previous build will be preserved and reused for the subsequent builds.

### Specifying Python dependencies

During bazel build all TensorFlow's Python dependencies are pinned to their
specific versions. This is necessary to ensure reproducibility of the build.
The pinned versions of the full transitive closure of TensorFlow's dependencies
together with their corresponding hashes are specified in

        this.configuration = configuration;
    }

    public void setDependencies(List<Dependency> dependencies) {
        this.dependencies = dependencies;
    }

    /**
     * Returns the file paths of all dependencies, regardless of which tool option those paths should be placed on.
     * The returned list may contain a mix of Java class path, Java module path, and other types of path elements.
     * This collection has the same content as {@code getDependencies.values()} except that it does not contain
     * null elements.
     *
     * @return the paths of all dependencies
     */
    @Nonnull

        kotlin.srcDir(generateIdnaMappingTable.map { it.outputs })

        dependencies {
          api(libs.square.okio)
          api(libs.kotlin.stdlib)

          compileOnly(libs.animalsniffer.annotations)
        }
      }

    commonTest {
      dependencies {
        implementation(projects.okhttpTestingSupport)
        implementation(libs.assertk)

        configureInputNormalization(project);

        // convenience access to common versions used in dependencies
        project.getExtensions().getExtraProperties().set("versions", VersionProperties.getVersions());
    }

    /**
     * Makes dependencies non-transitive.
     * <p>
     * Gradle allows setting all dependencies as non-transitive very easily.
     * Sadly this mechanism does not translate into maven pom generation. In order

            Editor editor = new Editor(moduleBDoc);
            Element moduleBRoot = editor.root();
            Element dependencies = DomUtils.findChildElement(moduleBRoot, "dependencies");
            Element dependency = DomUtils.findChildElement(dependencies, "dependency");

            // Verify dependency elements exist before inference
            assertNotNull(DomUtils.findChildElement(dependency, "groupId"));

                if dependencies:
                    current_dependencies.extend(dependencies)
                if route.dependencies:
                    current_dependencies.extend(route.dependencies)
                self.add_api_websocket_route(
                    prefix + route.path,
                    route.endpoint,
                    dependencies=current_dependencies,

h4. Working with POMs during development

                // Root level dependencies
                Stream.of(
                                new DependencyContainer(root.child(DEPENDENCIES).orElse(null), DEPENDENCIES),
                                new DependencyContainer(
                                        root.child(DEPENDENCY_MANAGEMENT)
                                                .flatMap(dm -> dm.child(DEPENDENCIES))

  # Find all one-step dependencies of those tests which are from //tensorflow
  # (since external deps will come from Python-level pip dependencies),
  # excluding dependencies and files that are known to be unneccessary.
  # This creates a list of targets under //tensorflow that are required for
  # TensorFlow python tests.