Why I will begin using Semicolons to terminate SQL Statements

Browsing along the SQL Server documentation today, I noticed something that scared me. As a spoiled non ANSI compliant SQL Server developer, I do not use a semicolon to terminate my statements. Well, here’s the excerpt that threw me a curveball:

Transact-SQL statement terminator. Although the semicolon is not required for most statements in this version of SQL Server, it will be required in a future version.

Here is the link: http://msdn.microsoft.com/en-us/library/ms177563.aspx

Holy Crap! My first thought was about all the code I have written and how I could automate inserting semicolons to the end of all the statements. While that would be almost futile, I suppose it could be done. I’m sure Microsoft will provide a tool that will append a semicolon for the future release however. If they did not, it would stop people from upgrading, which would cost them money.

This realization made me think more about how I code and how much more important it is to enforce ANSI standards into your work. Being shielded from these subtleties in the past has not been a benefit. You can now tell that this SQL Server developer team is really serious about their product. First they got rid of the NON ANSI joins, which was a big achievement in my view, now they are moving onto the semicolon. I can’t say it’s a big surprise because if you have ever used common table expressions, you’ll know that’s where the first indication of requiring a semicolon was put into place. When you create a CTE, if you have not terminated your last statement with a semicolon, then the CTE will give you an error. This has resulted in many of us writing CTE’s beginning with a semicolon, which is kind of funny and sad at the same time.