dd: fix ISO-8859-1 case conversion for conv=lcase/ucase (#10830)

2026-06-10 15:48:22 -07:00 · 2026-03-10 13:27:24 +01:00
parent b3f6e3b500
commit 36b5e59329
3 changed files with 198 additions and 38 deletions
@@ -2,6 +2,7 @@
 //
 // For the full copyright and license information, please view the LICENSE
 // file that was distributed with this source code.
+// spell-checker:ignore tolower toupper CTYPE

 // Note: Conversion tables are just lookup tables.
 // eg. The ASCII->EBCDIC table stores the EBCDIC code at the index
@@ -9,43 +10,55 @@

 pub type ConversionTable = [u8; 256];

-pub const ASCII_UCASE_TO_LCASE: ConversionTable = [
-    0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
-    0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
-    0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f,
-    0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f,
-    0x40, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f,
-    0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f,
-    0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f,
-    0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f,
-    0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
-    0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f,
-    0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf,
-    0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf,
-    0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf,
-    0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf,
-    0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef,
-    0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff,
-];
+/// Builds a lowercase conversion table using locale-aware libc::tolower
+/// This function builds the table dynamically based on the current locale.
+/// The nix crate doesn't provide safe wrappers for locale functions, so we use libc directly.
+pub fn build_lcase_table() -> ConversionTable {
+    // Initialize locale from environment if not already done
+    // SAFETY: setlocale is called with a valid C string and is used to initialize
+    // the locale for character conversion functions
+    unsafe { libc::setlocale(libc::LC_CTYPE, c"".as_ptr()) };

-pub const ASCII_LCASE_TO_UCASE: ConversionTable = [
-    0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
-    0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
-    0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f,
-    0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f,
-    0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f,
-    0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f,
-    0x60, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f,
-    0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f,
-    0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
-    0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f,
-    0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf,
-    0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf,
-    0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf,
-    0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf,
-    0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef,
-    0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff,
-];
+    let mut table = [0u8; 256];
+    for (i, item) in table.iter_mut().enumerate() {
+        // SAFETY: tolower is called with a valid byte value and returns a valid byte
+        *item = unsafe { libc::tolower(i as libc::c_int) } as u8;
+    }
+    table
+}
+
+/// Builds an uppercase conversion table using locale-aware libc::toupper
+/// This function builds the table dynamically based on the current locale.
+/// The nix crate doesn't provide safe wrappers for locale functions, so we use libc directly.
+pub fn build_ucase_table() -> ConversionTable {
+    // Initialize locale from environment if not already done
+    // SAFETY: setlocale is called with a valid C string and is used to initialize
+    // the locale for character conversion functions
+    unsafe { libc::setlocale(libc::LC_CTYPE, c"".as_ptr()) };
+
+    let mut table = [0u8; 256];
+    for (i, item) in table.iter_mut().enumerate() {
+        // SAFETY: toupper is called with a valid byte value and returns a valid byte
+        *item = unsafe { libc::toupper(i as libc::c_int) } as u8;
+    }
+    table
+}
+
+/// Gets the uppercase to lowercase conversion table using current locale
+pub fn get_ucase_to_lcase_table() -> &'static ConversionTable {
+    // For now, simply build a fresh table each time and leak it
+    // This is simpler and safer than trying to manage mutable static state
+    let table = build_lcase_table();
+    Box::leak(Box::new(table))
+}
+
+/// Gets the lowercase to uppercase conversion table using current locale
+pub fn get_lcase_to_ucase_table() -> &'static ConversionTable {
+    // For now, simply build a fresh table each time and leak it
+    // This is simpler and safer than trying to manage mutable static state
+    let table = build_ucase_table();
+    Box::leak(Box::new(table))
+}

 pub const ASCII_TO_EBCDIC: ConversionTable = [
    0x00, 0x01, 0x02, 0x03, 0x37, 0x2d, 0x2e, 0x2f, 0x16, 0x05, 0x25, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
@@ -552,8 +552,8 @@ fn get_ctable(
            Conversion::Ibm => &ASCII_TO_IBM,
        },
        (None, Some(case)) => match case {
-            Case::Lower => &ASCII_UCASE_TO_LCASE,
-            Case::Upper => &ASCII_LCASE_TO_UCASE,
+            Case::Lower => get_ucase_to_lcase_table(),
+            Case::Upper => get_lcase_to_ucase_table(),
        },
        (Some(conv), Some(case)) => match (conv, case) {
            (Conversion::Ascii, Case::Upper) => &EBCDIC_TO_ASCII_LCASE_TO_UCASE,
@@ -1945,3 +1945,150 @@ fn test_nocache_eof_fadvise_zero_length() {
        "Expected len=0 at EOF: {strace}"
    );
 }
+
+#[test]
+#[cfg(not(target_os = "openbsd"))]
+fn test_iso8859_1_case_conversion() {
+    use std::process::Command;
+    // Test ISO-8859-1 case conversion for accented characters
+    // Skip test if required locale is not available (common in CI environments)
+    let locale_test = Command::new("locale")
+        .arg("-a")
+        .output()
+        .ok()
+        .and_then(|output| String::from_utf8(output.stdout).ok())
+        .is_some_and(|locales| locales.contains("fr_FR"));
+
+    if !locale_test {
+        eprintln!("Skipping ISO-8859-1 test: French locale not available");
+        return;
+    }
+
+    let locale = "fr_FR";
+
+    // É (0xC9) should convert to é (0xE9) with lcase
+    let input = vec![0xC9, 0x0A]; // É\n in ISO-8859-1
+    let expected = vec![0xE9, 0x0A]; // é\n in ISO-8859-1
+    let result = new_ucmd!()
+        .args(&["conv=lcase", "status=none"])
+        .env("LC_ALL", locale)
+        .pipe_in(input)
+        .succeeds();
+    assert_eq!(result.stdout(), expected);
+
+    // é (0xE9) should convert to É (0xC9) with ucase
+    let input = vec![0xE9, 0x0A]; // é\n in ISO-8859-1
+    let expected = vec![0xC9, 0x0A]; // É\n in ISO-8859-1
+    let result = new_ucmd!()
+        .args(&["conv=ucase", "status=none"])
+        .env("LC_ALL", locale)
+        .pipe_in(input)
+        .succeeds();
+    assert_eq!(result.stdout(), expected);
+}
+
+#[test]
+fn test_locale_aware_case_conversion() {
+    // Test that case conversion respects different single-byte locales
+
+    // Test Turkish (ISO-8859-9) where 'I' has special behavior
+    // Turkish has İ (0xDD) ↔ i (0xFD) and I (0x49) ↔ ı (0xFD in some positions)
+    // For simplicity, test some basic accented characters that differ between locales
+
+    // Test with ISO-8859-9 (Turkish) - Ğ (0xD0) should convert to ğ (0xF0)
+    let input = vec![0xD0, 0x0A]; // Ğ\n in ISO-8859-9
+    let expected = vec![0xF0, 0x0A]; // ğ\n in ISO-8859-9
+    let result = new_ucmd!()
+        .args(&["conv=lcase", "status=none"])
+        .env("LC_ALL", "tr_TR.iso8859-9")
+        .pipe_in(input)
+        .succeeds();
+
+    // Note: This test may not work if the system doesn't have Turkish locale installed
+    // In that case, it should fall back to C locale behavior
+    if result.stdout() == expected {
+        println!("Turkish locale case conversion working correctly");
+    } else {
+        println!("Turkish locale not available, using fallback behavior");
+        // Test that it at least doesn't crash and produces some output
+        assert!(!result.stdout().is_empty());
+    }
+}
+
+#[test]
+fn test_french_locale_case_conversion() {
+    // Test French (ISO-8859-1) case conversion for French accented characters
+    // This test uses the same charset as the previous ISO-8859-1 test but with French locale
+
+    // Test French accented characters: À (0xC0) should convert to à (0xE0) with lcase
+    let input = vec![0xC0, 0x0A]; // À\n in ISO-8859-1
+    let expected = vec![0xE0, 0x0A]; // à\n in ISO-8859-1
+    let result = new_ucmd!()
+        .args(&["conv=lcase", "status=none"])
+        .env("LC_ALL", "fr_FR.iso8859-1")
+        .pipe_in(input)
+        .succeeds();
+
+    // Note: This test may not work if the system doesn't have French locale installed
+    // In that case, it should fall back to C locale behavior
+    if result.stdout() == expected {
+        println!("French locale case conversion working correctly for À -> à");
+    } else {
+        println!("French locale not available, using fallback behavior");
+        // Test that it at least doesn't crash and produces some output
+        assert!(!result.stdout().is_empty());
+    }
+
+    // Test reverse conversion: à (0xE0) should convert to À (0xC0) with ucase
+    let input = vec![0xE0, 0x0A]; // à\n in ISO-8859-1
+    let expected = vec![0xC0, 0x0A]; // À\n in ISO-8859-1
+    let result = new_ucmd!()
+        .args(&["conv=ucase", "status=none"])
+        .env("LC_ALL", "fr_FR.iso8859-1")
+        .pipe_in(input)
+        .succeeds();
+
+    if result.stdout() == expected {
+        println!("French locale case conversion working correctly for à -> À");
+    } else {
+        println!("French locale not available for reverse conversion, using fallback behavior");
+        assert!(!result.stdout().is_empty());
+    }
+
+    // Test another French character: Ç (0xC7) should convert to ç (0xE7) with lcase
+    let input = vec![0xC7, 0x0A]; // Ç\n in ISO-8859-1
+    let expected = vec![0xE7, 0x0A]; // ç\n in ISO-8859-1
+    let result = new_ucmd!()
+        .args(&["conv=lcase", "status=none"])
+        .env("LC_ALL", "fr_FR.iso8859-1")
+        .pipe_in(input)
+        .succeeds();
+
+    if result.stdout() == expected {
+        println!("French locale case conversion working correctly for Ç -> ç");
+    } else {
+        println!("French locale not available for Ç conversion, using fallback behavior");
+        assert!(!result.stdout().is_empty());
+    }
+}
+
+#[test]
+fn test_ascii_case_conversion_fallback() {
+    // Test that ASCII characters always convert correctly regardless of locale
+    let input = vec![b'A', b'B', b'C', 0x0A]; // ABC\n
+    let expected = vec![b'a', b'b', b'c', 0x0A]; // abc\n
+    let result = new_ucmd!()
+        .args(&["conv=lcase", "status=none"])
+        .env("LC_ALL", "C")
+        .pipe_in(input.clone())
+        .succeeds();
+    assert_eq!(result.stdout(), expected);
+
+    // Test reverse conversion
+    let result = new_ucmd!()
+        .args(&["conv=ucase", "status=none"])
+        .env("LC_ALL", "C")
+        .pipe_in(expected)
+        .succeeds();
+    assert_eq!(result.stdout(), input);
+}