Thread: Self-Deletion

Hello forums!

I've opened this thread to discusss "self deletion" in Java.

It is perfectly possible, but has a trick to it. Obviously, you can't delete a file that is running. We'd have to send a scheduled delete request to the shell of the OS. (though its bad architecture since you're program is dependant on an specific platform) anyways that can easily be forseen.

Anything that is being run cannot be deleted as it is a process in use. What you could do (and this is dirty, bad architecture) is use Runtime.exec() to (on windows) use the "AT" command to schedule a delete command on your java code files. Again, this is not the best idea and it's platform dependant (so you'd have to have a way to do this on linux too - cron would do it).

-from another forum

I'd say to use the ProcessBuilder instead of the Runtime.exec(), because it is very limmited.

Some people are using this

public static void main(String[] args) throws Throwable {
File f = new File("DeleteMe.jar");
f.deleteOnExit();
}

The reason why I've started to look at this option is because, the program I am working with, generates 2 copies of itself, and has one of the deletes the original, but it generates several unsyncronized problems, which thread and waiting would not prevent it to a 100%. So I rather have each program take care of itself, sending a delete request on itself upon conclusion of its instructions.

Ideas, advice, comment?

I appreciate your feedback. Meanwhile.. I'd be google-ing and trying to apply it.

The program simulates living cells (biology). When it "divides" there is no original. Nothing fishy, just being creative here.

Couldn't you do the same thing with objects in a single program without worrying about kludgy file manipulations? What benefit do you get by using files instead of Objects?

It is not intended to be a game by the way, for that I'd rather use objects.

Instead of having the object as "as a cell" contained in a program "as environment", I want the program "as a cell" contained in an OS environment "as environment".

Objects obviously occupy memory, but they don't have a "bodily representation" like a .class file to be able to exist.

If I'd like to use something fishy, I would've looked into quine programming instead.

These are just the baby steps. I've figured out how simulate "communication" between programs, assign them a purpose, read other .java fragments and use trial-and-error until code's prediction matches with the debugging the fragments of code, keep record of results, until it rewrittes the code for future use.

The logic is quite easy, just a little bit extense. I didn't know much about file manipulation in Java yet, I need to wait 2 weeks till I start learning that at the University.

I am just trying to simulate living cells. And I certainly need some safeguards to avoid infite looping from duplication.

I'll code another program that monitors the activity.

--- Update ---

So yeah, I need to know how to get a program delete itself. Instead of having one of their child programs to delete it, it causes some problems, files get deleted before another program's instructions get to use them. That's the sort of problem I am trying to avoid here. I'd rather have each program delete itself after its done.

I see. I don't know anything about lisp. I am using what I know, which is why I am using Java, it may be kind of a bottle neck but I rather do it, than wait long enough to lose interest. I don't know about any of these algorithms you are talking about (except quines).

Besides, Java has a broad community and platform support as well.

Do you have any resources of knowledge I can get my hands on? I would certainly start reading about it if I knew where to look and what to read. Any suggestions then?

I would look into temporary file execution. From the conversation here it seems to me Andrew that you are attempting to imitate biological processes to study them or learn in the process of creating something of equal value? But the best representation of genetic code is code, thus the physical would represent the physical. The combination of nucleotides is binary equivalent, and physical maintenance/input would be the computer (Organelles: monitor, essentially output/ Glycocalyx: inputs/ smaller components (transistors, capacitors) mirroring smaller portions of the cell (Cytoplasms' many smaller process that make things possible)).