[Bug #21842] Let rb_interned_str return US-ASCII if possible

Author: nobu

include/ruby/internal/intern/string.h

@@ -444,8 +444,8 @@ VALUE rb_str_to_interned_str(VALUE str);
  *                           terminating NUL character.
  * @exception  rb_eArgError  `len` is negative.
  * @return     A  found or  created instance  of ::rb_cString,  of `len`  bytes
- *             length, of  "binary" encoding,  whose contents are  identical to
- *             that of `ptr`.
+ *             length,  of US-ASCII  or "binary"  encoding, whose  contents are
+ *             identical to that of `ptr`.
  * @pre        At  least  `len` bytes  of  continuous  memory region  shall  be
  *             accessible via `ptr`.
  */

string.c

@@ -12709,7 +12709,15 @@ VALUE
 rb_interned_str(const char *ptr, long len)
 {
     struct RString fake_str = {RBASIC_INIT};
-    return register_fstring(setup_fake_str(&fake_str, ptr, len, ENCINDEX_US_ASCII), true, false);
+    int encidx = ENCINDEX_US_ASCII;
+    int coderange = ENC_CODERANGE_7BIT;
+    if (len > 0 && search_nonascii(ptr, ptr + len)) {
+        encidx = ENCINDEX_ASCII_8BIT;
+        coderange = ENC_CODERANGE_VALID;
+    }
+    VALUE str = setup_fake_str(&fake_str, ptr, len, encidx);
+    ENC_CODERANGE_SET(str, coderange);
+    return register_fstring(str, true, false);
 }
 
 VALUE

test/-ext-/string/test_interned_str.rb

@@ -9,4 +9,9 @@ def test_interned_str
     src << "b" * 20
     assert_equal "a" * 20, interned_str
   end
+
+  def test_interned_str_encoding
+    src = :ascii.name
+    assert_equal Encoding::US_ASCII, Bug::String.rb_interned_str_dup(src).encoding
+  end
 end