feat: add Conversion/base64 algorithm

[paul-leydier]

A basic implementation of the base64 encoding algorithm, based on RFC4648.

[paul-leydier]

Hi everyone!

You'll find in my pull request an implementation of base64 encoding/decoding according to RFC4648.

Here are the benchmark results on my computer:

goos: windows
goarch: amd64
pkg: github.com/TheAlgorithms/Go/cipher/base64
cpu: Intel(R) Core(TM) i5-9600K CPU @ 3.70GHz
BenchmarkEncode
BenchmarkEncode/Hello_World!
BenchmarkEncode/Hello_World!-6         	 7218757	       161.4 ns/op
BenchmarkEncode/Hello_World!a
BenchmarkEncode/Hello_World!a-6        	 4290879	       281.8 ns/op
BenchmarkEncode/Hello_World!ab
BenchmarkEncode/Hello_World!ab-6       	 4653878	       264.1 ns/op
BenchmarkEncode/Empty
BenchmarkEncode/Empty-6                	100000000	        10.60 ns/op
BenchmarkEncode/6
BenchmarkEncode/6-6                    	11310638	       104.9 ns/op
BenchmarkEncode/Lorem_ipsum
BenchmarkEncode/Lorem_ipsum-6          	  260875	      4933 ns/op
BenchmarkDecode
BenchmarkDecode/Hello_World!
BenchmarkDecode/Hello_World!-6         	10344987	       117.4 ns/op
BenchmarkDecode/Hello_World!a
BenchmarkDecode/Hello_World!a-6        	 8759992	       138.0 ns/op
BenchmarkDecode/Hello_World!ab
BenchmarkDecode/Hello_World!ab-6       	 8687486	       137.6 ns/op
BenchmarkDecode/Empty
BenchmarkDecode/Empty-6                	223022956	         5.307 ns/op
BenchmarkDecode/6
BenchmarkDecode/6-6                    	32433220	        38.15 ns/op
BenchmarkDecode/Lorem_ipsum
BenchmarkDecode/Lorem_ipsum-6          	  338247	      3498 ns/op
PASS

[tjgurwara99]

Other than that everything else seems to be okay.

[paul-leydier]

@tjgurwara99 Thanks for the review, useful feedback! I updated the pull request accordingly.

[tjgurwara99]

Thinking about this, I don't think base64 is a cipher. Its a conversion. Can you move this to the conversion package instead.

[paul-leydier]

Thinking about this, I don't think base64 is a cipher. Its a conversion. Can you move this to the conversion package instead.

Indeed I thought about it but saw on the The algorithms website that it was categorized as a Cipher for both C++ and Python (although it is categorized as "conversions" for js), so I used the cipher package for consistency.

But I can definitely move it to the conversion package if you confirm?

[tjgurwara99]

Yeah, move it to conversion package. Its more appropriate there.

[paul-leydier]

I moved the code to the conversion package and renamed the function to be more precise, since the package name is not "base64" anymore.

[raklaptudirm]

@tjgurwara99 The README generation is failing.

[tjgurwara99]

Yeah, its generating the documentation fine but is unable to push to the master branch. I'm waiting on the #434 to be merged and see if the changes work. If not we need to create a token specifically for the godocmd - I've asked @siriak for it but am waiting to hear back from him.