Base64 Encode Decode

Welcome to the Base64 Encode Decode page! Here, you'll find a description of the project as well as a list of sample programs written in various languages.

This article was written by:

• rzuckerm
• Zia

Description

Introduction

Base64 is a popular method of encoding strings and other data. It can encode images, text, JSON, and almost any other format as well. It is also URL-safe.

In this project you will be encoding normal text to Base64-encoded text and decoding Base64-encoded text into normal text.

Base64 Alphabet

From Base64 Table from Wikipedia, this is the Base64 Alphabet:

Index	Binary	Char		Index	Binary	Char		Index	Binary	Char		Index	Binary	Char
0	000000	A		16	010000	Q		32	100000	g		48	110000	w
1	000001	B		17	010001	R		33	100001	h		49	110001	x
2	000010	C		18	010010	S		34	100010	i		50	110010	y
3	000011	D		19	010011	T		35	100011	j		51	110011	z
4	000100	E		20	010100	U		36	100100	k		52	110100	0
5	000101	F		21	010101	V		37	100101	l		53	110101	1
6	000110	G		22	010110	W		38	100110	m		54	110110	2
7	000111	H		23	010111	X		39	100111	n		55	110111	3
8	001000	I		24	011000	Y		40	101000	o		56	111000	4
9	001001	J		25	011001	Z		41	101001	p		57	111001	5
10	001010	K		26	011010	a		42	101010	q		58	111010	6
11	001011	L		27	011011	b		43	101011	r		59	111011	7
12	001100	M		28	011100	c		44	101100	s		60	111100	8
13	001101	N		29	011101	d		45	101101	t		61	111101	9
14	001110	O		30	011110	e		46	101110	u		62	111110	+
15	001111	P		31	011111	f		47	101111	v		63	111111	/
													Padding	=

This alphabet is used for both encode and decode.

Encode

Base64 encode works as follows:

• Split the data up into 3-byte chunks (let's ignore the case where the last chunk smaller than 3 bytes; that will be discussed later):
• For each 3-byte chunk:
  • Convert each byte to 8-bit binary, 24 bits in total.
  • Divide the 24-bit value into 4 groups of 6 bits.
  • Use each 6-bit group as an index into the above table to get Base64 characters.

For example, let's use the string kitten. The 3-byte chunks are kit and ten. Let's focus on the first chunk: kit. Using 8-bit ASCII:

• k = 107 (01101011)
• i = 105 (01101001)
• t = 116 (01110100)

So, the 24-bit group is this:

01101011 01101001 01110100

Dividing this in 6-bit groups gives this:

011010 110110 100101 110100

Since it's easier to work with this as decimal, the table indices are these:

26 54 37 52

Looking these indices up in the table results a Base64 string of a2l0. Following the same procedure for ten (left as an exercise for the reader), the Base64 string is dGVu. Together, kitten encodes to a2l0dGVu.

For the case when the last chunk is not 3 bytes long, extend the last bits with zeros until it is 6 bits long, and add padding characters until the last Base64 string is 4 bytes long.

For example, let's consider the case where the last chunk is k:

• From the above example, k is 01101011.
• Dividing this into 6-bit groups results in 011010 11.
• Extending the last group results in 011010 110000.
• Converting this to decimal results in 26 48.
• Looking up those values in the table results in aw.
• Since this is only 2 bytes long, 2 padding characters are added: aw==.

For the final encode example, let's consider the case when the last chunk is ki:

• From the above example, ki is 01101011 01101001.
• Dividing this into 6-bit groups results in 011010 110110 1001.
• Extending the last group results in 011010 110110 100100.
• Convert this to decimal results in 26 54 36.
• Looking up those values in the table results in a2k.
• Since this is only 3 bytes long, 1 padding character is added: a2k=.

Decode

Before dividing into the algorithm for how to decode a Base64 string, let's talk about some rules for what constitutes a valid Base64 string:

• Its length must be evenly divisible by 4.
• Only the last 4-byte chunk may have padding characters.
• The number of padding characters must be either 1 or 2.
• Each byte in the Base64 string must be in the table or be a padding character.

Assuming the Base64 string is valid, and ignoring the case where the last 4-byte chunk has padding characters, decode works as follows:

• Divide the Base64 string into 4-byte chunks.
• For each 4-byte chunk:
  • Look up each byte in the table to get the corresponding index.
  • Convert the indices into 4 groups of 6-bit values, 24 bits in total.
  • Divide the 24 bits into 3 8-bit bytes.

Let's use the Base64 string a2l0dGVu. The 4-byte chunks are a2l0 and dGVu. Let's focus on the first chunk: a2l0.

Looking up each byte in the table results in this:

• a is index 26 (011010)
• 2 is index 54 (110110)
• l is index 37 (100101)
• 0 is index 52 (110100)

Converting the binary indices to 8-bit groups results in this:

01101011 01101001 01110100

Convert this back to ASCII results in this:

• 01101011 (107) = k
• 01101001 (105) = i
• 01110100 (116) = t

Following the same procedure for dGVu (left as an exercise for the reader), decodes to ten. Together, a2l0dGVu encodes to kitten.

For the case when the last chunk has pad characters, decode any complete 8-bit chunk, and ignore any chunk that is shorter than 8 bits.

For example, let's consider the case where the last chunk is aw==. Looking up the non-pad characters results in this:

• a is index 26 (011010)
• w is index 48 (110000)

Dividing this into 8-byte groups results in this:

01101011 0000

Ignoring the last 4 bits results in this:

• 01101011 (107) = k

Therefore, this decodes to k.

For the final decode example, let's consider the case where the last chunk is a2k=. Looking up the non-pad characters results in this:

• a is index 26 (011010)
• 2 is index 54 (110110)
• k is index 36 (100100)

Dividing this into 8-byte groups results in this:

01101011 01101001 00

Ignoring the last 2 bits results in this:

• 01101011 (107) = k
• 01101001 (105) = i

Therefore, this decodes to ki.

Further Reading

You can read more about Base64 in Wikipedia.

Requirements

Write a program that accepts two parameters: the mode (encode or decode) and some text.

• If the mode is decode, it should print the decoded Base64 text.
• If the mode is encode, it should print the encoded text.
• It should print an error message and exit with an error when the input is invalid.

$ ./base64-encode-decode.lang "encode" "hello world"
aGVsbG8gd29ybGQ=
$ ./base64-encode-decode.lang "decode" "aGVsbG8gd29ybGQ="
hello world

Acceptable language utilities include language features and built-in libraries. External dependencies are unacceptable. Remember, the goal is to show off language features and utilities.

In this project, the algorithm must handle ASCII strings. You don't need to worry about handling a string in the general case.

Testing

Every project in the Sample Programs repo should be tested. In this section, we specify the set of tests specific to Base64 Encode Decode. In order to keep things simple, we split up the testing as follows:

• Base64 Encode Valid Tests
• Base64 Decode Valid Tests
• Base64 Invalid Inputs Tests
• Base64 Invalid Encode Tests
• Base64 Invalid Decode Tests

Base64 Encode Valid Tests

Description	Input	Output
Lowercase String	"encode" "hello world"	"aGVsbG8gd29ybGQ="
Long String	"encode" "They swam along the boat at incredible speeds."	"VGhleSBzd2FtIGFsb25nIHRoZSBib2F0IGF0IGluY3JlZGlibGUgc3BlZWRzLg=="
Numbers	"encode" "1234567890"	"MTIzNDU2Nzg5MA=="
Symbols	"encode" "xyz!#$%&()*+,-./:;<=>?@[\]^_`{|}~"	"eHl6ISMkJSYoKSorLC0uLzo7PD0+P0BbXF1eX2B7fH1+"
All Base64 Characters	"encode" "! }gggIIT55;qqs!!Gjjb??=~~2$$+;;i::x..4kk,ppnoo"	"ISAgfWdnZ0lJVDU1O3FxcyEhR2pqYj8/PX5+MiQkKzs7aTo6eC4uNGtrLHBwbm9v"

Base64 Decode Valid Tests

Description	Input	Output
Lowercase String	"decode" "aGVsbG8gd29ybGQ="	"hello world"
Long String	"decode" "VGhleSBzd2FtIGFsb25nIHRoZSBib2F0IGF0IGluY3JlZGlibGUgc3BlZWRzLg=="	"They swam along the boat at incredible speeds."
Numbers	"decode" "MTIzNDU2Nzg5MA=="	"1234567890"
Symbols	"decode" "eHl6ISMkJSYoKSorLC0uLzo7PD0+P0BbXF1eX2B7fH1+"	"xyz!#$%&()*+,-./:;<=>?@[\]^_`{|}~"
All Base64 Characters	"decode" "ISAgfWdnZ0lJVDU1O3FxcyEhR2pqYj8/PX5+MiQkKzs7aTo6eC4uNGtrLHBwbm9v"	"! }gggIIT55;qqs!!Gjjb??=~~2$$+;;i::x..4kk,ppnoo"

Base64 Invalid Inputs Tests

Description	Input
No Input
Invalid Mode	"blue" "Oh look a Pascal triangle"

All of these tests should output the following:

Usage: please provide a mode and a string to encode/decode

Base64 Invalid Encode Tests

Description	Input
Missing String	"encode"
Empty String	"encode" ""

All of these tests should output the following:

Usage: please provide a mode and a string to encode/decode

Base64 Invalid Decode Tests

Description	Input
Missing String	"decode"
Empty String	"decode" ""
Length Number Not Multiple Of 4	"decode" "hello+world"
Invalid Characters	"decode" "hello world="
Too Many Pad Characters At End	"decode" "MTIzNDU2Nzg5M==="
Pad Characters In Middle	"decode" "MTIzNDU2=Nzg5M=="

All of these tests should output the following:

Usage: please provide a mode and a string to encode/decode

Articles

There are 8 articles:

• Base64 Encode Decode in Algol68
• Base64 Encode Decode in Beef
• Base64 Encode Decode in Commodore Basic
• Base64 Encode Decode in Euphoria
• Base64 Encode Decode in Go
• Base64 Encode Decode in Php
• Base64 Encode Decode in Python
• Base64 Encode Decode in Zig

---