fix versionsort chunk split on non-ASCII numerics

src/sort.rs

@@ -65,7 +65,7 @@ impl<'a> VersionChunkIter<'a> {
                 break;
             }
 
-            if !c.is_numeric() {
+            if !c.is_ascii_digit() {
                 continue;
             }
 
@@ -283,6 +283,10 @@ mod test {
                 source: "009"
             })
         );
+
+        // '๙' = U+0E59 THAI DIGIT NINE, General Category Nd
+        let mut iter = VersionChunkIter::new("x๙v");
+        assert_eq!(iter.next(), Some(VersionChunk::Str("x๙v")));
     }
 
     #[test]
@@ -297,6 +301,11 @@ mod test {
         input.sort_by(|a, b| version_sort(a, b));
         assert_eq!(input, expected);
 
+        let mut input = vec!["x๙x", "xéx", "x0x"];
+        let expected = vec!["x0x", "xéx", "x๙x"];
+        input.sort_by(|a, b| version_sort(a, b));
+        assert_eq!(input, expected);
+
         let mut input = vec!["applesauce", "apple"];
         let expected = vec!["apple", "applesauce"];
         input.sort_by(|a, b| version_sort(a, b));

tests/source/versionsort_non_ascii_numerics.rs

@@ -0,0 +1,33 @@
+use std::cmp::Ordering;
+use print๙msg::print as first_print;
+use print0msg::print as second_print;
+use printémsg::print as third_print;
+
+fn main() {
+    first_print();
+    second_print();
+    third_print();
+
+    assert_eq!("print๙msg".cmp("printémsg"), Ordering::Greater);
+}
+
+/// '๙' = 0E59;THAI DIGIT NINE;Nd;
+mod print๙msg {
+    pub fn print() {
+        println!("Non-ASCII Decimal_Number")
+    }
+}
+
+/// '0' = 0030;DIGIT ZERO;Nd;
+mod print0msg {
+    pub fn print() {
+        println!("ASCII Decimal_Number")
+    }
+}
+
+/// 'é' = 00E9;LATIN SMALL LETTER E WITH ACUTE;Ll;
+mod printémsg {
+    pub fn print() {
+        println!("Lowercase_Letter")
+    }
+}

tests/target/versionsort_non_ascii_numerics.rs

@@ -0,0 +1,33 @@
+use print0msg::print as second_print;
+use printémsg::print as third_print;
+use print๙msg::print as first_print;
+use std::cmp::Ordering;
+
+fn main() {
+    first_print();
+    second_print();
+    third_print();
+
+    assert_eq!("print๙msg".cmp("printémsg"), Ordering::Greater);
+}
+
+/// '๙' = 0E59;THAI DIGIT NINE;Nd;
+mod print๙msg {
+    pub fn print() {
+        println!("Non-ASCII Decimal_Number")
+    }
+}
+
+/// '0' = 0030;DIGIT ZERO;Nd;
+mod print0msg {
+    pub fn print() {
+        println!("ASCII Decimal_Number")
+    }
+}
+
+/// 'é' = 00E9;LATIN SMALL LETTER E WITH ACUTE;Ll;
+mod printémsg {
+    pub fn print() {
+        println!("Lowercase_Letter")
+    }
+}