CODEWITHSUNDEEP
javaunzipdata-loss
Jon Ellis
Jon Ellis

Reputation: 11

Java unzip from URL misses 2kb on file

I am trying to unzip a file from the internet using the following code. On one of the files("uq.class"), after it has been unzipped from the online source, is missing about 2kb of file size(the original file is 10,084, unzipped I get 8,261). All the other files seem to be completely fine, and when I copy the uq.class file from the zip and place it in manually, it functions perfectly. Can anyone explain whats going on and provide a fix? Below is the unzipping portions of the code.

public static File unpackArchive(URL url, File targetDir) throws IOException {
    if (!targetDir.exists()) {
        targetDir.mkdirs();
    }
    InputStream in = new BufferedInputStream(url.openStream(), 2048);
    // make sure we get the actual file
    File zip = File.createTempFile("arc", ".zip", targetDir);
    OutputStream out = new BufferedOutputStream(new FileOutputStream(zip),2048);
    copyInputStream(in, out);
    out.close();
    return unpackArchive(zip, targetDir);
}
public static File unpackArchive(File theFile, File targetDir) throws IOException {
    if (!theFile.exists()) {
        throw new IOException(theFile.getAbsolutePath() + " does not exist");
    }
    if (!buildDirectory(targetDir)) {
        throw new IOException("Could not create directory: " + targetDir);
    }
    ZipFile zipFile = new ZipFile(theFile);
    for (Enumeration entries = zipFile.entries(); entries.hasMoreElements();) {
        ZipEntry entry = (ZipEntry) entries.nextElement();
        File file = new File(targetDir, File.separator + entry.getName());
        if (!buildDirectory(file.getParentFile())) {
            throw new IOException("Could not create directory: " + file.getParentFile());
        }
        if (!entry.isDirectory()) {
            copyInputStream(zipFile.getInputStream(entry), new BufferedOutputStream(new FileOutputStream(file),2048));
        } else {
            if (!buildDirectory(file)) {
                throw new IOException("Could not create directory: " + file);
            }
        }
    }
    zipFile.close();
    theFile.delete();
    return theFile;
}

public static void copyInputStream(InputStream in, OutputStream out) throws IOException {
    byte[] buffer = new byte[1024];
    int len = in.read(buffer);
    while (len >= 0) {
        out.write(buffer, 0, len);
        len = in.read(buffer);
    }
    in.close();
    out.close();
}
public static boolean buildDirectory(File file) {
    return file.exists() || file.mkdirs();
}

Upvotes: 1

Views: 495

Answers (2)

Jon
Jon

Reputation: 1

I don't know why I cant sign in, but I figured out the issue. I did the whole cart before the horse thing. I extracted the proper file, then extracted the old file over it, so I kept re-integrating the older file. 5 hours of programming out the window. Remember, kiddies, proper programming architecture saves you A TON of headaches.

Upvotes: 0

Dorpsidioot
Dorpsidioot

Reputation: 474

Cannot directly see anything wrong with the code at first sight. What I would recommend you doing however is closing your streams more safely. In your current implementation you close the in and out streams at the same time, close statements can cause exceptions as can read and write statements! If any one of those fails, your files will be left open and in time your application will run out of file descriptors. You're better off doing the closing in a finally statement, that way you're sure they get closed.

Upvotes: 2

Related Questions