Re: [encoding] Big5 encoder treats a code point as error when both an HKSCS and non-HKSCS pointer exists for the code point (#9)

@annevk, seems good to me.

@inexorabletash, I think it would be good to have a framework that separated test sequences from harnesses and supported generating tests for at least TextDecoder in window context, TextDecoder in Workers, decoding in the HTML parser with various amounts of padding before the test sequence. Then implementors could also generate tests that hook into non-Web-exposed APIs like e.g. https://hg.mozilla.org/projects/htmlparser/file/0d906fb1ab90/test-src/nu/validator/encoding/test/Big5Tester.java#l33 .

---
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/encoding/issues/9#issuecomment-137417068