I have a scenario where I have to use a dynamic where condition in LINQ.
I want something like this:
public void test(bool flag)
{
开发者_运维问答 from e in employee
where e.Field<string>("EmployeeName") == "Jhom"
If (flag == true)
{
e.Field<string>("EmployeeDepartment") == "IT"
}
select e.Field<string>("EmployeeID")
}
I know we can't use the 'If' in the middle of the Linq query but what is the solution for this?
Please help...
Please check out the full blog post: Dynamic query with Linq
There are two options you can use:
Dynamic LINQ library
string condition = string.Empty;
if (!string.IsNullOrEmpty(txtName.Text))
condition = string.Format("Name.StartsWith(\"{0}\")", txtName.Text);
EmployeeDataContext edb = new EmployeeDataContext();
if(condition != string.empty)
{
var emp = edb.Employees.Where(condition);
///do the task you wnat
}
else
{
//do the task you want
}
Predicate Builder
Predicate builder works similar to Dynamic LINQ library but it is type safe:
var predicate = PredicateBuilder.True<Employee>();
if(!string.IsNullOrEmpty(txtAddress.Text))
predicate = predicate.And(e1 => e1.Address.Contains(txtAddress.Text));
EmployeeDataContext edb= new EmployeeDataContext();
var emp = edb.Employees.Where(predicate);
difference between above library:
- PredicateBuilder allows to build typesafe dynamic queries.
- Dynamic LINQ library allows to build queries with dynamic Where and OrderBy clauses specified using strings.
So, if flag
is false
you need all Jhoms, and if flag
is true you need only the Jhoms in the IT department
This condition
!flag || (e.Field<string>("EmployeeDepartment") == "IT"
satisfies that criterion (it's always true if flag is false, etc..), so the query will become:
from e in employee
where e.Field<string>("EmployeeName") == "Jhom"
&& (!flag || (e.Field<string>("EmployeeDepartment") == "IT")
select e.Field<string>("EmployeeID")
also, this e.Field<string>("EmployeeID")
business, smells like softcoding, might take a look into that. I guess
from e in employee
where e.EmployeeName == "Jhom"
&& (!flag || (e.EmployeeDepartment == "IT")
select e.EmployeeID
would be more compact and less prone to typing errors.
EDIT: This answer works for this particular scenario. If you have lots of this kinds of queries, by all means investingate the patterns proposed in the other answers.
You can chain methods :
public void test(bool flag)
{
var res = employee.Where( x => x.EmployeeName = "Jhom" );
if (flag)
{
res = res.Where( x => x.EmployeeDepartment == "IT")
}
var id = res.Select(x => x.EmployeeID );
}
from e in employee
where e.Field<string>("EmployeeName") == "Jhom" &&
(!flag || e.Field<string>("EmployeeDepartment") == "IT")
select e.Field<string>("EmployeeID")
You can call LINQ methods explicitly and chain them conditionally.
public IEnumerable<string> FilterEmployees (IEnumerable<Employee> source, bool restrictDepartment)
{
var query = source.Where (e => e.Field<string>("EmployeeName") == "Jhom");
if (restrictDepartment) // btw, there's no need for "== true"
query = query.Where (e => e.Field<string>("EmployeeDepartment") == "IT");
return query.Select (e => e.Field<string>("EmployeeID"));
}
精彩评论