@willbooster/tree-sitter-xml 3.0.1 → 3.0.2

package/README.md

@@ -23,6 +23,8 @@ upstream's 0.7.0 packages in these ways:
 - Where a NUL character cannot be part of any token, the `ERROR` node covers it as it covers any other unexpected
   character, without the unexpected-character node that upstream's parsers put inside it.
 
+- Names and enumeration tokens use the [XML 1.0 Unicode character ranges](https://www.w3.org/TR/xml/#NT-NameStartChar),
+  including supplementary characters, in attributes, references, and DTD declarations.
 - Generic processing instructions accept multiline content and question marks up to the closing `?>`, including
   when another instruction or markup immediately follows.
 - Nonempty attribute values (`AttValue` and `PseudoAttValue`) and entity values (`EntityValue`) expose their full

package/common/common.mjs

@@ -70,6 +70,9 @@ export const rseq = (...rules) => repeat(seq(...rules));
 /** @param {RuleOrLiteral[]} rules */
 export const rseq1 = (...rules) => repeat1(seq(...rules));
 
+const nameStartChars = String.raw`:A-Z_a-z\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u02FF\u0370-\u037D\u037F-\u1FFF\u200C-\u200D\u2070-\u218F\u2C00-\u2FEF\u3001-\uD7FF\uF900-\uFDCF\uFDF0-\uFFFD\u{10000}-\u{EFFFF}`;
+const nameChars = String.raw`${nameStartChars}\-.0-9\u00B7\u0300-\u036F\u203F-\u2040`;
+
 const O = optional;
 
 /** @type {Record<string, ($: GrammarSymbols<any>) => RuleOrLiteral>} */
@@ -309,9 +312,9 @@ export const rules = {
 
     _S: _ => /[ \t\r\n]+/,
 
-    Name: _ => /[a-zA-Z_][a-zA-Z0-9_:.·-]*/,
+    Name: _ => new RegExp(`[${nameStartChars}][${nameChars}]*`, 'u'),
 
-    Nmtoken: _ => /[a-zA-Z0-9_:.·-]+/,
+    Nmtoken: _ => new RegExp(`[${nameChars}]+`, 'u'),
 
     _Reference: $ => choice($.EntityRef, $.CharRef),
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@willbooster/tree-sitter-xml",
-  "version": "3.0.1",
+  "version": "3.0.2",
   "description": "XML & DTD grammars for tree-sitter",
   "keywords": [
     "incremental",

package/tree-sitter.json

@@ -36,7 +36,7 @@
     }
   ],
   "metadata": {
-    "version": "3.0.1",
+    "version": "3.0.2",
     "license": "MIT",
     "description": "XML & DTD grammars for tree-sitter",
     "authors": [