@Override
            public String[] getRoles(LdapUser user, String baseDn, String accountFilter, String groupFilter, Consumer<String[]> callback) {
                // Return roles with duplicates
                String[] roles = new String[] { "role1", "role2", "role1", "Uuser", "role2" };
                callback.accept(roles);
                return roles;
            }

            @Override

     * Associates {@code key} with {@code value} in the built bimap. Duplicate keys or values are
     * not allowed, and will cause {@link #build} to fail.
     */
    @CanIgnoreReturnValue
    @Override
    public Builder<K, V> put(K key, V value) {
      super.put(key, value);
      return this;
    }

    /**
     * Adds the given {@code entry} to the bimap. Duplicate keys or values are not allowed, and will

   *
   * <p>Some algorithms that operate on a {@code SuccessorsFunction} may produce undesired results
   * if the returned {@link Iterable} contains duplicate elements. Implementations of such
   * algorithms should document their behavior in the presence of duplicates.
   *
   * <p>The elements of the returned {@code Iterable} must each be:
   *
   * <ul>
   *   <li>Non-null

     * any of the algorithm name is not supported. The returned collection has equal count of elements as passed in
     * collection of names, and if names contains duplicated elements, the returned list of algorithms will have
     * duplicates as well.
     *
     * @throws ChecksumAlgorithmServiceException if any asked algorithm name is not supported.
     * @throws NullPointerException if passed in list of names is {@code null}.

   *
   * <p>Some algorithms that operate on a {@code PredecessorsFunction} may produce undesired results
   * if the returned {@link Iterable} contains duplicate elements. Implementations of such
   * algorithms should document their behavior in the presence of duplicates.
   *
   * <p>The elements of the returned {@code Iterable} must each be:
   *
   * <ul>
   *   <li>Non-null

   *
   * <p>Some algorithms that operate on a {@code SuccessorsFunction} may produce undesired results
   * if the returned {@link Iterable} contains duplicate elements. Implementations of such
   * algorithms should document their behavior in the presence of duplicates.
   *
   * <p>The elements of the returned {@code Iterable} must each be:
   *
   * <ul>
   *   <li>Non-null

  /**
   * Returns a {@code Collector} that accumulates the input elements into a new {@code
   * ImmutableSortedSet}, ordered by the specified comparator.
   *
   * <p>If the elements contain duplicates (according to the comparator), only the first duplicate
   * in encounter order will appear in the result.
   *
   * @since 33.2.0 (available since 21.0 in guava-jre)
   */

			success: false,
		},
		// no duplicates, shall be parsed properly.
		{

 *   <li>Deleting URL queues associated with a session.</li>
 *   <li>Deleting all URL queues.</li>
 *   <li>Offering a list of URLs to the queue, ensuring duplicates are not added.</li>
 *   <li>Polling (retrieving and removing) a URL from the queue.</li>
 *   <li>Saving the session (currently a no-op).</li>
 *   <li>Checking if a URL has already been visited.</li>

Then we can declare `tags` as a set of strings:

{* ../../docs_src/body_nested_models/tutorial003_py310.py hl[12] *}

With this, even if you receive a request with duplicate data, it will be converted to a set of unique items.

And whenever you output that data, even if the source had duplicates, it will be output as a set of unique items.

And it will be annotated / documented accordingly too.

## Nested Models { #nested-models }