*/
@GwtCompatible
@NullMarked
public abstract class TestStringListMultimapGenerator
    implements TestListMultimapGenerator<String, String> {

  @Override
  public SampleElements<Entry<String, String>> samples() {
    return new SampleElements<>(
        mapEntry("one", "January"),
        mapEntry("two", "February"),
        mapEntry("three", "March"),
        mapEntry("four", "April"),
        mapEntry("five", "May"));

 */
@GwtCompatible
@NullMarked
public abstract class TestStringSetMultimapGenerator
    implements TestSetMultimapGenerator<String, String> {

  @Override
  public SampleElements<Entry<String, String>> samples() {
    return new SampleElements<>(
        mapEntry("one", "January"),
        mapEntry("two", "February"),
        mapEntry("three", "March"),
        mapEntry("four", "April"),
        mapEntry("five", "May"));

            assertThrows(IllegalArgumentException.class, () -> new DummySmbResourceLocator(null));
        }

        @Test
        void unsupportedProtocolThrows() {
            String url = "http://example.com";
            assertThrows(MalformedURLException.class, () -> new DummySmbResourceLocator(url));
        }

        @Test
        void rootUrlIsRoot() throws Exception {

    # o sqlPackage: (NotRequired - Default all packages)
    #  The package of SQL file for outsideSql.
    #  This is basically for narrowing SQL-searching target,
    #  for example, when the project has SQL files for other framework.
    #  So basically you don't need this.
    #
    #  You can use variable '$$PACKAGE_BASE$$' that means 'packageBase'.

By using all the previous alternatives I had the chance to learn from all of them, take ideas, and combine them in the best way I could find for myself and the teams of developers I have worked with.

For example, it was clear that ideally it should be based on standard Python type hints.

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

 * annotated by {@link TestSubtype}.
 *
 * <p>These declaration methods rely on Java static type checking to make sure what we want to
 * assert as subtypes are really subtypes according to javac. For example:
 *
 * {@snippet :
 * class MySubtypeTests extends SubtypeTester {
 *   @TestSubtype(suppressGetSubtype = true, suppressGetSupertype = true)
 *   public <T> Iterable<? extends T> listIsSubtypeOfIterable(List<T> list) {

 * <a href="http://en.wikipedia.org/wiki/Decorator_pattern">decorator pattern</a>.
 *
 * <p><b>Warning:</b> The methods of {@code ForwardingSortedSet} forward <i>indiscriminately</i> to
 * the methods of the delegate. For example, overriding {@link #add} alone <i>will not</i> change
 * the behavior of {@link #addAll}, which can lead to unexpected behavior. In this case, you should

            String value = parser.getAttributeValue(i);

            if (name.indexOf(':') >= 0) {
                // just ignore attributes with non-default namespace (for example: xmlns:xsi)
            } else {
                checkUnknownAttribute(parser, name, tagName, strict);
            }
        }
        java.util.Set<String> parsed = new java.util.HashSet<String>();

    // Здесь ещё что-то
    "servers": [
        {
            "url": "/api/v1"
        },
        {
            "url": "https://stag.example.com",
            "description": "Staging environment"
        },
        {
            "url": "https://prod.example.com",
            "description": "Production environment"
        }
    ],
    "paths": {
            // Здесь ещё что-то
    }
}
```

        return this;
    }

    /**
     * Sets the delimiter for JavaBeans.
     * <p>
     * When copying between JavaBeans and {@link Map}, you can change the delimiter used in property names.
     * For example, if you specify an underscore as the delimiter for JavaBeans and a period as the delimiter for {@link Map},
     * the property names for the source and destination will be as follows:
     * </p>
     * <table border="1">