Fix double-negatives with MatchCharacterClass

[stephentoub]

Fixes#62622
cc: @joperezr

[ghost]

Tagging subscribers to this area: @dotnet/area-system-text-regularexpressions
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Fixes #62622
cc: @joperezr

Author:	stephentoub
Assignees:	-
Labels:	area-System.Text.RegularExpressions
Milestone:	-

[joperezr]

LGTM, generated code does look much better.

Separate comment: even though I know the purpose of this is purely for readability purposes (therefore not very applicable to Compiled) would it make sense to make the changes in MatchCharacterClass on Compiled engine as well just for consistency purposes?

[joperezr]

BTW it's been a bit since I last saw that many XORs in C# 😄

[stephentoub]

BTW it's been a bit since I last saw that many XORs in C#

If you need it, flaunt it 😄

Separate comment: even though I know the purpose of this is purely for readability purposes (therefore not very applicable to Compiled) would it make sense to make the changes in MatchCharacterClass on Compiled engine as well just for consistency purposes?

Sure, I can make that change.

[stephentoub]

Separate comment: even though I know the purpose of this is purely for readability purposes (therefore not very applicable to Compiled) would it make sense to make the changes in MatchCharacterClass on Compiled engine as well just for consistency purposes?

Sure, I can make that change.

Actually, I took a look at it, and best case it saves us a single ceq operation, while adding additional complexity. This particular function is one with a signature that already differs from what's in RegexCompiled, e.g. to be able to pass out additional declarations, so I don't think this really hampers the consistency much, and I'd prefer not to augment RegexCompiler with it.