printf: Improve support for printing multi-byte values of characters

Author: sylvestre

src/uucore/src/lib/features/format/argument.rs

@@ -58,7 +58,18 @@ impl<'a, T: Iterator<Item = &'a FormatArgument>> ArgumentIter<'a> for T {
         };
         match next {
             FormatArgument::UnsignedInt(n) => *n,
-            FormatArgument::Unparsed(s) => extract_value(u64::extended_parse(s), s),
+            FormatArgument::Unparsed(s) => {
+                // Check if the string is a character literal enclosed in quotes
+                if s.starts_with(['"', '\'']) && s.len() > 2 {
+                    // Extract the content between the quotes safely
+                    let chars: Vec<char> =
+                        s.trim_matches(|c| c == '"' || c == '\'').chars().collect();
+                    if chars.len() == 1 {
+                        return chars[0] as u64; // Return the Unicode code point
+                    }
+                }
+                extract_value(u64::extended_parse(s), s)
+            }
             _ => 0,
         }
     }

tests/by-util/test_printf.rs

@@ -1385,3 +1385,13 @@ fn float_arg_with_whitespace() {
         .fails()
         .stderr_contains("expected a numeric value");
 }
+
+#[test]
+fn mb_input() {
+    for format in ["\"á", "\'á"] {
+        new_ucmd!()
+            .args(&["%04x\n", format])
+            .succeeds()
+            .stdout_only("00e1\n");
+    }
+}