Make charCategorizer only use the highest-precedence set of word chars

marijnh · marijnh · commit 386d55d3c243 · 2025-12-22T15:47:55.000+01:00
FIX: `EditorState.charCategorizer` now only uses the highest-precedence set of word characters from the language data, to allow overriding these. See https://discuss.codemirror.net/t/dash-separated-kebab-case-selection-too-inclusive/9619
diff --git a/src/state.ts b/src/state.ts
@@ -371,7 +371,8 @@ export class EditorState {
   ///  - Space (contains only whitespace)
   ///  - Other (anything else)
   charCategorizer(at: number): (char: string) => CharCategory {
-    return makeCategorizer(this.languageDataAt<string>("wordChars", at).join(""))
+    let chars = this.languageDataAt<string>("wordChars", at)
+    return makeCategorizer(chars.length ? chars[0] : "")
   }
 
   /// Find the word at the given position, meaning the range

Original file line number	Diff line number	Diff line change
`@@ -371,7 +371,8 @@ export class EditorState {`
`371`	`371`	`/// - Space (contains only whitespace)`
`372`	`372`	`/// - Other (anything else)`
`373`	`373`	`charCategorizer(at: number): (char: string) => CharCategory {`
`374`		`- return makeCategorizer(this.languageDataAt<string>("wordChars", at).join(""))`
	`374`	`+ let chars = this.languageDataAt<string>("wordChars", at)`
	`375`	`+ return makeCategorizer(chars.length ? chars[0] : "")`
`375`	`376`	`}`
`376`	`377`
`377`	`378`	`/// Find the word at the given position, meaning the range`