Weird bugs in Eclipse and Java logic

Development Code
1

I'm in a computer science class, and am making a Monopoly program on my own time for fun. All was going well until about an hour ago, when I came across some bugs I cannot explain.

The first has to do with Eclipse. If I try to debug my program, it crashes with a NullPointerException. However, I cannot for the life of me figure out why that NullPointerException ever occurs. If I export the project to a JAR file and run it, I have no trouble at all with it. The line in question is:
cPrinter.println("╔═════════╦═════════╤═════════╤═════════╤═════════╤═════════╤═════════╤═════════╤═════════╤═════════╦═════════╗", Attribute.NONE, FColor.WHITE, BColor.NONE);

(cPrinter is an instance of the class print.Color.ColoredPrinter, from the JCDP library.)
Google only returns results for the opposite - programs running in Eclipse, but not in JAR files.
I am using proper parameters for the constructor, and it works in Windows command prompt as a JAR file if I change the type to ColorPrinterWIN (although since command prompt does not display UTF-8 characters it shows up as a bunch of ?'s).

Edit: On time, Eclipse gave me this debugging info:
Exception in thread "main" java.lang.NullPointerException
at MonopolyBoardDraw.drawBoard(MonopolyBoardDraw.java:35)
at MonopolyLaunch.main(MonopolyLaunch.java:76)
FATAL ERROR in native method: JDWP on checking for an interface, jvmtiError=JVMTI_ERROR_WRONG_PHASE(112)
JDWP exit error JVMTI_ERROR_WRONG_PHASE(112): on checking for an interface [util.c:1313]

A fatal error has been detected by the Java Runtime Environment:

EXCEPTION_ACCESS_VIOLATION (0xc0000005) at pc=0x0000000072cf2868, pid=14860, tid=13652

JRE version: Java(TM) SE Runtime Environment (8.0_66-b18) (build 1.8.0_66-b18)

Java VM: Java HotSpot(TM) 64-Bit Server VM (25.66-b18 mixed mode windows-amd64 compressed oops)

Problematic frame:

Edit 2: Sorry that this bolds for some reason. I do not know why, and I removed the code tags since they did not work properly. There are supposed to be # pound signs at the start of each line that is bolded (besides this one).

The second problem I have come across is a really disturbing logic error.
I could not catch a NullPointerException no matter how many different ways I tried, and I have no clue why. I was trying to determine how many players were in the game.
Each player is an instance of the class MonopolyPlayerManager. Since you cannot create variables not originally written (that is, I cannot create X instances of a class because the user typed X players will be playing), I defined 4 players, but only have 1 actually created according to the constructor's parameters, like this:

public static MonopolyPlayerManager p1;
public static MonopolyPlayerManager p2;
public static MonopolyPlayerManager p3;
public static MonopolyPlayerManager p4;
...
public static void main(String[] args){
...
//createPlayers(players);
System.out.println("[Debug] For debug purposes, there is 1 player, name \"Test\", piece ♜.");
p1 = new MonopolyPlayerManager("Test", "♜", 0); //For testing purposes ONLY! Remove in final builds.
...
}

The constructor is as follows:
public MonopolyPlayerManager(String playerName, String playerSymbol, int currentPosition){
....
pID = totalPlayers;
totalPlayers++;
}

The totalPlayers is important, as it is the only way I can determine how many players are actually active without throwing a NullPointerException when trying to check any player's properties (i.e. p2.getName() will return null because p2 was not given a name since it was not constructed).

In the MonopolyPlayerManager class is this method:
public static String getPlayersOnSpace(int space){
String r = "";
if(totalPlayers == 1 && MonopolyLaunch.p1.getPosition() == space)
r += MonopolyLaunch.p1.getPiece();
if(totalPlayers == 2 && MonopolyLaunch.p2.getPosition() == space)
r += MonopolyLaunch.p2.getPiece();
if(totalPlayers == 3 && MonopolyLaunch.p3.getPosition() == space)
r += MonopolyLaunch.p3.getPiece();
if(totalPlayers == 4 && MonopolyLaunch.p4.getPosition() == space)
r += MonopolyLaunch.p4.getPiece();
for(int i = 1; i <= 9 - r.length(); i++){ //Fills the rest of the space with spaces.
r += "x";
}
return r;
}
What this does is it makes a 9 character length string, "r". If a player is at that spot, their symbol/piece (a UTF-8 character) is placed there, and the rest is filled with spaces.

Since there is only one player active (p1), and p1 is at position 0 (Go/collect $200), not position 20 (free parking), it should fail for all of these and I should have just 9 spaces, right?
Well, it printed 5 spaces, but just to make sure that my terminal was not resizing the spaces, I changed it to X's. 5 X's were printed.
Even if all 4 players were properly instantiated, they would still be at position 0 (Go) because this is when the board is being drawn for the first time, before they could have moved. Yet it is filling in 5 X's as if there were 4 characters there already. Even if it was inserting null characters, that still returns a space to the console, which I learned when working on a tic-tac-toe program.

Have I found a crazy logic bug or have I managed to miss something big. I can link you to my BitBucket repo, if you want to examine my code further.

Edit: I formatted my code on here with the spaces the way they should be, but I have no indentations. Sorry that it looks so sloppy, is it more than 4 spaces or is indentation broken?

2

A link to the code would help :).

3

[EDIT: BitBucket link redacted because I deleted the repo, and my username was my real name. The issue has been solved, so it is no longer needed here.]
Sorry that it's undocumented, I've just started to build stuff out. It's not very complex yet, though, so it should be fairly easy to understand.

In MonopolyBoardDraw.java (original, I know), line 113 references MonopolyPlayerManager.getPlayersOnSpace(int space) (MonopolyPlayerManager.java, line 44).
Each of those tests should fail because even though only 1 player "exists" (as defined in `MonopolyLaunch.java'), it is not even on space 20 as nobody has moved yet (so it has to still be on space 0).

If I got "xxxxxx" (6 X's), that would likely indicate that the first test failed (as there are 3 non-existent characters) but for some reason the last 3 passed. Instead, all 4 tests "pass" and I get "xxxxx" (5 X's).

Note that this uses JCDP (which is in the main folder since I develop on multiple machines and dislike having multiple class paths), and the board is drawn in UTF-8, not ASCII. To run this in Windows, you have to change the printer from ColorPrinter to ColorPrinterWIN, and set command prompt to use UTF-8 (keep in mind this can break batch files). For a Windows build, I am going to make a different branch where I write it in ASCII.
Also of note is that I have only populated 1 card set (Monopony) since I cannot find some of my Monopoly cards (they must have fell out of the box in my closet), so using the Original Monopoly card set will likely have a bunch of null entries on the board.

4

for(int i = 1; i <= 9 - r.length(); i++) //Fills the rest of the space with spaces.
{
r += "x";
}

Your intent here is to loop between 5 and 9 times, depending on how many players occupy a particular space. The exact number of times you perform the loop is dictated by the length of 'r'. However, inside the loop you modify 'r' by adding an 'x' to it each time - therefore changing the length of 'r' and therefore your loop exit condition.

This has the effect of only allowing the loop to run 5 times. After the 5th time i == 5 and r.length() == 5, and your exit condition is i <= 9 - r.length, or with the current values, 5 <= 9 - 5, which obviously evaluates to false.

Try capturing the length of r outside of the loop:

int currentLength = r.length();
for(int i = 1; i <= 9 - currentLength ; i++) //Fills the rest of the space with spaces.
{
r += "x";
}

5

Ok, that makes sense.
I never realized that it rechecked the length each time, I assumed the length() got stored as a final.
But it does seem more logical that since length() is a method, it would recheck every time.
Thanks!