Ruby > Ruby code snippets >

This code was tested and works on Ruby 1.9.2

http://ruby-doc.org/core/classes/Array.html

As with everything, there are multiple ways to do something. When faced with multiple choices, I just picked one method I felt was straightforward and went with it and ignored the rest.

Arrays ∞

To avoid any messy gotchas, initialize your arrays with Array.new. For example:

a = Array.new

A "1d" (one-dimensional) array with two elements looks like this:

a = [ "0x", "1x" ]

Instead of having all elements in one row, it's possible to arrange it as a column:

a = [
  "0x",
  "1x",
]

Notice how it looks a lot nicer.

I'm now able to comment in a straightforward way:

 a = [
  "0x",   # The left side
  "1x",   # The right side.
]

I ended the final entry with a comma. This is optional, but helpful for those of us who like to cut-and-paste to move entries up and down in that array.

Examples ∞

Creating an array / adding elements ∞

a = Array.new

a << "one"
a << "two"

puts a.inspect
# =>
# ["one", "two"]

Retrieving elements ∞

a = [
  "one",
  "two",
  "three",
]

puts a[0].inspect
# =>
# "one"

puts a[-1].inspect
# =>
# "three"

puts a[0..1].inspect
# =>
# ["one", "two"]

puts a[1..-1].inspect
# =>
# ["two", "three"]

2d Arrays ∞

If you kneel and look at the very edge of your desk, you can only see the front corners. It looks like a line.

If you stand and look at the top of your desk, you can see the full desktop with its four corners.

Let's make a desktop-like array with four elements arranged like a 2x2 square.

We can put four elements into an array like this:

a = [
  "one",
  "two",
  "three",
  "four",
]

That isn't four points at the corners of a square, that's four points along a single line, because it is still a regular 1d array. A 2d array looks like:

a = [         # Imagine that this is the top-left corner of the 2d array's "box".
  [           # This begins the first row of the box:
    "0x,0y",  # row 1, left side.
    "0x,1y",  # row 1, right side.
  ],
  [           # A new, second, row:
    "1x,0y",  # row 2, left side.
    "1x,1x",  # row 2, right side.
  ], 
]             # Imagine that this is the bottom-right corner of the 2d array's "box".

You can imagine a 2x2 square with that. To make it into a "taller" 2x3 rectangle, you can add more rows on to the bottom.

a = [
  [
    "0x,0y",  # row 1, left side.
    "0x,1y",  # row 1, right side.
  ],
  [           # A new row:
    "1x,0y",  #   row 2, left side.
    "1x,1x",  #   row 2, right side.
  ],
  [           # A new row.
    "2x,0y",  # row 3, left side.
    "2x,1x",  # row 3, right side.
  ],        
]

To add more "width" to our box, you would add more items like this.

a = [
  [           # This begins the first row of the box:
    "0x,0y",  # row 1, left side.
    "0x,1y",  # row 1, right side.
    "0x,2y",  # row 1, right side.  "Widening" it.
  ],
  [           # A new row:
    "1x,0y",  # row 2, left side.
    "1x,1x",  # row 2, right side.
    "1x,2x",  # row 2, right side.  "Widening" it.
  ],
]

Examples ∞

Creating an array / adding elements ∞

row1 = Array.new
row2 = Array.new

row1 << "one"
row1 << "two"
puts row1.inspect
# =>
# ["one, "two"]

row2 << "three"
row2 << "four"
# =>
# ["three", "four"]

array2d = [ row1, row2 ]
# We can't call it "2darray" because you cannot start a variable with a number.
puts array2d.inspect
# =>
# [["one", "two"], ["three", "four"]]

Retrieving elements ∞

a = [
  [
    "one",
    "two",
  ],
  [
    "three",
    "four",
  ]
] 
puts a[0].inspect
# =>
# ["one", "two"]

puts a[0][0].inspect
# =>
# "one"

TODO - Retrieving a "box" of elements ∞

a = [
  [
    "one",
    "two",
    "three",
  ],
  [
    "four",
    "five",
    "six",
  ]
  [
    "seven",
    "eight",
    "nine",
  ]
]

puts a[0..1].inspect
# =>
# [["one", "two", "three"], ["four", "five", "six"]]

a = [
  [             # This is 0
    "one",   
    "two",   
    "three", 
  ],         
  [            # This is 1
    "four", 
    "five", 
    "six",  
  ]         
  [
    "seven",
    "eight",
    "nine",
  ]
]

Missing elements ∞

It's perfectly fine to have missing elements in our array-box. Let's imagine we have a 3x3 array:

a = [
  [
    "0x,0y",  # row 1, left side.
    "0x,1y",  # row 1, middle.
    "0x,2y",  # row 1, right side.
  ],
  [
    "1x,0y",  # row 2, left side.
    "1x,1x",  # row 2, middle.
    "1x,2x",  # row 2, right side.
  ],
  [
    "2x,0y",  # row 3, left side.
    "2x,1x",  # row 3, middle.
    "2x,2x",  # row 3, right side.
  ],
]

If we imagine it as a bunch of Xs, it looks like this:

XXX
XXX
XXX

But it's ok to have an array with any of those removed. I'll represent them with a middot ( · ).

XX·
X·X
XXX

a = [
  [
    "0x,0y",  # row 1, left side.
    "0x,1y",  # row 1, middle.
              # row 1, right side.  Removed.
  ],
  [
    "1x,0y",  # row 2, left side.
              # row 2, middle.      Removed.
    "1x,2x",  # row 2, right side.
  ],
  [
    "2x,0y",  # row 3, left side.
    "2x,1x",  # row 3, middle.
    "2x,2x",  # row 3, right side.
  ],
]

If we simply cut out those entries, then the array will collapse. Elements will "shift over".

XXX    XX·    XX
XXX => X·X => XX
XXX    XXX    XXX

As the array collapses in on itself, one of its elements will change so that it is accessed in a different way from before. It's easier to explain with numbers.

You might be tempted to think of it like this:

012    01·
012 => 0·2
012    012

But it's actually doing this:

012    01·    01
012 => 0·? => 01
012    012    012

This is because the "columns and rows" concept does not really apply to arrays! It's an easy convention to think of 1d arrays as being parts of a line, and to think of 2d arrays as being parts of a square or rectangle, but these are not how things actually work.

To explain better, let's show the above code again and comment it with actual Ruby code.

a = [
  [
    "1",   # a[0][0]
    "2",   # a[0][1]
    "3",   # a[0][2]
  ],
  [
    "4",   # a[1][0]
    "5",   # a[1][1]
    "6",   # a[1][2]
  ],
  [
    "7",   # a[2][0]
    "8",   # a[2][1]
    "9",   # a[2][2]
  ],
]
puts a.inspect
# =>
# [["1", "2", "3"], ["4", "5", "6"], ["7", "8", "9"]]

a[0][2] = nil
a[1][1] = nil
puts a.inspect
# =>
# [["1", "2", nil], ["4", nil, "6"], ["7", "8", "9"]]
# Setting an element to "nil" is not removing them, it's just
# emptying that element.

# Delete the element at a[1][1]
a[1].delete_at(1)
puts a.inspect
# =>
# [["1", "2", "3"], ["4", "6"], ["7", "8", "9"]]

# Our existing numbering concept doesn't really work anymore.
# Any time we fetch an element that has no content we will get nil.
puts a[1][2]
# =>
# nil
puts a[1][1]
# =>
# "6"

An element in an array can have some value. So a[1][1] can be the number 6. But its value has nothing to do with its actual "position" in the array.

So removing an element shifts everything over.

• a[1].delete_at(1) deleted a[1][1].
• a[1][2] used to be 6 but it became nil.

• a[1][1] used to be 5 but it became 6.

Fetching nonexistent elements ∞

Any time we fetch an element that has no content we will get nil. This is true for a 1d array:

a = [
  1,
  2,
]
puts a[99]
# =>
# nil

This is not entirely true for a 2d array:

a = [
  [
    1,
    2,
  ],
  [
    3,
    4,
  ],
]
puts a[99]
# =>
# nil
puts a[99][99]
# =>
# NoMethodError: undefined method `[]' for nil:NilClass

The a[99] works the same as a 1d array would. But doing the equivalent of a[nil][99] won't work, because nil isn't actually an array!

Think of a 1d array as having infinite nils:

[1,2,nil,nil,nil,...]

But a 2d array has no such concept:

[
[1,2,nil,nil,nil,...]
[1,2,nil,nil,nil,...]
]

[
[1,2,nil,nil,nil,...]
[1,2,nil,nil,nil,...]
···
]

There is no concept of "infinite nil-arrays" underneath there.