hi i have table called products with columns
product_id
prodcut_name
prodcut_price( values like 1200,2000,3000
from p in dbcontext.products
join c in dbcontext.category on p.category_id equals c.category_id
where p.product_price > 500 && p.product_price < 10000 && c.Name == "a"
select p
This will return you an IQueryable with the filtered results.
Based on Vittore's answer below, if there is a foreign key constraints this would be the more appropriate version:
from p in dbcontext.products
where p.category.name =='a' && p.product_price > 500 && p.product_price < 10000
select p;
You are looking for logic operators, and these are the ones you are going to use in c#:
&&
logical AND
||
logical OR
!
logical NOT
^
logical (bitwise) XOR
COND ? TRUE-STATEMENT : FALSE-STATEMENT
ternary operator So your composite condition in pseudo code going to look like:
product_price > 500 AND product_price < 10000
Now, if you have no foreign key in DB, when you've created your context it only going to have DBSets, and they are not going to have navigation properties. So your only option is to use Cubicle.Jockey's answer.
If you have foreign keys in DB, you will have navigation properties on your entity objects and you will be able to do the following:
var query = from p in dbcontext.products
where p.category.name == 'a'
&& p.product_price > 500
&& p.product_price < 10000
select p;
Alternatively you can use LINQ extension methods directly:
var query = dbcontext.Products.Where(p => p.category.name == 'a'
&& p.product_price > 500
&& p.product_price < 10000);
If you need a list or array and want close dbcontext you are calling either ToList
or ToArray
on query
:
var products = query.ToList();
There is alternative way of doing it with Entity Framework (EF) - build-in EntitySQL query language. You are going to have similar expression with it:
var query = ProductsQuery.Where("it.Category.Name == 'a' AND it.Product_Price BETWEEN 500 AND 10000");
where ProductsQuery
is ObjectQuery<Product>
.
join the product and categories tables on category_id
and then use a where
to filter out product_price
.