Add support for v flag to regexp/no-misleading-unicode-character rule (#584)

Author: ota-meshi

.changeset/smart-chefs-poke.md

@@ -0,0 +1,5 @@
+---
+"eslint-plugin-regexp": minor
+---
+
+Add support for v flag to `regexp/no-misleading-unicode-character` rule

lib/rules/no-misleading-unicode-character.ts

@@ -35,7 +35,11 @@ function getProblem(grapheme: string, flags: ReadonlyFlags): Problem | null {
         (grapheme.length === 2 && !startsWithSurrogate(grapheme))
     ) {
         return "Multi"
-    } else if (!flags.unicode && startsWithSurrogate(grapheme)) {
+    } else if (
+        !flags.unicode &&
+        !flags.unicodeSets &&
+        startsWithSurrogate(grapheme)
+    ) {
         return "Surrogate"
     }
     return null
@@ -84,6 +88,13 @@ function getGraphemeProblems(
 ): GraphemeProblem[] {
     let offset = cc.negate ? 2 : 1
 
+    const ignoreElements = cc.elements.filter(
+        (element) =>
+            element.type === "CharacterClass" || // Nesting CharacterClass
+            element.type === "ExpressionCharacterClass" || // Nesting ExpressionCharacterClass
+            element.type === "ClassStringDisjunction",
+    )
+
     const graphemes = splitter.splitGraphemes(cc.raw.slice(offset, -1))
     const problems: GraphemeProblem[] = []
 
@@ -93,6 +104,14 @@ function getGraphemeProblems(
             const start = offset + cc.start
             const end = start + grapheme.length
 
+            if (
+                ignoreElements.some(
+                    (ignore) => ignore.start <= start && end <= ignore.end,
+                )
+            ) {
+                continue
+            }
+
             problems.push({
                 grapheme,
                 problem,
@@ -113,6 +132,7 @@ function getGraphemeProblems(
 function getGraphemeProblemsFix(
     problems: readonly GraphemeProblem[],
     cc: CharacterClass,
+    flags: ReadonlyFlags,
 ): string | null {
     if (cc.negate) {
         // we can't fix a negated character class
@@ -131,21 +151,25 @@ function getGraphemeProblemsFix(
     }
 
     // The prefix of graphemes
-    const prefix = problems
-        .map((p) => p.grapheme)
-        .sort((a, b) => b.length - a.length)
-        .join("|")
+    const prefixGraphemes = problems.map((p) => p.grapheme)
 
     // The rest of the character class
     let ccRaw = cc.raw
     for (let i = problems.length - 1; i >= 0; i--) {
         const { start, end } = problems[i]
         ccRaw = ccRaw.slice(0, start - cc.start) + ccRaw.slice(end - cc.start)
     }
+
+    if (flags.unicodeSets) {
+        const prefix = prefixGraphemes.join("|")
+        return `[\\q{${prefix}}${ccRaw.slice(1, -1)}]`
+    }
+
     if (ccRaw.startsWith("[^")) {
         ccRaw = `[\\${ccRaw.slice(1)}`
     }
 
+    const prefix = prefixGraphemes.sort((a, b) => b.length - a.length).join("|")
     let fix = prefix
     let singleAlternative = problems.length === 1
     if (ccRaw !== "[]") {
@@ -238,8 +262,7 @@ export default createRule("no-misleading-unicode-character", {
                         start: problems[0].start,
                         end: problems[problems.length - 1].end,
                     }
-
-                    const fix = getGraphemeProblemsFix(problems, ccNode)
+                    const fix = getGraphemeProblemsFix(problems, ccNode, flags)
 
                     const graphemes = problems
                         .map((p) => mention(p.grapheme))

tests/lib/rules/no-misleading-unicode-character.ts

@@ -3,7 +3,7 @@ import rule from "../../../lib/rules/no-misleading-unicode-character"
 
 const tester = new RuleTester({
     parserOptions: {
-        ecmaVersion: 2020,
+        ecmaVersion: "latest",
         sourceType: "module",
     },
 })
@@ -46,6 +46,13 @@ tester.run("no-misleading-unicode-character", rule as any, {
 
         // Ignore escaped symbols because it's obvious they aren't together
         `/[\\uD83D\\uDC4D]/`,
+
+        // ES2024
+        "var r = /[👍]/v",
+        String.raw`var r = /^[\q{👶🏻}]$/v`,
+        String.raw`var r = /[🇯\q{abc}🇵]/v`,
+        "var r = /[🇯[A]🇵]/v",
+        "var r = /[🇯[A--B]🇵]/v",
     ],
     invalid: [
         {
@@ -262,5 +269,58 @@ tester.run("no-misleading-unicode-character", rule as any, {
             options: [{ fixable: true }],
             errors: [{ messageId: "characterClass" }],
         },
+
+        // ES2024
+        {
+            code: String.raw`/[[👶🏻]]/v`,
+            output: String.raw`/[[\q{👶🏻}]]/v`,
+            options: [{ fixable: true }],
+            errors: [{ messageId: "characterClass" }],
+        },
+        {
+            code: String.raw`/[👶🏻[👨‍👩‍👦]]/v`,
+            output: String.raw`/[\q{👶🏻}[👨‍👩‍👦]]/v`,
+            options: [{ fixable: true }],
+            errors: [
+                { messageId: "characterClass", column: 3 },
+                { messageId: "characterClass", column: 8 },
+            ],
+        },
+        {
+            code: String.raw`/[👶🏻👨‍👩‍👦]/v`,
+            output: String.raw`/[\q{👶🏻|👨‍👩‍👦}]/v`,
+            options: [{ fixable: true }],
+            errors: [{ messageId: "characterClass" }],
+        },
+        {
+            code: String.raw`/[👶🏻&👨‍👩‍👦]/v`,
+            output: String.raw`/[\q{👶🏻|👨‍👩‍👦}&]/v`,
+            options: [{ fixable: true }],
+            errors: [{ messageId: "characterClass" }],
+        },
+        {
+            code: String.raw`/[^👶🏻&👨‍👩‍👦]/v`,
+            output: null,
+            options: [{ fixable: true }],
+            errors: [{ messageId: "characterClass" }],
+        },
+        {
+            code: String.raw`/[^👨‍👩‍👦]/v`,
+            output: null,
+            options: [{ fixable: true }],
+            errors: [{ messageId: "characterClass" }],
+        },
+        {
+            code: String.raw`new RegExp("[👨‍👩‍👦]", "v")`,
+            output: String.raw`new RegExp("[\\q{👨‍👩‍👦}]", "v")`,
+            options: [{ fixable: true }],
+            errors: [{ messageId: "characterClass" }],
+        },
+        {
+            code: `/👨‍👩‍👦+/v`,
+            output: `/(?:👨‍👩‍👦)+/v`,
+            options: [{ fixable: true }],
+            errors: [{ messageId: "quantifierMulti" }],
+        },
     ],
 })