Question

I’ve read that good programming practice claims that one should use a function if a block is to be used repeatedly in a program. But what if you have a procedure with clear conceptual boundaries, but it is only called once. Should you immediately write it in the main code or should you put it into a function and invoke it in the main code?

Answer 1

Write-once is one reason to use functions. However, any block of code with clear conceptual boundaries is a great candidate for a function.

The resulting logic flow of your script will be more straightforward and easier to read/debug.

For example, consider you have a database script that runs some basic audits before continuing with the main work. This happens once in the script, but the resulting logic flow would be:

dbConnect()
dbAudit()
...
do main loop
...

Nice and clean, easy for others to read and maintain.