Question

I made a program and I created an export function for documents (entries).
The procedure: The user has favorited documents. There exist 2-3 strategies
(BibTex,RIS,HTML) a user can choose to export his documents.
For each strategy a new .zip file is being generated with all documents inside.
The created .zip archives are sent to the user via email.

For me ( Windows ) it works great. I can extract those archives without any problems.
But a friend of my, who is using Mac, gets errors while extracting them and I do not know why.

Here the important code:

for ( String strategy : strategies ) {
// Coderedundanz
// Jede Strategie benötigt eigene Parameter
if (strategy.equals("BibTex")) {
    _zipName = "ezdl_export_bibtex";
    _fileExtension = ".bib";
    _strategy = csf.bibTex;
}
else if (strategy.equals("RIS")) {
    _zipName = "ezdl_export_ris";
    _fileExtension = ".ris";
    _strategy = csf.ris;
}
else if (strategy.equals("HTML")) {
    _zipName = "ezdl_export_html";
    _fileExtension = ".html";
    _strategy = csf.html;
}
else {
    _zipName = _zipExtension = "";
    _fileExtension = "";
    _strategy = null;
}

// Gibt es eine korrekte Strategie?
if ( !_zipName.equals("") && !_fileExtension.equals("") && _strategy != null) {
    // 1. .zip Datei generieren
    // 2. Für jedes TextDocument eine eigene Datei erstellen
    // 3. Datei in die .zip Datei einfügen
    // 4. .zip Datei schließen und in die E-Mail hinzufügen
    File file = File.createTempFile(_zipName + _zipExtension, ".tmp");
    ZipOutputStream out = new ZipOutputStream(new FileOutputStream(file));
    out.setLevel(6);
    for ( TextDocument document : documents ) {
        out.putNextEntry(new ZipEntry( document.getOid() + _fileExtension));

        String temp = _strategy.print ( (TextDocument) document).asString().toString();

        out.write( temp.getBytes() );
        out.closeEntry();
    }

    out.finish();
    out.close();

    PreencodedMimeBodyPart part_x = new PreencodedMimeBodyPart("base64");
    part_x.setFileName(_zipName + _zipExtension);
    part_x.setContent(new String(Base64Coder.encode( getBytesFromFile (file))), "text/plain");
    multi.addBodyPart(part_x);

    if (file.exists())
        file.delete();

You see for each strategy an own archive is being created.
The program loops through the documents (TextDocument)
and with _strategy.print you get a String as output.

As I said.. for me it works great, but not on Mac.
Are there any differences? I guess.. .zip is .zip.
Or should I create tarballs (.tar.gz) for Mac?

EDIT:

serena:tmp3 alex$ unzip ezdl_export_bibtex.zip 
Archive:  ezdl_export_bibtex.zip
  End-of-central-directory signature not found.  Either this file is not
  a zipfile, or it constitutes one disk of a multi-part archive.  In the
  latter case the central directory and zipfile comment will be found on
  the last disk(s) of this archive.
note:  ezdl_export_bibtex.zip may be a plain executable, not an archive
unzip:  cannot find zipfile directory in one of ezdl_export_bibtex.zip or
        ezdl_export_bibtex.zip.zip, and cannot find 

Here is a screen: http://img3.imageshack.us/i/ziperror.png. It shows the error: “Unable to unarchive – Error – 1 – Operation not permitted”

I also changed my code to:

out.write( temp.getBytes() );
out.flush();
out.closeEntry();

But still the same problem.

Answer 1

Although it doesn’t address your problem directly, you can verify the integrity of the zip file using the -t option on the command line.

$ unzip -t java-puzzlers.zip | tail -1
No errors detected in compressed data of java-puzzlers.zip.

In addition, you can examine the parent directory’s permissions, walking up the path until you see a problem.

$ ls -ld ..
drwxr-xr-x@ 26 trashgod  staff  884 Jan 17  2010 ..
$ ls -ld ../..
drwx------+ 23 trashgod  staff  782 Dec 17 17:15 ../..

Addendum: If it “has something to do with encoding,” I always start with Joel Spolsky’s oft-cited article on the subject. This answer may be helpful, too.