Creates an array of substrings by splitting +self+
at each occurrence of the given field separator +field_sep+.

With no arguments given,
splits using the field separator <tt>$;</tt>,
whose default value is +nil+.

With no block given, returns the array of substrings:

  'abracadabra'.split('a') # => ["", "br", "c", "d", "br"]

When +field_sep+ is +nil+ or <tt>' '</tt> (a single space),
splits at each sequence of whitespace:

  'foo bar baz'.split(nil)          # => ["foo", "bar", "baz"]
  'foo bar baz'.split(' ')          # => ["foo", "bar", "baz"]
  "foo \n\tbar\t\n  baz".split(' ') # => ["foo", "bar", "baz"]
  'foo  bar   baz'.split(' ')       # => ["foo", "bar", "baz"]
  ''.split(' ')                     # => []

When +field_sep+ is an empty string,
splits at every character:

  'abracadabra'.split('') # => ["a", "b", "r", "a", "c", "a", "d", "a", "b", "r", "a"]
  ''.split('')            # => []
  'こんにちは'.split('')   # => ["こ", "ん", "に", "ち", "は"]

When +field_sep+ is a non-empty string and different from <tt>' '</tt> (a single space),
uses that string as the separator:

  'abracadabra'.split('a')  # => ["", "br", "c", "d", "br"]
  'abracadabra'.split('ab') # => ["", "racad", "ra"]
  ''.split('a')             # => []
  'こんにちは'.split('に')    # => ["こん", "ちは"]

When +field_sep+ is a Regexp,
splits at each occurrence of a matching substring:

  'abracadabra'.split(/ab/) # => ["", "racad", "ra"]
  '1 + 1 == 2'.split(/\W+/) # => ["1", "1", "2"]
  'abracadabra'.split(//)   # => ["a", "b", "r", "a", "c", "a", "d", "a", "b", "r", "a"]

If the \Regexp contains groups, their matches are included
in the returned array:

  '1:2:3'.split(/(:)()()/, 2) # => ["1", ":", "", "", "2:3"]

Argument +limit+ sets a limit on the size of the returned array;
it also determines whether trailing empty strings are included in the returned array.

When +limit+ is zero,
there is no limit on the size of the array,
but trailing empty strings are omitted:

  'abracadabra'.split('', 0)  # => ["a", "b", "r", "a", "c", "a", "d", "a", "b", "r", "a"]
  'abracadabra'.split('a', 0) # => ["", "br", "c", "d", "br"]  # Empty string after last 'a' omitted.

When +limit+ is a positive integer,
there is a limit on the size of the array (no more than <tt>n - 1</tt> splits occur),
and trailing empty strings are included:

  'abracadabra'.split('', 3)   # => ["a", "b", "racadabra"]
  'abracadabra'.split('a', 3)  # => ["", "br", "cadabra"]
  'abracadabra'.split('', 30)  # => ["a", "b", "r", "a", "c", "a", "d", "a", "b", "r", "a", ""]
  'abracadabra'.split('a', 30) # => ["", "br", "c", "d", "br", ""]
  'abracadabra'.split('', 1)   # => ["abracadabra"]
  'abracadabra'.split('a', 1)  # => ["abracadabra"]

When +limit+ is negative,
there is no limit on the size of the array,
and trailing empty strings are omitted:

  'abracadabra'.split('', -1)  # => ["a", "b", "r", "a", "c", "a", "d", "a", "b", "r", "a", ""]
  'abracadabra'.split('a', -1) # => ["", "br", "c", "d", "br", ""]

If a block is given, it is called with each substring and returns +self+:

  'foo bar baz'.split(' ') {|substring| p substring }

Output :

  "foo"
  "bar"
  "baz"

Note that the above example is functionally equivalent to:

   'foo bar baz'.split(' ').each {|substring| p substring }

Output :

  "foo"
  "bar"
  "baz"

But the latter:

- Has poorer performance because it creates an intermediate array.
- Returns an array (instead of +self+).

Related: see {Converting to Non-String}[rdoc-ref:String@Converting+to+Non--5CString].