# Experimental filesystem C APIs for TensorFlow.
# Will be moved in proper place once all filesystems are converted to the
# modular framework.
load("//tensorflow:tensorflow.bzl", "tf_cc_test")
load("//tensorflow/core/platform:rules_cc.bzl", "cc_library")

package(
    # copybara:uncomment default_applicable_licenses = ["//tensorflow:license"],
    licenses = ["notice"],
)

# This is only for plugins
cc_library(
    name = "filesystem_interface",

     * If the JAR is modular, its module information are ignored.
     * This type is new in Maven 4.
     */
    String CLASSPATH_JAR = "classpath-jar";

    /**
     * Artifact type name for a JAR file to unconditionally place on the module-path.
     * If the JAR is not modular, then it is loaded by Java as an unnamed module.
     * This type is new in Maven 4.
     */

     *
     * @see #getModuleInfo(Path)
     */
    private final Map<Path, PathModularization> moduleInfo;

    /**
     * Whether JAR files are modular. This map is redundant with {@link #moduleInfo},
     * but cheaper to compute when the module names are not needed.
     *
     * @see #getPathType(Path)
     */
    private final Map<Path, PathType> pathTypes;

     * @return an identification of the patch-module path for the given module.
     *
     * @see Modular#moduleName()
     */
    @Nonnull
    public static Modular patchModule(@Nonnull String moduleName) {
        return PATCH_MODULE.new Modular(moduleName);
    }

    /**
     * The {@code javax.tool} enumeration value corresponding to this {@code JavaPathType}, or {@code null} if none.

Java modules can also be runnable and you can use the application plugin to run and package such a modular application.
For this, you need to do two things in addition to what you do for a non-modular application.

First, you need to add a `module-info.java` file to describe your application module.

* Create a modular software project by combining multiple subprojects
* Share build configuration logic between subprojects using convention plugins in `buildSrc`
* Run similar named tasks in all subprojects
* Run a task in a specific subproject
* Build, bundle and run the application

== Next steps

When your project grows, you might be interested in more details how to configure JVM projects, structuring multi-project builds and dependency management:

            if (type instanceof JavaPathType.Modular) {
                type = ((JavaPathType.Modular) type).rawType();
            }
            if (JavaPathType.PATCH_MODULE.equals(type)) {
                return true;
            }
        }
        return false;
    }

    /**
     * Returns whether at least one previously added modular dependency contains a module of the given name.
     *

 */
class PathModularization {
    /**
     * A unique constant for all non-modular dependencies.
     */
    public static final PathModularization NONE = new PathModularization();

    /**
     * Name of the file to use as a sentinel value for deciding if a directory or a JAR is modular.
     */
    private static final String MODULE_INFO = "module-info.class";

    /**

/// This file builds classes needed to hold a filesystem implementation in the
/// modular world. Once all TensorFlow filesystems are converted to use the
/// plugin based approach, this file will replace the one in core/platform and
/// the names will lose the `Modular` part. Until that point, the `Modular*`
/// classes here are experimental and subject to breaking changes.

"checker". For example, the printf checker reports mistakes in
fmt.Printf format strings.

A "modular" analysis is one that inspects one package at a time but can
save information from a lower-level package and use it when inspecting a
higher-level package, analogous to separate compilation in a toolchain.
The printf checker is modular: when it discovers that a function such as
log.Fatalf delegates to fmt.Printf, it records this fact, and checks