All posts
T-SQL Tuesday

T-SQL Tuesday #156 - Wrap Up

The wrap-up for T-SQL Tuesday #156: 15 community bloggers on what turns ordinary SQL into production-grade code, from testing to naming to alerting.

Tom · 3 min read
T-SQL Tuesday #156 - Wrap Up

Foreword

Thank you everyone for participating in November's T-SQL Tuesday! There is a total of 15 submissions and thanks to them, I've widened my perspective.

The order of the posts is chosen at random.

Greg Moore

Greg Moore

Greg has a shortlist of best practices he wants to have in production code. I especially liked the one about alerting being actionable.

Read the contribution

Aaron Bertrand

Aaron Bertrand

Aaron has a list of 5 examples with both the bad and good alternatives. I'm in complete agreement, and I will fail any code review that uses shorthand dateparts.

Read the contribution

Olivier Van Steenlandt

Olivier Van Steenlandt

Olivier has a practical checklist of steps he takes to make the code production worthy. Having a checklist is great, so you don't accidentally forget something.

Read the contribution

Kenneth Fisher

Kenneth Fisher

Kenneth's answer is in true DBA fashion: It depends. Read on to find out which criteria affect the answer.

Read the contribution

Deepthi Goguri

Deepthi Goguri

Deepthi has a list of several tips for you. Many of those points relate to the readability of the code, which is extremely important.

Read the contribution

Damien Jones

Damien Jones (amazonwebshark)

Damien went all out and used three different programming languages. Each example has a different scenario, problem and resolution.

Read the contribution

Ajay Dwivedi

Ajay Dwivedi

Ajay has a list of practices and other SOLID advice - from naming conventions to testing and tools used to build the project.

Read the contribution

David Wiseman

David Wiseman

I agree with everything on David's list. I love the points about which things specifically prevent the code from being production-grade.

Read the contribution

Josh Darnell

Josh Darnell

Josh invites you to think about scalability and maintenance, among other things. Each task can have different criteria based on how vital it is.

Read the contribution

Deborah Melkin

Deborah Melkin

Deborah warns us that there is nothing more permanent than temporary code. Therefore, you should have some level of standards and require it for all the code.

Read the contribution

Kevin Chant

Kevin Chant

I like to have fun during work, but I agree with Kevin that there are better mediums for jokes than code comments.

Read the contribution

Chad Callihan

Chad Callihan

Chad doesn't let "perfect" be the enemy of "good" but still has a list of several quality gate requirements.

Read the contribution

Andy Yun

Andy Yun

Andy requires all code to be tested. But there are exceptions to every rule when production is at stake.

Read the contribution

Rob Farley

Rob Farley

I caught Rob by surprise with this invitation. But, of course, "production-grade" can mean different things to different people. Still, Rob has some advice on where to look for inspiration.

Read the contribution

Tom Zíka

Tom Zíka (StraightforwardSQL)

My submission is about the importance of testing. But I wrote the post on the last day of the deadline and had to publish a minimal viable blog post (MVP). So that's an unintended meta-commentary on the production code.

Read the contribution

Thank you for reading

Tom
Tom, TSQL Dev

SQL Server consultant from Czechia.

Give me a problem where the answer isn't obvious and the evidence doesn't add up. That's my idea of a good time.