MapDifference<Integer, String> original = Maps.difference(left, right);
    MapDifference<Integer, String> same = Maps.difference(left, right);
    MapDifference<Integer, String> reverse = Maps.difference(right, left);
    MapDifference<Integer, String> diff2 = Maps.difference(left, right2);

    new EqualsTester()
        .addEqualityGroup(original, same)

    Map<K, V> onBoth = new LinkedHashMap<>();
    Map<K, ValueDifference<V>> differences = new LinkedHashMap<>();
    doDifference(left, right, valueEquivalence, onlyOnLeft, onlyOnRight, onBoth, differences);
    return new MapDifferenceImpl<>(onlyOnLeft, onlyOnRight, onBoth, differences);
  }

  /**
   * Computes the difference between two sorted maps, using the comparator of the left map, or

    Map<K, V> onBoth = new LinkedHashMap<>();
    Map<K, ValueDifference<V>> differences = new LinkedHashMap<>();
    doDifference(left, right, valueEquivalence, onlyOnLeft, onlyOnRight, onBoth, differences);
    return new MapDifferenceImpl<>(onlyOnLeft, onlyOnRight, onBoth, differences);
  }

  /**
   * Computes the difference between two sorted maps, using the comparator of the left map, or

Also, the best approach was to use already existing standards.

So, before even starting to code **FastAPI**, I spent several months studying the specs for OpenAPI, JSON Schema, OAuth2, etc. Understanding their relationship, overlap, and differences.

## Design { #design }

Then I spent some time designing the developer "API" I wanted to have as a user (as a developer using FastAPI).

All the data conversion, validation, documentation, etc. will still work as normally.

That way, we can declare just the differences between the models (with plaintext `password`, with `hashed_password` and without password):

{* ../../docs_src/extra_models/tutorial002_py310.py hl[7,13:14,17:18,21:22] *}

## `Union` or `anyOf` { #union-or-anyof }

* And it could also have a declaration of the response it should return, e.g. `response_model=InvoiceEventReceived`.

{* ../../docs_src/openapi_callbacks/tutorial001_py310.py hl[14:16,19:20,26:30] *}

There are 2 main differences from a normal *path operation*:

* It doesn't need to have any actual code, because your app will never call this code. It's only used to document the *external API*. So, the function could just have `pass`.

{* ../../docs_src/sql_databases/tutorial001_an_py310.py ln[1:11] hl[7:11] *}

The `Hero` class is very similar to a Pydantic model (in fact, underneath, it actually *is a Pydantic model*).

There are a few differences:

* `table=True` tells SQLModel that this is a *table model*, it should represent a **table** in the SQL database, it's not just a *data model* (as would be any other regular Pydantic class).

            assertTrue("Result for 7 days should be greater than 1 day", result7 > result1);
            assertTrue("Result for 1 day should be greater than 0 days", result1 > result0);

            // Verify that the differences are approximately correct
            long diff1 = result1 - result0;
            long diff7 = result7 - result0;

 *
 * <p>Starting with Java 8, the core Java class libraries provide a new "Streams" library (in {@code
 * java.util.stream}), which is similar to {@code FluentIterable} but generally more powerful. Key
 * differences include:
 *
 * <ul>
 *   <li>A stream is <i>single-use</i>; it becomes invalid as soon as any "terminal operation" such
 *       as {@code findFirst()} or {@code iterator()} is invoked. (Even though {@code Stream}

 * Iterables} class.
 *
 * <p><a id="streams"></a>
 *
 * <h3>Comparison to streams</h3>
 *
 * <p>{@link Stream} is similar to this class, but generally more powerful, and certainly more
 * standard. Key differences include:
 *
 * <ul>
 *   <li>A stream is <i>single-use</i>; it becomes invalid as soon as any "terminal operation" such
 *       as {@code findFirst()} or {@code iterator()} is invoked. (Even though {@code Stream}