The `split()` function breaks the given string into multiple substrings,
using the given separator value.

The separator may be either a string or a regular expression value.

Returns an array containing the resulting parts.

Returns `null` if the given input value is not a string or if the separator
argument is neither a string nor a regular expression.


-- Testcase --
{%
	print(join("\n", [
		// split by string
		split("foo|bar|baz", "|"),

		// split by regexp
		split("apples, bananas and strawberries are fruits", /, | and | are | /),

		// splitting an empty string yields an array containing one empty string
		split("", "|"),
		split("", ""),
		split("", /\s+/),

		// splitting with an empty string as separator yields an array containing
		// all characters individually
		split("foo|bar|baz", ""),
		split("foo|bar|baz", /()/),

		// splitting on a separator not found within the string will yield an
		// array containing the entire string as sole element
		split("foo|bar|baz", "xxx"),
		split("foo|bar|baz", /\d+/),

		// subsequent separators are not coalesced
		split("abc|||def", "|"),
		split("foo1bar23baz", /[[:digit:]]/),

		// leading and trailing empty substrings are retained
		split("|abc|def|", "|"),
		split(",foo;bar:", /[,;:]/),
	]), "\n");
%}
-- End --

-- Expect stdout --
[ "foo", "bar", "baz" ]
[ "apples", "bananas", "strawberries", "fruits" ]
[ "" ]
[ "" ]
[ "" ]
[ "f", "o", "o", "|", "b", "a", "r", "|", "b", "a", "z" ]
[ "f", "o", "o", "|", "b", "a", "r", "|", "b", "a", "z" ]
[ "foo|bar|baz" ]
[ "foo|bar|baz" ]
[ "abc", "", "", "def" ]
[ "foo", "bar", "", "baz" ]
[ "", "abc", "def", "" ]
[ "", "foo", "bar", "" ]
-- End --


Supplying an invalid input string value will yield `null`.

-- Testcase --
{%
	printf("%.J\n", split(true, "u"));
%}
-- End --

-- Expect stdout --
null
-- End --


Supplying a non-string, non-regexp separator will yield `null`.

-- Testcase --
{%
	printf("%.J\n", split("null true false", true));
%}
-- End --

-- Expect stdout --
null
-- End --