Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Mephisto]

And lo, we have discovered rubber duck debugging.

[head]

That feeling when you catch yourself having a conversation with yourself via comments.

C#: How would I go about loading all the images from a folder into an ArrayList without knowing the filenames or the amount of files in the content folder?

http://pastebin.com/W2TuAYhg

Something like that.

[Thendash]

Thanks Head and Mephisto, I'll try those out.

[Neonivek]

I did a lot of PM work for my problem and it ended hillariously. I will say thanks to the thread anyhow even if it didn't really pan out.

I guess I'll just have to try harder on my own, or just hunt down someone who did what I did (which is no one on this forum)

---

So I am guessing no one still knows about to make a graphical Adventure game (Point and Click adventure game) like Kings Quest the way I want it to be. So I'll continue on the endless quest.

[dizzyelk]

So, I've spent the past good while trying to get A* implemented. It works, usually. However sometimes it stops, and when I look at what its doing during these hangs, I see its looking for a cell that doesn't even come close to being on the map, like -6534x-5276 (made up values, several thousand, anyways, and not the same when it happens).

Spoiler (click to show/hide)

It happens somewhere in here, where current_node is the last square it calculates the costs around, c in my little picture

Code: [Select]

xxx
xcx
xxx

The functions:

//sets current node to a node at a given coordinate, returns true if it exists, false otherwise
bool pathing_list::find( coord search )
{
        //make sure there are nodes in the list
if( root_node == NULL )
return false;

pathing_node * temp_node;
current_node = root_node;
do
{
if( current_node->get_coord() == search )
return true;

else if( current_node == end_node )
{
current_node = root_node;
return false;
}

else
{
temp_node = current_node->get_next();
current_node = temp_node;
}
} while( true );
}


//sets current node, and returns a pointer to it, to the one with the lowest F cost
pathing_node * pathing_list::lowestF()
{
coord delta;
delta.X = current_node->get_coord().X - current_node->get_parent()->get_coord().X;
delta.Y = current_node->get_coord().Y - current_node->get_parent()->get_coord().Y;

delta.X += current_node->get_coord().X;
delta.Y += current_node->get_coord().Y;

find( delta );
pathing_node * starting_node = current_node;
pathing_node * lowest_node = current_node;

double cur_lowest = current_node->get_F();
current_node = current_node->get_next();

while( current_node != starting_node )
{
if( current_node->get_F() < cur_lowest )
{
cur_lowest = current_node->get_F();
lowest_node = current_node;
}

current_node = current_node->get_next();
}

current_node = lowest_node;
return current_node;
}

Any thoughts on what the problem might be?

[Starver]

I can't tell from the code you give, and with the made up values representing ones that are "several thousands, anyways" adrift, but it sounds like a classic underflow/overflow error.

If not, could it instead be some obscure type-conversion that's (say) taking a value 0..65536 and either 1:1 mapping it to -32768..32767?  Either 0 to -32768 all the way to 65536 becoming 32768 (or vice-versa), or interpreting the bits of a signed (2's compliment) value as an unsigned value (or, again, vice-versa[1]).

Either way, perhaps there's some way you can work out what value should be seen, and see if its any 'power of two' value (maybe plus or minus one) away from what it should be?  Otherwise, put some debug statements in that check (and display/log, somewhere handy) the possibly erroneous variables' defined types at various stages.


There's some other possibilities, and it's strange that it doesn't always error, so the problem might be in some conditional code[2], but I'd rule the above out first.


[1] That'd only affect the top half of the value range, though, and I can't see raw coordinate values going that high, unless your ambitions or implementation give you such huge array location values...

[2] I got something like this in some "wrap-around" code I did, once.  When on one edge (e.g. having a coordinate of zero, in one/more dimension), it was supposed to translate a "-1" offset to the "max" number, and "-2" to "max-1", etc.  I doubt you're doing that, either, but you may still have something of a similar ilk, where the error yet-to-be-identified lies mostly dormant, except when it isn't.

[Kofthefens]

I'm trying to make a program that will check a yahoo email. However, when I do emailChecker.recieve(), it gives me an error message. I am using JavaMail. The code looks like this:

Spoiler: Code (click to show/hide)

Code: [Select]

public static void receive(String popServer, String popUser, String popPassword){
Store store = null;
Folder folder = null;
try {
// Get hold of the default session
Properties props = System.getProperties();
Session session = Session.getDefaultInstance(props, null);
// Get hold of a POP3 message store, and connect to it
store = session.getStore("pop3");
store.connect(popServer, popUser, popPassword);

// Try to get hold of the default folder
folder = store.getDefaultFolder();
if (folder == null) throw new Exception("No default folder");

// ...and its inbox
folder = folder.getFolder("INBOX");
if (folder == null) throw new Exception("No POP3 INBOX");

// Open the folder
folder.open(Folder.READ_ONLY);

// Get the message wrappers and process them
Message[] msgs = folder.getMessages();
for (int msgNum = 0; msgNum < msgs.length; msgNum++) {
printMessage(msgs[msgNum]);
}
} catch (Exception ex) {
ex.printStackTrace();
} finally {
// Close down
try {
if (folder!=null) folder.close(false);
if (store!=null) store.close();
} catch (Exception ex2) {
ex2.printStackTrace();
}
}
}

And the error message looks like this:

Spoiler (click to show/hide)

Code: [Select]

javax.mail.MessagingException: Connect failed;
  nested exception is:
java.net.ConnectException: Connection timed out: connect
at com.sun.mail.pop3.POP3Store.protocolConnect(POP3Store.java:210)
at javax.mail.Service.connect(Service.java:295)
at javax.mail.Service.connect(Service.java:176)
at src.Checker.receive(Checker.java:52)
at src.Checker.update(Checker.java:140)
at src.Launcher.main(Launcher.java:16)
Caused by: java.net.ConnectException: Connection timed out: connect
at java.net.DualStackPlainSocketImpl.connect0(Native Method)
at java.net.DualStackPlainSocketImpl.socketConnect(Unknown Source)
at java.net.AbstractPlainSocketImpl.doConnect(Unknown Source)
at java.net.AbstractPlainSocketImpl.connectToAddress(Unknown Source)
at java.net.AbstractPlainSocketImpl.connect(Unknown Source)
at java.net.PlainSocketImpl.connect(Unknown Source)
at java.net.SocksSocketImpl.connect(Unknown Source)
at java.net.Socket.connect(Unknown Source)
at java.net.Socket.connect(Unknown Source)
at com.sun.mail.util.SocketFetcher.createSocket(SocketFetcher.java:319)
at com.sun.mail.util.SocketFetcher.getSocket(SocketFetcher.java:233)
at com.sun.mail.pop3.Protocol.<init>(Protocol.java:111)
at com.sun.mail.pop3.POP3Store.getPort(POP3Store.java:261)
at com.sun.mail.pop3.POP3Store.protocolConnect(POP3Store.java:206)
... 5 more


Thanks in advance for the help.

[Derekristow]

Well, it seems to be failing at

Code: [Select]

store.connect(popServer, popUser, popPassword);due to a time out.  Are you sure the server exists and is responding to you?  Google will be able to help more than me, as I've never even heard of that library, but looking up what will cause that exception in that function is where I'd start.

[Lompocus]

I have a decision to make, both options have their pros and cons.  I'm trying to decide if I should learn win32 GUI programming so I can make a 2d map editor for a game I'm making, but the thing is I'm planning on ending up making the levels procedurally generated, but I need a map or two made for when I am doing tests and making the game other than the world / level gen.

Someone already answered your question, but I'll add some stuff. The win32 API is something you should know if you're using windows. In practice, for any GUI (that's not a game's GUI), you will _not_ write code using the win32 API to open windows (nor will you use MFC!); you'll use something else like Qt or Gtk if you want to use C++, or WinForms (C# and other stuff... I'm not so sure) or Tk (Perl) if you're using a different language. However, it comes in handy at the weirdest times, e.g. SFML's current stuff from its repo has a bug in it where it renders window titles incorrectly, something which is fixed in 1 win32 api function call whenever you want to change the window title. Buy the Petzold WinAPI book (5th edition). Go over most of the first section (short at ~340 pages) and parts of the 3rd section as needed.

And, for some extra gui stuff, here are a few more GUI libraries:
-SFGUI (or whatever it's called) -- It's a GUI library built for integration with SFML.
-GLUI (which provides a look and feel just like regular windows 95) -- Another OpenGL GUI. Very simple.

Here's something else I stumbled across when trying to remember what GLUI was called (GLConsole); not a GUI, but it looks like a useful OpenGL console system:
http://www.robots.ox.ac.uk/~gsibley/GLConsole/

*edit: Here's a linky with ~180 Qt tutorials. Excellent ones. Here's a case where you want to build a gui for a tool for building your game, but you don't want to learn another language (though Tk w/ Perl doesn't take more than a few days to pick up at a simple level): http://www.youtube.com/watch?v=Id-sPu_m_hE&feature=plcp

[Sergius]

Ok, here's a open-source licensing question. It's not about coding so I hope it's not off-topic.

I'm spoilering it just in case it's the wrong topic and people can skip it.

Spoiler (click to show/hide)

I'm making a game engine (yeah, I know, make a game, not an engine. I'm actually making a game, and it runs on an engine, which I'm going to program, along with the game, to handle the use cases that the game needs). Basically, it's going to be open source, so that anyone can use/improve it. The idea is that, without editing any hard code, anyone can modify scripts/resources/config files/whatever (basic enginey stuff here), but the Source Code will also be modifiable, so if the executable can't handle a special case (like, game maker wants a completely different "combat" or whatev) they can just add more code, which in turn probably reads some new kind of config/resource file.

SOOO, anyhow. There's no question that the "gamefiles" themselves are no subject to any copyleft licensing. If someone wants to sell them, make some sort of proprietary thingy, whatever, it's their right. Their "files", however they choose to use them. And of course, include the unmodified executables needed to run it. No charge.

But here's the question: the source code itself for the hard exe files. I'm not a big fan of "copyleft" (CC:Share-Alike, GPL) and forcing non-commercial use on people, as long as there's attribution. If this was just some source that HAS to be modified to turn into a newfangled game, I'd rather use a "permissive" license like BSD or some other that doesn't force people to open their code. But I'm thinking, well, it being a shared ENGINE in the first place, letting people make their own private changes and not sharing them with the rest of the community, as the original program was, doesn't seem too fair. It's not like they're giving away secret quests or something, since those are in resource files and private(ish). I've seen how that ends up with things like MOOs: any improvement is never shared, and the only way people can learn to do things like combat engines is PRAY that someone has made a sample, or start from scratch.

So, would it be a good idea to use a copyleft/Share Alike license for the source code to the engine itself? Would that limit too much use of it? Would people "rather not use it if they have to share"? Should I care that "those" people don't use the engine? (after all, my primary motivation is for ME to use it for the games I'm planning/already planned to do, not that it's a popular engine that everyone uses for commercial projects).

Opinions?

BTW, java + libgdx + android SDK RULES! Multiplatform development FTW!

[Shades]

I'm not a big fan of "copyleft" (CC:Share-Alike, GPL) and forcing non-commercial use on people, as long as there's attribution.

Not sure what you mean here, GPL doesn't force non-commercial use, just that any modification is available as source. I think the LGPL would work better in this case, or the GPL with linking exceptions (like a lot of compiler runtimes use) as it gives more flexibility to the user.

However personally I'm not a big fan of the GPL, as useful as it has been, and would instead suggest either BSD or MIT licences and just assume any good change will be pushed back to you. Seems to work well enough for most github projects, and companies are getting better and better and pushing back.

[Sergius]

I'm not a big fan of "copyleft" (CC:Share-Alike, GPL) and forcing non-commercial use on people, as long as there's attribution.

Not sure what you mean here, GPL doesn't force non-commercial use, just that any modification is available as source.

I meant those two as separate things. Share-Alike (copyleft) and Non-Commercial. Like two of the CC licenses.

I think the LGPL would work better in this case, or the GPL with linking exceptions (like a lot of compiler runtimes use) as it gives more flexibility to the user.

Well, thing is, this isn't a library, so linking isn't really an issue (it's a program, it runs "datafiles" or "resource packs" or whatever). Maybe "engine" isn't the right description. Game interpreter?

However personally I'm not a big fan of the GPL, as useful as it has been, and would instead suggest either BSD or MIT licences and just assume any good change will be pushed back to you. Seems to work well enough for most github projects, and companies are getting better and better and pushing back.

I know what you mean, just that I'm having problems imagining a case with this specific program where I'd like someone to make changes to the engine and keep them proprietary. Except, yeah, some people MIGHT make the changes BECAUSE they can keep them proprietary, then have a change of heart later and release the code... changes which might not be made otherwise. It's a tricky issue...

I've set the project to GPL 3 for now and put the license text (I'm working on SourceForge right now), I assume I can later change it to something more permissive when I make my mind?

[Chattox]

C++: To use a struct in a function outside of main(), can you use forward declaration and define it in main(), or does it have to be defined outside of a block?

[lorb]

C++: To use a struct in a function outside of main(), can you use forward declaration and define it in main(), or does it have to be defined outside of a block?

Is there any specific advantage you're getting from defining it inside main when you're using it outside of main?

[Chattox]

C++: To use a struct in a function outside of main(), can you use forward declaration and define it in main(), or does it have to be defined outside of a block?

Is there any specific advantage you're getting from defining it inside main when you're using it outside of main?

Other than being retarded, nope. Turns out I'd completely misunderstood how they work xD

EDIT: How do you define a range for rand()? Example; if I wanted to generate a number between 1 and 100, what would I do?