java.io.File file = new java.io.File("myfile.txt");
file.length();
This returns the length of the file in bytes or 0
if the file does not exist. There is no built-in way to get the size of a folder, you are going to have to walk the directory tree recursively (using the listFiles()
method of a file object that represents a directory) and accumulate the directory size for yourself:
public static long folderSize(File directory) {
long length = 0;
for (File file : directory.listFiles()) {
if (file.isFile())
length += file.length();
else
length += folderSize(file);
}
return length;
}
WARNING: This method is not sufficiently robust for production use. directory.listFiles()
may return null
and cause a NullPointerException
. Also, it doesn't consider symlinks and possibly has other failure modes. Use this method.
Using java-7 nio api, calculating the folder size can be done a lot quicker.
Here is a ready to run example that is robust and won't throw an exception. It will log directories it can't enter or had trouble traversing. Symlinks are ignored, and concurrent modification of the directory won't cause more trouble than necessary.
/**
* Attempts to calculate the size of a file or directory.
*
* <p>
* Since the operation is non-atomic, the returned value may be inaccurate.
* However, this method is quick and does its best.
*/
public static long size(Path path) {
final AtomicLong size = new AtomicLong(0);
try {
Files.walkFileTree(path, new SimpleFileVisitor<Path>() {
@Override
public FileVisitResult visitFile(Path file, BasicFileAttributes attrs) {
size.addAndGet(attrs.size());
return FileVisitResult.CONTINUE;
}
@Override
public FileVisitResult visitFileFailed(Path file, IOException exc) {
System.out.println("skipped: " + file + " (" + exc + ")");
// Skip folders that can't be traversed
return FileVisitResult.CONTINUE;
}
@Override
public FileVisitResult postVisitDirectory(Path dir, IOException exc) {
if (exc != null)
System.out.println("had trouble traversing: " + dir + " (" + exc + ")");
// Ignore errors traversing a folder
return FileVisitResult.CONTINUE;
}
});
} catch (IOException e) {
throw new AssertionError("walkFileTree will not throw IOException if the FileVisitor does not");
}
return size.get();
}
You need FileUtils#sizeOfDirectory(File)
from commons-io.
Note that you will need to manually check whether the file is a directory as the method throws an exception if a non-directory is passed to it.
WARNING: This method (as of commons-io 2.4) has a bug and may throw IllegalArgumentException
if the directory is concurrently modified.
In Java 8:
long size = Files.walk(path).mapToLong( p -> p.toFile().length() ).sum();
It would be nicer to use Files::size
in the map step but it throws a checked exception.
UPDATE:
You should also be aware that this can throw an exception if some of the files/folders are not accessible. See this question and another solution using Guava.
public static long getFolderSize(File dir) {
long size = 0;
for (File file : dir.listFiles()) {
if (file.isFile()) {
System.out.println(file.getName() + " " + file.length());
size += file.length();
}
else
size += getFolderSize(file);
}
return size;
}
For Java 8 this is one right way to do it:
Files.walk(new File("D:/temp").toPath())
.map(f -> f.toFile())
.filter(f -> f.isFile())
.mapToLong(f -> f.length()).sum()
It is important to filter out all directories, because the length method isn't guaranteed to be 0 for directories.
At least this code delivers the same size information like Windows Explorer itself does.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With