}
      }

    fun username(username: String) =
      apply {
        this.encodedUsername = username.canonicalize(encodeSet = USERNAME_ENCODE_SET)
      }

    fun encodedUsername(encodedUsername: String) =
      apply {
        this.encodedUsername =
          encodedUsername.canonicalize(
            encodeSet = USERNAME_ENCODE_SET,
            alreadyEncoded = true,
          )
      }

      }
    }

  private fun testParseAlreadyEncoded(
    codePoint: Int,
    encoding: Encoding,
    component: Component,
  ) {
    val expected = component.canonicalize(encoding.encode(codePoint))
    val urlString = component.urlString(expected)
    val url = urlString.toHttpUrl()
    val actual = component.encodedValue(url)
    if (actual != expected) {

   *       extends Enum<?>>} canonicalizes to {@code Enum<?>}, which is a supertype (if we disregard
   *       the upper bound is implicitly an Enum too).
   *   <li>If {@code canonicalize(A) == canonicalize(B)}, then {@code Foo<A>.isSubtypeOf(Foo<B>)}
   *       and vice versa. i.e. {@code A.is(B)} and {@code B.is(A)}.
   *   <li>{@code canonicalize(canonicalize(A)) == canonicalize(A)}.
   * </ol>
   */

			}
		}

		if hexes != "" {
			hexByLine[fmt.Sprintf("%s:%d", input, lineno)] = hexes
		}

		// Canonicalize spacing in printed form.
		// First field is opcode, then tab, then arguments separated by spaces.
		// Canonicalize spaces after commas first.
		// Comma to separate argument gets a space; comma within does not.
		var buf []byte
		nest := 0

</p>

<h2 id="Source_code_representation">Source code representation</h2>

<p>
Source code is Unicode text encoded in
<a href="https://en.wikipedia.org/wiki/UTF-8">UTF-8</a>. The text is not
canonicalized, so a single accented code point is distinct from the
same character constructed from combining an accent and a letter;
those are treated as two code points.  For simplicity, this document