user_id, currency_id, and transaction_amount are all defined as NOT
NULL columns in dbo.transactions
It looks to me that SQL Server has a blanket assumption that an aggregate can produce a null even if the field(s) it operates on are not null. This is obviously true in certain cases:
create table foo(bar integer not null);
select sum(bar) from foo
-- returns 1 row with `null` field
And is also true in the generalized versions of group by like cube
This simpler test case illustrates the point that any aggregate is interpreted as being nullable:
CREATE VIEW dbo.balances
with schemabinding
AS
SELECT
user_id
, sum(1) AS balance_amount
FROM dbo.transactions
GROUP BY
user_id
;
GO
IMO this is a limitation (albeit a minor one) of SQL Server - some other RDBMSs allow the creation of certain constraints on views that are not enforced and exist only to give clues to the optimizer, though I think 'uniqueness' is more likely to help in generating a good query plan than 'nullability'
If the nullability of the column is important, perhaps for use with an ORM, consider wrapping the indexed view in another view that simply guarantees the non-nullability using ISNULL:
CREATE VIEW dbo.balancesORM
WITH SCHEMABINDING
AS
SELECT
B.[user_id],
B.currency_id,
balance_amount = ISNULL(B.balance_amount, 0),
transaction_count = ISNULL(B.transaction_count, 0)
FROM dbo.balances AS B;
Please don't use the UI for this. It's a confusing mess.
It sounds to me like what you want is to create a user in a database, for a specific login, who only has permissions to select from one view. So, since you already have the login created:
USE your_db;
GO
CREATE USER username FROM LOGIN username;
GO
GRANT SELECT ON dbo.MyViewName TO username;
GO
EDIT here is an example of a script that will lead to the error you mention.
First, create some table in the unrelated_db:
CREATE DATABASE unrelated_db;
GO
USE unrelated_db;
GO
CREATE TABLE dbo.foo(bar INT);
GO
Now create a relatively restricted login:
USE [master];
GO
CREATE LOGIN username WITH PASSWORD='foo', CHECK_POLICY = OFF;
GO
Now create a database where the view will live, and add the login as a user:
CREATE DATABASE velojason;
GO
USE velojason;
GO
CREATE USER username FROM LOGIN username;
GO
Now create a function that will reference the table in the other database, and a synonym to the other table:
CREATE FUNCTION dbo.checkbar()
RETURNS INT
AS
BEGIN
RETURN
(
SELECT TOP (1) bar
FROM unrelated_db.dbo.foo
ORDER BY bar
);
END
GO
CREATE SYNONYM dbo.foo FOR unrelated_db.dbo.foo;
GO
Now create a local table:
CREATE TABLE dbo.PaymentDetails
(
PaymentID INT
);
GO
Now create a view that references the table, the function and the synonym, and grant SELECT to username:
CREATE VIEW dbo.SomeView
AS
SELECT
p.PaymentID,
x = dbo.checkbar(), -- function that pulls from other DB
y = (SELECT bar FROM dbo.foo) -- synonym to other DB
FROM dbo.PaymentDetails AS p;
GO
GRANT SELECT ON dbo.SomeView TO username;
GO
Now try to execute as username and select only the local column from the view:
EXECUTE AS USER = 'username';
GO
-- even though I don't reference any of the columns
-- in the other DB, I am denied SELECT on the view:
SELECT PaymentID FROM dbo.SomeView;
GO
REVERT;
GO
Result:
Msg 916, Level 14, State 1, Line 3
The server principal "username" is
not able to access the database "unrelated_db" under the current
security context.
Now change the view to not reference any external objects, and run the above SELECT again, and it works:
ALTER VIEW dbo.SomeView
AS
SELECT
p.PaymentID
--x = dbo.checkbar(),
--y = (SELECT bar FROM dbo.foo)
FROM dbo.PaymentDetails AS p;
GO
Short of showing us the scripts for the Payment Details, Account Details and MyView objects, maybe you can let us know if this query returns any results. You can find references to various objects through the catalog view sys.sql_expression_dependencies, but this view is not perfect - I believe it depends on all the views being refreshed (in the case where views reference other views, for example, or underlying schema has changed) in order to be accurate.
DECLARE
@dbname SYSNAME = N'unrelated_db',
@viewname SYSNAME = N'dbo.SomeView';
SELECT DISTINCT
[This object] =
OBJECT_SCHEMA_NAME([referencing_id])
+ '.' + OBJECT_NAME([referencing_id]),
[references this object] =
OBJECT_SCHEMA_NAME([referenced_id])
+ '.' + OBJECT_NAME([referenced_id]),
[and touches this database] = referenced_database_name,
[and is a(n)] = o.type_desc,
[if synonym, it references] = s.base_object_name
FROM sys.sql_expression_dependencies AS d
LEFT OUTER JOIN sys.objects AS o
ON o.[object_id] = d.referenced_id
LEFT OUTER JOIN sys.synonyms AS s
ON d.referenced_id = s.[object_id]
AND s.base_object_name LIKE '%[' + @dbname + ']%'
WHERE OBJECT_ID(@viewname) IN (
referenced_id,
referencing_id,
(SELECT referencing_id FROM sys.sql_expression_dependencies
WHERE referenced_database_name = @dbname)
) OR referenced_database_name = @dbname;
SQL Server isn't just going to try to access unrelated_db for the fun of it... there must be some tie to that database from the view you're trying to use. Unfortunately if we can't see the view definition and more details about the objects it touches, all we can do is speculate. The two main things I can think of are synonyms or functions that use three-part names, but seeing the actual scripts will give us a much better idea instead of guessing. :-)
You may also want to check sys.dm_sql_referenced_entities, however this function returns nothing useful in the example above.
Best Answer
It should be possible to write this as a query without using the variable. Assuming the following setup (slightly simplified - column "trans_ref" omitted), tested with MySQL 5.7 and MariaDB 10.2, respectively:
Using your approach (with the variable) we get:
Suggestion: create 2 views, one for getting the "running totals" encompassing all debtor_ids, and another one for obtaining "specific" balances for one or more debtor_ids.
1 running totals:
Testing the view:
2 view for single/multiple debtor_ids
Testing (single debtor_id):
Testing (multiple debtor_ids):
See dbfiddle here.