Question

It’s been a while since I’ve done Java I/O, and I’m not aware of the latest “right” ways to work with Zip and GZip files. I don’t necessarily need a full working demo – I’m primarily looking for the right interfaces and methods to be using. Yes, I could look up any random tutorial on this, but performance is an issue (these files can get pretty big) and I do care about using the best tool for the job.

The basic process I’ll be implementing:

• Download a bunch of files (that might be zipped, gzipped, or both) to a temp folder.
• Add all the extracted files to a new zip file in a temp folder.

The input files might be compressed and archived more than once. For example, the “full extraction” should take any of the following inputs (I’m not in control of these), and leave behind foo.txt:

• foo.txt.gz
• foo.txt.zip
• foo.txt.gz.zip
• foo.txt.zip.gz
• …
• foo.txt.gz.gz.gz.zip.gz.zip.zip.gz.gz
• …

Then, I might be left with foo.txt, bar.mp3, baz.exe – so I would just add them all to a new zip file with some generic name.

Questions:

• With file size being a potential concern, which (interfaces/classes/methods) should I use to quickly:
  • extract zip files?
  • extract gzip files?
  • write zip files?
• Am I better off keeping the individual extracted files in memory before writing back to the disk? Or,
• Do potentially large files make that a bad idea?

Answer 1

Note that TrueZip, the library suggested below, has been superseded by TrueVFS.

---

I’ve found the TrueZIP library useful. It allows you to treat archive files as if they’re just another file system and use the familiar Java I/O APIs.

Unlike the java.util.zip API, TrueZIP provides random access to the contents of the archive, so file size should not be a concern. If I remember correctly, it will detect archive files and not try to redundantly compress them when you put them into an archive.

Quoting the TrueZIP page:

The TrueZIP API provides drop-in replacements for the well-known classes File, FileInputStream and FileOutputStream. This design makes TrueZIP very simple to use: All that is required to archive-enable most client applications is to add a few import statements for the package de.schlichtherle.io and add some type casts where required.

Now you can simply address archive files like directories in a path name. For example, the path name "archive.zip/readme" addresses the archive entry readme within the ZIP file archive.zip. Note that file name suffixes are fully configurable and TrueZIP automatically detects false positives and reverts back to treat them like ordinary files or directories. This works recursively, so an archive file may even be enclosed in another archive file, like in outer.zip/inner.zip/readme.