Is it possible to do the following in linq, where in my select new I use the value from TotalOrderItems as part of the calculation for TotalInStockItems?
var Order = (from o in orderItems
select new ShippingOrder { OrderId = orderItems.OrderId
TotalOrderItems = orderItems.GroupBy(x => x.Sku).Count(),
TotalInStockItems = TotalOrderItems - orderItems.Where(x => x.InStock =='F').GroupBy(x => x.Sku).Count(),
}).ToList();
try using the let clause (C# Reference) to store the value for use later in the query
(from o in orderItems
let totalOrderItems = orderItems.GroupBy(x => x.Sku).Count()
select new ShippingOrder {
OrderId = o.OrderId
TotalOrderItems = totalOrderItems,
TotalInStockItems = totalOrderItems - orderItems.Where(x => x.InStock =='F').GroupBy(x => x.Sku).Count(),
}).ToList();
Related
I'm converting from EF6 to EF Core 3.1 and this LINQ query is failing with a runtime exception stating 'The LINQ expression ... could not be translated.
The group by is what is causing the issue, but I'm not sure how to rewrite it to work with EF Core and keep the result in a nested list.
Notification notification = new Notification()
{
ProductReminders = new List<List<ProductNotification>>(),
ProductStats = new List<StatResult>()
};
var profileCode = 123;
notification.ProductReminders =
(from ng in ProductNotification
where ng.UserProfileCode == profileCode
orderby ng.EndDate ?? DateTime.MaxValue
group ng by ng.GroupGUID into groupG
select (from pn in ProductNotification
join p in Product on pn.ProductID equals p.ProductID
where pn.UserProfileCode == profileCode
&& pn.GroupGUID == groupG.Key
orderby pn.EndDate ?? DateTime.MaxValue
select new ProductNotification()
{
ProductDetail = new ProductDetail()
{
ProductId = pn.ProductID ?? 0,
Upc = p.UPC,
Brand = p.Description,
Manufacturer = p.Name,
ProfileCode = p.ProfileCode,
},
EndDate = pn.EndDate,
NotificationId = pn.NotificationID,
Status = pn.Status,
GroupGuid = pn.GroupGUID
})
.ToList())
.ToList();
Since grouping operator has limitations, I would suggest to read all needed data and provide grouping on the client side. Query in your case will be much effective:
// select only needed data from database
var minimalRequiredData =
from pn in ProductNotification
join p in Product on pn.ProductID equals p.ProductID
where pn.UserProfileCode == profileCode
select new ProductNotification
{
ProductDetail = new ProductDetail
{
ProductId = pn.ProductID ?? 0,
Upc = p.UPC,
Brand = p.Description,
Manufacturer = p.Name,
ProfileCode = p.ProfileCode,
},
EndDate = pn.EndDate,
NotificationId = pn.NotificationID,
Status = pn.Status,
GroupGuid = pn.GroupGUID
};
// materialize result
var materialized = minimalRequiredData.ToList();
// form required result shape using IEnumerable<T>
var resultQuery =
from m in materialized
orderby ng.EndDate ?? DateTime.MaxValue
group m by new m.GroupGUID into g
select g.Orderby(x => x.EndDate ?? ng.EndDate).ToList();
notification.ProductReminders = resultQuery.ToList();
I got a linq lambda select code that works before I added the Select index overload. Before, I got the list of records but I need the index which I use to assign a unique Id to each record. When I add with ToList(), I get an exception with no error/inner exception. Only way I can get the code to not throw an error is to use .AsEnumberable() but I need a list. I read many post that .ToList() works with the overload but I have been unsuccessful.
Here is my code and my attempt to fix this
var emps = this.DbContext.Employees
.GroupJoin(this.DbContext.Depts,
employee => employee.EmployeeId,
dept => dept.EmployeeId,
(employee, dept) => new { employee, dept }
)
.SelectMany(
employee_dept_left => employee_dept_left.dept.DefaultIfEmpty(),
(employee_dept_left, dept) => new { employee_dept_left, dept }
)
.Join(this.DbContext.Divs,
emp_emp_dept => emp_emp_dept.employee_dept_left.employee.DivId,
division => division.DivId,
(emp_emp_dept, division) => new { emp_emp_dept, division }
)
.Where(s => !string.IsNullOrEmpty(filter.selectedDiv))
.GroupBy(grouped => new
{
grouped.emp_emp_dept.employee_dept_left.employee.EmployeeId,
grouped.emp_emp_dept.employee_dept_left.employee.LastNm,
grouped.emp_emp_dept.employee_dept_left.employee.FirstNm,
grouped.emp_emp_dept.employee_dept_left.employee.DivId
})
.Select((joined, index) => new EmployeeViewModel
{
Id = index,
EmployeeId = joined.Key.EmployeeId,
LastNm = joined.Key.LastNm.Trim(),
FirstNm = joined.Key.FirstNm.Trim(),
DivisionId = joined.Key.DivId,
}).ToList();
The error message says
Could not be translated. Either rewrite the query in a form that can be translated, or switch to client evaluation explicitly by inserting a call to 'AsEnumerable', 'AsAsyncEnumerable', 'ToList', or 'ToListAsync'. See https://go.microsoft.com/fwlink/?linkid=2101038 for more information.
I tried using .AsEnumerable() instead of .ToList():
List<EmployeeViewModel> test = emps.Cast<EmployeeViewModel>().ToList();
but this throws an exception.
Any help is greatly appreciated.
Thanks in advance
Problem that this Select is not currently translatable to the SQL. You can make additional Select to solve issue with AsEnumerable().
...
.Select(joined => new
{
EmployeeId = joined.Key.EmployeeId,
LastNm = joined.Key.LastNm.Trim(),
FirstNm = joined.Key.FirstNm.Trim(),
DivisionId = joined.Key.DivisionId,
})
.AsEnumerable()
.Select((x, index) => new EmployeeViewModel
{
Id = index,
EmployeeId = x.EmployeeId,
LastNm = x.LastNm,
FirstNm = x.FirstNm,
DivisionId = x.DivisionId,
}).ToList();
And note that query is more readable in Query syntax when there are joins.
var query =
from employee in this.DbTracsContext.Employees
join dept in his.DbTracsContext.Depts on employee.EmployeeId equals dept.EmployeeId into employee_dept_left
from dept in employee_dept_left.DefaultIfEmpty()
join division in this.DbTracsContext.Depts on employee.DivisionId equals division.DivisionId
where string.IsNullOrEmpty(filter.DivisionSelection) || filter.DivisionSelection == "0" || employee.DivisionId == filter.DivisionSelection
group employee by new { employee.EmployeeId, employee.LastNm, employee.FirstNm, employee.DivisionId } into g
select new
{
EmployeeId = g.Key.EmployeeId,
LastNm = g.Key.LastNm.Trim(),
FirstNm = g.Key.FirstNm.Trim(),
DivisionId = g.Key.DivisionId,
};
var emps = query
.AsEnumerable()
.Select((x, index) => new EmployeeViewModel
{
Id = index,
EmployeeId = x.EmployeeId,
LastNm = x.LastNm,
FirstNm = x.FirstNm,
DivisionId = x.DivisionId,
}).ToList();
i have the following database-model: http://i.stack.imgur.com/gRtMD.png
the many to many relations for Kunde_Geraet/Kunde_Anwendung are in explicit Mapping-Table with additional Information.
i want to optimize the following LINQ-query:
var qkga = (from es in db.Eintrag_Systeme.Where(es => es.Eintrag_ID == id)
from kg in db.Kunde_Geraet.Where(kg => es.Geraet_ID == kg.Geraet_ID)
select new { Kunde = kg.Kunde, Geraet = es.Geraet, Anwendung = es.Anwendung })
.Union(
from es in db.Eintrag_Systeme.Where(es => es.Eintrag_ID == id)
from ka in db.Kunde_Anwendung.Where(ka => es.Anwendung_ID == ka.Anwendung_ID)
select new { Kunde = ka.Kunde, Geraet = es.Geraet, Anwendung = es.Anwendung })
.GroupBy(kga => kga.Kunde, kga => new {Geraet = kga.Geraet, Anwendung = kga.Anwendung});
it would be better, when the result is a IEnumerable(Kunde, IEnumerable(Geraet), IEnumerable(Anwendung)) without the null-Values for the union.
i try it as SQL command
select Count(es.Geraet_ID), null as Anwendung_ID
from Eintrag_Systeme es cross join Kunde_Geraet where es.Geraet_ID = Kunde_Geraet.Geraet_ID AND es.Eintrag_ID = #id
union
select null as Geraet_ID, Count(es.Anwendung_ID)
from Eintrag_Systeme es cross join Kunde_Anwendung where es.Anwendung_ID = Kunde_Anwendung.Anwendung_ID AND es.Eintrag_ID = #id
group by Kunde_ID
but don´t get the Count() of Anwendungen(Apps)/Geraete(Devices) to Lists grouped by Key Kunde(Client)
Don't use join but navigation properties:
from k in context.Kunden
select new
{
Kunde = k,
Geraete = k.Kunde_Geraete.Select(kg => kg.Geraet),
Anwendungen = k.Kunde_Anwendungen.Select(ka => ka.Anwendung)
}
Now you have a basis from which you get counts, etc.
The EntityModel is defined as:
Personnel has a link to a Country
When executing this code in LinqPad, I see that the SQL which is generated is not optimized (all fields are returned) in the first query ? What am I missing here or doing wrong ?
Query 1 LINQ
var Country = Countries.FirstOrDefault(o => o.Id == 100000581);
var personnelIds = Country.Personnels.Select(p => p.Id).ToArray();
personnelIds.Dump();
Query 1 SQL
exec sp_executesql N'SELECT [t0].[Id], [t0].[Version], [t0].[Identifier], [t0].[Name], , [t0].[UpdatedBy] FROM [Personnel] AS [t0] WHERE [t0].[Country_Id] = #p0',N'#p0 bigint',#p0=100000581
Query 2 LINQ
var Country = Countries.FirstOrDefault(o => o.Id == 100000581);
var personnelIds2 = Personnels.Where(p => p.Country == Country).Select(p => p.Id).ToArray();
personnelIds2.Dump();
Query 2 SQL
exec sp_executesql N'SELECT [t0].[Id] FROM [Personnel] AS [t0] WHERE [t0].[Country_Id] = #p0',N'#p0 bigint',#p0=100000581
The database used is SQL Express 2008. And LinqPad version is 4.43.06
//var Country = Countries.FirstOrDefault(o => o.Id == 100000581);
var personnelIds = context.Personnels
.Where(p => p.Country.Id == 100000581)
.Select(p => p.Id)
.ToArray();
personnelIds.Dump();
Try this, it should be better.
Personnels collection will be populated via lazy loading when accessed, hence retrieving all of the fields from the DB. Here's what's happening...
// retrieves data and builds the single Country entity (if not null result)
var Country = Countries.FirstOrDefault(o => o.Id == 100000581);
// Country.Personnels accessor will lazy load and construct all Personnel entity objects related to this country entity object
// hence loading all of the fields
var personnelIds = Country.Personnels.Select(p => p.Id).ToArray();
You want something more like this:
// build base query projecting desired data
var personnelIdsQuery = dbContext.Countries
.Where( c => c.Id == 100000581 )
.Select( c => new
{
CountryId = c.Id,
PersonnelIds = c.Personnels.Select( p => p.Id )
}
// now do enumeration
// your example shows FirstOrDefault without OrderBy
// either use SingleOrDefault or specify an OrderBy prior to using FirstOrDefaul
var result = personnelIdsQuery.OrderBy( item => item.CountryId ).FirstOrDefault();
OR:
var result = personnelIdsQuery.SingleOrDefault();
Then get the array of IDs if not null
if( null != result )
{
var personnelIds = result.PersonnelIds;
}
Try can also try grouping personnel into a single query
var groups =
(from p in Personnel
group p by p.CountryId into g
select new
{
CountryId = g.Key
PersonnelIds = p.Select(x => x.Id)
});
var personnelIds = groups.FirstOrDefault(g => g.Key == 100000581);
Do you have the ForeignKey explicitly defined in your POCO for Personnel? It's common to leave it out in EF, but adding it would massively simplify both this code and the resulting SQL:
public class Personnel
{
public Country Country { get; set; }
[ForeignKey("Country")]
public int CountryId { get; set; }
. . .
}
> update-database -f -verbose
var ids = db.Personnel.Where(p => p.CountryId == 100000581).Select(p => p.Id).ToArray();
I'm trying to translate an SQL query into LINQ, but after numerous attempts, I'm still not there... Can anyone help ?
This is my working SQL statement
SELECT Users.email, SUM(Skills.level) AS SkillLevel
FROM Skills INNER JOIN
SkillsPerUser ON Skills.pk_skill_id = SkillsPerUser.fk_skill_id INNER JOIN
Users ON SkillsPerUser.fk_user_id = Users.pk_userid
GROUP BY Users.email
ORDER BY SkillLevel DESC
This is what I came up with so far, but it lacks a sum() where I hard coded the number 3, that should be the sum of Skills.level:
var allSkillsPerUser = from u in dc.Users
join spu in dc.SkillsPerUsers on u.pk_userid equals spu.fk_user_id
join s in dc.Skills on spu.fk_skill_id equals s.pk_skill_id
select new { Email = u.email, Level = s.level } into su
group su by su.Email into gsu
select new { Email = gsu.Key, SkillLevel = gsu.Sum(su => su.Level) };
ViewBag.spu = allSkillsPerUser.ToList();
The view bag gives the following error (Email can't be found, while in the variables below you can see that they do exist...):
context.Skills
.Join(context.SkillsPerUser, s => s.pk_skill_id, spu => spu.fk_skill_id, (s, spu) => new { Skill = s, SkillToUser = spu })
.Join(context.Users, sspu => sspu.SkillToUser.fk_userId, u => u.pk_userid, (sspu, u) => new { Email = u.Email, SkillLevel = sspu.Skill.level })
.GroupBy(su => su.Email)
.Select(g => new { Email = g.Key, SkillLevel= g.Sum(su => su.Level) })
.OrderByDescending(g => g.SkillLevel)
It's a bit simpler if you have navigation properties set up on your entities:
context.SkillsPerUser
.Select(spu => new { Email = spu.User.email, Level = spu.Skill.level }) // guessing at the navigation property names here
.GroupBy(su => su.Email)
.Select(g => new { Email = g.Key, SkillLevel = g.Sum(su => su.Level) })
.OrderByDescending(g => g.SkillLevel)
Or, using the query syntax
from u in dc.Users
join spu in dc.SkillsPerUsers on u.pk_userid equals spu.fk_user_id
join s in dc.Skills on spu.fk_skill_id equals s.pk_skill_id
select new { Email = u.email, Level = s.level } into su
group su by su.Email into gsu
select new { Email = gsu.Key, SkillLevel = gsu.Sum(su => su.Level) }
To order by the sum you can do this:
from u in dc.Users
join spu in dc.SkillsPerUsers on u.pk_userid equals spu.fk_user_id
join s in dc.Skills on spu.fk_skill_id equals s.pk_skill_id
select new { Email = u.email, Level = s.level } into su
group su by su.Email into gsu
select new { Email = gsu.Key, SkillLevel = gsu.Sum(su => su.Level) } into userSkills
orderby userSkills.SkillLevel descending