Category: Syntax

Lukas Eder rises to a challenge:

SQL is a really cool language. I can write really complex business logic with this logic programming language. I was again thrilled about SQL recently, at a customer site:

Writing some nifty SQL queries for a customer. Wondering why anyone would even consider using a 3GL for any business logic at all??

— Lukas Eder (@lukaseder) August 21, 2017

But whenever I tweet something like the above, the inevitable happened. I was nerd sniped. Oleg Šelajev from ZeroTurnaround challenged me to prove that SQL is so awesome:

You have a String s (of length N), find all (i; j), i < j, such that s[i:j] is a palindrome. For bonus points, do it in O(n) or O(n*log(n))

— Oleg Šelajev (@shelajev) August 21, 2017

Given a string, find all substrings from that string, which are palindromes. Challenge accepted! (For the moment, let’s forget about algorithmic complexity.)

His answer is in Postgres syntax, and a commenter includes Oracle syntax.  T-SQL is left as an exercise for the reader.

Daniel Janik shows a pair of non-standard functions you can use to replace NULL values:

It’s Wednesday and that means another SQL/Oracle post. Today we’ll be discussing NULL Values, which can sometimes be a real pain. Don’t worry though there’s a simple solution. Simply replace the NULL value with another.

Comparing a column with NULL and replacing with another value is really simple. There are built in functions for replacing NULL values. I’m not going to discuss the ANSI standard COALESCE here. If you want to know more about it you can find it on Bing.

I provide no comment on Daniel’s claim regarding being able to find something on Bing…  Click through to see the custom NULL replacement functions in SQL Server versus Oracle.

Daniel Janik compares Oracle and SQL Server date conversion functions:

There are many ways to create a date from a string. First you’ve got the CONVERT() function and the CAST() function. Next you’ve got DATEFROMPARTS(), DATETIMEFROMPARTS(), DATETIME2FROMPARTS(), SMALLDATETIMEFROMPARTS(), TIMEFROMPARTS(), and DATETIMEOFFSETFROMPARTS().

That’s a lot of functions for one simple task isn’t it? To be fair, it’s really more than 1 simple task. Each of these functions is meant to be paired with the matching data type so you get just what you want. To go along with these you also have the ISDATE() function which tests the value to be sure it’s a date.

I never liked the verbosity of the Oracle TO_DATE() function…but I am biased.

Daniel Janik is currently running an Oracle versus SQL Server series, looking at how the two database systems expose common functionality.  His latest topic is pagination:

Today’s topic is Pagination. Paging is a really important feature for web pages and applications. Without it you’d be passing large amounts of data to the application and expecting the application code to figure out which rows it needed to display.

Thankfully, someone smart came up with a way to do this on the database so you’re not returning gigs and gigs of data to the webserver to sort through.

Read on to see how the two platforms do this.

Denis Gobo doesn’t like RIGHT JOIN:

Do you use RIGHT JOINs? I myself rarely use a RIGHT JOIN, I think in the last 17 years or so I have only used a RIGHT JOIN once or twice. I think that RIGHT JOINs confuse people who are new to databases, everything that you can do with a RIGHT JOIN, you can also do with a LEFT JOIN, you just have to flip the query around

So why did I use a RIGHT JOIN then?

Don’t be lazy; switch out those right joins.  The trick is that for every RIGHT JOIN statement, there is an equivalent statement which does not use RIGHT JOIN.  The percentage of the time that you might benefit from RIGHT JOIN is so low that the fixed costs of mentally processing what’s going on tend to overwhelm the slight benefit of that style of join.

Andrew Pruski shows off a new feature in SQL Server 2017:

Now I can run the SELECT…INTO statement using the new ON FILEGROUP option. I’m going to run an example SELECT statement to capture Sales in the UK: –

We are one step closer to CTAS on-prem…  Being able to select into a specific filegroup is nice when you want to segregate tables by filegroup to make recovery of the most critical tables faster:  having a primary filegroup, and then a filegroup for the critical tables for your application, followed by the history tables and other large tables that the app doesn’t need immediately.