Does null ever equal anything?
Yesterday’s discussion of nulls got me to thinking about this common bit of syntax in C#:
[code:c#]if (myObject == null)
{
…
[/code]
But here’s the thing: null is not data. It’s metadata. That is, it describes the state of a variable, not its contents.
Null indicates the absence of a value. So the above code says, is myObject’s value equal to not having a value? Seems strange.
Other languages seem to make this distinction. VB (if I recall) has the construct “Is Nothing”. SQL uses “IS NULL” (as opposed to “= NULL”). These syntaxes make more sense to me — they differentiate testing a value and testing the existence of a value.
I imagine the “== null” operator is allowed in C# as a syntactic convenience. I mean, it’s readable and easy to remember. No complaints.
But I bet this requires the compiler to extract those statements and send them to a different evaluation algorithm. It’s a different kind of test. Curious if any C# language designers would have an insight.