Category: Data Types

Kenneth Fisher sorts out data types:

With the expectation that you would get back a decimal. Did you wonder why it converted the integers to decimals rather than the other way around? If not you probably should have. I’ll give you a hint. It’s not because it’s the first value in the equation, although that does make a difference.

Thought about it? In case you didn’t figure it out T-SQL (and again, I’m going to say all programing languages where implicit conversions are allowed) has a conversion precedence order. In other words in any calculation where an implicit conversion is required the datatype with a value higher on the list is converted to the datatype with the lower value on the list.

The parenthetical statement above is correct. This makes things really convenient until you have an implicit conversion you weren’t expecting. There’s an advantage to explicit-only languages, though those do involve more typing.

Comments closed

Kenneth Fisher talks types:

Did you know that 1/2=0? If you didn’t you could be in for some interesting surprises.

1/2 returns 0 because 1 and 2 are both integers. When you divide two integers you get back an integer. Now, we, as human beings, know that what you probably wanted was to get back 0.5. Unfortunately, computers really aren’t all that smart and we have to be very explicit with them. 

Read on for more details, as well as a chart of which conversions may be implicit and which must be explicit.

Comments closed

Dave Saunders describes the concept of linear data structures:

As we said in the introduction, a data structure is ‘linear’ if the elements form a sequence.

That means that the data structure has a first and last element, and each element is connected to its previous and next element.

– An ‘array’ is a linear data structure; the items are stores sequentially.

– A ‘graph’ is not a linear data structure; any node can be linked to any other node in the graph – there is no fixed ‘sequence’.

Click through for several examples of linear data structures.

Comments closed

Steve Jones wants to know if people are still using CHAR:

I ran across an article, titled When to use CHAR, VARCHAR, or VARCHAR(MAX), which struck me as something I never do. I mean, I do use varchar (and nvarchar), but I can’t remember the last time I actually created a char column. The article is worth a read, and it sets the stage for you to think about your database design process and the choices you make.

It’s Friday, and if you take a few minutes and think about the last few times you’ve added columns to a table have you used CHAR as a data type? Or do you default to varchar of any size as a general rule?

Yeah, using CHAR (or NCHAR, which you’d probably want to use instead) is pretty rare. This is a consequence of storage becoming much less of a constraint over the past couple decades, as well as considerable gains in efficiency for data platform products. (N)CHAR might be a little faster than (N)VARCHAR for certain operations and may be a bit more efficient for single-character columns, but I agree with Steve’s premise.

Comments closed

Greg Larsen compares CHAR, VARCHAR, and VARCHAR(MAX):

In every database, there are different kinds of data that need to be stored. Some data is strictly numeric, while other data consists of only letters or a combination of letters, numbers, and even special symbols. Whether it is just stored in memory or on disk, each piece of data requires a data type. Picking the correct data type depends on the characteristics of the data being stored. This article explains the differences between CHAR, VARCHAR, and VARCHAR(MAX).

Click through for Greg’s explanation. My official rule of thumb is as follows:

• If you have a fixed-length code which you display to customers, use NCHAR.
• If you have a fixed-length code which you only use internally and you know that the code will never include characters outside of your SQL Server installation’s code page and you know that the code page will never change…probably still use NCHAR, though if you twist my arm enough I’d say fine, CHAR.
• Otherwise, use NVARCHAR.

Three decades ago, the choice was a lot trickier given performance differences between the two. Today? Unless you’re hunting for microseconds I don’t think you’ll see a practical difference. And if you are hunting for microseconds, you probably want more than rules of thumb.

Comments closed