[COMPRESS-96] Corrupt zip files can cause infinite loop

• Type: Bug
• Status: Resolved
• Priority: Critical
• Resolution: Duplicate
• Affects Version/s: 1.0
• Fix Version/s: 1.1

Environment

OS/X java 1.6

• Assignee: Unassigned
• Reporter: Matthew Bellew
• Created: 11/Feb/10 01:52
• Updated: 05/May/10 10:23
• Resolved: 12/Feb/10 10:11

Description

I have a corrupt .zip file (in my Download directory) that causes ZipArchiveInputStream to enter an infinite loop on the second call to getEntry(). This was discovered while testing a Tika/Lucene based search application. Obviously, it would be preferable to detect and throw an exception. The short explanation is that closeEntry() calls skip(Long.MAX_VALUE). skip() calls read() which returns 0 and inf.finished() remains false, so no progress is made and skip() spins.

If it helps, entry.getSize() returns -1 for the first entry. I can provide the file to repro (775800 bytes).

UPDATE: I just hit he same error scanning a truncated commons-compress-1.0-bin.zip (437687 bytes) with this code.

package main;

import org.apache.commons.compress.archivers.ArchiveEntry;
import org.apache.commons.compress.archivers.zip.ZipArchiveInputStream;

import java.io.File;
import java.io.FileInputStream;
import java.io.IOException;

public class Main
{
public static void main(String[] args) throws IOException
{
File f = new File(args[0]);
FileInputStream fis = new FileInputStream(f);
ZipArchiveInputStream zip = new ZipArchiveInputStream(fis);
ArchiveEntry entry;

entry = zip.getNextEntry();
while (null != entry)

{ if (entry.isDirectory()) System.out.printf("%s/\n", entry.getName()); else System.out.printf("%s %d\n", entry.getName(), entry.getSize()); entry = zip.getNextEntry(); }

System.out.print("DONE\n");
}
}

Issue Links

• duplicates: COMPRESS-87 ZipArchiveInputStream doesn't report the end of a truncated archive