Address TXT records > 255 characters; make language more consistent; separate section for root records

[decentralgabe]

Fix #94

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 83.72093% with 7 lines in your changes are missing coverage. Please review.

Project coverage is 55.63%. Comparing base (98c10ef) to head (6ae2a1e).

❗ Current head 6ae2a1e differs from pull request most recent head 5db77ba. Consider uploading reports for the commit 5db77ba to get more accurate results

Files	Patch %	Lines
impl/internal/did/did.go	83.72%	4 Missing and 3 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #162      +/-   ##
==========================================
+ Coverage   55.33%   55.63%   +0.29%     
==========================================
  Files          29       29              
  Lines        2230     2245      +15     
==========================================
+ Hits         1234     1249      +15     
  Misses        879      879              
  Partials      117      117              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[decentralgabe]

added a test vector for a chunked record

[andresuribe87]

Sweet, great work! One suggestion on handling other charsets.

[andresuribe87]

TIL (thanks chad!) that len(str) returns the number of bytes. This may not be equivalent to the number of chars. For instance, len("£") returns 2.

Given the above, if you want to actually do this into 255 chars, then see the function below.

// SplitString255 splits a string into substrings of up to 255 characters.
func SplitString255(input string) []string {
	var chunks []string
	runeArray := []rune(input) // Convert to rune slice to properly handle multi-byte characters

	for len(runeArray) > 0 {
		if len(runeArray) <= 255 {
			chunks = append(chunks, string(runeArray))
			break
		}

		chunks = append(chunks, string(runeArray[:255]))
		runeArray = runeArray[255:]
	}

	return chunks
}

But the real question is whether the library should allow setting a value in a DID Document which contains values outside the ASCII range. I think the library should. And thus, I believe the value should be base64url encoded before doing chunking, so there are no issues with existing DNS systems. If you agree, then my suggestion doesn't matter, as all ASCII chars are 1-byte when encoded in UTF-8.

[decentralgabe]

this is a great catch about bytes vs character sizes - thank you
I am opposed to b64 encoding because of size increases

[decentralgabe]

I am going to merge, but please feel free to open an issue to support non-ASCII characters.