Category: Syntax

Rajendra Gupta has a backup plan in case of NULL:

NULL is a special marker that indicates a missing or undefined value in a column. It is different from zero or an empty string. Handling NULL values is essential for accurate data analysis, data integrity, and error avoidance. This tip explores how to handle NULL values in SQL Server using the COALESCE() function using various queries and reviewing the results.

Click through for a primer on the COALESCE() function, a few use cases for COALESCE(), and how it differs from ISNULL().

Constantine Kokkinos provides an explanation:

I was talking to a friend of mine and they are learning some SQL and they said something that I have seen come up multiple times in learning SQL.

They said “Yeah, I need to study the join types more. They make sense to me but I want to be able to not reference my notes” and also “I don’t really get the point of a right join if your can do the same thing with a left join by just switching the table name.”

These are great points, and common questions that occur when first learning SQL.

I won’t steal CK’s thunder (too much) about how we express joins in set theory, though I think when he mentions “OUTER” as a type of join, perhaps that’s supposed to be FULL OUTER JOIN?

Regardless, my take: there is a good reason to use INNER JOIN. There is a good reason to use LEFT OUTER JOIN. There is a good reason to use CROSS JOIN. There is a good reason to use FULL OUTER JOIN. The frequency in which you should use each is in descending order, meaning that there are relatively few circumstances in which you should use a FULL OUTER JOIN, but they do exist.

There are no good circumstances for a RIGHT OUTER JOIN. The concept logically exists, but has no practical value to us.

Simon Frazer rolls back:

If you’ve been working with T-SQL scripts for a while, you’ve likely encountered variables. These are essential for writing scripts that go beyond basic SELECT, INSERT, UPDATE, or DELETE statements.

Variables can serve multiple purposes: they can act as parameters passed into stored procedures, hold the results of one query to use in another, or even help debug scripts during testing (you do test your scripts before running them on real data, right?).

One particularly interesting aspect of variables is how they behave in transactions, often in ways we might not expect.

Read on for the demo. This also applies to table variables, which is why they can be extremely important for diagnosing issues before performing a rollback. You can write error results to the table variable first, then access them post-rollback.

Joe Fleming is not a fan of cursors:

So what are cursors, and why are they so bad?  A cursor is a construct in SQL Server that lets you define a query which produces a multi-row dataset and allows you to step through it one row at a time. You declare a cursor in this way:

DECLARE Performance_killer1 CURSOR FOR SELECT OrderNumber, OrderLineNumber, ItemID, ItemDescription, Quantity, Price FROM Orders

You must then OPEN the cursor and FETCH the next value from it into a set of variables, then perform your calculations and updates, repeating until you’ve reached the end of your dataset. From the standpoint of someone unfamiliar with database processing, this method is fairly intuitive. You do one thing at a time. From the database professional’s perspective, it’s very painful to see. There are a few reasons why.

I generally agree with this, though there are specific queries that you cannot write in a set-based context, particularly administrative queries that have to run over each table or database in an instance. I also found that I was unable to write a proper leaky bucket algorithm implementation in T-SQL without using a cursor. But the other 99.x% of the time? No cursors needed.

Also, if you do need a cursor, use a cursor instead of a WHILE loop. It’s more to remember but you can performance tune cursors, whereas WHILE loops are about as dog-slow as they get and never get better.

Erik Darling has a new video. Erik mentions the best use case of this being for archival tables, but I’ll add one more: if you’re using a queue table (ignoring how good or bad of an idea this is), you have multiple processes operating on this queue table, and you want to reduce the likelihood of two processes picking up the same value, you can perform the equivalent of popping off of a queue: delete the first element(s) from the queue table and output it into a temp table. From there, you can operate on that data at your leisure, and the next process will grab some other batch of data. And if everything goes mildly wrong, re-insert that data back into the queue and let some other sucker try it. I’ve used this a few times for data warehousing processes and it works out pretty well.

The only thing I’m unsure about is how he figured out that I’m CommonTableExpressionLover11357.

Erik Darling has a pair of videos covering 3 1/2 set operations. First up is UNION and UNION ALL. These are the set operations that most people know about and use fairly regularly, and Erik explains the difference between the two, including the performance difference between the two.

Then, Erik hits upon the two lesser-known set operations: INTERSECT and EXCEPT. These are extremely useful in certain circumstances, and tend to perform much better than other alternatives. For example, to figure out if two datasets are exactly the same, it’s really hard to go wrong with the following pair of queries:

SELECT a.* FROM a EXCEPT SELECT b.* FROM b;
SELECT b.* FROM b EXCEPT SELECT a.* FROM a;

You need both queries because the former tells you if there are any records in A that do not exist in B, either because the record simply is not there or because there is a difference in one or more values in B’s matching record. But then, you also have to check the opposite side, because there might be a record in B that does not exist in A and the first query will not expose it.

This is typically the way I’d write test cases, ensuring that both queries return 0 results. Granted, you could always just check that the count of the intersection equals the count of records:

DECLARE @c1 INT, @c2 INT;
SELECT @c1 = SELECT COUNT(*) FROM (SELECT a.* FROM a INTERSECT SELECT b.* FROM b);
SELECT @c2 = SELECT COUNT(*) FROM a;

In this case, @c1 and @c2 should be the same number.

Chad Callihan engages the limit breaker:

I’m familiar with using the OVER clause and don’t think it’s too uncommon to see it used for including row numbers by using ROW_NUMBER() and aggregating data. But even though they’ve been around since SQL Server 2012, I’m not too familiar with using the OVER clause with the UNBOUNDED PRECEDING and UNBOUNDED FOLLOWING to affect the window being queried.

Let’s take a look at a couple of examples using UNBOUNDED PRECEDING and UNBOUNDED FOLLOWING.

Click through for those examples. The default ranges for window functions usually make a lot of sense, but it’s good to understand your options for frames: ROWS vs RANGE, as well as the frame values (UNBOUNDED PRECEDING, {N} PRECEDING, CURRENT ROW, {N} FOLLOWING, and UNBOUNDED FOLLOWING).