Subroutine is the provisional term I’ll use for a chunk of reused logic in programming. In most languages this will consist of some kind of name and signature involving parameters and return value or values, along with a body which contains the implementation, though this term is also seems general enough to cover reuse without that type of interface (such as in assembly or Forth).
Potentially the most common name for such a construct is “function”, though I’m reluctant to use that too generally due to the utility of the mathematical definition which doesn’t apply to “impure functions”.<p>
In object-oriented languages the term "method" may be used
which is appropriate when working with an instance but gets
stretched a bit when there's a desire to define functions that
then masquerade as "static methods" even if they don't
interact with their surrounding namespace (i.e. operating on
class-wide data)...but more of warts with overuse of
object-oriented programming will be covered separately.
</p>
<p>
Finally potentially on the opposite end of "function" would be
"procedure" which is not particularly in fashion but at least
seems to imply action/side effects which may not always be
appropriate.
</p>
</section>
<section>
</section>
</article>C supports definition of functions using the production:
<return type> <function name>(<parameters>) { <body> }The definition itself can get fairly verbose given both the types and compiler advice. At the moment I’m not sure whether the braces are required for the body or incidental to the expectation of a block, but I’d use them consistently regardless.
Functions in JavaScript can be invoked using the production:
<name>(<args>*);