Documentation

Walking the File Tree
Trail: Essential Classes
Lesson: Basic I/O
Section: File I/O (Featuring NIO.2)

Walking the File Tree

Do you need to create an application that will recursively visit all the files in a file tree? Perhaps you need to delete every .class file in a tree, or find every file that hasn't been accessed in the last year. You can do so with the FileVisitor interface.

This section covers the following:

The FileVisitor Interface

To walk a file tree, you first need to implement a FileVisitor. A FileVisitor specifies the required behavior at key points in the traversal process: when a file is visited, before a directory is accessed, after a directory is accessed, or when a failure occurs. The interface has four methods that correspond to these situations:

If you don't need to implement all four of the FileVisitor methods, instead of implementing the FileVisitor interface, you can extend the SimpleFileVisitor class. This class, which implements the FileVisitor interface, visits all files in a tree and throws an IOError when an error is encountered. You can extend this class and override only the methods that you require.

Here is an example that extends SimpleFileVisitor to print all entries in a file tree. It prints the entry whether the entry is a regular file, a symbolic link, a directory, or some other "unspecified" type of file. It also prints the size, in bytes, of each file. Any exception that is encountered is printed to the console.

The FileVisitor methods are shown in bold:

import static java.nio.file.FileVisitResult.*;

public static class PrintFiles
    extends SimpleFileVisitor<Path> {

    // Print information about
    // each type of file.
    @Override
    public FileVisitResult visitFile(Path file,
                                   BasicFileAttributes attr) {
        if (attr.isSymbolicLink()) {
            System.out.format("Symbolic link: %s ", file);
        } else if (attr.isRegularFile()) {
            System.out.format("Regular file: %s ", file);
        } else {
            System.out.format("Other: %s ", file);
        }
        System.out.println("(" + attr.size() + "bytes)");
        return CONTINUE;
    }

    // Print each directory visited.
    @Override
    public FileVisitResult postVisitDirectory(Path dir,
                                          IOException exc) {
        System.out.format("Directory: %s%n", dir);
        return CONTINUE;
    }

    // If there is some error accessing
    // the file, let the user know.
    // If you don't override this method
    // and an error occurs, an IOException 
    // is thrown.
    @Override
    public FileVisitResult visitFileFailed(Path file,
                                       IOException exc) {
        System.err.println(exc);
        return CONTINUE;
    }
}

Kickstarting the Process

Once you have implemented your FileVisitor, how do you initiate the file walk? There are two walkFileTree methods in the Files class.

The first method requires only a starting point and an instance of your FileVisitor. You can invoke the PrintFiles file visitor as follows:

Path startingDir = ...;
PrintFiles pf = new PrintFiles();
Files.walkFileTree(startingDir, pf);

The second walkFileTree method enables you to additionally specify a limit on the number of levels visited and a set of FileVisitOption enums. If you want to ensure that this method walks the entire file tree, you can specify Integer.MAX_VALUE for the maximum depth argument.

You can specify the FileVisitOption enum, FOLLOW_LINKS, which indicates that symbolic links should be followed.

This code snippet shows how the four-argument method can be invoked:

import static java.nio.file.FileVisitResult.*;

Path startingDir = ...;

EnumSet<FileVisitOption> opts = EnumSet.of(FOLLOW_LINKS);

Finder finder = new Finder(pattern);
Files.walkFileTree(startingDir, opts, Integer.MAX_VALUE, finder);

Considerations When Creating a FileVisitor

A file tree is walked depth first, but you cannot make any assumptions about the iteration order that subdirectories are visited.

If your program will be changing the file system, you need to carefully consider how you implement your FileVisitor.

For example, if you are writing a recursive delete, you first delete the files in a directory before deleting the directory itself. In this case, you delete the directory in postVisitDirectory.

If you are writing a recursive copy, you create the new directory in preVisitDirectory before attempting to copy the files to it (in visitFiles). If you want to preserve the attributes of the source directory (similar to the UNIX cp -p command), you need to do that after the files have been copied, in postVisitDirectory. The Copy example shows how to do this.

If you are writing a file search, you perform the comparison in the visitFile method. This method finds all the files that match your criteria, but it does not find the directories. If you want to find both files and directories, you must also perform the comparison in either the preVisitDirectory or postVisitDirectory method. The Find example shows how to do this.

You need to decide whether you want symbolic links to be followed. If you are deleting files, for example, following symbolic links might not be advisable. If you are copying a file tree, you might want to allow it. By default, walkFileTree does not follow symbolic links.

The visitFile method is invoked for files. If you have specified the FOLLOW_LINKS option and your file tree has a circular link to a parent directory, the looping directory is reported in the visitFileFailed method with the FileSystemLoopException. The following code snippet shows how to catch a circular link and is from the Copy example:

@Override
public FileVisitResult
    visitFileFailed(Path file,
        IOException exc) {
    if (exc instanceof FileSystemLoopException) {
        System.err.println("cycle detected: " + file);
    } else {
        System.err.format("Unable to copy:" + " %s: %s%n", file, exc);
    }
    return CONTINUE;
}

This case can occur only when the program is following symbolic links.

Controlling the Flow

Perhaps you want to walk the file tree looking for a particular directory and, when found, you want the process to terminate. Perhaps you want to skip specific directories.

The FileVisitor methods return a FileVisitResult value. You can abort the file walking process or control whether a directory is visited by the values you return in the FileVisitor methods:

In this code snippet, any directory named SCCS is skipped:

import static java.nio.file.FileVisitResult.*;

public FileVisitResult
     preVisitDirectory(Path dir,
         BasicFileAttributes attrs) {
    (if (dir.getFileName().toString().equals("SCCS")) {
         return SKIP_SUBTREE;
    }
    return CONTINUE;
}

In this code snippet, as soon as a particular file is located, the file name is printed to standard output, and the file walking terminates:

import static java.nio.file.FileVisitResult.*;

// The file we are looking for.
Path lookingFor = ...;

public FileVisitResult
    visitFile(Path file,
        BasicFileAttributes attr) {
    if (file.getFileName().equals(lookingFor)) {
        System.out.println("Located file: " + file);
        return TERMINATE;
    }
    return CONTINUE;
}

Examples

The following examples demonstrate the file walking mechanism:


Previous page: Links, Symbolic or Otherwise
Next page: Finding Files