WebUtility.HtmlDecode does not decode all HTML5 character entities

[devatwork]

The current character entity replace dictionary used inside WebUtility uses the entity set as defined in HTML4, see https://github.com/dotnet/corefx/blob/master/src/System.Runtime.Extensions/src/System/Net/WebUtility.cs#L757.

However, the HTML5 spec defines additional entities, see https://www.w3.org/TR/html5/syntax.html#named-character-references.

I'm willing to send a PR to include the new named character references, if this that would be an acceptable change to the behavior of WebUtility.HtmlDecode. Any specific guidance with respect to making this change? Are there specific areas that need to be addressed?

[karelz]

Sounds like reasonable thing to do. @davidsh @CIPop any suggestions?

[karelz]

Feel free to submit a PR.
Guidance:

• We should avoid breaking compat -- i.e. do not change existing 'wrong' code (if you find any). Such changes should be discussed explicitly upfront.
• Add test coverage

[karelz]

Continued design discussion from the PR dotnet/corefx#13152 (comment):
@DamianEdwards do you have suggestion how to go from here?

The safe route is to introduce just yet-another class for HTML5 only.

[karelz]

@DamianEdwards ping?

[karelz]

Triage: Very little interest in last 3 years (just 2 upvotes). Let's close it. If it has higher demand, we can reconsider.