numfmt: find the suffix by byte, not by character

src/uu/numfmt/src/format.rs

@@ -57,22 +57,24 @@ fn find_valid_number_with_suffix(s: &str, unit: Unit) -> Option<&str> {
     let accepts_suffix = unit != Unit::None;
     let accepts_i = [Unit::Auto, Unit::Iec(true)].contains(&unit);
 
-    let mut characters = s.chars().skip(numeric_part.len());
+    // What follows the number begins where the number ends, which is a
+    // position in bytes: the decimal separator of a locale such as ar-SA is
+    // two bytes wide, so skipping that many *characters* instead lands past
+    // the suffix, and the byte index taken from it lands inside a character.
+    let mut characters = s[numeric_part.len()..].chars();
     let potential_suffix = characters.next();
     let potential_i = characters.next();
 
     if !accepts_suffix {
         return Some(numeric_part);
     }
 
+    // Every suffix is one ASCII character, so the byte it ends on is known.
     match (potential_suffix, potential_i) {
-        (Some(suffix), None) if RawSuffix::try_from(&suffix).is_ok() => {
-            Some(&s[..=numeric_part.len()])
-        }
         (Some(suffix), Some('i')) if accepts_i && RawSuffix::try_from(&suffix).is_ok() => {
             Some(&s[..numeric_part.len() + 2])
         }
-        (Some(suffix), Some(_)) if RawSuffix::try_from(&suffix).is_ok() => {
+        (Some(suffix), _) if RawSuffix::try_from(&suffix).is_ok() => {
             Some(&s[..=numeric_part.len()])
         }
         _ => Some(numeric_part),

tests/by-util/test_numfmt.rs

@@ -1614,6 +1614,35 @@ fn test_locale_fr_rejects_period() {
         .stderr_contains("invalid");
 }
 
+/// The suffix begins where the number ends, and ar-SA's decimal separator is
+/// two bytes wide, so counting that end in characters walked into the middle
+/// of the one after it and aborted the process.
+#[test]
+#[cfg_attr(wasi_runner, ignore = "WASI: locale env vars not propagated")]
+fn test_locale_multibyte_separator_before_a_multibyte_char() {
+    for (unit, input) in [
+        ("si", "1٫€K"),
+        ("si", "1٫€Kx"),
+        ("auto", "1٫€Ki"),
+        ("iec-i", "1٫€K"),
+        ("none", "1٫€K"),
+    ] {
+        new_ucmd!()
+            .env("LC_ALL", "ar_SA.UTF-8")
+            .args(&[format!("--from={unit}"), input.into()])
+            .fails_with_code(2)
+            .stderr_contains("invalid suffix in input");
+    }
+
+    // The suffix that is there is still found: '€' is what is wrong with this
+    // one, not 'K'.
+    new_ucmd!()
+        .env("LC_ALL", "ar_SA.UTF-8")
+        .args(&["--from=si", "1٫Ki"])
+        .fails_with_code(2)
+        .stderr_is("numfmt: invalid suffix in input '1٫Ki': 'i'\n");
+}
+
 #[test]
 fn test_locale_c_uses_period() {
     // C locale should still use '.' as usual