Rule Proposal: require-unicode-regexp

[mysticatea]

Please describe what the rule should do:

This rule enforces people to use RegExp u flag.

Because of Annex B.1.4 Regular Expressions Patterns, regular expressions have different between u flag or not. For example:

/a{/.test("a{12}")     //→ true
/a{/u.test("a{12}")    //→ Incomplete quantifier
/a{1/.test("a{12}")    //→ true
/a{1/u.test("a{12}")   //→ Incomplete quantifier
/a{12/.test("a{12}")   //→ true
/a{12/u.test("a{12}")  //→ Incomplete quantifier
/a{12}/.test("a{12}")  //→ false
/a{12}/u.test("a{12}") //→ false
/\x/.test("x")         //→ true
/\x/u.test("x")        //→ Invalid escape

This is similar to "use strict".
The u flag makes clearer syntax and helps to catch mistakes early.

What category of rule is this? (place an "X" next to just one item)

[X] Enforces code style
[ ] Warns about a potential error
[ ] Suggests an alternate way of doing something
[ ] Other (please specify:)

Provide 2-3 code examples that this rule will warn about:

//✘ BAD
/foo/

//✓ GOOD
/foo/u

Why should this rule be included in ESLint (instead of a plugin)?

We have strict rule for "use strict".
This rule is similar to the strict rule, disabling legacy syntax to catch mistakes early.

[ljharb]

I like the rule, but I'm not sure the name is appropriate - "strict" is pretty broad.

What about require-unicode-regexp?

[mysticatea]

Sounds good to me.

[platinumazure]

@eslint/eslint-team Can we get one more supporter for this proposal? Otherwise, I think we need to close this.

[not-an-aardvark]

This issue is now accepted.