[izpack-devel] Self-Extracting and Self Running Jars -- Anyone Want Info On Them?
miraodb at hotmail.com
Mon May 15 17:05:04 CEST 2006
All ideas are more than welcome !!! And god knows you have a lot ;-)
I believe that what you found is quite useful.
But i don't think it would match the platform independance that IzPack is so
We started a few months ago to write down cookbooks in all kinds of tricks
or workarounds that couldn't fit directly into the core of the system.
This would be a great way to start for your idea.
I could integrate that fairly easily and quickly as soon as you send me the
links and docu you came up with.
A simple word doc is fine, or html even better... then i'll translate it
into the requested format for our Docu.
I vote for putting it in the cookbooks to start with and if it happens to be
something worth it to develop a plugin or new feature than we can decide
something later on.
PS: If anyone has another opinion..... well write it here.
----- Original Message -----
From: "Hal Vaughan" <hal at thresholddigital.com>
To: <izpack-devel at lists.berlios.de>
Sent: Sunday, May 14, 2006 9:01 PM
Subject: [izpack-devel] Self-Extracting and Self Running Jars -- Anyone Want
Info On Them?
> The program I'm using IzPack for is for Windows and Linux only. Since
> it's the first experience my clients will have with my company, my
> services, and my software, I want it to be as effortless, easy, and
> idiot-proof as possible. I've found that, for people on some operating
> systems, even telling them "right click, pick 'run with' and pick Java"
> can be a pain because they're not used to context menus or thinking
> about what they do on computers.
> I found out how to create self-installing archives for both Linux and
> Windows. The Linux archives are easy to make and basically involve
> using "cat" to add the archive to a bash script and later "tail" from
> within the script to extract the archive. It's a bit more complex on
> Windows and requires 7-Zip, which is also open-source and the developer
> was quick to respond to my questions. The method for making Linux
> self-extracting archives should work on any Unix system that uses bash
> and I would think it would work on any Mac with OSX on it, since it
> uses only script commands and no natively compiled code.
> When clicked on or run, these archives will automatically self-extract
> and run a specified program (in this case, I have it run "java -jar
> Install.jar") and clean up the temp files when done. As usual, this
> script won't run directly from a CD in Linux unless "sh" is used to run
> it (which also means running it from a browser often won't work).
> However, I think it's about as much as can be done to make a jar file
> idiot proof, since a programmer or program distributor has no control
> over how the MIME types are defined on someone else's computer.
> The draw back is that IzPack is multiplatform and these archives are
> platform dependent. However, the same IzPack Install.jar can be used
> to create self-extracting and self-running archives for all different
> platforms. I have a script set up that will use two of these archives
> to install Java, then run my IzPack install. On a web install, only
> the first jar needs to be self-extracting. All the pack jars can be
> left as they are.
> Is there any interest in this among other programmers? I am not at a
> point where I can actually integrate this into the compile process
> because I have some other work that has to be done, but I will be glad
> to post all the instructions and links I have on it and work with
> anyone who wants to either standardize into a script or two so it's
> easy to do or who wants to do some of the programming.
> At the very least, I can write it up as documentation for IzPack as "How
> to Make Your Archives Self-Extracting and Self-Running".
> I'm not going to do it, though, if there's no interest.
> Any comments?
> izpack-devel mailing list
> izpack-devel at lists.berlios.de
More information about the izpack-devel