You can easily run the experiment locally with a stripped path prefix using [Traefik](https://docs.traefik.io/).

[Download Traefik](https://github.com/containous/traefik/releases), it's a single binary, you can extract the compressed file and run it directly from the terminal.

Then create a file `traefik.toml` with:

```TOML hl_lines="3"
[entryPoints]
  [entryPoints.http]
    address = ":9999"

            assertNotNull(results);
            assertEquals(0, results.size());
        }
    }

    /**
     * Test with searcher that returns single document.
     */
    @Test
    public void test_searcherReturnsSingleDocument() throws Exception {
        try (RankFusionProcessor processor = new RankFusionProcessor()) {

    return incorporateRequirements(clonedClassRequirements, declaredRequirements, testerMethod);
  }

  /**
   * Find all the constraints explicitly or implicitly specified by a single tester annotation.
   *
   * @param testerAnnotation a tester annotation
   * @return the requirements specified by the annotation
   * @throws ConflictingRequirementsException if the requirements are mutually inconsistent.

   * if necessary. The returned map contains entries that were already cached, combined with newly
   * loaded entries; it will never contain null keys or values.
   *
   * <p>Caches loaded by a {@link CacheLoader} will issue a single request to {@link
   * CacheLoader#loadAll} for all keys which are not already present in the cache. All entries
   * returned by {@link CacheLoader#loadAll} will be stored in the cache, over-writing any

	// MinIOSnowballIgnoreDirs will skip creating empty directory objects.
	MinIOSnowballIgnoreDirs = "X-Amz-Meta-Minio-Snowball-Ignore-Dirs"
	// MinIOSnowballIgnoreErrors will ignore recoverable errors, typically single files failing to upload.
	// An error will be printed to console instead.
	MinIOSnowballIgnoreErrors = "X-Amz-Meta-Minio-Snowball-Ignore-Errors"

        final String clusterName = UUID.randomUUID().toString();
        runner.onBuild((number, settingsBuilder) -> {
            settingsBuilder.put("http.cors.enabled", true);
            settingsBuilder.put("discovery.type", "single-node");
        }).build(newConfigs().clusterName(clusterName).numOfNode(1));

        // wait for yellow status
        runner.ensureYellow();

        runner = new OpenSearchRunner();
        runner.onBuild((number, settingsBuilder) -> {
            settingsBuilder.put("http.cors.enabled", true);
            settingsBuilder.put("discovery.type", "single-node");
        })
                .build(newConfigs().clusterName("SuggesterIndexLifecycleTest")
                        .numOfNode(1)
                        .pluginTypes("org.codelibs.opensearch.extension.ExtensionPlugin"));

            val destFile = rootBuildDir.resolve("report$projectPathName-${it.name}.zip")
            zip(destFile, it)
        }

        // Zip all files in project build directory into a single zip file to avoid publishing too many tiny files
        reports.filter { it.isFile && it.toPath().startsWith(projectBuildDirPath) }
            .map { projectBuildDirPath.relativize(it.toPath()).toString() to it }

SAM Conversions
---------------

When you use Java APIs from Kotlin you can operate on Java interfaces as if they were Kotlin
lambdas. The [feature][java_sams] is available for interfaces that define a Single Abstract Method
(SAM).

But when you use Kotlin APIs from Kotlin there’s no automatic conversion. Code that used SAM lambdas
with OkHttp 3.x: must use `object :` with OkHttp 4.x:

Kotlin calling OkHttp 3.x:

    @DisplayName("Supports various wildcard values without throwing (edge cases)")
    void wildcardEdgeCases(String wildcard) throws Exception {
        // minimal successful first response with a single entry
        when(tree.send(any(Trans2FindFirst2.class), any(Trans2FindFirst2Response.class))).thenAnswer((InvocationOnMock inv) -> {
            Trans2FindFirst2Response resp = inv.getArgument(1);