Improve QName+NCName regexps

Author: tvdijen

src/XML/Assert/NCNameTrait.php

@@ -11,7 +11,21 @@
  */
 trait NCNameTrait
 {
-    private static string $ncname_regex = '/^[a-z_][\w.-]*$/Dui';
+    private static string $ncname_regex = '/
+(?(DEFINE)
+  (?<NCNameStartChar>
+    [A-Z_a-z\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{2FF}\x{370}-\x{37D}\x{37F}-\x{1FFF}\x{200C}-\x{200D}\x{2070}-\x{218F}\x{2C00}-\x{2FEF}\x{3001}-\x{D7FF}\x{F900}-\x{FDCF}\x{FDF0}-\x{FFFD}\x{10000}-\x{EFFFF}]
+  )
+  (?<NCNameChar>
+    (?&NCNameStartChar)
+    |[-.0-9\x{B7}\x{0300}-\x{036F}\x{203F}-\x{2040}]
+  )
+  (?<NCName>
+    (?&NCNameStartChar)(?&NCNameChar)*
+  )
+)
+^(?&NCName)$
+/uxD';
 
 
     /**

src/XML/Assert/QNameTrait.php

@@ -11,7 +11,21 @@
  */
 trait QNameTrait
 {
-    private static string $qname_regex = '/^([a-z_][\w.-]*)(:[a-z_][\w.-]*)?$/Di';
+    private static string $qname_regex = '/
+(?(DEFINE)
+  (?<NCNameStartChar>
+    [A-Z_a-z\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{2FF}\x{370}-\x{37D}\x{37F}-\x{1FFF}\x{200C}-\x{200D}\x{2070}-\x{218F}\x{2C00}-\x{2FEF}\x{3001}-\x{D7FF}\x{F900}-\x{FDCF}\x{FDF0}-\x{FFFD}\x{10000}-\x{EFFFF}]
+  )
+  (?<NCNameChar>
+    (?&NCNameStartChar)
+    |[-.0-9\x{B7}\x{0300}-\x{036F}\x{203F}-\x{2040}]
+  )
+  (?<NCName>
+    (?&NCNameStartChar)(?&NCNameChar)*
+  )
+)
+^(?:(?<prefix>(?&NCName)):)?(?<local>(?&NCName))$
+/uxD';
 
 
     /**

tests/XML/Assert/NCNameTest.php

@@ -46,6 +46,7 @@ public static function provideValidNCName(): array
             'valid contains dashes' => [true, '_1950-10-04_10-00'],
             'valid contains dots' => [true, 'Te.st'],
             'valid contains diacriticals' => [true, 'fööbár'],
+            'valid contains ideographs' => [true, '東京'],
             'valid prefixed v4 UUID' => [true, '_5425e58e-e799-4884-92cc-ca64ecede32f'],
         ];
     }

tests/XML/Assert/QNameTest.php

@@ -46,8 +46,11 @@ public static function provideValidQName(): array
             '1st part containing dash' => [true, 'som-e:Test'],
             '2nd part containing dash' => [true, 'some:T-est'],
             'both parts containing dash' => [true, 'so-me:T-est'],
+            'single character unprefixed' => [true, 'a'],
             // A single NCName is also a valid QName
             'no colon' => [true, 'Test'],
+            'unicode' => [true, 'café:élément'],
+            'ideographs' => [true, '東京:品'],
         ];
     }
 
@@ -61,8 +64,11 @@ public static function provideInvalidQName(): array
             'start 2nd part with dash' => [false, 'some:-Test'],
             'start both parts with dash' => [false, '-some:-Test'],
             'start with colon' => [false, ':test'],
+            'end with colon' => [false, 'test:'],
+            'empty string' => [false, ''],
             'multiple colons' => [false, 'test:test:test'],
             'start with digit' => [false, '1Test'],
+            'colon' => [false, ':'],
             'wildcard' => [false, 'Te*st'],
             // Trailing newlines are forbidden
             'trailing newline' => [false, "some:Test\n"],