Java SE (Java Platform, Standard Edition)

I have what may sound like a bizzare request but I need to know how to decode the format of the .idx files (naming and content) that are created in the JRE cache. The reason for this is I have devices that are located over extremely slow WAN links. When I manage these devices via web interface, they download a .jar file which always times-out no matter what. Since I have the exact devices on my local LAN with the exact same .jar file, what I want to do is take an existing cache entry for the local device and "fool" the JRE it into thinking it's the cached entry for the remote device over the WAN link. I noticed that inside the .idx file is the URL and it seemed like all I needed to do was change it for the remote URL and I'd be done. But, whenever I change anything the .idx file, it no longer is a valid cache entry and won't show up in the JRE cache viewer. I'm guessing there is some sort of integrity check being performed. This would solve a HUGE problem for me.

Does anyone know how I can accomplish what I'm trying to do?

I'm using JRE version 5 update 1 and here is a sample of the name of .idx file I have: myjar.jar-497a1330-51b2cd4e.idx

Here would be the accompanying zip file: myjar.jar-497a1330-51b2cd4e.zip

The contents of the idx file are binary but the URL portion is in clear text. It would be awesome if there is a way to generate this file or change the URL the cache entry points to so it thinks it already has the jar in cache and won't try to download it.

- Mike