Merge pull request #82 from Chia-Network/20240417-fix-zero-pad-repr

20240417 fix zero pad repr

resources/tests/test_binutils_api.py

@@ -13,7 +13,10 @@
     "((4 5 6) (7 8 9))",
     "(i (q . 1) (q . 2) (q . 3))",
     "((0x33 . i))",
-    "(g . (138935604050210 ()))"
+    "(g . (138935604050210 ()))",
+    "0x0001",
+    "0x00",
+    "0x5c4859"
 ]
 
 def extend_atom():

src/classic/clvm/__type_compatibility__.rs

@@ -33,7 +33,7 @@ pub fn vec_to_string(r: &[u8]) -> String {
  * @see https://github.com/python/cpython/blob/main/Objects/bytesobject.c#L1337
  * @param {Uint8Array} r - byteArray to stringify
  */
-pub fn pybytes_repr(r: &[u8], dquoted: bool) -> String {
+pub fn pybytes_repr(r: &[u8], dquoted: bool, full_repr: bool) -> String {
     let mut squotes = 0;
     let mut dquotes = 0;
     for b in r.iter() {
@@ -59,7 +59,7 @@ pub fn pybytes_repr(r: &[u8], dquoted: bool) -> String {
 
     for b in r.iter() {
         let c = *b as char;
-        if c == quote || c == '\\' {
+        if c == quote || (c == '\\' && full_repr) {
             s = (s + "\\") + char_to_string(c).as_str();
         } else if c == '\t' {
             s += "\\t";
@@ -172,11 +172,11 @@ impl Bytes {
     }
 
     pub fn to_formal_string(&self) -> String {
-        pybytes_repr(&self._b, true)
+        pybytes_repr(&self._b, true, false)
     }
 
     pub fn pybytes(&self) -> String {
-        pybytes_repr(&self._b, false)
+        pybytes_repr(&self._b, false, true)
     }
 
     pub fn hex(&self) -> String {
@@ -209,7 +209,7 @@ impl Bytes {
 
 impl Display for Bytes {
     fn fmt(&self, fmt: &mut std::fmt::Formatter<'_>) -> Result<(), std::fmt::Error> {
-        fmt.write_str(&pybytes_repr(&self._b, false))?;
+        fmt.write_str(&pybytes_repr(&self._b, false, true))?;
         Ok(())
     }
 }

src/classic/clvm_tools/binutils.rs

@@ -50,12 +50,14 @@ pub fn assemble_from_ir(
 }
 
 fn has_oversized_sign_extension(atom: &Bytes) -> bool {
+    let data = atom.data();
+
     // Can't have an extra sign extension if the number is too short.
+    // With the exception of 0.
     if atom.length() < 2 {
-        return false;
+        return data.len() == 1 && data[0] == 0;
     }
 
-    let data = atom.data();
     if data[0] == 0 {
         // This is a canonical value.  The opposite is non-canonical.
         // 0x0080 -> 128

src/tests/classic/smoke.rs

@@ -789,7 +789,7 @@ fn test_io_err_from_syntax_err() {
 fn test_bytes_to_pybytes_repr_0() {
     let b = b"\x11\x01abc\r\ntest\ttest\r\n";
     assert_eq!(
-        pybytes_repr(b, false),
+        pybytes_repr(b, false, true),
         "b'\\x11\\x01abc\\r\\ntest\\ttest\\r\\n'"
     );
 }