Support non-ASCII characters in EuiSearchBar & Query

CHANGELOG.md

@@ -3,6 +3,10 @@
 - Added `onBlur` prop to `EuiComboBox` ([#1400](https://github.com/elastic/eui/pull/1400))
 - Added `initialFocus` prop typedefs to `EuiModal` and `EuiPopover` ([#1410](https://github.com/elastic/eui/pull/1410))
 
+**Bug fixes**
+
+- Support extended characters (e.g. non-latin, unicode) in `EuiSearchBar` and `EuiQuery` ([#1415](https://github.com/elastic/eui/pull/1415))
+
 ## [`6.2.0`](https://github.com/elastic/eui/tree/v6.2.0)
 
 - Added `logoCodesandbox` and updated `apmApp` icons ([#1407](https://github.com/elastic/eui/pull/1407))

src/components/search_bar/query/default_syntax.js

@@ -129,6 +129,16 @@ wordChar
   = alnum
   / [-_*:]
   / escapedChar
+  / extendedGlyph
+
+// This isn't _strictly_ correct:
+// for our purposes, a non-ascii word character is considered to
+// be anything above \`Latin-1 Punctuation & Symbols\`, which ends at U+00BF
+// This allows any non-ascii character, including the full set of unicode characters
+// even those in the supplementary planes (U+010000 → U+10FFFF) as those will be seen individually
+// in their surrogate pairs which are of the format /[\uD800-\uDBFF][\uDC00-\uDFFF]/
+extendedGlyph
+  = [\u00C0-\uFFFF]
 
 escapedChar
   = "\\\\" reservedChar

src/components/search_bar/query/default_syntax.test.js

@@ -68,6 +68,28 @@ describe('defaultSyntax', () => {
     expect(clause.value).toBe('dash-3');
   });
 
+  test('unicode field and term values', () => {
+    const query = 'name:👸Queen_Elizabeth 🤴King_Henry';
+    const ast = defaultSyntax.parse(query);
+
+    expect(ast).toBeDefined();
+    expect(ast.clauses).toBeDefined();
+    expect(ast.clauses).toHaveLength(2);
+
+    let clause = ast.getSimpleFieldClause('name', '👸Queen_Elizabeth');
+    expect(clause).toBeDefined();
+    expect(AST.Field.isInstance(clause)).toBe(true);
+    expect(AST.Match.isMustClause(clause)).toBe(true);
+    expect(clause.field).toBe('name');
+    expect(clause.value).toBe('👸Queen_Elizabeth');
+
+    clause = ast.getTermClause('🤴King_Henry');
+    expect(clause).toBeDefined();
+    expect(AST.Term.isInstance(clause)).toBe(true);
+    expect(AST.Match.isMustClause(clause)).toBe(true);
+    expect(clause.value).toBe('🤴King_Henry');
+  });
+
   test('escaped chars as default clauses', () => {
     const query = '-\\: \\\\';
     const ast = defaultSyntax.parse(query);

yarn.lock

@@ -13720,6 +13720,11 @@ underscore@~1.4.4:
   resolved "https://registry.yarnpkg.com/underscore/-/underscore-1.4.4.tgz#61a6a32010622afa07963bf325203cf12239d604"
   integrity sha1-YaajIBBiKvoHljvzJSA88SI51gQ=
 
+unicode-11.0.0@^0.7.8:
+  version "0.7.8"
+  resolved "https://registry.yarnpkg.com/unicode-11.0.0/-/unicode-11.0.0-0.7.8.tgz#5eecdd6c2802fdd6b260661c57921d0294aeea98"
+  integrity sha512-O/7kwPxe1ZOiwbLr4/OleBnUDwDRldjubW5SqcQvz1b2EMPp5mQotOdf4L1z/5CNzSDVmWM9cFmseOI6L8vb6g==
+
 unicode-canonical-property-names-ecmascript@^1.0.4:
   version "1.0.4"
   resolved "https://registry.yarnpkg.com/unicode-canonical-property-names-ecmascript/-/unicode-canonical-property-names-ecmascript-1.0.4.tgz#2619800c4c825800efdd8343af7dd9933cbe2818"