letting users try out your program

[vlad1950]

I wrote a small java program using Gel, it compiles and works perfectly but how do you export the file so you could let other users try out my program if they dont have software like BlueJ, Gel etc

(Im only a beginner btw)

here is my program:

import javax.swing.JOptionPane;
public class HelloWorld
{
public static void main(String[] args)
{
JOptionPane.showMessageDialog(null, "Hello");
String name = JOptionPane.showInputDialog("What is your name?");
JOptionPane.showMessageDialog(null, "Hello " + name);
System.exit(0);
}
}

[Jessii]

You should be able to compile it with that program. If not then you can search for a java compiler on google.

[vlad1950]

It compiles and works fine, what I'd like to do is send the file to a friend and let him run it, but when I send the java file all he gets is a text file.
Any way he can run the file if he doesnt have a programming program & compiler?

if I'm not making any sense make sure to say it :\

[elbekko]

The result of a compile action should be an exe file, which you can send to everywone you want.

[CronoMan]

At 10/30/06 01:15 PM, elbekko wrote: The result of a compile action should be an exe file, which you can send to everywone you want.

Would be true for any real language, and not the joke which is java ;)

[elbekko]

At 10/30/06 01:32 PM, CronoMan wrote:

At 10/30/06 01:15 PM, elbekko wrote: The result of a compile action should be an exe file, which you can send to everywone you want.

Would be true for any real language, and not the joke which is java ;)

Oh right, Java.

*runs in fear*

[Jordan]

Good old Jordan did it for you, send this zip folder to ya mates and tell them to run the bat file.

http://www.doodle-fl..co.uk/HelloWorld.zip

[CronoMan]

At 10/30/06 01:44 PM, Jordan wrote: Good old Jordan did it for you, send this zip folder to ya mates and tell them to run the bat file.

http://www.doodle-fl..co.uk/HelloWorld.zip

I was going to post that as a suggestion, but I'm not sure that java.exe is on computers without the sdk installed? not sure.. but perhaps...

[Jordan]

At 10/30/06 01:50 PM, CronoMan wrote: I was going to post that as a suggestion, but I'm not sure that java.exe is on computers without the sdk installed? not sure.. but perhaps...

I use a program called JavaLauncher and with it i can just click on .class files and they open, including applets.

With it i can also make .exes out of class files, if this fails.

Search for Java Launcher on download.com

[CronoMan]

At 10/30/06 01:54 PM, Jordan wrote: With it i can also make .exes out of class files, if this fails.

That's probably exactly what he's looking for ;)

[vlad1950]

At 10/30/06 01:54 PM, Jordan wrote:

Java Launcher

Ok I got the Java Launcher but not sure how to work it. I selected my .java file for the "Main Class", but what do I put for "Output file"?

[vlad1950]

At 10/30/06 01:32 PM, CronoMan wrote:

At 10/30/06 01:15 PM, elbekko wrote:

Would be true for any real language, and not the joke which is java ;)

Why is java the least favourite here? I'm forced learning it in college this year but will be moving on to C++ next year

also got JavaLauncher working so ignore my last post

[whatthedeuce]

At 10/30/06 06:56 PM, vlad1950 wrote: Why is java the least favourite here? I'm forced learning it in college this year but will be moving on to C++ next year

Because there are a lot of people on this forum who ignore logic in favour of strong baseless opinions. Java isn't the best language out there, but it's not horrible either. Without a doubt, it's a better language than C++ (there are some situations where C++ is better, but they are not the majority).

[seel]

At 10/30/06 07:03 PM, whatthedeuce wrote:

At 10/30/06 06:56 PM, vlad1950 wrote: Why is java the least favourite here? I'm forced learning it in college this year but will be moving on to C++ next year

Because there are a lot of people on this forum who ignore logic in favour of strong baseless opinions. Java isn't the best language out there, but it's not horrible either. Without a doubt, it's a better language than C++ (there are some situations where C++ is better, but they are not the majority).

Rofl just stop your nonesense already. Java is slow. That's it. Also you can make console games in C++, java can't.

lmao iRux

[amaterasu]

vlad,

this is the easiest way. send your friend the .class file, with these instructions.

1. Save the .class file to your desktop
2. Open up a console
3. Locate the directory where his/her jre is stored. Type into the command prompt "set path=<address>, where <address> is the address of the jre. Ex- C:\Program Files\Java\jre\bin. It needs to refer to the bin folder. This sets the environment.
4. Then, type set classpath=<address>, where <address> is the location of the .class file.
5. Type java <file>, and it should run. Don't add the .class extension.

For Java in general,

I really enjoyed learning the basics with Java, but there's one thing that turned me off to it for serious development. Not speed, but security. It takes almost no effort to decompile a java bytecode file (.class), and it is way too much of a hassle to secure java bytecode so it's unreadable. Usually it involves adding tricks to disable decompilers, or adding redundant code to confuse the reader. Why go through the trouble when you can use natively compiled languages and gain much more security? Any compiled files can be decompiled, but it's much more difficult with native code. Anyway, that's just my argument, and I think it's a much more useful fact than just saying "it's slow," or "it gives you less control."

[whatthedeuce]

At 10/30/06 07:18 PM, seel wrote:

At 10/30/06 07:03 PM, whatthedeuce wrote: (there are some situations where C++ is better, but they are not the majority).

Rofl just stop your nonesense already. Java is slow. That's it. Also you can make console games in C++, java can't.

Way to demonstrate a complete lack of reading comprehension. Speed is of very little concern in many applications, and Java isn't even that slow with JIT compiling.

[RageOfOrder]

I'll put it this way.

Java is like your mom.

Also:
"Knock knock"
"Who's there?"
....
...
....
.....

.....

......

...

.....

.....

"Java."

[amaterasu]

At 10/30/06 10:07 PM, RageOfOrder wrote: I'll put it this way.

Java is like your mom.

Also:
"Knock knock"
"Who's there?"
....
...
....
.....

.....

......

...

.....

.....

"Java."

get out of the programming forum.

[White-Rhyno]

At 10/30/06 09:52 PM, whatthedeuce wrote:

At 10/30/06 07:18 PM, seel wrote:

At 10/30/06 07:03 PM, whatthedeuce wrote: (there are some situations where C++ is better, but they are not the majority).

Rofl just stop your nonesense already. Java is slow. That's it. Also you can make console games in C++, java can't.

Way to demonstrate a complete lack of reading comprehension. Speed is of very little concern in many applications, and Java isn't even that slow with JIT compiling.

I understand what you're saying, but it shouldn't take 2 minues to load a Hello World program. The language is quite nice, though if I want a cross port OOP language then I'll head towards Python.

[CronoMan]

At 10/30/06 10:07 PM, RageOfOrder wrote: I'll put it this way.

Java is like your mom.

Also:
"Knock knock"
"Who's there?"
....
...
....
.....

.....

......

...

.....

.....

"Java."

rofl

imo java has a lot of problems, and when it comes to GUI applications, java is _better_ than C++ in many ways, but still there are alot better alternatives than java.

And talking about speed : if you're java applications runs as fast or faster than C++ applications, you should really reconsider your profession.

C++ can be extensively optimised, and in most compilers you can also use an inline assembler to make it even faster.

take this simple iteration :

int[] da = new int[5000];

for(int i=0;i5000;i++)
do something with da[i];

this will look almost identical in both languages if you're a c++ noob.
What you might or might not know, is that each time you retrieve da[i], the framework or library has to iterate all items from 0 to i.
This can be solved in C++ to make it much faster, like this :

int da[5000];
int *da_ptr = da;
for(int i=0;i<5000;i++ && da_ptr++)
do something with *da_ptr;

this is one of the reasons to how and why C++ is so much more powerful than java
(you can avoid this in java by using a linked list, but that would alot of overhead for a little part, and not very memory-efficient)

[thoughtpolice]

At 10/30/06 07:18 PM, seel wrote: Rofl just stop your nonesense already. Java is slow. That's it. Also you can make console games in C++, java can't.

Get a clue and come back. Until then you offer no substantial contribution to this forum. Believe it or not, you really have got little to no idea what you're talking about.

It's not the fact that people might not KNOW this stuff that makes me sick, that's just human nature not to know things, it's wannabe know-it-all's who make me want to vomit. There's an implication there, if you're sharp enough to catch it.

At 10/31/06 02:14 AM, White-Rhyno wrote: I understand what you're saying, but it shouldn't take 2 minues to load a Hello World program. The language is quite nice, though if I want a cross port OOP language then I'll head towards Python.

Pardon me for saying, but you're kind of overexaggerating a bit. It takes less than about 3 seconds for me to load a text-based class file on my machine, and large systems like ArgoUML or Frostwire which are substantially large applications take less than 15. On low-end systems, or 10 years ago, it may have taken 2 minutes to load a file, but these days, it's just a complete misconception that's been carried over.

At 10/31/06 02:39 AM, CronoMan wrote: And talking about speed : if you're java applications runs as fast or faster than C++ applications, you should really reconsider your profession.

Go away. I've explained this point a myriad of times on this forum, go back and look it up if you're so inclined, because I think you're just ignoring real life situations that I've explained many times before.

C++ can be extensively optimised, and in most compilers you can also use an inline assembler to make it even faster.

Java can optimize programs in way compilers cannot. Yes, it can. Go look it up, it's pretty interesting.

this is one of the reasons to how and why C++ is so much more powerful than java

I think the previous example was a very, very bad example to push a point of "power." You may perhaps be talking about 'efficiency.' But then again, every langauge is efficient in different points, so it's not a good point anyway.

[thoughtpolice]

At 10/30/06 09:07 PM, amaterasu wrote: It takes almost no effort to decompile a java bytecode file (.class), and it is way too much of a hassle to secure java bytecode so it's unreadable.

Most of those obfuscators don't do much other than mess around with class names and variable names. Any sort of data-flow analysis program can easily rename these to meaningful things (such as IDA Pro does.)

Usually it involves adding tricks to disable decompilers, or adding redundant code to confuse the reader. Why go through the trouble when you can use natively compiled languages and gain much more security? Any compiled files can be decompiled, but it's much more difficult with native code.

Actually with native code it's an undecidable problem and not to be trusted, Java can be decompiled only because it's run inside a virtual machine. Also, you don't gain any "security" through using native code - anybody with an intelligence level above that of your µP will be way ahead of the curve that native binaries have. You don't gain any safety, it's just a matter of knowing how and where to look. I'll give you that at first, binary analysis seems second to none in complexity, but with little time you can read and find issues, bugs in native binaries almost as easily as source code.

[amaterasu]

At 10/31/06 02:56 PM, kb-resin wrote:

At 10/30/06 09:07 PM, amaterasu wrote: It takes almost no effort to decompile a java bytecode file (.class), and it is way too much of a hassle to secure java bytecode so it's unreadable.

Most of those obfuscators don't do much other than mess around with class names and variable names. Any sort of data-flow analysis program can easily rename these to meaningful things (such as IDA Pro does.)

Usually it involves adding tricks to disable decompilers, or adding redundant code to confuse the reader. Why go through the trouble when you can use natively compiled languages and gain much more security? Any compiled files can be decompiled, but it's much more difficult with native code.

Actually with native code it's an undecidable problem and not to be trusted, Java can be decompiled only because it's run inside a virtual machine. Also, you don't gain any "security" through using native code - anybody with an intelligence level above that of your µP will be way ahead of the curve that native binaries have. You don't gain any safety, it's just a matter of knowing how and where to look. I'll give you that at first, binary analysis seems second to none in complexity, but with little time you can read and find issues, bugs in native binaries almost as easily as source code.

You know, kbresin, it's really funny. You try to correct me, but all you basically did was take my point and restate it a different way. I never said it is impossible to decompile and understand native code. I said it is much harder than decompiling and analyzing java bytecode. What do you think is more difficult? Decompiling all of ArgoUML's .class files and figuring out exactly how it works (which is pointless because it's open source, but pretend it isn't), or to decompile all of ZSNES's binaries and figuring out exactly how it works (again, pretend it's not open source). That is what I mean by security. You can reverse engineer your way into anything, so security acts as a deterrent rather than a preventer.

Maybe you should re-think your post about being a "wannabe know it all", because you my friend have been the poster child of that recently.

[thoughtpolice]

At 10/31/06 03:21 PM, amaterasu wrote: You know, kbresin, it's really funny. You try to correct me, but all you basically did was take my point and restate it a different way.

Actually I wasn't correcting you at all, your post was pretty much totally correct. I was only elaborating on certain points - I wasn't (at least not trying) being hostile or critical twords your post.

I never said it is impossible to decompile and understand native code.

It's elaboration, and I didn't say decompilation was impossible either, it's undecidable. There's a -large- difference between the two. I also didn't mean for my post to imply that you said it's impossible to understand native code.

I said it is much harder than decompiling and analyzing java bytecode. What do you think is more difficult? Decompiling all of ArgoUML's .class files and figuring out exactly how it works (which is pointless because it's open source, but pretend it isn't), or to decompile all of ZSNES's binaries and figuring out exactly how it works

Well, considering you typically when reverse engineering aren't trying to understand everything unless you're trying to do a clean-room approach to perhaps write a new program (such as perhaps writing a newer more modern TASM clone,) I'd have to say you are correct when you imply it'd be harder in the ZSNES case than Argo. I was only elaborating, once again. Generally when you reverse engineer you're looking for something specific, not to understand the whole case study. Even when you have source code, it still takes much time before you can generally understand many works. Albeit, much longer for native code, so you are correct, I wasn't correcting you.

That is what I mean by security. You can reverse engineer your way into anything, so security acts as a deterrent rather than a preventer.

So what are you preaching? Sounds a little like security through obscurity, which doesn't work. But I might be wrong, oh well. I get what you're saying, I just think security might be slightly the wrong word.

Maybe you should re-think your post about being a "wannabe know it all", because you my friend have been the poster child of that recently.

I'll be sure to note that.

[thoughtpolice]

@amaterasu: In case this was apart of the confusion any:

At 10/31/06 02:56 PM, kb-resin wrote: but with little time you can read and find issues, bugs in native binaries almost as easily as source code.

I apologize, considering I accidentally went into a tangent on things like finding bugs, which we weren't talking about. In the whole, just to figure out something, a native binary will generally take much longer to analyze to find what you're looking for, so yes, you are correct there, I just went into the area of finding problems on accident, so that's a bad on my part.

[Jordan]

MAJOR E-drama over a simple question that i solved in 1 post about 10 posts ago, no need to keep on flaming eachother over whether java is more 1337 that c++.

Java rulz.

[whatthedeuce]

At 10/31/06 02:14 AM, White-Rhyno wrote: I understand what you're saying, but it shouldn't take 2 minues to load a Hello World program. The language is quite nice, though if I want a cross port OOP language then I'll head towards Python.

Python is a great language (I use it more than anything else), but it's definitely not faster than Java. I'm not disputing than Java has a slow startup (that's a well known fact) but if you write any reasonably sized program, Java will run circles around Python.

Again, speed really isn't that important. Many programs spend more time waiting for input than they do actually working.

[CronoMan]

At 10/31/06 02:07 PM, kb-resin wrote:

At 10/31/06 02:39 AM, CronoMan wrote: And talking about speed : if you're java applications runs as fast or faster than C++ applications, you should really reconsider your profession.

Go away. I've explained this point a myriad of times on this forum, go back and look it up if you're so inclined, because I think you're just ignoring real life situations that I've explained many times before.

Believe me, I know java, I've been working with it as a proffessional for 1 1/2 years now.

And you, my friend, you're living in disneyland ;)

C++ is ALL about what you know about compilers, what you know about processors, memory etc. Like for instance : I bet you'd write int a = b / 2, right?

Well, that's wrong. int a = b >> 1 is the correct answer. It's all about how you write the code, and how much effort you put into making it optimized. Of course, you could live in Java the rest of your life, and think that Java is the ultimate, doesn't get any better than that! But if you would have a discussion with... say... anyone with knowledge of both C++ and Java, you would soon discover that they would say exactly the same I say. C++ is the most powerful development tool currently availble. Few will argue on that. But of course, you always have the Visual Basic, Delphi and Java fanatics which will say otherwise, but they usually don't have a clue.

And : Machine code is as fast as it gets. period. If you're using the right libraries, and the correct amount of inline assembler, you will gain speed unimaginable for a java application.

Could you make a full-screen 3D software renderer in Java?
No.
Could you make a full-screen 3D software renderer in C++?
Yes. Quake 1 was written in C for 486's. You won't even be able to make one polygon fill itself with a texture with more than 5fps in java with today's computers. And that's a fact.
Go ahead, try it.

Imo, programming is an art, not just a tool to make cash. Java is ruining the art by taking away the freedom and extensibility. Now every other douche claims they're good at programming because they can make a console application that takes input, in java.

[thoughtpolice]

At 11/1/06 09:30 AM, CronoMan wrote: C++ is ALL about what you know about compilers, what you know about processors, memory etc. Like for instance : I bet you'd write int a = b / 2, right?

Well, that's wrong. int a = b >> 1 is the correct answer.

Yes, I've known about that for a while (read "Hackers Delight.") In this case, it doesn't really matter because the compiler will unroll that optimization for you anyway (using a shr/shl instruction instead of div/mul,) so the point is moot. In fact almost ALL of those little 'bit shift tricks' you can pull for optimization, the compiler is probably already way ahead of you and you might actually hinder your code by getting in the way (for example, inline assembler can typically be *much* less efficient than say, just linking in an ASM-based .o file, as the compiler and assembler will have to optimize 'around' your inlining to preserve registers properly, etc. So you can actually lose performance here rather than gain it.)

Of course, you could live in Java the rest of your life, and think that Java is the ultimate, doesn't get any better than that!

Which, uh, I don't. I like Java, I don't live in it, but I acknowledge its existance and that it has its advantages for being used.

But if you would have a discussion with... say... anyone with knowledge of both C++ and Java, you would soon discover that they would say exactly the same I say. C++ is the most powerful development tool currently availble.

Yes, yes. I know, C++ is the ultimate, blah blah blah. I've heard it before and I don't believe it, same way I didn't years ago. While you and others may consider C++ a superior language, that doesn't mean that Java is non-existant or non-useful. There is a REASON those other tools are available. Perhaps you just haven't discovered the reason. If C++ is just so darn powerful, why does any other language even exist?

Oh yeah, to help distinguish and solve problems easier and quicker. Which is what programming is all about. There's quite a bit of reading on this subject, go look it up.

Few will argue on that.

Well, someone has to play devil's advocate, yes?

But of course, you always have the Visual Basic, Delphi and Java fanatics which will say otherwise, but they usually don't have a clue.

So I guess I'll just never employ another VB, Delphi or Java programmer again since they "don't have a clue." Right, I'll take your word on it. From now on I'll do nothing but employ arrogant C++ programmers, no matter how much longer it takes the system to be completed, no matter how much more inflexible my architecture is, and no matter how much less control I might have over certain parts of the language (such as the object-model, which you can control much easier in, say, Ruby.) Right, that sounds like the good choice to make.

And : Machine code is as fast as it gets. period. If you're using the right libraries, and the correct amount of inline assembler, you will gain speed unimaginable for a java application.

I generally don't care about this point much anymore. You're living ten years ago - it's not all about speed. I'm probably never going to argue the point of speed again because it's stupid and there're plenty of statistics otherwise to show that given particular cases, you're wrong. If all you do in programming is optimize with inline assembler and use minimalistic speed based libraries to try to beat the Java croud, I'm suprised you're employed.

Could you make a full-screen 3D software renderer in Java?
No.

Yes, you can. It's just a matter of how practical it would be. I'm pretty sure you're just confusing "possibility" with "practicality." Might not be practical, but that doesn't mean it's not possible. Seriously, knowing a little theory and a little logic is useful before you go making blatant, unsupported claims.

Yes. Quake 1 was written in C for 486's. You won't even be able to make one polygon fill itself with a texture with more than 5fps in java with today's computers. And that's a fact.
Go ahead, try it.

A fact? If it's a fact, show me evidence. It's YOUR job to prove that what you say is correct, not my job to try and disprove or prove it for you.

Now every other douche claims they're good at programming because they can make a console application that takes input, in java.

Seems that there're a lot of C++ programmers who do that too. Perhaps not in your part of the woods, but I seem to find people who just because they can program a bubblesort or multiply some numbers think they're the shit and know everything there is about the language.

[White-Rhyno]

At 10/31/06 02:07 PM, kb-resin wrote:

At 10/31/06 02:14 AM, White-Rhyno wrote: I understand what you're saying, but it shouldn't take 2 minues to load a Hello World program. The language is quite nice, though if I want a cross port OOP language then I'll head towards Python.

Pardon me for saying, but you're kind of overexaggerating a bit. It takes less than about 3 seconds for me to load a text-based class file on my machine, and large systems like ArgoUML or Frostwire which are substantially large applications take less than 15. On low-end systems, or 10 years ago, it may have taken 2 minutes to load a file, but these days, it's just a complete misconception that's been carried over.

If the VM is already loaded then the difference between loading a Java program and a C program are discerable only by timestamps. But loading all that shit from scratch? Not 2 minutes, but long enough to make me wonder if anything is happening.