressOK && s.Data == r.DataOK ));
foreach (var item in diff)
{
Console.WriteLine("{0} {1}", item.Address, item.Data);
}
}
}
public class Sent
{
public string Address;
public string Data;
}
public class Result
{
public string AddressOK;
public string DataOK;
}
Output:
040004 0720
040037 31
Live Code: https://dotnetfiddle.net/ZVuiPd
The types Sent and Result are distinct types, but sent.Except(res) expects them to be the same. That's your first mistake.
The following is a simple (but incorrect) fix:
var diff =
sent
.Except(res.Select(x => new Sent() { Address = x.AddressOK, Data = x.DataOK }))
.ToList();
Even though this compiles, and runs, it doesn't remove the duplicates because your Sent doesn't override GetHashCode and Equals, hence it only compares references and not the actual properties.
You can either implement GetHashCode and Equals, or create an IEqualityComparer<Sent> to get this to work.
An IEqualityComparer<Sent> implementation might look like this:
public class SentEqualityComparer : IEqualityComparer<Sent>
{
public int GetHashCode(Sent sent)
{
return sent.Address.GetHashCode() ^ sent.Data.GetHashCode();
}
public bool Equals(Sent left, Sent right)
{
return (left.Address == right.Address) && (left.Data == right.Data);
}
}
And you would use it like so:
var diff =
sent
.Except(
res.Select(x => new Sent() { Address = x.AddressOK, Data = x.DataOK }),
new SentEqualityComparer())
.ToList();
This works as you expect.
The other option, to override GetHashCode and Equals, comes with an additional hurdle. The result of GetHashCode should not ever change throughout the lifetime of the object otherwise you can't use the object in a dictionary or any other data structure that relies on the hash code.
So, to make it work, you need to change Address & Data to be read-only.
Here is an implementation of your Sent class that will work correctly:
public sealed class Sent : IEquatable<Sent>
{
private readonly string _Address;
private readonly string _Data;
public string Address { get { return _Address; } }
public string Data { get { return _Data; } }
public Sent(string Address, string Data)
{
_Address = Address;
_Data = Data;
}
public override bool Equals(object obj)
{
if (obj is Sent)
return Equals((Sent)obj);
return false;
}
public bool Equals(Sent obj)
{
if (obj == null) return false;
if (!EqualityComparer<string>.Default.Equals(_Address, obj._Address)) return false;
if (!EqualityComparer<string>.Default.Equals(_Data, obj._Data)) return false;
return true;
}
public override int GetHashCode()
{
int hash = 0;
hash ^= EqualityComparer<string>.Default.GetHashCode(_Address);
hash ^= EqualityComparer<string>.Default.GetHashCode(_Data);
return hash;
}
}
If you are comfortable using an AOP component to automate the manual code of implementing IEquatable, another approach would be is to use Equals.Fody:
using System;
using System.Collections.Generic;
using System.Linq;
public class Program
{
public static void Main()
{
var a = new Sent { Address = "04004C", Data = "55AA55" };
var b = new Sent { Address = "04004C", Data = "55AA55" };
Console.WriteLine(a.Equals(b)); // True with use of an AOP, False with no AOP
var sent = new List<Sent>() {
new Sent { Address = "04004C", Data = "55AA55" },
new Sent { Address = "040004", Data = "0720" },
new Sent { Address = "040037", Data = "31" },
new Sent { Address = "04004A", Data = "FFFF" }
};
var res = new List<Result>() {
new Result { AddressOK = "04004C", DataOK = "55AA55" },
new Result { AddressOK = "040004", DataOK = "0721" },
new Result { AddressOK = "040038 ", DataOK = "31" },
new Result { AddressOK = "04004A", DataOK = "FFFF" }
};
var diff =
sent.Except(
res.Select(r => new Sent { Address = r.AddressOK, Data = r.DataOK })
);
foreach (var item in diff)
Console.WriteLine("{0} {1}", item.Address, item.Data);
}
}
[Equals]
public class Sent
{
public string Address;
public string Data;
[CustomEqualsInternal]
bool CustomLogic(Sent other)
{
return other.Address == this.Address && other.Data == this.Data;
}
}
public class Result
{
public string AddressOK;
public string DataOK;
}
Output:
True
040004 0720
040037 31
If you'll do map Result to Sent very often, you can further shorten your Linq query code to..
var diff = sent.Except(res.Select(r => (Sent)r));
..by automating the mapping of Result to Sent, use implicit operator:
[Equals]
public class Sent
{
public string Address;
public string Data;
[CustomEqualsInternal]
bool CustomLogic(Sent other)
{
return other.Address == this.Address && other.Data == this.Data;
}
public static implicit operator Sent(Result r)
{
return new Sent { Address = r.AddressOK, Data = r.DataOK };
}
}
#Kurisuchin
Suppose you have 2 list and in both you have ID property based on which you want to compare both list and want to store non matching item in third list.
In this Situation following Linq Query can help out.
var result = List2.Where(p => !List1.Any(p2 => p2.ID == p.ID)).ToList();
We are using entity framework codefirst approach
I am new to entity framework and I am facing error while trying to do "New transaction is not allowed because there are other threads running in the session.
public class DatabaseBackup : IDataBackup
{
private readonly IMonarchDbContext m_db;
public DatabaseBackup(IMonarchDbContext podb)
{
if (podb == null)
throw new ArgumentNullException("podb");
m_db = podb;
}
public DBBackupHistory GetLatestBackupHistory(DBBackupFrequency backupFrequency = DBBackupFrequency.Periodic)
{
DBBackupHistory result = null;
// get the backup history of the given backuptype and populate the objects
var configId = m_db.DBBackupConfigurations.Where(c => c.ScheduleType == (int)backupFrequency && c.BackupStatus == 1).Distinct().Select(c => c.ConfigurationId).DefaultIfEmpty(-1).First();
if (configId > 0)
{
result = m_db.DBBackupHistorys.Where(b => b.Status == 1 && b.ConfigurationId == configId).OrderByDescending(lb => lb.BackupDatetime).FirstOrDefault();
}
return result;
}
public IEnumerable<DBBackupConfiguration> GetAllConfiguration()
{
var result = m_db.DBBackupConfigurations.Where(c => c.BackupStatus == 1).OrderByDescending(c => c.ConfigurationId);
return result;
}
public void Backup(DBBackupConfiguration config, int fileIndex)
{
Console.WriteLine("Running DB Backup type {0} to device {1}", (DBBackupType)config.BackupType, fileIndex);
m_db.StoredProc.SPBackup(config, fileIndex);
}
I am calling the below methods in another class as follows
private readonly IDataBackup m_dataBackup;
public int PerformBackup(int defaultPollIntervalInMinutes = 15)
{
// polling interval in Minutes
int pollInterval = defaultPollIntervalInMinutes;
int fileIndex = getCurrentDumpFileIndex();
// check for the backup configuration
var configurations = m_dataBackup.GetAllConfiguration();
foreach (var config in configurations)
{
var lastBackup = m_dataBackup.GetLatestBackupHistory(DBBackupFrequency.Weekly);
if (lastBackup == null)
{
m_dataBackup.Backup(config, fileIndex + 1);
break;
}
Here is the Db Context class is as below
public class MonarchDbContext:DbContext,IMonarchDbContext
{
private IStoredProcedure m_storedProc;
private static object m_dbIntializerSet;
public MonarchDbContext(string nameOrConnectionString)
: base( nameOrConnectionString )
{
//-- Set the DB initializer only once.
System.Threading.LazyInitializer.EnsureInitialized( ref m_dbIntializerSet,()=>{
Database.SetInitializer<MonarchDbContext>(null);
//-- Give debug builds a chance to overwrite the above.
_SetInitializerForDebugBuilds();
return new object();
});
Configuration.LazyLoadingEnabled = false;
Configuration.ProxyCreationEnabled = false;
var csb = new SqlConnectionStringBuilder( this.Database.Connection.ConnectionString );
csb.MultipleActiveResultSets = true;
this.Database.Connection.ConnectionString = csb.ToString();
var objectContext = ( this as IObjectContextAdapter ).ObjectContext;
objectContext.CommandTimeout = 3600;
}
#region Public "Tables"
public IDbSet<DBBackupConfiguration> DBBackupConfigurations { get; set; }
public IDbSet<DBBackupHistory> DBBackupHistorys { get; set; }
public IStoredProcedure StoredProc
{
get
{
return System.Threading.LazyInitializer.EnsureInitialized(ref m_storedProc, () => new BackupStoredProc(this.Database));
}
}
#endregion
please let me know how can i solve the issue.
I found the issue
I need to add toList() at the end of the Linq code and it just worked for me.
public IEnumerable<DBBackupConfiguration> GetAllConfiguration()
{
var result = m_db.DBBackupConfigurations.Where(c => c.BackupStatus == 1).OrderByDescending(c => c.ConfigurationId).ToList();
return result;
}
Just add the List to Ienumerbale types
I have a base class called BaseStatus which looks like this:
public class BaseStatus
{
public int UnitId { get; protected set; }
public UInt16 StatusValue { get; protected set; }
public string StatusCode { get; protected set; }
public string StatusDescription { get; protected set; }
public BaseStatus()
{
this.UnitId = -1;
this.StatusValue = 0;
this.StatusCode = null;
this.StatusDescription = null;
}
}
Furthermore i have two or more other base classes which derive from BaseStatus and define a other unit id. For example the two classes
public class BaseGlobalStatus : BaseStatus
{
public BaseGlobalStatus()
{
base.UnitId = -1;
}
}
public class BaseGcmGdmStatus : BaseStatus
{
public BaseGcmGdmStatus()
{
base.UnitId = 2;
}
}
public class BaseCcuStatus : BaseStatus
{
public BaseCcuStatus()
{
base.UnitId = 1;
}
}
The Background is that i want to derive from for example BaseCcuStatus and have the correct UnitId in the derived class.
Now i define my correct status classes for example:
public class StatStErrDefinition : BaseGlobalStatus
{
public StatStErrDefinition()
: base()
{
base.StatusDescription = "Kommando nicht zulässig, unit im state ERROR";
base.StatusCode = "STAT_ST_ERR";
base.StatusValue = 3;
}
}
public class GcmStErrDefinition : BaseGcmGdmStatus
{
public GcmStErrDefinition()
: base()
{
base.StatusDescription = "Kommando nicht zulässig, unit im state ERROR";
base.StatusCode = "STAT_ST_ERR";
base.StatusValue = 3;
}
}
public class CcuStErrDefinition : BaseCcuStatus
{
public CcuStErrDefinition()
: base()
{
base.StatusDescription = "Kommando nicht zulässig, unit im state ERROR";
base.StatusCode = "STAT_ST_ERR";
base.StatusValue = 3;
}
}
For my understading, the three classes StatStErrDefinition, GcmStErrDefinition and CcuStErrDefinition should have the UnitId which is set in the derived BaseClass?
Now that i have defined my three Status Classes i want to get them into a registry. Currently im using this piece of code to try get them. Problem is that the result has no items.
registry = new StatusDictionary<UInt16, BaseStatus>();
var unitStatus = typeof(BaseStatus)
.Assembly.GetTypes()
.Where(x => x.BaseType == typeof(BaseStatus))
.Select(x => new
{
StatusType = x,
UnitId = x.GetProperty("UnitId", BindingFlags.Public)
StatVal = x.GetProperty("StatusValue", BindingFlags.Public)
}
)
.Where(x => x.StatVal != null && x.UnitId != null)
.Select(x => new
{
UnitId = (int)x.UnitId.GetValue(null, null),
StatusValue = (UInt16)x.StatVal.GetValue(null, null),
Factory = (Func<BaseStatus>)(() => ((BaseStatus)Activator.CreateInstance(x.StatusType)))
});
try
{
foreach (var status in unitStatus)
{
if (status.UnitId == unitId
|| status.UnitId < 0)
registry.Register(status.StatusValue, status.Factory);
}
}
catch (Exception ex)
{
string temp = ex.Message;
}
After the LINQ expression the var unitStatus is empty...
Later, the registry call looks like that to get the specific class but that is unimportant at this point:
stat = StatusContainer.GetRegistry(this.unitTypeId).GetInstance(this.StatusValue);
For information:
I want to get the status class which should be in the registry by the unittypeid and the specific status value.
Currently my registry method does not work because he is not able to find any class. So there has to be a mistake somewhere. Thanks in advance
#Update 1
I changed my functionality a little bit:
registry = new StatusDictionary<UInt16, BaseStatus>();
//get all types of cucrent assembly
var allAssemblyTypes = Assembly.GetCallingAssembly().GetTypes();
//get all types from base status
var baseStatusTypes = allAssemblyTypes.Where(x => x.BaseType == typeof(BaseStatus));
//Place all concrete types in the foundtypes
List<Type> foundTypes = new List<Type>();
foreach (Type item in baseStatusTypes)
{
var temp = allAssemblyTypes.Where(x => x.BaseType == item)
.Select(x => new
{
StatusType = x,
UnitId = x.GetProperty("UnitId", BindingFlags.Public),
StatVal = x.GetProperty("StatusValue", BindingFlags.Public),
}
);
}
Temp contains now the correct type.
Problem is that if temp is type of StatStErrDefinition the StatusValue and UnitId Property is null.
The fact is that these members are instance members. Is there a way to get the values out of them?
First thing first : your LINQ query is pretty long.
divide it in different step and store them in different variables (or make properties out of them, whatever you prefer)
This is
easy to read / maintain
easy to debug
With this given I think you are able to solve your problem :)
To check if the class is of a certain type you could use the method .OfType
Use this method to get the value. Notice that you must make an instance in your case because the value change in your constructor.
public static object GetPropValue(Type src, string propName)
{
var prop = src.GetProperty(propName);
var instance = Activator.CreateInstance(src);
var value = prop.GetValue(instance);
return value;
}
Instead of
UnitId = x.GetProperty("UnitId", BindingFlags.Public),
use
UnitId = GetPropValue(x,"UnitId"),