How Base64 Encoding Works

Base64 encoding is used to convert a sequence of bits to a sequence of alphanumeric characters.

There are 64 possible values that each encoded character can take, hence the name base64 encoding: 26 lowercase letter, 26 uppercase letters, 10 numbers, and the + and / characters.

Actually, there is an extra character used for padding (the = character). So technically, a 65-character set is used for base64 encoding.

How it works

Since the base 64 character set has 64 characters (excluding the padding character), we need 6 bits to represent each character.

Since most computers deal with bytes (8 bits), we take 3 bytes at a time (24 bits) and break them down in a sequence of four 6 bit characters. We can then use a character from the base 64 alphabet to represent each group of 6 bits.

Table 1: The Base 64 Alphabet
 Value Encoding  Value Encoding  Value Encoding  Value Encoding
     0 A            17 R            34 i            51 z
     1 B            18 S            35 j            52 0
     2 C            19 T            36 k            53 1
     3 D            20 U            37 l            54 2
     4 E            21 V            38 m            55 3
     5 F            22 W            39 n            56 4
     6 G            23 X            40 o            57 5
     7 H            24 Y            41 p            58 6
     8 I            25 Z            42 q            59 7
     9 J            26 a            43 r            60 8
    10 K            27 b            44 s            61 9
    11 L            28 c            45 t            62 +
    12 M            29 d            46 u            63 /
    13 N            30 e            47 v
    14 O            31 f            48 w         (pad) =
    15 P            32 g            49 x
    16 Q            33 h            50 y

Source: https://tools.ietf.org/search/rfc4648

Example

Let's say we want to encode the word "foo".

First, we need to convert this string to a sequence of bits. For simplicity, let's convert each character into the binary representation of its corresponding ASCII value:

foo = 01100110 01101111 01101111

We then take this sequence of 3 bytes and break it down into a sequence of four 6-bit groups:

011001 | 10 0110 | 1111 01 | 101111

We then convert each 6-bit group into the corresponding printable character from the base 64 alphabet. Therefore, the base64 encoded representation of "foo" is: Zm9v

Since base64 encoding works on any sequence of bits, it can be used to encode non-textual data like images as well.

The padding token

When processing a sequence of bits in groups of 3 bytes, it is possible that the last sequence does not contain all 3 bytes. In this case, the sequence is padded with 0s to make the number of bits a multiple of 6.

If the last sequence has only 2 bytes, the base64 encoded output will be 3 characters from the base 64 alphabet followed by the = character to signify padding.

If the last sequence has only 1 byte, the base64 encoded output will have 2 characters from the base 64 alphabet followed by == to signify padding.