Error Message When Running Program

User 28 said: Hey all,

I'm using my laptop at home (it is a UVA computing initiative laptop if that makes a difference) and I'm getting a strange error message when I try to run my Hello World program. When I type "java HelloWorld" I get this error message:


Exception in thread "main" java.lang.NoClassDefFoundError: HelloWorld
Caused by: java.lang.ClassNotFoundException: HelloWorld
at java.net.URLClassLoader$1.run(Unknown Source)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
at sun.misc.Launcher$AppClassLoader.loadClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
at java.lang.ClassLoader.loadClassInternal(Unknown Source)



I have installed Java successfully (I can type java -version, and it comes up as Java 1.6), followed the configuring Windows to use Java from the command line instructions carefully, and my program compiles fine.

I did find somewhat of a solution. I googled the error message and I found a suggestion to type: "java -classpath . HelloWorld". I did this and it works fine. I was wondering what the problem is and if this will affect my programs when I submit them. Thanks!


Professor Humphreys said: The classpath is a list of directories where Java will look for the program to run.

So when you say "java HelloWorld", java says "okay, where might I find a HelloWorld program? Let's start looking around".

The current directory should be in the default classpath, but it looks like your computer has a customized classpath. You can verify this by typing the command "set" at the command line and looking for a variable called "CLASSPATH".

If you find one, it's probably there for a reason (some other software package needs it, or something like that). However, all is not lost! Edit the CLASSPATH environment variable just like you edited the PATH environment variable when you were setting up Java. CLASSPATH (on Windows) is a SEMICOLON-SEPARATED-LIST of directories.

So add the string ".;" to the beginning of it.

If you have trouble getting this to work, the TAs should be able to do it for you.


User 14 said: Hi Mark,

This might have something to do with Java not knowing what to use for the classpath. I know this happened to some people last semester who accidentally changed their "classpath" environment variable instead of "path". Try going back to your environment variables (where you changed your "path"), find "classpath" and add a simple "." to it. Let us know if that helps at all.

-- Eric

101E and the J drive, and simple Java syntax question

User 39 said: In following the setup instructions, I had a question relating to setting up our network J drive. As I am a 101E student and we don't participate in the actual in class lab sections, I was uncertain as to whether or not mapping a J drive on my home computer was necessary/recommended, or if it wouldn't make any difference over my just storing it on a local drive.

My other question is - is it possible in java to have both a text output as well as an args[...] in the same system.out statement. In other words, in C you could say printf("blah blah, %i", x); and have no worries, but when I tried to mix the args with "....." I'd get compile errors.

Thanks for the help


Lesley said:

The concatenation character in Java is a plus sign. You can concatenate whatever you wish (inline strings, variables, mathematical expressions...) inside the print statement (Java also has a printf should you decide to go that route). Please ctrl+f "System.out" on http://www.cs.princeton.edu/introcs/12types/index.php --which is assigned reading for sometime soon if not now-- should you need more help with that, and for printf see http://www.cs.princeton.edu/introcs/15inout/.

I took 101E last semester and found it imperative to have my J drive mapped. If, for some reason, there is a submission error, you need a version of your code on the UVA server so that there is no dispute over the time and honor code.

error?

User 26 said:

i set up my computer with java etc and when i type java -version into the command, it does list a version.

i put the helloworld program into notepad and saved it as HelloWorld.java

when i put javac HelloWorld.java in command this is what I get:

javac: file not found: HelloWorld.java

Usage: javac <options> <source files>

use -help for a list of possible options

whats up? and if this was a setup error on my part and TAs can help with setup - do they have office hrs today (thrs) or ever? I'm ur typical Liberal Arts major - I can't follow directions to save my life.

Professor Humphreys said: You need to make sure you have changed to the same directory that contains the .java file. This is shown in the videos; look for the part about the "cd" command.

Lesley said: Also, make sure you use quotes around the file name when saving or Notepad will sneakily save the file as HelloWorld.java.txt - see attached image.

User 167 said: Think of the command prompt a lot like Windows Explorer - at any given time, you're "in" a particular folder. Typing commands like "javac " will call javac on the file of the given name *in whatever directory you're in*. The error message you got indicates that you were in a folder that didn't contain a file called 'HelloWorld.java' To change directories, you use the 'cd' command: type 'cd ' to use this. Thus, 'cd workspace/HelloWorld/src/' will take you to the src dir for your HelloWorld project (assuming that you've started off in the folder containing your workspace, and your project is called HelloWorld). Also be sure that the file really is called 'HelloWorld.java' - type 'dir' at the command line to see a list of files in the current directory, or just open the folder normally through Windows Explorer or whatever to see what's in it. As Lesley pointed out, some programs try to be "helpful" and screw up your filenames.

Moving through directories in command prompt

User 200 said: Is it possible to go back through directories in the command prompt without closing and reopening it entirely? I went through the command prompt to compile and test my Hello World program, so I was at:

C:\student\workspace\Hello World\src\

and I wanted to go back to my workspace folder and test my UseThree program right afterwards. Is it possible to get to my folder

C:\student\workspace\Use Three\src

without closing and reopening the command prompt?


Professor Humphreys said: cd ..

Mapping the J drive

User 240 said: I was wondering if someone could explain how to set up the J-drive on my personal laptop. I am in CS 101E, so I don't have a lab time, but I figure it would be useful to be able to save things to the network so that I can go to Thornton Stacks and get help if necessary. Thank you.

Professor Humphreys said: I believe ITC has instructions about how to install Home Directory Services on personal machines.

submission

User 26 said:

now I have a HelloWorld.java txt file and a HelloWorld.class file...

which am i submitting?

Professor Humphreys said: Every assignment will have a list of filenames at the bottom. In this case, it's the .java file.

One long line of code vs. many short lines?

User 240 said: Which is better, to have one long line of code or to have about half a dozen shorter lines of code?

Professor Humphreys said: Generally, the latter, although this is a style issue and sometimes it's just personal preference..

Command Prompt

User 200 said: Sorry to repost this in an entirely new thread, but I didn't see a reply option on my old post:

Is it possible to go back through directories in the command prompt without closing and reopening it entirely? I went through the command prompt to compile and test my Hello World program, so I was at:

C:\student\workspace\Hello World\src\

and I wanted to go back to my workspace folder and test my UseThree program right afterwards. Is it possible to get to my folder

C:\student\workspace\Use Three\src

without closing and reopening the command prompt?

When I tried using "cd" to move backwards, I kept getting a message saying
"The system cannot find the file path specified"


Professor Humphreys said: cd ..

CS 101-E Lab Submissions

User 449 said: For submitting the CS 101-E labs, it says that you may re-submit files as many times as you like. Does that mean that the following scenario is legit?:

- Submit program X.
- RoboGrade.
- Bad program, so credit is not received.
- Submit program X again (after fixing).
- RoboGrade.
- Credit received!

I am wary of the RoboGrader.

-Joel


User 167 said: Yes, that is exactly right. This is also true for students in 101, but for 101 students you can only keep doing this as long as you do it from a lab computer during your assigned lab time. 101E students can do it any number of times before the lab is due.

Professor Humphreys said: Yes. You can re-robograde as much as you like.

The pattern is:

submit
robograde
curse
fix
submit
robograde
curse
...
submit
robograde
HOORAY!

javac

User 355 said:

When I type the javac command in, it says that javac is not a recognizable command...what's wrong??

Professor Humphreys said: It's probably not in your path. You need to follow the installation instructions on the course website, or see a TA with your laptop.

Invalid Directory in Command Prompt

User 106 said: When I use the command prompt window to compile my program, I use "cd workspace," but it returns with "Invalid Directory." When I use "dir" to search the directories in my home directory, it shows my workspace to be there, but I can't get to the program.

Professor Humphreys said: So you type "dir", and see the "workspace" directory in the output listing, but when you then type "cd workspace" you get an error?

That is quite mysterious -- I can't think of why that would happen. You might be running into some transient errors with ITC's home directory service; we had a few issues with that in the lab this morning.


User 167 said: Makes sure you're using the right name - if it's called 'workspace', then "cd workspace" should work, but if you've given the workspace folder a different name, then you have to use that name instead (for example, if it's called 'myWorkspaceFolder', type 'cd myWorkspaceFolder').

Lesley said: When you're in the directory where your workspace is, and you do 'dir' and it shows up there, try next typing 'cd ' and then pressing the tab key. This will magically make the first directory alphabetically be fully typed out after cd. Continue to press the tab key to make your way through the directories alphabetically until you see your workspace directory's name, then you can press enter to continue. You can also do 'cd w' and then press tab to shorten up however many to go through if your workspace directory starts with a w.

how do i reply

User 355 said: how do i respond to a post?? (i feel like an idiot...)

User 14 said: Click "reply" next to edit and delete right underneath the text of the message. So you should see Edit | Delete | Reply. Click reply

eclipse

User 221 said: In the video that we were supposed to watch before the lab they used eclipse to write the program. I downloaded drjava but I can't find eclipse. Everyone else is talking about typing it into notepad... Since I'm in CS101 can I just wait until lab and they'll tell me what to do? Do I even need to bring my laptop to lab? Thanks in advance.

Professor Sullivan said: Hi Julianne,

Please sett the instructions on our web site for "Setting up your computer." Everything that you need is there. You (and your colleagues) should all be using Eclipse. If you get stuck, bring your laptop to the 101 Office Hours for help from the TAs.

KS

Using a Mac

User 124 said: I've downloaded Eclipse, but I don't know where or how to use the command prompt as seen on Windows. Professor Sullivan said that the java application was already installed on a Mac, but I can't find it. So, what should I do about this?

Professor Sullivan said: None of the Windows command line stuff applies (at least not in the same way) to the Mac.

Just run Eclipse and try to write an compile the HelloWorld program. It should just work! If not, see a TA during office hours Sunday, Monday or Tuesday eve, or see my in my office hours for help.

KS


Professor Humphreys said: You can do all the same stuff from the "Terminal" program. Instead of "dir" use "ls".

lab0 - compile

User 546 said:

I'm trying to compile the program putting "javac HelloWorld.java" in the prompt, but it is saying that 'javac' is not recognized as an internal or external command, operable program or batch file.

Do you know why? The rest of the things before are equal to the assignment video. And I clicked Run in the Eclipse just to test and it worked (the Hello, World showed up in the window...

Thanks!

Professor Sullivan said: Hi Camila, Please see the directions for "Setting Up Your Own Computer," on the web site, and specifically the part about setting your search path.

Question about grading

User 71 said: Is it true that as long as it can be compile and runs properly and has the desired function it will be graded with 100, or the program has to be exactly the same as it appears in the book?

Also, is it cool if I use Jdrjava(instructed by the book) instead of Eclipse, so I don't have to compile and run somewhere else?


Professor Humphreys said: As long as the program behaves as specified, it will get credit.

You need to use Eclipse.


Professor Sullivan said: You said ... so I don't have to compile and run somewhere else?

You can use Eclipse anywhere you can bring your computer. Just install it on your machine following the instructions on our web site: setting up your own computer.

lab0 - compile 2

User 546 said:

I couldn't reply the other message...

I did what is written in the "Setting up my computer" but it didn't work. It is saying there that we should put in the path: C:\Program Files\Java\jdk1.6.0_04\bin;c:\Program Files\Java\jre1.6.0_04\bin

But I only have the "jre" ... I dont have the "jdk"...

Do you know why?

Thanks!

Lesley said:

It is very important to follow the directions exactly as they appear on this page #10 specifically:

Visit http://java.sun.com

Under popular downloads (on the right) click on Java SE

Next to JDK 6 Update 4 click on Download

Professor Sullivan said: You must not have downloaded the JDK properly. Try again. Be sure to dowload the JDK.

Subscribing problems

User 337 said: I can't subscribe the announcements and technical help programs.

Professor Sullivan said: You'll need to provide more information in order that we might diagnose the problem you're having.

Question about replying

User 344 said: Can only teachers/TA's reply to posts in the technical help forum? Because I don't have a reply button here, but do have one in the social forum.

User 14 said: Ah, you're probably right about this, and that might explain the earlier post about not knowing how to reply. It might be that way by design, but one of the professors would know more about this.

User 14 said: Can one of the professors and/or someone familiar with the set up of the site address this question?

Professor Sullivan said: We currently have the technical help forum set up for faculty/TA replies only. We can revisit this decision if there's a push for that, or we might consider setting up a student help-each-other forum. Let us know what you think!

Run the Programme "UseThree"~

User 161 said:

When I ran the programme "UseThree", all the words and punctuations lied in different lines,

it appears like:

Hi,

Alice

,

Bob

,

Lucy

.

Why not "Hi, Alice, Bob, Lucy."

Thanks!!

Professor Humphreys said: You probably used "println" instead of "print". Please read the portion of the book that describes the difference between these functions.

C:\ to J:\

User 100 said: How do you switch from the C:\ to the J:\ in the command prompt?

Professor Sullivan said: Just type J: at the command prompt

RoboGrader

User 449 said: My question is similar to one posted under the thread title, "Question about Grading", though I am still mildly flummoxed by this RoboGrading beast, and I quest for clarity:

How precise does the output have to be?

For example, one of the programs we have to write for groupwork requires computation of rise, run, slope, and distance. Would the three following outputs be received the same by RoboGrader? (assuming the values were correct):

Output 1:
4
3
1.3
5

Output 2:
4 3 1.3 5

Output 3:
Rise: 4
Run: 3
Slope: 1.3
Distance: 5

-Joel


Professor Humphreys said: Once you write the program, you can just run the robograder and see what it does!




Professor Sullivan said: Joel,

This is a great question, and precisely the kind of question that a really good software developer would ask. The problem that you've identified is that the specification is incomplete: it doesn't specify the output format. Problems of this kind often lead to conflicts (e.g., lawsuits), where the developer think he or she has done the work that was required, and the customer begs to differ. It's very important in such cases to clarify the specification before too much time has been invested in coding.

We will now clarify the specification. Each output item should appear alone on separate, adjacent lines, in the order indicated in the assignment. I'll go update the assignment now.

Thanks for asking!
Kevin Sullivan

Lab Quiz 1

User 380 said:

If you put 0 0 0 into the program you get 0/0 for the values of cyan magenta and yellow and the program displays NaN. Is this okay, or do we need to make it output 0s for these values?

Professor Humphreys said: 0 0 0 represents pure black; your program needs to get this right.

This is covered in the "Testing Your Program" portion of the assignment.

Group Assignment question

User 31 said: Directions for the fifth program are written as follows:
Write a program QuadraticEval.java that reads in three double values, a, b, c, and x, from the command line and that prints the result of evaluating the expression ax²+bx+c.

My question is whether all the program is required to do is to have three inputs from the command line (a, b, and c) and output (ax²+bx+c) in the terminal window. Seems almost too easy.


Professor Sullivan said: Oops! The specification should have said, "reads in four double values, a, b, c, and x, from the command line ..." Yes, it's not a hard program to write! --KS

Need help with the output's format.

Professor Humphreys said: can you please tell us which ones you think are ambiguous? Also you can always run the robograde script after you have submitted if you are not sure.

Professor Sullivan said: English-language specifications are notorious for their ambiguity and incompleteness. In practice, it's usually impossible to make them satisfactoryily precise and complete (but we try). For this reason, in the industrial engineering of software (as opposed to amateurish hacking), natural language (e.g., English) specification are supplemented by other approaches. In particular, software testing is an important practice. Testing means that you run a program, giving it known inputs, and you check to see whether its actual outputs match those known to be correct for the given inputs.

(Question: If a program produces the correct output for one test case, can you conclude that it produces the correct output for all test cases?)

Test cases constitute a partial but precise specification of the required behavior of a program. The incompleteness and imprecision in our CS101 specifications (whether intended or not ) will give you an excellent sense of the kinds of ambiguities that you'll face when you're developing software in whatever field you end up working in.

Our automated on-line testing mechanism, by contrast, will give you a good sense of the power of testing to help developers to develop programs that satisfy precise, documented specifications. These specifications are not written in English, but in code -- test code that runs your code to see if it behaves correctly in certain cases!

Assignment 1

User 71 said: As for the Histogram, should we print out the amount of random values in a certain sub-integral or its probability of percentage?


Professor Humphreys said: Good question! The assignment page says to output "the values in the array", so don't reinterpret the histogram counts as probabilities, just print them out.

Professor Sullivan said: Right. It's just a histogram: a count of the number of elements in each interval. (What would you do if you wanted to output an estimate of the probability of a random draw falling in any given interval?)

Converting Strings to booleans

User 200 said: I tried using the following syntax:

[source code removed by admin]

to cast a string from the command line into a boolean. When I compile the code on Eclipse, it doesn't display any error messages, but when I try compiling it on the command prompt, I get the following error message:


"cannot resolve symbol

[code information removed by admin]

1 error"

Did I write the code incorrectly, or is my computer just not recognizing the Boolean class?

(Edited by Greg Humphreys - original submission Saturday, 19 January 2008, 04:52 PM)

Professor Humphreys said: I edited your post to remove code that you really should NOT be posting on the internet. Please re-read the course collaboration policy.

In repsonse to your specific question, the solution code looks identical to the code I cut out, so it's hard to say what's going on. It might be possible that your version of Java is not correct, but it's really hard to say.

Try running your same code on other computers (like the ones in stacks) and see if it works there. If so, then your installation of Java needs to be redone with the latest versions of everything.


Lesley said:

This might also be a result of an old version of javac even if java is version 1.6 (typing java version produces a 1.6 version but javac still gives this error).

Instructions for solving this problem on Vista, as done right now in the Stacks:

Go to Control Panel (classic view) > System, then click Advanced settings on the side menu

Click on the "environmental variables" button in the window that pops up (bottom of window)

Go to the SYSTEM variables, not the user ones (these are in the lower half of the window that pops up)

In the PATH variable, put the directory to the JDK you have installed in there (ie. edit the variable and at the BEGINNING add C:\Program Files\Java\jdk1.6.0_04\bin;)

Exit and restart the command prompt

What does it mean?

User 71 said: When I try to run the StdGaussianHistogram.java I typed "java StdGaussianHistogram 1000" and it said this:"Exception in thread 'main' java.lang.ArrayIndexOutofBoundsException:100 at StdGaussianHistogram.main<StdGaussianHistogram.java:25>",
what does it mean?


Professor Humphreys said: It means you tried to access element 100 of some array, but the array is smaller than that.

easy way to generate an error like that:

public class MakeError {
public static void main( String[] args ) {
int some_array[] = new int[100];
some_array[100] = 5;
}
}


Oops, some_array is a 100-element array, so only slots 0,1,2,...,97,98,99 are legal.

Assignment 1-again

User 71 said: Because there are two arrays in this program, and basically I want to to do the problem by using "for" loop with two variables of "i" and "j". But because those two arrays don't have same amount of elements, if I still use the for loop and if say the first arrays has 1000 elements and the second has 100, the loop command basically doesn't work as I want it to do. Because the second arrays has 100 elements, I don't want to write 99 lines of "else if". What else can I do?


Professor Humphreys said: I apologize, but I'm not sure I understand your question.

The assignment says to loop twice. The first loop populates the array (it will run N times) and the second loop will print out the array (it will run Number-Of-Array-Elements times).

The standard output

User 189 said: What does "on the standard output" means in the team assignment? Is System.out.print()" good for the homework? I typed in "StdOut.print(d)" and the system said "StdOut can not be resolved".

Thank you.


Professor Sullivan said: System.out.print() and System.out.println() do indeed output data to what we call the "standard output."

By default "standard output" is connected to the console. In Eclipse output to the console is presented in a panel down at the bottom of the screen. If you're using command line commands, standard output appears in the command window (on Windows) or the terminal window (Mac) in which you're running the program.

When we learn (shortly) about input/output "redirection" you'll see that it's very easy to connect the "standard output" of a program to other outputs, e.g., to send the standard output of a program not to the console but to a file so that the data can be saved for later use.


The short answer: System.out.print() and System.out.println() are just right!

Java Library Documentation

Professor Sullivan said: Here's a link to all the relevant on-line documentation for the Java libraries.

It's worthwhile early in the semester to have a look at the following classes:

• Math
• Integer
• Double
• Boolean
  

Lab Quiz 1

User 199 said: During the quiz will we be able to look at the instructions for the quiz, and therefore the formulas to convert between the different colors? Or, do we need to memorize the formulas?

Professor Sullivan said: You don't have to memorize the quiz. Accessing the quiz web page and Eclipse, and you should be able to write, compile, run, test and submit the program.

TA office hours

User 364 said:

I couldn't find on the website a schedule for ta office hours. Does anyone know what time they are today and tommorrow?

Professor Humphreys said: They are 6-10 Sunday, Monday, and Tuesday, in Thornton stacks.

MixedNumberTypes

User 393 said: The program will not let me submit MixedNumberTypes and I believe it is because it is not listed under the list of the finals that is supposed to be submitted.

Professor Humphreys said: Whoops! Fixing, check back in a minute or two.

Thanks.

Lab Quiz 1

User 223 said: In the directions it states that the program should take 3 intergers (RGB) and convert them to 4 reals (CMYK). Does this mean that your program should only work if intergers are entered. Technically if it works if three real numbers are entered is that ok too as long as it also works for intergers.

Professor Humphreys said: The program only needs to work when integers are supplied. If it behaves reasonably when other types of input are supplied (like reals, as you point out), that's basically a bonus, and good for you for realizing this.

We won't be testing anything outside of the specification, though, so it's unnecessary.


Professor Sullivan said: This is a great question. It goes to the question of what we mean when we say that a program satisfies its specification. One possible answer is that a program does at least what the specification requires (and if it does more that's okay). A different answer is that it does at least what the specification requires and no more. For purposes of the current assignment, your program must do at least what the specification demands. If it can do more, that's okay. In some systems, it's important to ensure that a program can do no more than it's asked to do. For example, if software is controlling a safety-critical device (such as a fast-moving train full or a radiation treatment machine for cancer patients), it is important to know precisely what the software will do for all possible inputs: it should do no more than has been explicitly stated. One can handle this in the specification by saying something like this: any other input is considered erroneous, and any such case should be handled as follows: ... details

changing directories

User 128 said: This is a general question. I know you the command "cd" will move you to a more specific directory, but what is the command if you want to move back "up" the hierarchy to a more general folder?

Professor Sullivan said: This question was answered in an earlier post on the technical help forum. Thanks, everyone, for checking earlier posts.

Groupwork 1

User 71 said: Will we lost score if we put (or omit) names before the values in the printout while the assignment does not ask us to do so.


Professor Humphreys said: Yes.

The assignment specifies what your program should print out. You should follow the directions in the assignment.

If you feel that the directions are ambiguous, by all means, let us know.

Also, if you're ever confused about precisely what the output should look like, you can <i>always</i> submit your program and run the robograde test!

Groupwork 1 Question 9

User 99 said:

I've tried to figure out what this problem means for a while and I've gotten nowhere.

Write a program DeMorgan.java that reads two boolean values from the command line, stores them in two boolean variables b1 and b2, and prints out true if the negation of the conjunction of the values is equal to the disjunctions of the negations of the values, and that prints false otherwise. The bolded text is primarily where the confusion lies.

Professor Humphreys said: Somehow, this question does use a lot of logic jargon. Here's a glossary: conjunction: "and" disjunction: "or" negation: "not"

Re-submitting Corrected Programs

User 179 said: I first submitted one version of Distance.java that was incorrect, so I ran through Robograde to test and took it off. I fixed the program so that now when tested in the command prompt it works as directed, but when I submit the new program and check on Robograde, it still uses the first program I submitted. What should I do?

Professor Humphreys said: When you resubmit, it overwrites the previous program, so my guess is that you didn't resubmit properly.

Professor Sullivan said: It's possible that you didn't save the file in Eclipse. In that case, the version on disk might be old, even though you're working on a newer version in Eclipse. Hit Control-S (hold down Ctrl while pressing S) to save.

boolean

User 166 said: In problem #3 from the group work my group in unsure about the format for typing in a boolean value. Also, is the answer supposed to print out true and false?

Professor Humphreys said: You should pass the boolean to System.out.println(). And just type "true" or "false".

Uploading Group programs

User 113 said: We seem to be having trouble uploading Echolnt.java. Although it is listed as one of the programs we need to include, after we have selected it and hit the upload button, we receive the message:

I'm sorry, but Echolnt.java is not one of the files for this assignment, so I'm skipping that one.

Professor Humphreys said: Try replacing that lowercase ell with an uppercase eye.

RoboGrader

Professor Humphreys said: Hmm. It probably means I broke the robograde script this morning. Please bear with me...

Professor Humphreys said: Okay, this has been fixed. Sorry about that.

Quiz 1

User 91 said: Are we allowed to consult TAs for help on the quiz prior to Thursday?

Professor Humphreys said: You are *encouraged* to do so.

Group Assignment #3 - Boolean

User 52 said: I'm having a problem with the command prompt for the boolean exercise.

for some reason the command prompt cannot resolve the symbol "parseBoolean"

but in the previous exercise the command prompt worked fine in resolving the symbol "parseDouble". I had no problem using command prompt for the previous two exercises.


User 13 said: See previous post "Converting Strings to booleans" please

Lesley said: Please see the other thread on this (titled something like Strings and booleans). Your java and or javac version is out of date and you will need to go update your path variable.

arguments

User 555 said:

when ever i use a write a program on my computer that needs an argument to be entered i get the following error mesage.

java.lang.NoClassDefFoundError: 3

Caused by:

at java.net.URLClassLoader$1.run(Unknown Source)

at java.security.AccessController.doPrivileged(

at java.net.URLClassLoader.findClass(Unknown Source)

at java.lang.ClassLoader.loadClass(Unknown Source)

at sun.misc.Launcher$AppClassLoader.loadClass(Unknown Source)

at java.lang.ClassLoader.loadClass(Unknown Source)

at java.lang.ClassLoader.loadClassInternal(Unknown Source)

Exception in thread "main"

java.lang.ClassNotFoundException: 3Native Method)

Professor Sullivan said: How are you editing, saving, compiling and running your program?

User 167 said: You need to type "java " For example, java EchoInt 3 I'll bet you typed something like java 3 which would produce exactly the error you got above. This is because 'java' is itself a program that takes several arguments. The first is the name of the java program to run (in this case "EchoInt"), and the remainder of the arguments go to that program. So "java 3" tells java to look for a program called "3", which doesn't work; but "java EchoInt 3" tells java to run the program called "EchoInt" and pass it an argument of "3".

Group Assignment problem #7

User 151 said: I have tried writing a program And.java, but no matter what I give as arguments, even if both arguments are the same, the output is false. I am at a loss as to why this is happening, as I have tried several things with no result.

Lesley said:

There are a number of things you could do to determine what is going on.

Have you recently saved and recompiled the proper program? You need to do both if you want your program to do something different than the last time you compiled it.

Is your algorithm correct? Try printing out the arguments, and seeing if they are what you expect. If they're correct, then you will need to inspect your algorithm to see why it isn't doing what you want it to do.

If you still have problems, you might consider coming to TA office hours tonight. It's hard to diagnose a problem with an algorithm without seeing the algorithm in use, and as you can't post it here, that leaves you with few options.

Home Directory and Mac

User 124 said: Eclipse won't allow me to set my Home Directory as my workspace... any ideas?

User 167 said: Eclipse has a bug (registered as bug 44735 that screws up the file locking for some NFS file systems. I haven't seen this bug manifest except when a Mac user attempts to set the workspace to a directory on an NFS filesystem such as the one ITC uses to host home directories. Essentially, the filesystem doesn't support a particular form of file locking that Eclipse wants to use to lock the workspace.

The error message you'll get is either that "the workspace is in use or cannot be created" or a long error message about file locking.

I'm looking for an elegant solution to this; so far I haven't found one, but here's an ugly workaround if you have this problem:

---

Ugly Hack for Disabling Eclipse Locking on Mac OS X

Find your Eclipse.app (probably in /Applications/eclipse).
Right-click (or control-click) on it and select "Show package contents".
This pops up a window containing a single folder called "Contents". Go into this folder, and into the sub-folder called "MacOS".
In "MacOS" you'll find a file called "eclipse.ini".
Open eclipse.ini in your favorite text editor, and add the following line to the end of the file:

-Dosgi.locking=none



---

This disables the locking, and should allow you to use the workspace in your Home Directory.

This is a fix for MAC users only, and since it involves hacking Eclipse's configuration files you should ONLY; do this if you have the error above.

I'll keep looking for a better way to do this.

Robograde DeMorgan

User 130 said: In our DeMorgan program, the robograde tests b1 is true and b2 is false twice. The program said the first test was false and the second test was true. The same happens when b1 is false and b2 is true on the robograde progam. Testing the program in Eclipse and on the Command Prompt, we get the right answer in all cases.

Professor Humphreys said: The DeMorgan robograder tests the following combinations:

false, false
false, true
true, false
false, false

error message

User 240 said: What does this error message (when trying to run the program through Command Prompt) mean?

Exception in thread "main" java.lang.NoSuchMethodError: main


User 14 said: This means that java can't find your main method, which is where it starts executing code in your program. Make sure you typed correctly "java (programName) (arguments)" in the command prompt. You can get this error if you type in "programName.class" instead of just "programName". You can also get this error if you didn't compile your program before running it.

error from command prompt

User 210 said: when i try to run the program "HelloWorld" in command prompt, an error message appears: exception in thread "main" java.lang.unsupportedClassVersionError: bad version number in .class file



User 14 said: Yufei,
This happens when your jdk version (the program you use to compile your code) is different from your jre version (the program you use to run your code). Please see the directions for "Setting Up Your Own Computer," on the web site, and make sure you follow them exactly. Particularly, make sure your java jdk and jre are the same version. Any older version of the java jre should be uninstalled. If you still need help setting this up, come to the TA office hours.

RoboGrader

User 194 said:

After submitting the programs and using the RoboGrade program, the program first said that our programs were incorrect because the output didn't match. However, after resubmitting it over and over again- we eventually got full credit on the programs submitted (without changing anything in our original programs). A TA suggested we mention this on the forum because many other students have had similar problems where RoboGrade first says that the programs are incorrect, but then after resubmitting they are correct.

Professor Humphreys said: When robograde said that the output didn't match, was it right? It shows you the output of your program and the solution program, so you should easily be able to look at the output that it's comparing.




User 13 said: The problem is that, when the students run the program on the desktops, it works. But when they submit to the RoboGrader, it does NOT. And they submit it many times without changing the file, finally accepted by the RoboGrader. (This is the situation the students described)

Another group of students mentioned the same problem.


Professor Humphreys said: I understood the post, but without information about the nature of the failure it's very hard to figure anything out.

For example, when robograde says that the outputs don't match, what are the outputs? If someone can reproduce this error and post the output on the forums I might be able to see what's going on.

Mixed Number Types

User 240 said: I started writing my code for the MixedNumberTypes.java program (#13 of the group assignment). I copied the line that they gave us exactly, compiled and ran the program, just to make sure everything was in working order before I continued. However, instead of printing out 1/3, like it's supposed to, my program printed out 0. What did I do wrong?

Professor Humphreys said: What did you do wrong? Nothing.

1/3 in java is zero. Welcome to the world of integer arithmetic.

Try 1.0/3.0 and see if you like that answer better :)


Professor Sullivan said: Jennifer,

What you're seeing is the point of the exercise. The expression 1/3 denotes integer division of 1 by 3. 1 divided by 3 in the integers is 0 remainder 3, which is why the answer is 0. If you write 1.0/3.0, a different kind of division is indicated: namely floating point division. In this case, you'll get an answer that looks like one third approximately, as a decimal number.

As a programming you have to be able to recognize when integer operations will be performed and when floating point operations will be performed. In brief, if all of the operands in an expresssion are of type int then in most cases, you'll get integer operations. If any of the operands are of a floating point type, then all of the operands are typically converted to floating point values and floating point operations are used.

Kevin Sullivan

difference between"print" and "println"

User 210 said:
first time i tried to type "print"
2nd time i typed "println"
surprisingly, the outcomes were the same

I'm wondering what are the differences between "print" and"println"

Thanks




Professor Humphreys said: "print" prints.
"println" prints, and then begins a new line.

Mixed Number Types (2)

User 323 said: Referring to the same question asked in 'Mixed Number Types':
Can you possibly say "1/3" instead of 1.0/1.3? That solved my problem (Using quotation marks), but I was wondering if that was the intention of the question.


Lesley said:

"1/3" is a String. It is not equal to 1.0/3.0 which is a number.

1/3 is the int 1 divided by the int 3. Please see Professor Sullivan's detailed explanation on why this is 0.

1.0/3.0 is one-third the floating point (decimal) number as you expect it.

Try adding "1/3" + "2/3", ie. System.out.print("1/3" + "2/3") - I think you won't get 3/3 or 1 like you seem to want!! What will (1/3 + 2/3) yield? What will (1.0/3.0 + 2.0/3.0) yield? This is the point of the exercise.

"javac" problems

User 105 said: I have quadruple checked that I added the info into my PATH like the instructions on the website, and when I enter "java -version" into the command prompt I get a java version number, however whenever I try to compile anything I am told that "javac is not recognized as an internal or external command..." even when I'm sure I'm in the right directory.

Help?! Thanks.


Lesley said:

It's possible that you don't have the JDK installed but rather just the JRE (which doesn't contain javac). If your directory that you added to the PATH variable doesn't have JDK in it, then you'll need to reinstall the correct version of Java accordingly.

If you do have JDK in your PATH variable, make sure it points to /jdk_(version number)/bin because javac lives in that bin directory.

After modifying the PATH variable, you will need to exit and restart the command prompt.

Professor Sullivan said: If anyone is struggling too much with installing the software, see the TA in office hours for help

reply button...

User 550 said:

Is it just me, or is there no way to reply to someone's post in the technical forum?

thanks

Andrey

User 167 said: Only TAs and the profs can reply to the technical forum. If there's a lot of demand for letting students reply as well, that can probably be changed.

Errors for min in color program

User 282 said:

I don't know what I'm doing wrong, but eclipse doesn't like the way I'm using the min method. Has anyone else gotten these error messages?

Syntax error on token "min", { expected RGBtoCMYK/src RGBtoCMYK.java line 11 1200976283187 49

Syntax error on tokens, Expression expected instead RGBtoCMYK/src RGBtoCMYK.java line 11 1200976283187 50

User 167 said: Make sure that you have the right number of opening and closing curly braces ({}) and that they're all in the right places. If that doesn't work, and you still can't figure out what's wrong, bring your code into office hours, and we'll take a look at it.

Eclipse Won't Let me Access My Workspace Anymore

User 53 said: I was writing a while loop and I think I may have accidently created a program that runs forever. I can no longer open up the workspace that I saved the java file in. Windows won't let me delete the program file either. Help. I've tried restarting my computer.

User 167 said: When Eclipse opens a workspace, it creates a file /.metadata/.lock that marks the workspace as "in use". This prevents multiple copies of Eclipse from accidentally editing the same workspace. When Eclipse stops using the workspace, it deletes this file. However, if Eclipse crashes or otherwise gets forced to quit, it frequently leaves the lockfile behind, and so it will think that the workspace is still in use. If the error you get when you try to use the workspace is that the workspace "is already in use", then the above is most likely what's happened. To fix it, go to the ".metadata" folder in your workspace and delete the file called ".lock" In the future, if you run a program from Eclipse that goes into an infinite loop, press the little red square over the Eclipse console to force your program to stop.

Mac Terminal

User 323 said: I am not sure how to switch to my J: drive when using Terminal on my Mac.

User 167 said: Instead of mounting as a separate drive, on a Mac the home directory appears as a networked drive in the /Volumes folder. Type "cd /Volumes/" in the Terminal, where is your UVa network ID, to move to your home directory.

Lab1 EC

User 343 said: What exactly is the "more difficult edge case mentioned above" that gives us 5 pts extra credit on the colors lab? What must our program do differently as opposed to just going for the basic requirements?


Professor Humphreys said: (r,g,b)=(0,0,0)

Professor Sullivan said: That's right: (0,0,0) is a legal input according to the specification, but the formulae provided do not work for this value. Why? What color does (0,0,0) represent? What is the correct CMYK equivalent? You will get an extra 5 points for handling this case correctly. Hint: You probably need to read ahead about "conditionals" (if statements).

Lab Quiz

User 95 said: For the lab quiz, will we be given the formulas for cyan, magenta, etc.? And will it only be the program we have to create and run or will there be an actual pen and paper quiz?

Professor Humphreys said: You will have access to the lab web page during the lab. There is no written component to the quiz.

Professor Sullivan said: You have to write the program :-) We're just looking to be sure that you can write a program that implements simple arithmetic conversions without help.

Assignment 1

User 251 said: For part two of the assignment, is there a specific method for splitting the interval [-5.0, 5.0] into equal sub-intervals?

Professor Humphreys said: Are you asking if Java has some sort of built-in support for this? In that case, the answer is no - the splitting is just conceptual. You will just create an array with the right number of slots, one for each sub-interval.

Laboratory Quiz

User 90 said: In the instructions for the first lab quiz it says "The program should assume (and not check) that the inputs are between 0 and 255." Does this mean it assumes you input numbers between 0 and 255, or should we create a way for an error message to be output when a number greater than 255 is entered? Thank you!

Nic


User 14 said: The specification says that you should assume (and not check) that the inputs are between 0 and 255. This means that when your program is run, you can assume that the user gives it an integer value between 0 and 255, and you don't need to check this within your code.

Group Assignment # 5

User 432 said:

When I run my program for QuadraticEval.java through Command Prompt, I get a message saying, "Exception in thread "main" java.lang.NoSuchMethodError: main." Just wondering what the means and how to fix the code.

User 14 said: See the earlier post titled "error message" about 15 posts down to answer why this can happen.

Assignment 1: Intervals

User 380 said: When we break up the interval [-5,5] into the smaller intervals, does it matter which of the endpoints we exclude and which one we include (i.e [-5,-4.9) or (-5,-4.9])?

User 181 said: This is a very good question. I think there is a small problem with this range. We'd better consider [-5,5] as [-5,5), in which each interval is like this: [-5,-4.9), [-4.9,-4.8),......,[4.9,5).

Professor Sullivan said: >We'd better consider [-5,5] as [-5,5), in which each interval is like this: [-5,-4.9), [-4.9,-4.8),......,[4.9,5).

That's almost right. The interval [-5,5] is the interval [-5,5] not the interval [-5, 5). In other words, -5 is included in the interval on the left and 5 is included on the right.

But the question remains: what are the intermediate intervals. The specification was indeed ambiguous in this regard. Zhiheng has given a fine resolution. Let's disambiguate as follows. The sub-intervals are like this:

[-5,-4.9), [-4.9,-4.8) ... [4.8, 4.9), [4.9, 5.0]. The last sub-interval on the right is different than the rest, being "closed" on both ends.


Professor Humphreys said: [Post removed because of complete and utter incorrectness]

Professor Humphreys said: Please ignore my previous post about ranges and so forth, it demonstrates a complete lack of understanding about math and how numbers work. Sorry.

Professor Sullivan said: Hi Everyone: This is exactly how software development works in practice. We talk through problems and work out understandings of (1) precisely what is to be done, and (2) how do get it done, e.g., by programming in Java. In many cases, in fact in general in practice, (1) is actually much harder than (2).

The current discussion is an example. In the problem at hand, we ask you to use the basic Box-Muller formula, which takes uniformly distributed pseudo-random numbers in the interval [0.0, 1.0) as inputs (and these are indeed the numbers that Math.random() generates for you). But then you run these numbers through the Box-Muller formula, and what you get out are (pseudo-) random numbers from (approximately) a normal or Gaussian distribution. Such numbers cluster around 0.0 but in principle can be arbitrarily large negative or positive numbers. The probability of picking such outliers is low, but it is certainly possible. (In fact, in applications such as options pricing, it's important to calculate the probability distribution on the selection of such outliers -- Google for order statistics). So, indeed, the random picks that your program will generate can be >= 5. Unless you generate a very large number of samples, you're unlikely to see very many numbers outside the interval [-5,5]. But there can be some. When you create your histogram, we're asking you to account for such outliers in the [-5.0, 4.9) and in the [4.9, 5.0] intervals, respectively.

Assignment 1

User 343 said: The directions ask that the values in the array be outputted on a single line in the command promt, but my outputs (for all my programs) always wrap down the next line in the command prompt once they reach the end of one line. Is there java code that will make it all appear on a single line or would I have to change something in my command prompt settings in some other way?

Professor Sullivan said: See the difference between print and println

computer lab

User 406 said: last time i went to computer lab at Olson i did not have a user name and a password, i signed in with the temporary user name and password that they had. i asked one of the TA s and he told me it should have been emailed to me, but i have not gotten any email concerning this issue.

User 35 said: please wait with patient, we will solve that, thank you.

Lab quiz error

User 188 said:

Hi:
For the most part my code seems to be working, but for some reason whenever I put in an input other than 255 255 255, it comes up with:

cyan = NaN
magenta = NaN
yellow = NaN
black = 1.0

When I put in 255 255 255 everything is 0.0.

Can anyone tell me what NaN is supposed to mean? I can't see what I'm doing wrong!

Thanks,
Becca

Professor Humphreys said: NaN stands for "Not a Number". It's the sort of thing you would get, for example, if you tried to divide by zero. That would be the first thing I would check -- look for anywhere you are doing division and PRINT OUT the denominator. See if any of them are zero or otherwise goofy.

One more question about the intervals

User 161 said:

Are we asked to count and print out the number of "random numbers" falling in each subinterval?

If they are intervals with 0.1 width, then there will be 100 intervals...Then when we output the values in the array, there will be an array with length 100. But why are we asked to print out an array "from index 0 to index n-1"?

Or should we divide [-5,5] to n subintervals?

Thanks!

Professor Sullivan said: There are n numbers in the range 0 ... n-1. Check it out. Suppose n=3. An array , a, with n=3 entries will have elements a[0], a[1], a[2]. See? The indices go from 0 to n-1.

Wrapping

User 343 said:

This forum doesn't let me reply to my own post, but this is a continuation of my previous "Assignment 1" post. I understand the difference between print and println. I'm having the trouble that even when i use "print", my text output from codes still wraps down to the following line in the command prompt. It's all technically the same line, it just gets carried down to the next line when there's no more room. Is this ok (i.e. still considered one line) or is there something i can do with my code to correct this? For example, is there a way to make a horizontal scroll bar appear in the command prompt so that my output will continue forever on one long line if I wish to do so?

Professor Sullivan said: Oh, I see. It is actually one line. There's no embedded "carriage return" or "line feed" characters. The output window, however, does its own thing and wraps long lines. Here's a good way to see what's really going on.

If you usually run your program like this, at the command line

java myprogram

Do this instead

java myprogram > filename

Instead of printing on the console, the program output will be "redirected" into the file given by filename. You can then look at that file using any old text editor, e.g., on Windows, WordPad. In a program like WordPad I believe it'll be clear that it's really just one long line of text.

KS

Lab Quiz Extra Credit Question

User 270 said: Is "the more difficult edge" mentioned in the quiz 255, 255, 255? What kind of output from the program do we need to have for that input?

Professor Sullivan said: First, ask yourself, what color does this represent? What if I add the brightest red, the brightest green, and the brightest blue light together? What do you get? Next, run those numbers through the formulae and see if the result makes sense. Remember, what the CMYK numbers indicate is how much of each color of ink to put on the paper to reproduce the given RGB color. Does the result make any sense? If so, then that's probably not the edge case we're talking about. There are values for which the formulae "blow up." What are they?

Re: Assignment 1:Intervals

User 380 said:

"So, indeed, the random picks that your program will generate can be >= 5. Unless you generate a very large number of samples, you're unlikely to see very many numbers outside the interval [-5,5]. But there can be some. When you create your histogram, we're asking you to account for such outliers in the [-5.0, 4.9) and in the [4.9, 5.0] intervals, respectively."

Does this mean you want the 'outliers' in the first and last indexes of the array? I thought we had to account for the outliers in separate variables 'leftTail' and 'rightTail'.

Professor Sullivan said: Oh good grief. You're right. I should have re-read the assignment. Sorry. Yes, account for the outliers in the outlier variables. That's exactly right. --KS

quiz 1

User 473 said: How to express min{c,m,y} in Java? Thank you!

Lesley said:

Taken from the quiz page:

Hint: Math.min(x, y) returns the minimum of x and y.

The min function will not take 3 arguments, but you can think of a way to use it as it is (taking 2 arguments) to do what you want.

Requirements for Assignment One

User 200 said: I've been reading through posts and the specifics for Assignment 1, and I'm still confused about what the exact requirements are. The requirements specify to output:

• the values in the array, from index 0 to index n-1, all on one line, separated by spaces
  
• the values of the leftTail and rightTail variables, each on its own line
  
• the average of the generated numbers, on its own line

Unless I interpreted these directions incorrectly, I see that we have to output the randomly generated numbers, the number of times it fell outside the [-5.0, 5.0] range, and the average of the numbers, but I don't see anything about outputting the histogram itself. Can someone please clarify?



Lesley said:

"then account for it by incrementing the corresponding value in the array"

Thus, the array keeps track of how many numbers fall in that range, not the actual numbers themselves... From what I understand.

Professor Sullivan said: Lesley's right.

Lab Quiz EC Question

User 270 said: After looking at the formulae for 1 hour, I still don't see what numbers will "blow up" the formulae. Only numbers I could think of are numbers outside of the range 0 to 255. I tried numerous inputs, and outputs came out fine. I still don't see the edge case mentioned in the quiz. Can you tell me what is that edge case?

Professor Sullivan said: What might cause one of the operations to have an undefined result? Work backwards through the formulae to figure out what inputs lead to that result.

Professor Humphreys said: r=0 g=0 b=0

Converting double to string

User 459 said: Is there a way to convert a double, X, into a String with the same name, X?
(i.e. this is done so in some cases X displays a number and in others - a message).
Thanks,
Vladimir


User 14 said: I'm not sure if I quite understand your question, but if you want to convert a double into a string representation of that value, you can use String.valueOf(x) to do it.

Lesley said:

"displays" -> You can print a double without turning it into a String;

double d = 3.14;

System.out.println(d);

is valid code.

You can also concatenate as you wish...

double d = 3.14;

String s = "Pi is approximately " + d;

s now contains "Pi is approximately 3.14"

The only time you would have to worry about this is when you want to use a function that only takes Strings as arguments. Then you need to convert the double to a string. This is most easily done like this:

double d = 3.14;

String s = d + "";

Attempting to do String s = d; is a compile time error.

It is going the other way that is more problematic.

double d = 3.14;

String s = "1";

s += d;

s now contains "13.14" and not 4.14 like you wanted. d += s; is a compile time error. This is why you use Double.parseDouble().

User 167 said: Or, are you trying to use the name of a variable as data? That is to say, you want a way to do: double myDouble = 3.14; System.out.println(); so that it prints the exact string "myDouble"? The short answer is that this is impossible, because Java doesn't support that sort of introspection. The long answer is that A) there might be a way to do it, but I'm sure if it exists it's very obscure and hard to use, and B) in practice you will NEVER need to do this. For example, in the above, you'd be typing out the characters in the name myDouble anyway, so you might as well just type System.out.println("myDouble"). I hope one of the TA responses answers your question.

quiz??

User 448 said: i was trying to figure out the quiz today, and i finished it, but i just realized that out of habit i saved it on my J: drive. I went to delete it, but it was already gone. Im wondering what happened, and how i can get the program i wrote back, without saving it to the J: drive.

Professor Sullivan said: John,

If you wrote the program in Eclipse, with your workspace set to J:\workspace, then your program should be there.

KS

Eclipse Freezing

User 380 said: So I have been talking to some people and we have all noticed that Eclipse tends to "freeze up" periodically while you're in the middle of typing and you have to wait a few seconds for it to unfreeze and catch up with what you typed. Is there any way to fix this?

User 14 said: Joseph,
I believe this is not an Eclipse problem, but a home directory problem. Unfortunately, the home directory is often slow or unresponsive. But it is the best place for you to keep your work, so for this semester all we can do is deal with it and hope it doesn't become more than a small annoyance.

practicing the quiz

User 221 said:

For this lab quiz and any other quiz, I understand that its ok to work them out ahead of time. However, are we required to delete them from our workspace before we go into the quiz or are we on our honor to not look at our practice java program while we do the quiz? Just making sure :-)

Lesley said:

You're on your honor - just like when a professor gives a take home, closed notes quiz: the notes are in your room (you don't have to burn them or something else ridiculous), but you're on your honor not to look at them while taking the test, and your pledge and signature signifies you did not cheat.

You should not even mount your J: drive during lab quizzes, either.

RoboGrade for Color Lab?

User 268 said: Do we get to have the automatic grader check our color program?

I just want to be sure I have exactly the right output.


Professor Humphreys said: No, we are going to robograde after all the labs are in. You should match the format of the sample output given on the web page.

RoboGrade of Group Work

User 331 said: Is robograde disabled for anyone else besides me? Just wanted to see what was going on.

Thanks,
Daniel McCarron


Professor Sullivan said: Yes. The grading robot is broken. Sorry about that. Will be back ASAP. --KS

I need help with loops

User 440 said: I am having trouble with the HelloFor and HelloWhile problems. I know its group work but I don't understand how to make it say hello world n times. I don't know how to define that.



Matt Cassi


Professor Sullivan said: Matt,

Study carefully the concept of the for loop. The basic idea is this: whenever you need to make something happen some number of times (let's say n times), then you use a for loop. In particular, you do the following:

• initialize to zero a variable (often simply called i or j) that will serve to count the number of times that you've already executed the loop body;
  
• at the top of the loop you use a a boolean expression involving the count variable (e.g., i or j) to see whether you've already executed the loop enough times; and each time after you execute the loop you increment the value of the counter (i or j) to properly reflect the number of times you've been through the loop.
• Inside the loop body you take the required action (e.g., printing Hello)

If you're having trouble we strongly recommend that you download, run, and experiment with some of the code examples in the book. You should (all) be doing that, anyway.

Ok?

Kevin Sullivan


Professor Sullivan said: By the way, we also strongly recommend that everyone use groups to work out these sorts of questions. In computing, there's often nothing so useful as having someone to talk to about a particular issue. It's really almost miraculous. I have often struggled for hours, sometimes days, to figure out what's wrong with a program. As soon as I ask someone to come have a look, voila, before they say a word, I see the problem. I'm not sure why it is this way, but that's very much my experience. I guess we think differently sometimes when we're with someone. It can really help. Use your groups. Explain things to each other: the problems you're having, and what you know about how to solve problems. Go beyond the work we've assigned. Do little programming experiments together. Have fun.

Waitlist Protocol for Stacks Office Hours

Professor Sullivan said: Hi Folks,

In the past, once class gets underway, the demand for help from TA's tends to get a little out of hand sometimes. In these circumstances it's important that we serve students in a fair way. So, the best thing to do is to have a signup list in the stacks, and to have TA's serve students in the order in which they've signed up for help. It's best to limit the attention give to any one student to 5 minutes or so, if there's a waiting line.

Kevin

Fatal Grading Error on quiz!

User 54 said: Hi, When I checked my grade for my Lab quiz today I recieved an error message. The message said, "FATAL GRADING SCRIPT ERROR, PLEASE REPORT THIS IMMEDIATELY AND WE WILL FIX IT." If you could please help me with this it would be greatly appreciated.

Professor Humphreys said: sorry, I'm in the middle of doing the lab grading, those things will be in flux. Very sorry about the panic.

fatal grading script error

User 91 said: my quiz also has an error

Professor Humphreys said: this should be fixed very soon, if not already.

Problems with the quiz grade

User 337 said: When I finished my quiz, I checked it on command prompt and had the correct answers, but the robograde gave me a zero out of 105 points. I'm really confused and frustrated. Can you tell me whats going on?

Professor Humphreys said: Robograde was being a bit temperamental.

Reading the results now, you definitely don't have a zero.

float versus double

User 236 said: So, I have an interesting question. In the lab quiz 1 document, it says such things as "This scheme represents a single color as a four-tuple, (c,y,m,k), of floating point values between 0.0 and 1.0" and "mixed integer and floating point expressions."

When I was coding my program for practice and when I went into lab to code, I assigned type float to the variables. I know good and well of type double. In fact, I ran and compiled the program using double or float for the type. However, the reason why I used type float for this quiz was that is what seemed to be indicated in the quiz document (see excerpts in first paragraph), which is deceiving. And, yes I realized that type float did yield accuracy out to a certain decimal place. Given this, I found the use of type float (given the last sentence) justified in the first paragraph of this post and problem 20 of the group work.
In order to make the lab quiz document clearer, it would have behooved him to indicate a type of 64-bit or 32-bit.
Given this, and given that everything in my program was correct (even the extra credit) , I feel that the following is not justified:

"Testing the command 'java RGBtoCMYK 75 0 130'...
Your program produced the following output:

Line 0: cyan = 0.42307687
Line 1: magenta = 1.0
Line 2: yellow = 0.0
Line 3: black = 0.49019605

Running our sample solution with the same command line...
Testing the command 'java RGBtoCMYK 75 0 130'...
Our solution produced the following output:

Line 0: cyan = 0.423076 9230769229
Line 1: magenta = 1.0
Line 2: yellow = 0.0
Line 3: black = 0.4901960 784313726"

I received an 85/105 on this assignment. Twenty points were subtracted from my grade because of the "fuzziness" of specifying double and float in the lab document. I feel that if I am not able to get full compensation, then I should at most get 7-10 points off my grade (especially since they are correct to a specific decimal place: 7th for black and 6th for cyan).

And how my grade for this stands now: it is 85/105
Shouldn't it be 85/100?

Thank You
Best Regards,
Brandon




User 181 said: Any one knows the answer? I have no idea of this.

Professor Sullivan said: Brandon, we agree that the specification was ambiguous in terms of the issue that you have identified. You analysis is spot-on. We will give you full credit. What's most important are the technical lessons:

• it's really easy to write ambiguous specifications
• ambiguous specifications can lead to all kinds of problems/disagreements
• tests can be important in disambiguating natural language specifications
  

Assignment 1

User 513 said: For the StdGaussianHistogram, are we suppose to output the number of values that fall into each subinterval or are we outputting the values themselves?

Lesley said: #208

How Precisely do We Subdivide Intervals in Assmt #1

User 86 said:

When we subdivide the interval [-5,5], which values are included in the sub-intervals. For example, is the first sub-interal supposed to be [-5,-4.9) or (-5,-4.9] ?

(Edited by Kevin Sullivan - original submission Friday, 25 January 2008, 11:24 PM)

Lesley said: #194

Output of No.2 assignment 1

User 254 said: "the values in the array, from index 0 to index n-1, all on one line, separated by spaces"
What is the space requirement of the last number? can I put a space after it?

"the values of the leftTail and rightTail variables, each on its own line"
do we need to type "leftTail" or "rightTail" before the value?



Professor Sullivan said: We'll accept a trailing space.

If we do not specify that text should be printed then it should not be printed.

KS

Double inaccuracy

User 449 said: Hey,

What is one way to deal with the problem of Java returning values as exampled in a partial program below?:

double x = 0;
for (int i=0; i<5; ++i)
{
x = x + .1;
System.out.println(x);
}

Output:
.1
.2
.30000000000000004
.4
.5

The third line gives me pause. I realize it is perhaps a trivial inaccuracy for the programs we are writing, but I would like to know how to correct it nonetheless.

As a temporary solution, I have modified the example program to:

double x = 0;
for (int i=0; i<5; ++i)
{
x = x + .1;
x = (int) Math.round(x*10);
x = x/10;
System.out.println(x);
}

...which fixes the problem in this particular case, but is there a more general solution?

-Joel



Professor Humphreys said: Fantastic question. Unfortunately, there is no general solution. When we come to the portion of the course on scientific computing, it will be more clear why. For now, remember that the "double" type is trying to use 64 bits to SOMEHOW represent a real number. Therefore there are at MOST 2^64 different possible real numbers that a "double" can represent precisely. All other real numbers are just approximations. Yes, this introduces error. Yes, this error can propagate through a calculation and be multiplied by other such errors, making the problem worse. Yes, people have lost their lives because of this.

Lesley said:

You might also find this interesting:

http://mindprod.com/jgloss/floatingpoint.html

Professor Sullivan said: Yup: Great example illustrating the point that I made in my class the other day: NEVER FORGET: Computer arithmetic is not the arithmetic that you learned in school. It's close, in many circumstances, but there are many subtleties.

Assigning Sub-intervals to Elements

User 236 said: I know that the process of dividing the interval into sub intervals is conceptual, but how do I go about assigning these subintervals to the actual elements?

Professor Sullivan said: The interval from -5 to 5 is to be divided into 100 sub-intervals. If you generate the random number -4.99, for example, that would have be counted in the first sub-interval (i.e., zero'th array element). If you generate the number 4.99, that would have to be counted in the 100'th sub-interval (what's the array index?). What you need is a formula for figuring out which array index to use for any give number in the range [-5,5]. Just work it out. This kind of problem solving is a big part of programming: it's not so much about how to express something Java as it is figuring out precisely what to express! That's part of the job.

Question on Part 2, Assignment 1

User 52 said: it seems like so many people are struggling with this, and so am i.

I am wondering, if this assignment is possible with all the knowledge we had been given (from lecture, group assignment #1, lab, and lab quiz). Or if not, which parts of our book should we read, in preparation for this.

I've been trying to figure out this histogram part for the longest time. I still have not gotten anywhere.


Professor Humphreys said: Everyone should have read sections 1.1-1.4 in the book. Everything you have done so far is definitely adequate to do this assignment. If you're struggling, ask for help!

Professor Sullivan said: The group work is intended to ensure that you've practiced with the basic building blocks needed to do the individual assignments. If you haven't finished the group work, start there. There are some problems there intended to give you practice with arrays. We will be covering arrays in lecture shortly. In the meantime, read about them, do simple exercises before trying harder ones, and build up slowly to the hard problems. The only tricky part in the assignment is to derive a formula for the array index to use to account for a give random pick.

RoboGrade for Assignment 1?

User 270 said: I don't see the RoboGrade button on the Assignment 1 page. Does that mean we can't RoboGrade for Assignment 1?

User 181 said: Kevin or Greg,

This question has not been answered yet. So do you know what the problem is?


Professor Humphreys said: Only group works (and possibly lab quizzes) will be robograded.

Nit-picky question

User 449 said: Hey,

I'm pretty sure the answer to my question is obviously implied, but I'm trying to be very exact, so I thought I'd ask (in reference to Part 2 of Individual Assignment #1):

Quote 1: "...produces a histogram of n pseudo-random values..."
Quote 2: "[Output] the values in the array, from index 0 to index n-1..."

Am I correct to assume that the two n's are unrelated?

-Joel


User 167 said: Yes. In the second quote, it refers to the number of elements in the array, which is different from the n used earlier.

Professor Sullivan said: It's very strange: I'm certain that I corrected this issue on the web site last week. I just did it again. Generate m random numbers and account for them in n intervals.

can't compile

User 71 said: I can still run java program in cmd, but when I type in "javac" it says it is not either a internal or external command, what can I do?

User 167 said: Most likely, you don't have java installed properly. See $@FORUMVIEWBYID*12@$ for instructions for setting up java on your machine, and be sure you check your PATH variable.

stdlib

User 449 said: I downloaded the stdlib file to my home directory on J:\ and followed the instructions for including it in a Java project. Here is what happened:

- Eclipse recognized the new library and no longer underlined the statement StdOut.println() in red.
- When compiling the program from cmd, however, I was greeted with errors telling me:

symbol: variable StdOut
location: class HelloWorld
StdOut.print("HelloWorld");
^

Any suggestions?

-Joel

Individual Assigment 1 question

User 196 said:

Hey,

I uploaded the first part of the individaul assigment 1 to robograde it and see if what I did was correct except I do not see a place to robograde it...am i looking in the worng place or is there no Robograde for this part?. If there is not is ther anyway I can take off my first upload and take a better look at it so that I can make sure it was right?...If we cannot robograde it, then I need to get it taken off because I did not know you can only submit it once. Can anyone help?

Thanks

Nathan Wethington

Professor Humphreys said: You can submit as many times as you want. Individual assignments are not auto-graded.

standard library

User 90 said: Every time I try to download the standard library I get an error message saying: Failed to load Main-Class manifest attribute from C:\Users\Owner\Desktop\stdlib.jar. What is the problem?

Professor Sullivan said: It looks like you're trying to "run" the jar file.

In fact, it's possible to package up a Java program as a runnable jar file, but that's not what we're trying to do here.

You need to save the file to a location on disk and then include it as a library, following the directions on the class web site.

KS

stdlib problems

User 77 said: I was having the same problem as Joel Gillespie, as he explained in his question labelled "stdlib". First, don't you have to mention the stdlib.jar file while compiling the program?, using in the cmd:
javac -cp j:/stdlib.jar HelloWorld.java


I tried this and it compiled the program without complaining, but then when I tried in cmd: java -cp j:/stdlib.jar HelloWorld
I received this message:
Exception in thread "main" java.lang.NoClassDefFoundError: HelloWorld
Caused by: java.lang.ClassNotFoundException: HelloWorld
at java.net.URLClassLoader$1.run(Unknown Source)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
at sun.misc.Launcher$AppClassLoader.loadClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
at java.lang.ClassLoader.loadClassInternal(Unknown Source)


I added stdlib.jar as an external jar to the java project, and it shows up in eclipse in referenced libraries. Did I miss something?


Professor Sullivan said: You should include the "-cp" thing when compiling. Once the code is compiled, you don't need to specify the jar file.

KS

can't compile anymore

User 71 said: My java was running fine, but today when I type in "javac" it says it is not either a internal or external command, I can still run java program in cmd with"java filename", and I've reinstalled the java, check the path, it still doesn't work, what else can I do?

Professor Sullivan said: Double check your path. That's almost certainly where the problem is. If that doesn't help, see a TA during office hours tomorrow night. --KS

Lesley said: I would suggest bringing your laptop in during TA hours sometime this week then.

Groupwork #18

User 90 said: I'm confused about how our output should look. If the input is 3 should the output be something such as : 0 1 2?

Professor Sullivan said: Almost right. The problem says in reverse order. --KS

ASSIGNMENT 1!!

User 253 said: I just submitted my standardgaussian.java and it appears there is no Robograde. I am not sure if I want to submit it then, but it will not let you unsubmit it even though it says we can submit it multiple times. Help!

Professor Humphreys said: You can submit as many times as you want, just resubmit the file.

How do I use StdLib at Command Line (was "stdlib problems continued")

User 77 said: As I mentioned in stdlib problems, stdlib was not working for me after compiling a program, as cmd would give me an error message when trying to run the program. I tried, after compiling the program with the -cp stuff, and then the regular java HelloWorld, as recommended, and I essentially get the same error message, except the first few lines are different and read:

Exception in thread "main" java.lang.NoClassDefFoundError: StdOut
at HelloWorld.main(HelloWorld.java:4)
Caused by: java.lang.ClassNotFoundException: StdOut

(and it continues with the rest of the error messages as before) What is going on?

(Edited by Kevin Sullivan - original submission Sunday, 27 January 2008, 12:23 AM)

Lesley said:

please try running your program as:

javac -cp j:\workspace\stdlib.jar World.java

java -cp .;j:\workspace\stdlib.jar World

Assignment 1 - Edited?

User 52 said: I just recently looked at the Assignment #1 and it seems to have been edited? Or am I wrong? :

2. Write a Java program called StdGaussianHistogram.java that produces a histogram of m pseudo-random values using the Box-Muller formula, where the value m is given as a command-line argument. Do this in the following way. First, divide the interval [-5.0, 5.0] into n sub-intervals of width 0.1.

it says... "m" pseudo-random values...while it says "n" sub-intervals of width 0.1.
If its going from -5.0 to 5.0..then..shouldn't there be 100 sub-intervals...instead of..n sub-intervals. I'm just confused by all the variables now.

I looked at it every day for the past ~5 days or so, and it seemed to have made sense, I'm just confused now, because of everything I've read from the forums, and from reading the assignment over and over again. Has the assignment recently been changed?


Professor Sullivan said: Hi Jae,

Yes we tried to clarify it. Here's the deal:

You generate m random numbers, where the value m is given as a command line argument. For example, when you run your program, you could define m to be 1 million. Your program would thus generate m=1000000 numbers, each on picked at random from a standard normal distribution. I can tell you that it very unlikely that you'll pick more than a very small number that are outside the range [-5,5].

Each time you generate a number, we want you to count it as falling into one of n bins. The bins are intervals of size 0.1 between -5 and 5. You can easily figure out how many there are. This is the value of n -- the number of bins.

KS

assignment 1

User 91 said: should the output for assignment 1, part 1 have any particular format? For instance should the output line read only the randomly generated number or should it also include "z = "?

Professor Sullivan said: Hi Kara,

Again, if we do not specify that you should add text, then you should not. In this case, we just want to see the numerical output, exactly as specified. Thanks!

KS

Group Assignment # 18

Professor Sullivan said: It's impossible to say what's causing this behavior without looking at your program, except that it's safe to say your program is doing exactly what it's programmed to do!

One of the most important of all skills for a program designer / implementer is the ability to mentally simulate the execution of a program. What I suggest is that you take a pencil and paper, start at the beginning of your program, and trace its execution statement by statement. If you do this you are very likely to discover with it is about the program you've written that's causing the behavior that you're seeing.

KS

Why is 1/0 undefined while 1.0/0.0 is legal in java?

User 49 said: Why is 1/0 undefined while 1.0/0.0 is legal in java? Anyone knows the answer? Thanks

Professor Humphreys said: That is a very strange asymmetry. 1.0/0.0 gives you the value "Infinity", which is not terribly likely to be useful in subsequent calculations, but 1/0 does give a runtime error. I had previously thought (wrongly) that it gave NaN. The answer to "why" is almost surely "the designers of Java made a choice".

Professor Sullivan said: Virtually all modern computer programming languages and computer hardware implementations implement part or all of the so-called IEEE 754 specification the behavior of floating point (but not integer) arithmetic. Java is among the many languages that are designed to conform to this standard. The standard requires the behavior that you see for floating point operations. Rather than exceptions, this standard uses special values, Inf and NaN, to indicate pathological results. The designers of Java chose to make integer division raise an exception.

Ok, everyone, let's recite the mantra again: computer arithmetic is weird -- it's not what you learned in school.

You can find more about Java's implementation of IEEE 754 here.

For those of you who read this far, here's some grist for a possible extra credit question on some test. Is it possible for a floating point number to be neither greater nor less than or equal to 0.0? What floating point number has this bizarre property (if there is one)?

And a final bit of trivia: Before IEEE 754, calculators and computers tended to produce vastly different results for the same floating point computations. That was a real problem. Kahan won the Turing Award (Nobel Prize" of computer science) for his work developing IEEE 754.


User 167 said: Also, integer division can't return infinity or NaN because the specification of an integer doesn't provide a way to represent either of these values, whereas the IEEE specification for floats does.

Assignment 1

User 268 said: Is my StdGaussian.java supposed to yield any values above or below the interval [-5.0, 5.0]?

I never seem to get any values outside that interval even after 100 numbers are generated.


User 14 said: James,

You're right. You should very rarely get numbers outside this range because you are simulating a normal distribution with a standard deviation of 1. Look at your numbers close to the edges and you will probably even have all zeros outside of -4 to 4. About 99.9999% of the numbers you generate should lie between -5 and 5. For more information, here is a relevant wikipedia page: http://en.wikipedia.org/wiki/Normal_distribution#Standard_deviation_and_confidence_intervals

Professor Sullivan said: Yup, Eric's right. I generally don't see numbers outside of [-5,5] unless I generate a million or more samples. A normal distribution has this nice property: anything is possible but most everything is close to average!

Question about the while command

User 33 said: So, when coding a while loop in java (like, say while c !=2), does java quit out of the while loop as soon as c = 2, or does it have to go to the end of the set of commands in the {brackets} before it ends the loop?

Thanks in advance


User 14 said: Java will run to the end of the curly braces {} before checking the condition again. So yes, it has to go to the end of the set of commands in the {brackets} before it ends the loop.

Professor Sullivan said: Right. See the graphical representation of the while loop in the lecture notes.

Assignment 1 - Command Line Argument

User 52 said: Hey

I got mine to work, but every once in a while, it says I have an error, but I just type the same exact command line argument again, and it works. Is there a reason? (I'm afraid the same might happen while its being graded, therefore I'll lose points). I'm 99% sure my code is right, but I don't know why it has those failures from time to time.

And also, there is an obvious limit on how high your command line argument can be, but I've compared with someone who already finished it as well, but that person's code could allow much higher command line arguments than I could.

What could be the problem? And would it cost me a lot during the grading process?


Professor Sullivan said: Remember that you're generating random numbers. So certain "bad" things might happen in one run of your program that don't happen in other runs. It's often the "edge" cases that are problematical. I suggest that you mentally simulate a run of your program to mentally test the "edge" cases -- e.g.,. where the generated numbers are right up against the edge of the [-5,5] interval.

Kevin Sullivan

Histogram Question

User 268 said: When placing the 100 randomly generated numbers do I just put them in a one dimensional array in the order they are generated? Or do I have to put them in the correct sub-intervals of 0.1 thus possibly having multiple values for the same interval? The word histogram scares me.

Professor Sullivan said: The program is to generate random numbers and count the number of generated numbers that fall into each subinterval.

Assignment 1 output

User 200 said: For the histogram assignment, does it matter whether or not there's a space after the last number of the histogram, or is the grading program going to be that picky?

Lesley said: I don't believe the assignments are graded with Robograder (think about the implications of saying your random number program could be graded by a robograder), but it's only one line of code to make it not print a space after the last number...

Professor Sullivan said: Take a few minutes and see if you can figure out what it would take to do it "right," i.e., with spaces between elements but not before the first one or after the last one.

Here's a hint:

If you repeat the following, it's not right (you get a space before):

Nor is it right to repeat the following pattern (you get a space after):

Assuming that you have at least one element to print, what pattern works to get you what you want? Working through such problems builds intellectual capacity for computational problem solving!

Robograde on random numbers

User 200 said: I was just curious as to how RoboGrade grades programs that have to produce random numbers. Since each set of random numbers will always be different, it can't compare the output to that of another program. Does Robograde just look at the output of the submitted program to see if it makes general sense, or can it actually examine the program's code and see if there's any possibility for logic or run-time errors?

For example, with the histogram assignment, it's very unlikely to produce any values outside of the [-5.0, 5.0] range, but it IS possible. Hypothetically then, someone could just be lazy with their program and decide to leave the leftTail and rightTail variables at 0 without actually checking for outliers. How does RoboGrade detect a situation like that?


Professor Sullivan said: Good question. How do you test programs that by their very definition have non-deterministic behaviors? The real answer is this: testing is just one of quite a few ways to determine whether a program satisfies its specification or not. Other techniques include human inspection of code and use of mathematical proof techniques (where you literally formulate a putative theorem about the behavior of a program and then attempt to prove the theorem based on axioms about how various programming constructs work). In this case we'll combine testing (robograde) and human inspection (using TAs/graders).


Intentional attempts to "fool" our robograder will result in negative increments to grades. (For example, it would not be a good idea to implement a program that simply prints out a long list of numbers, hoping that they'll pass our tests. We're smarter than that -- we hope -- and would in such an instance assign a negative grade :-)

Robograde for Groupwork #17

User 347 said:

When Robograde tests our program, the outputs match perfectly, but it is not awarding us the five points.

Lesley said:

If it is not awarding you 5 points, then your outputs do not match.

"in ascending order with n numbers on each line and with single spaces between numbers on a given line. Do not include a space after the last number on any line."

I would suggest making sure you've fulfilled that last requirement.

Professor Sullivan said: Double check as Lesley has indicated. If you still disagree, submit a regrade request per our regrade policy. Thanks!

Running programs within eclipse

User 68 said: Running programs from the command prompt works perfectly fine. Whenever I try to run a program from within eclipse, no matter what program I try, eclipse returns the following message: Exception in thread "main" java.lang.UnsupportedClassVersionError: Bad version number in .class file at java.lang.ClassLoader.defineClass1(Native Method) at java.lang.ClassLoader.defineClass(ClassLoader.java:620) at java.security.SecureClassLoader.defineClass(SecureClassLoader.java:124) at java.net.URLClassLoader.defineClass(URLClassLoader.java:260) at java.net.URLClassLoader.access$100(URLClassLoader.java:56) at java.net.URLClassLoader$1.run(URLClassLoader.java:195) at java.security.AccessController.doPrivileged(Native Method) at java.net.URLClassLoader.findClass(URLClassLoader.java:188) at java.lang.ClassLoader.loadClass(ClassLoader.java:306) at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:268) at java.lang.ClassLoader.loadClass(ClassLoader.java:251) at java.lang.ClassLoader.loadClassInternal(ClassLoader.java:319)

Professor Sullivan said: You've got a configuration error in your Eclipse and/or Java installation. The problem, technically, is that you're compiling with a late version of Java but trying to run the file on a Java Virtual Machine for an earlier version of Java. What computer are you running on? Did you follow our directions for installing Java and then Eclipse?

KS

Error Error

User 236 said: If I am on my account and I go to grades up at the upper right hand corner, and then scroll down to my groupwork grades, in the columns for "grade" and "percent" it has the word "Error" in them. But it has my group's actual grade present in the Bold lettering.

Professor Humphreys said: I logged in as you, clicked on "grades", and it looks like I would expect; no "error" anywhere.

Opening an in-progess project

User 88 said: While working on a project (any of them) I can compile and run it within Eclipse without problems. However, when I close Eclipse and then come back to work on it later and try to run it I get the message "URLClassLoader$1.run()line: not available" even though I see all the code the way it was before.



User 35 said: Actually, I think you'd better first reinstall the Eclipse and Java.

Assignment 1

User 120 said: Is this assignment meant to be really hard because I can't get number 2 at all and I have spent many days trying to figure it out. I read the book, went over my notes, looked up terms on wikipedia, and I'm still not getting it. I've never done programming before and I was under the impression that we didn't need prior experience to be in this class. Is there anything else I can do? Thank you.

Lesley said:

Here is what I have been telling people at stacks hours:

Try to do the program like this, to get started:

Make it generate ONE random number. This part you already did in part 1 of the assignment.

Now make an array that covers subintervals, but make them of width = 1. So these intervals are [-5.0, -4.0) [-4.0, -3.0)... [3.0, 4.0) [4.0, 5.0]. This array will be an array of ints (like the assignment specifies) and what it is doing is literally acting like some boxes. Every time you get a random number, you want to put it in the box of the interval it falls into - but the thing the array keeps track of are how MANY numbers you put into each box. If your random number were 4.3, it would go in the [4.0, 5.0] box. If it were -1.2, it would go in the [-2.0, -1.0) box. But how do you figure this out? Well, 4.0 < 4.3 < 5.0, and -2.0 < -1.2 < -1.0 and how does this translate into computer code?? That's a big part of the assignment to figure out.

See if you can get the computer to figure out which interval your ONE random number goes in, and print out the resulting histogram: it will be a bunch of 0's and one 1 for the bin you have put the number in. Now see if you can make it keep track of 2 numbers, then m numbers. You also need to remember that the end result needs an interval width of .1, not 1.

The TAs have office hours tomorrow should you get lost and need more help.

Professor Sullivan said: Hi Boxi,

First, thanks for asking. We appreciate that. There are surely others "in the same boat." We want to make sure that everyone succeeds. So we do want to know when people are having trouble. For some help, please see the note that I just posted on the main web site. (Click to download the PDF file.) Read it over and please let me know if you need more help.

Kevin Sullivan

Using values defined in a for loop, outside the loop

User 427 said: Is it possible to use a variable defined in a for loop, outside to the loop. Because my program does not recognize them. If not, what would be the best way for me to print the values determined in the for loop, outside the loop?

Professor Sullivan said: Great, great question.

You can use variables only within the code blocks in which they are declared. So, for example, if you say int i within a for loop, then you can't use that i outside of the loop. You can always declare a variable outside of a code block in which a value for it is calculated and assigned, and in that case, the variable will be in scope outside of that block.

KS

Formatting Errors

User 560 said: on the groupwork assignment, I have submitted everything and robograde has awarded me 5 points for each but took off 10 for "formatting errors". can anyone explain?

Question about Individual problem

User 189 said: For problem 2 if it's ok that I create a 102-slot array, say a[102], and set a[0] as the lefttail and a[101] as the right tail. I think this will work but i wonder if it fulfill the requirement since the instruction says "from index 0 to index n-1" for the first line.

Thank you


Professor Sullivan said: Shanshan,

In practice what you propose would be a fine solution, because a specification would generally leave the choice of implementation variables and strategies up to you, the program designer. However, the specification here is very clear. It tells you not only to use two separate tail variables, but even what to call them. The general rule is that you must satisfy the specification. You can argue about whether a specification is a good specification, but whatever it ends up being, that's what your program must satisfy.

KS

problem with results

User 440 said: I can only get my program to use up to m = 100 and nothing more. I'm not sure where I went wrong.

Professor Sullivan said: Matt,

I have two pieces of advice:

(1) Make sure you understand exactly what the program is required to do. For example, without thinking about programming at all, simulate in your mind the process of generating a large number of random numbers and counting how many fall in each of a set of intervals.

(2) Make the effort to understand precisely what your program actually does. There are several ways to do this:

• mentally simulate/trace its execution
• use the debugger to watch it run, line by line
• add code to your program to print out certain intermediate results

KS

Average for part 2?

User 331 said: In part 2 of assingment one it say "the average of the generated numbers, on its own line"

Does it want the average of the z ( random generated number) or does it want the average of the data in the histogram( 0 0 1 2 0 3 2 0)?

Thanks


Professor Sullivan said: The average of the generated random numbers (the z's).

individual assignment part 2 output

User 543 said: I had a question about the output for the individual assignment part 2. For the left tail, right tail, and average, do you want the output to say like "left tail: 0" or do you just want the number(s) alone? And also, the directions say that the
left tail, right tail should be one their own separate line. So I put an
System.out.println(). before printing out the left tail, and when my command
line argument is 1000. it prints fine, but when it's like 100, theres an
extra line between the left tail and they regular array output. Does this extra line matter? Will I get points taken off for it?

Professor Sullivan said: Hi. We continue to receive questions of this form. There are several of them with answers already posted on this help forum. It's a good practice to read over the questions that have already been asked and answered.

As we've said before, yes, your output must conform to the specification. If we do not specify that certain text elements should be included in your output (such as "left tail:") then they should not be included. Blank lines are not specified and so should not be included.

Note: System.out.println(...) can take arguments just like System.out.print(...). They have identical behaviors except that the former includes a carriage return at the end of the specified output (whether an empty blank line or not), and System.out.print does not include a carriage return at the end of the output.

KS

How Do I Use StdLib at Command Line (was "Trouble with Stdlib")

User 200 said: I downloaded stdlib.jar off the website and added it as an additional library on Eclipse. I used StdOut on my Hello World program just to make sure the library was added correctly, and it worked fine when I ran it on Eclipse. However, the command prompt won't recognize the Stdlib methods. How do I fix that?

(Edited by Kevin Sullivan - original submission Wednesday, 30 January 2008, 12:55 PM)

Professor Humphreys said: #267

Failed to load Main-Class manifest? (Was How to download stdlib.jar)

User 309 said: When I try to download the stdlib.jar, I get an error that says "Failed to load Main-Class manifest attribute from ...(file name)." Am I downloading incorrectly?

(Edited by Kevin Sullivan - original submission Wednesday, 30 January 2008, 01:36 PM)

Professor Sullivan said: Don't try to run the program. Just save it to your disk.

RoboGrade? Assignment 1

User 136 said: I just submitted both of my programs for Assignment 1, are we not using Robograde? Also it won't let me unsubmit...

Histogram with DrawDistribution

User 200 said: I'm getting an error when I try to compile my StdGaussianHistogram program with the DrawDistribution line of code at the end. Like the directions said, I saved DrawDistribution as a .java file to the src folder of my Histogram project, I downloaded, configured, and tested the stdlib.jar file, so I know that it works.

The error I get when I compile StdGaussianHistogram.java is this:

"cannot find symbol:
symbol: variable DrawDistribution
location: class StdGaussianHistogram"

I guess the problem is that my computer thinks that DrawDistribution is supposed to be a variable instead of an actual class. How do I fix this?


Professor Sullivan said: If you're running your program from within Eclipse (as you should be), the file needs to appear in the "src" folder of the Eclipse project. To get it there is as simple as dragging it from a Windows explorer window and dropping it into the "src" folder.

I'm sorry but I can't trace variables in Debug menu

User 71 said: Things turned out fine in the Console, but nothing showed up in the Variables window, am I doing anything wrong?


Professor Sullivan said: There's not enough information here to provide a good answer. Please see a TA for additional help.

Compiling Lab Study #1 error

User 215 said:

I get this error when trying to compile Lab Study #1:

StdGaussianHistogram.java:73: cannot find symbol

symbol : variable DrawDistribution

location: class StdGaussianHistogram

DrawDistribution.drawMyHistogramAsProbabilityDistribution(histogram,-5, 5, leftTail, rightTail);

And there's a little arrow underneath the "a" in "leftTail".

Any ideas on how to fix this?

Another Lab Study #1 error

User 249 said:

Here is another confused person adding to this forum. I have stdlib.jar and DrawDistribution where they should be, and am trying to compile the program with -cp and everything but I get the following error...

StdGaussianHistogram.java:448: cannot find symbol

symbol: variableDrawDistribution

location: class StdGaussianHistogram

And then it has the line of code I was told to download with the carrot under the first D. None of the other solutions on this forum work, and frustration is setting in.

HELP!

Professor Sullivan said: Sean,

Please try to compile and run the program from within Eclipse. Have you done that?

KS

I need help

User 368 said: I was wondering if any TA could possible meet with me or let me know when their office hours are to work out some kinks in my computer. A lot of things are concerning eclipse are not working correctly and I think I have made it worse trying to fix it myself. Thanks in advance

Lesley said:

TA's have official office hours 6-10pm Sun, Mon, Tues.

If something is due before then, consider going down to Olsson 001 and doing the work on one of those computers with your login that was emailed to you.

There are labs all day tomorrow so if you bring your computer and ask a TA nicely they might be able to debug your computer for you before or after the lab section they teach (times on COD). Consider offering candy as a thank you.

You could also ask another student to help you fix your computer (that would not be a violation of the honor code) if you know someone in 101 or 101E or another CS class that is familiar with Eclipse. Again, consider bribery.

Lab Study #1- command line

User 200 said: Judging by the responses to questions about this assignment, I would say it's safe to assume that we need to be able to run the revised Histogram program only through Eclipse and not necessarily through the command prompt. That being the case, what should we do about the command line argument (where we input "n" for the number of random numbers)? Do we leave the command line input statement as is (because I don't know how to run the program through Eclipse with that statement in there), do we need to change it to use something from the stdlib.jar library, or do we just set a fixed value for n?

Professor Humphreys said: Please read this post to see how to run from the command line: #267

Professor Sullivan said: Kiran,

How do you run a program that uses the standard libraries?

Eclipse

If you want to run your program entirely in Eclipse, you have to specify your command line arguments in Eclipse and you have to have added stdlib.jar as an external jar file, as covered in an earlier posting.

Command Line

The key here is that stdlib.java needs to be on your CLASSPATH. The -cp <file> argument to the Java compiler can be used here, where <file> is a path to a .jar file. So the command javac -cp .;<path> DrawDistribution.java will compile DrawDistribution.java and will access stdlib.jar to resolve symbols not defined within the java file itself. Again, <path> is a path to the jar file.

You can add that "-cp ..." stuff every time you type javac if you want, but there's a much more convenient approach. If you're on Windows, here's what we strongly recommend. Create or edit a CLASSPATH system variable, in just the same way that you had to set a PATH variable so that windows could find the javac and java commands. Here's how.

• Right click on My Computer, then click Advanced. The click Environment Variables. In the system variables area look for CLASSPATH. If it's there, edit it and add to the end the following;
  
  ;<path-of-your-jar-file>
  
  Here the semicolon separates the new part of the CLASSPATH from what was there before. Be very careful not to remove or change what was there before.
• If there is no CLASSPATH variable, create one and edit it giving it the following value
  .;<path-of-jar-file>
  The period before the semicolon means search the current directory in which the compiler is running before searching files on the rest of the classpath. My CLASSPATH looks like this:
  .;C:\Program Files\Java\jre1.6.0_04\lib\ext\QTJava.zip;C:\Kevin\workspace\stdlib.jar
  The material before ;C:\Kevin\workspace\stdlib.jar was already there. I just added that part to include the jar file in the set of files that will be searched for symbols that are undefined in the Java files you're actually compiling.
  

I hope this helps. If not, let me know, and we'll make it work for you.

Kevin Sullivan

Command Line Compiling Follow-up

User 449 said: I have successfully mastered the ability to compile from the command line using -cp .; etc., but this is terribly inconvenient. I followed the instructions on creating a CLASSPATH variable, but I am still receiving errors. Here is some background:

- stdlib.jar is located in j:\workspace
- I have edited the Java build path for my histogram project to include stdlib.jar
- Under system variables->CLASSPATH, I have: .;C:\Program Files\Java\jre1.5.0_06\lib\ext\QTJava.zip;j:\workspace\stdlib.jar
- It compiles fine, but when I try to run it, I get:

Exception in thread "main" java.lang.NoClassDefFoundError: StdDraw
at DrawDistribution.initialize(DrawDistribution.java:43)
...
Caused by: java.lang.ClassNotFoundException: StdDraw

The program works fine when I do the -cp thing, and I get the graph and everything, so I don't know what the problem is.

-Joel


Lesley said:

You can always write yourself a batch program to do it for you.

put this in a file called 'run.bat' in the directory where your .java file is - you can do this by saving it as run.txt and then renaming it should Windows not want to let you edit a file ending in .bat

javac -cp j:\workspace\stdlib.jar %1.java
java -cp .;j:\workspace\stdlib.jar %1

then run it as 'run World' if your file were World.java for example

Should you want to avoid copying this file a bunch of times, you can reference it as 'j:\workspace\run.bat World' ...You could probably even get away with putting j:\workspace in your path variable and then being able to run it as 'run World' regardless of which directory you're in relative to run.bat's location (though remember to be in the appropriate src directory when you run it so Java can find your code).

In addition to my previous post

User 449 said: I should also note that DrawDistribution.java is in its proper place in the src folder of my histogram project.

-Joel

Problems with Home Directory Login

User 282 said:

Hey,

Sorry this isn't about programming, but all of a sudden my home directory login client doesn't work. An error screen pops up saying that the client has "encountered an error and needs to close" even though it wasn't running in the first place. Has anyone else run into this problem?

Thanks,

Scott

Lesley said: I believe ITC could best answer your question. Please send them an email. I didn't see your error on their FAQ page at least...

StdDraw.picture problems - finding picture file

User 188 said:

Hi:

I'm trying to practise for the assignment, using deluxe bouncing ball, but I can't seem to make it find the earth.gif file. I've put a copy of the picture in the directory with the .java file, and have added nbody.zip to the libraries, but every time I try to run the program it comes up saying 'image earth.gif not found'.

What am I doing wrong?

Thanks,

Becca

Professor Humphreys said: It needs to be in the same directory as the class file. If you are running from inside eclipse, this will be the "bin" directory, not the "src" directory.

Professor Sullivan said: If you drag and drop earth.gif (for example) to your Eclipse project, then Eclipse will put the file in the right place to be found when you run your program from within Eclipse.

KS

How to Read In a File Name using StdIn

User 25 said: For Assignment 2, I understand that we must use StdIn.readDouble(); and all those methods for the various parts of the program. However, how do I specify to the scanner being used in StdIn to read a certain file (i.e. planets.txt).

do I make an object constructor to tell it to read planets.txt?

please help!

Rohan


Professor Humphreys said: In order to make the contents of StdIn come from a file, instead of from the keyboard, do the following from the command line: java YourProgram < the_input_file If you're trying to use Eclipse, you can't do this. You will have to copy and paste the contents of the file into the input window where you would normally type input.

Groupwork2 formatting error

User 86 said:

I submitted uploaded the first 4 programs for the second groupwork assignment and for problem 4 part of the Robograde output was as follows:

Running our sample solution with the same command line...
Testing the command 'java -cp .:/home/moodle/robograde/cs101-spr08/stdlib.jar Gravity1 0.0 0.0 1.9889999999999999e+30 1496300000.0 0.0 5.9700000000000003e+24'...
Our solution produced the following output:

Line 0: -3.5375093490093405E26 0.0
Line 1: 3.5375093490093405E26 -0.0

WHOOPS! YOUR OUTPUT IS CLOSE, BUT NOT QUITE RIGHT! (line 1)!.

Also, robograde deducted 10 points for formatting (which I assume is due to the error above. How can I fix this?

Thanks!

User 14 said: You showed the correct output, but what was your group's output for this problem?

Lab study 1 Grading

User 343 said:

Are labs graded by robograder or by a person like the individual assignments? I ask because like other people on here, I can get the histogram and picture to show up fine in Eclipse, but I can't make it work from the command line. Looking at robograder from other assignments, it seems that it uses the command prompt as opposed to running each program in eclipse. Basically, if my program works in Eclipse, is that good enough?

p.s. Someone asked about a red line in the 101E class showing up on the histogram and the professor made it sound like that was not normal. I too have a red line on my graph so i'm wondering if that is/isn't supposed to be there.

User 14 said: When your labs/assignments are graded (either by roboGrader or a human), they are compiled and run with the appropriate support files needed. So I believe if you followed the correct procedure for using these support files (like the jar or the DrawDistribution.java), and your program runs in Eclipse, then you should be fine. And yes, that red line is supposed to be there.

Troubling Red Xs

User 449 said: Everything compiles and runs fine from the command line, but I am still getting these annoying red Xs in Eclipse. Any thoughts? Below is a link to a screenshot... I couldn't get the Insert Image function of the forum to work properly.

http://i25.tinypic.com/2sb9yme.jpg

-Joel


Professor Sullivan said: DrawDistribution.java isn't in your project! Save the file to disk then drag and drop it into the Eclipse src directory.

GroupWork submission

User 200 said: For the groupwork #2, it says that we're allowed to submit a program as many times as we like, but it won't let me unsubmit the programs I already uploaded so that I can do that. Can someone fix this?

Professor Humphreys said: just submit the file(a) again. Unsubmit is not, and never was, necessary.

JAR Extension FIle

User 268 said: How do i extract the stdlib file as an extension to my java file?

I extracted all the files to my src folder... and I got the image of the distribution to work, but I'm worried that my submission won't work if it's not in my src folder.


Professor Sullivan said: James,

Don't extract the files. Save the jar file then do

Project>Properties>Java Build Path>LIbraries>Add External Jar, browse to the jar file, and click ok.

KS

Compile and Run in Eclipse - Lab 1

User 199 said:

I have not been able to compile my program through the command prompt window. I tried using -cp and also editing the classpath. Whenever I try to use the -cp command I get an error that reads:

javac: invalid flag: Libraries\stdlib.jar

Usage: javac <options> <source files>

use -help for a list of possible options

Do you have any idea what I should do?

I would also like to know how to compile and run within eclipse and input the command line argument so that I can see if my program works. How do I do this?

Thanks.

User 14 said: To run your program in Eclipse and use command line arguments, click Run --> Open Run Dialog... Make sure the project and Main class match the project and main class you want to run, then click the "Arguments" tab. In the "Program arguments" box, enter the command line arguments you want the program to use. As an example, for the StdGaussianHistogram program, you would enter a number like 1000 here.

In addition to my last post

User 199 said:

When I downloaded stdlib.jar it downloaded as a zip file, I have tried to download it again as a jar file but when I do I get an error message saying:

Failed to load Main-Class manifest attribute from J:\private\CS Standard Libraries\stdlib.jar

I really don't understand what zipped files are, so I have no idea if this is part of the reason that I can't get my lab 1 program to compile. Is it? And if so, how do I fix this?

Thanks

Professor Sullivan said: It sounds like you're trying to Open rather than Save the file to disk. You need to save the file on disk.


Lesley said: #313

Red line on grapth (Lab work)

User 531 said: Ive got the gaussian distribution to output to a grapth that looks like, but it has a red line that starts out following the data then just runs off up the graph, looks a little like a cubic graph. What is that, and how do I fix it/make it go away?

Professor Sullivan said: That's interesting. It's definitely not what we're expecting. I can't diagnose the problem remotely. See a TA or show me after class / in office hours.

User 14 said: Nick,

I believe that red line is supposed to be there. It has been there on every student's graphs that I've looked at. It represents the cumulative frequency distribution. http://en.wikipedia.org/wiki/Cumulative_frequency

Professor Sullivan said: Eric's right. The red represents the cumulative probability, but it should not diverge upwards. Rather it should start at zero, rise quickly where the prob. density is the highest, and then flatten out to the right, at the value one.

Please disregard my previous inquiry

User 449 said: The red Xs referenced in my previous post (Troubling Red Xs) have gone away upon reboot and re-mounting the j:\ drive. Regardless of the source of their disappearance, I am rather pleased with this development.

-Joel

Lab Study #1

User 344 said: If my original histogram program had a problem with it, and thus the graphical output for this lab does not display exactly right, should I attempt to fix the program or leave it how it was when I turned it in the first time?

User 14 said: you can use the solution file from the lab page

Variable precision

User 449 said: Is there a way to account for variable precision using printf()?

Pretend that I have a variable called places_past_decimal, which accounts for the following:
Example #1
Input: 4.5
places_past_decimal = 1

Example #2
Input: 2.3109000
places_past_decimal = 4

If I wanted to print out 4.5 instead of 4.500000 (let's store this number in a variable x) in Example 1, I would do:
StdOut.printf("%.1f\n", x);

Similarly with Example 2, I would do:
StdOut.printf("%.4f\n", x);

But say I wanted to more generally truncate zeroes from an arbitrary double, how would I go about doing this? A statement like:
StdOut.printf("%.places_past_decimalf\n", x)
yields an error, which I fully expected it to do, because the "f" is mixed up with the variable name.

-Joel

P.S. Is this what you are asking from us in VariableArgs3 (based on the sample output)?


User 14 said: Joel,
If you want to print a variable precision in your floating point values, remember that the first argument to printf is a string. Thus, you can concatenate things together to create this string. Something like "%." + places_past_decimal + "f\n" would do what you describe in example 3. However, VariableArgs3 is not asking you to change the precision of the floating point arguments.


Lesley said:

VariableArgs3 isn't asking you to code the kind of formatting you are requesting. You just have to print out the -double-, however it makes its way through your program. You should get the kind of behavior you wish "for free" (ie. without additional work).

I would like to quote Professor Humphreys: "Convince yourself: Either you understand the code, in which case, you’re on the road to greatness... Or... The rabbit hole goes deeper than you wish to go, in which case, you’re on the road to redemption."

Basically, your goal (do I have to code something that formats numbers like this to get credit) is accomplishable without answering your question specifically. If you want a solution to YOUR question (how to format a STRING that looks like a number into nice output), I would suggest regular expressions over printf mainly because: exactly how are you going to calculate your places_past_decimal variable? With regular expressions you can eliminate the need to calculate it at all. This code at least works with your test cases and a few others I tried.

import java.util.regex.*;

public class Precision {
public static void main (String[] args) {
String regex = "^0*((0|[1-9][0-9]*)(\\.([0-9]*[1-9]|0))?)0*$";
Matcher m;

m = Pattern.compile(regex).matcher(args[0]);
m.find();
System.out.println(m.group(1));
}
}

Lab Study 1

User 166 said: What if we can't get the graph to work/print out, but we edited our program to your specifications? If we submit that would that be ok? I still can't seem to get it to work and I don't really understand your directions becuase I'm mostly computer illiterate. Thank you.

User 14 said: You should really test your code before submitting it, and at the very least run it to make sure it works. Otherwise you won't know if you've made a small mistake that could cause your program to do something unintended (resulting in a lower grade than you could have gotten). My advice would be to start from scratch and make sure you follow the directions for the lab very carefully, especially noting where to put the code that draws the graph and how to correctly import the jar file.


Professor Sullivan said: Lynna, You can also see me after class today and I can have a look at what's up.

Running in Eclipse

User 368 said: When I try to run my program in Eclipse, I go to Run As and it says (none applicable). What can I do?

User 14 said: That could be a workspace issue caused by not using the correct workspace. I saw the same thing in lab last week. First make sure you're using the correct workspace. Otherwise, what worked in that case was just creating a new project and copying all the source files into that new project.

Another lab study 1 question

User 33 said: I know I've already turned it in and all, this is more for my curiosity. I'm pretty sure my histogram program worked (seemed to for me, at least), but when DrawDistribution made the graph, it always had horrendously inaccurate values for leftTail and rightTail in the picture. For example, I ran the histogram with an input of 2 billion (yay core2duo!), and had it print the two tail values. They were 576 and 578 respectively. Yet, the drawn image said they were 2.88E-7 and 2.89E-7. Why is that?

User 14 said: if you look closely at the drawn image, the leftTail and rightTail are displayed as P[leftTail] and P[rightTail]. These values are the probabilities that a random number in your sample fell in the left tail or right tail. And so these numbers are what you'd expect, very small.

Assignment2: "

User 71 said: two questions about A2:
1.
I finished the program and try to run it, it will only work if I copy the data out from the txt into the console window after press debug/run, if I try put "< planets.txt"(brackets not included) in the arguments, it will start but is not doing anything (debug said the Thread(main) is running), until I type "< planets.txt" into the console and it will said source not found(in debug,error when run). I thought it was something wrong with the file's directory but I put it in that project's root directory which works for Blueplanet that I did for Groupwork, so it should not be a problem. So I trace the variable and found out that it read "< planets.txt"as two string instead of a special command(I try without the space in between and didn't work either). Since the "PlotFilter" file in the book can use the < command without doing anything else special, I don't know what the problem is here.

2.
When running, my planets all flying away from the sun outside the window. Since there's nothing run with the program I don't see a way to solve it by debugging. And it is a lot of work trying to find out the problem by tracing the variables...any advice based on former experience? Or I should just trace the variables and calculating them one by one :( ?


Lesley said:

1. #337

2. From the checklist:

My planets repel each other. Why don't they attract each other? Make sure that you get the sign right when you apply Newton's law of gravitation: (x[j] - x[i]) vs. (x[i] - x[j]). Note that Δx and Δy can be positive or negative so do not use Math.abs(). Do not consider changing the universal gravitational constant G to patch your code!

Something for Groupwork 2

User 71 said: I discover that the one of the test that are used the Robograde for Gravity1 is requiring a result of "-0.0" or negative zero, which will never be achieve unless you do the calculating within the printout command.
I discover it when trying to debug my program, and I guess that is the reason many people can pass 2 test in Robograde but just fail the one with negative zero.


Professor Sullivan said: We'll have another look at the test case and fix it if necessary (which seems likely).

Groupwork2 formatting error

User 86 said:

Oops! Sorry. Our output is missing the "-" on the 0.0.

Compiling Gravity1...
Testing the command 'java -cp .:/home/moodle/robograde/cs101-spr08/stdlib.jar Gravity1 0.0 0.0 1.9889999999999999e+30 1496300000.0 0.0 5.9700000000000003e+24'...
Your program produced the following output:

Line 0: -3.5375093490093405E26 0.0
Line 1: 3.5375093490093405E26 0.0

Running our sample solution with the same command line...
Testing the command 'java -cp .:/home/moodle/robograde/cs101-spr08/stdlib.jar Gravity1 0.0 0.0 1.9889999999999999e+30 1496300000.0 0.0 5.9700000000000003e+24'...
Our solution produced the following output:

Line 0: -3.5375093490093405E26 0.0
Line 1: 3.5375093490093405E26 -0.0

WHOOPS! YOUR OUTPUT IS CLOSE, BUT NOT QUITE RIGHT! (line 1)!.

Thanks!

Professor Sullivan said: Ann,

I believe this is an example that illustrates the point that "floating point" arithmetic is complex in sometimes unexpected ways. A negative sign before a zero exponent is mathematically meaningless but it does cause our test to fail. In this case, the test should be changed (or ignored). Please submit a regrade to the TAs in the usual way, with a copy of this note, and they'll take care of it.

KS

Individual Assignment file addresses for grading

User 77 said: For the Galaxy Simulation individual assignment, where should we assume the the external files required (.gif and .jpg files) are located, since we have to call those files in the program?

Professor Humphreys said: #334

GroupWork Gravity2

User 200 said: Part of the description for Gravity2 reads as follows:

Your program should store this data in arrays: one for the x positions, one for the names, one for the x positions, one for the y positions, and one for the masses

I assume that the redundancy is a typo, so is that supposed to be omitted, or is one of those statements supposed to say something else?


Professor Sullivan said: Thank you. We're sorry for the typo. It is now fixed. --KS

Gravity2 Output error

User 14 said: It's hard to say what happened without looking at your code. Make sure you're using doubles everywhere instead of "float". Floats will give you less precision. Otherwise, the best thing to do is show a TA during office hours.

Professor Sullivan said: Mathematically equivalent expressions can produce subtly different results when using floating point arithmetic. This case might well be an example of a subtlety in floating point computation. In general it's a bad idea to compare floating point numbers using == (which is probably, in essence, what our test case is doing). I'd have the TAs look at your code, and call us with us if you've done everything right but are getting and answer that differs in the last bits. In that case, we'll probably fix our test case.

So, if you generally should not compare floating point numbers using ==, how should you compare such numbers for "equality"?

KS

Audio For Bouncing Ball

User 251 said: I am able to compile and run my DeluxeBouncingBall program by taking out the audio parts with lazer and pop. I can not find any audio files in the n-body file. Any help on where to find the files?

User 14 said: 2001.mid is an audio file located in the nbody folder of nbody.zip

Professor Sullivan said: When you find the audio file you want to use, store it somewhere on your disk other than in the workspace/project directory. Then drag and drop the file to the project icon in Eclipse. That'll add the file to the project in a place where your running program will find it.

Gravity2 Output error (in addition to previous post)

User 449 said: The previous post of this same name quoted rounding problems in Gravity2 (and presumably in Gravity3).

I think it is worth noting that my output was identical to Kiran's. RoboGrader says to me:

Your program produced the following output:
Line 0: Earth -3.541410008204977E22 0.0 -0.005928038179117805 0.0
Line 1: Mars -1.639662166651638E21 0.0 -0.0025543887936620006 0.0
Line 2: Mercury -1.3067081912709674E22 0.0 -0.03957323413903596 0.0
Line 3: Sun 1.0529518828292188E23 0.0 5.293875730664751E-8 0.0
Line 4: Venus -5.51743441215108E22 0.0 -0.011331760961493284 0.0

Our solution produced the following output:
Line 0: Earth -3.5414100082049774E22 0.0 -0.005928038179117806 0.0
Line 1: Mars -1.639662166651638E21 0.0 -0.0025543887936620006 0.0
Line 2: Mercury -1.3067081912709674E22 0.0 -0.03957323413903596 0.0
Line 3: Sun 1.0529518828292188E23 0.0 5.293875730664751E-8 0.0
Line 4: Venus -5.517434412151081E22 0.0 -0.011331760961493286 0.0

ERROR! YOUR OUTPUT DOESN'T MATCH OURS! (line 0)!.

The exact same statement is true of Gravity3 (considering they are basically the same problem). I suspect that once people get around to doing the groupwork, this will be a common error message. I am using doubles everywhere, so that is not the problem. I doubt the TAs want to process Regrade Requests from every student, so if the test case were changed, that would make things much easier for everyone. Also, if the case is not changed, would it be acceptable to change my program to fit the RoboGrade output in this particular instance?


Professor Sullivan said: Dude, you're off by a quintillionth. What do you expect?



I'll ask the good Dr. Humphreys to fix the test case.

Thanks for the diagnoses.

And remember, everyone: In general it's a bad idea to try to compare floating point values for exact equality. Numbers that should be equal are often off by just a tad. == will return false in such cases, even though the numbers should be considered equal (as here). It's best to compare floating point numbers a and b like this: if (Math.abs(a-b)<EPSILON) { ... }. That is, take the absolute value of the difference and make sure it's sufficiently small. EPSILON here is a floating point variable with a very small value, e.g., 1.0e-10 or something. Picking good values for EPSILON is an application-specific problem. The bottom line is that floating point numbers are problematical. They really don't act like real real numbers. We'll talk later in the semester about conditions under which they are and are not reasonable substitutes for real reals.

Read the input file - assignment 2

User 546 said:

How can I make my program read the input file "planets.text"?

I am typing in the command prompt java NBody.java < planets.text and it is saying that "The system cannot find the file specified". But my file is the right directory...

I read some advices in the forum and saw that we could write in the Eclipse in the input window the things that are in the input file. Where is this inpu window? I can't find it... is it the console? I couldn't write in it...

Thanks!

Camila.

Lesley said:

• DOS: java NBody < planets.txt (not java NBody.java or planets.text)
• Eclipse: #346

BluePlanet image not found

User 380 said: When we run our BluePlanet program, it says that the file earth.gif cannot be found. We have added the nbody.zip and the stdlib.zip files to the libraries being used and it still won't work. Eclipse recognizes the file because when we add the nbody.zip to the library the red squiggly line disappears, but when we run the file (from eclipse or the command prompt) it can't find it. what are we doing wrong?

Lesley said: #334

how many points for each program in group assignment 2?

User 49 said: well, we have 7 programs totally. I submitted one and got 14. Is that full credits for one program? Cuz some of my answers are "pretty close to" the keys. I was not sure if I lost some credits for minor differences.

Professor Sullivan said: 100/7=14.3, so that looks like it's close to full credit. That said, small differences are generally not ok, unless they're very small differences in floating point math (where small differences are to be expected). That is, we define equality a bit more loosely for floating point (doubles, floats), than we do for String, int, etc.

Changing Input within Eclipse

User 41 said: After some experimentation, it seems that attempting to redirect input from within the Eclipse arguments box (an option of Run Dialog) has no effect. In fact, the attempted command is simply stored within args[].

My command was "< planets.txt" (without the quotation symbols, of course).

Wondering over the logic of this, I ran the same program from within my terminal (with the same arguments), and args[] was empty. Presumably, there my command had the desired effect.

I would greatly appreciate any advice to aid in the circumvention of this issue.


Professor Sullivan said: Ethan,

There's no way to do it (that we know of). There is however a work-around: copy and paste the contents of the file you want to read from into the Eclipse console window. It's not great, but it works. Be sure to add a ctrl-Z return at the end.

KS

end- of -file sequence, Book example Program 1.5.3 Averaging a stream of numbers

User 305 said:

end- of -file sequence, Book example Program 1.5.3 Averaging a stream of numbers

I get an error when i type <ctrl-d> or <ctrl-z> as an end-of-file sequence and according to pg 131 in the book there are different conventions for this sequence.

so, am i doing something wrong? i cut and paste example onto eclipse, compiled it, all the normal stuff, but i get an error after typing in some numbers and using the end of file sequence.

Professor Sullivan said: Are you holding down the CTRL key while typing Z?

How to run a program in eclipse by using standard input instead of the command line?

User 49 said: Is that exactly the same as by using command line? type data in the argument?

Professor Sullivan said: No, copy and paste the data to the "console" window. When running a program within Eclipse, Standard Input is connected to the Console Window.

Running Bouncing Ball from command line

User 200 said: I got the Bouncing Ball program to work from Eclipse when I use audio and picture files, and it compiles fine on the command prompt, but when I try to run it on the command prompt, it gives me a message saying that the audio file wasn't found and that my picture file was corrupt. What do I do to fix this?

Professor Sullivan said: The files should be in the directory where you're running the code. When you compile from the command line, the .class files end up in the src subdirectory. When you compile in eclipse, they go into the bin directory.

Image Earth.gif not found

User 199 said:

I am trying to run the DeluxeBouncingBall program but I get an error (whether I run it in the command prompt or through Eclipse) saying that the image earth.gif cannot be found.

I have added stdlib.zip and nbody.zip to my libraries and have tried to make sure the earth.gif file is in the same directory as the class file which is what you said to do in an earlier post.

In the explorer on the left side of the Eclipse screen there is no directory labeled bin; however I tried putting the earth.gif file under "default package" which is where the class file is. When I do this I get an error saying, "cannot place contents into the selected elements". I also tried to simply drag the earth.gif file into the project, but when I do this my pointer turns into a "no" sign (circle with line through it) and nothing happens.

What am I doing wrong??

Professor Sullivan said: That's strange. I just dropped Earth.gif into the top-level project (in Eclipse), and after that everything worked fine. See a TA or see me. I'm around tomorrow (in my office). If you want to stop by shoot an e-mail to make sure I'm here.

stdlib

User 53 said:

I don't think that my standard library is working properly because everytime I run a program using standard library voids nothing happens. I went through the whole process of adding a JAR to my workspace and stdlib is listed under reference libraries.

Professor Sullivan said: See the TA in the stacks tonight. It's probably just a little glitch somewhere.

placement of external files

User 77 said: This isn't a problem per se (at least I hope it's not), but I've tried dragging external files (*.mid, *.gif, etc) into the project folder shown on eclipse (as shown in class), and I've tried to run the programs like that, but it never works. However, I've found that if I place the files in the src folder of the program, it runs perfectly, with no errors (at least at the command line). I know we were explicitly told that that wouldn't work, and that they had to be in the bin folder, but if it works this way, I guess there's no harm, right?

Professor Humphreys said: They need to be in the bin folder to run from within eclipse. They need to be in the src folder to run from the command line.

re: Lab quiz this thursday

User 505 said: Will we have access to our textbooks during the lab quiz this thursday?

Professor Sullivan said: Yes

lab quiz

User 250 said:

Are there any examples to practice where we have to "read an integer value n in from StdIn then use a for loop to read in n additional values from StdIn (ints, doubles, Strings)" which was from lab description

Question

User 368 said: I can't run programs through eclipse. The option to run as java app is not even there. What should I do?

Professor Sullivan said: See a TA in the stacks before 10PM tonight. Last chance before next weekend.

Robograde

User 124 said: After submitting Gravity1.java, I had it Robograded and it said I received 14 points for it; however, at the very bottom of the page the program said, "MINUS 10 POINTS FOR FORMATTING"... why's it doing this?

Professor Humphreys said: In general, the robograde output tells you exactly what LINE it's taking off points for. However, looking at your output, it's clear that robograde isn't properly handling the case when one of us produces 0.0 and the other produces -0.0 (that's right, NEGATIVE ZERO). This has been fixed, and your grade has changed. Sorry about the panic.

More stdlib problems

User 89 said: I've changed my CLASSPATH so that it now includes the location for the stdlib.jar file, yet when I use the cmd line to compile I receive the following error:

javac: invalid flag: File\stdlib.jar
Usage: javac <options> <source files>
use -help for a list of possible options

The same error comes up even when I use the -cp function.


Professor Humphreys said: Does the stdlib.jar directory path (whatever you added to your CLASSPATH) contain any spaces? If so, you should put the entire directory path in quotation marks.

Individual Assignment 2 - LeapFrog Method?

User 52 said: Hey

How are we suppose to make Force, (and all the other stuff in relation to time)? I have no clue how time gets factored into this (formula-wise). I get it conceptually, but I have a hard time writing it down mathematically. Could someone please elaborate?


Lesley said: If your Gravity2 program works correctly, you can just repeat the process you have already successfully implemented - forever (to quote the checklist, "Add an outer loop to repeat Steps 2 and 3 forever"). If your Gravity2 program is not working yet, you NEED to do that first. Please try following the steps on the checklist if you get lost, or see the section on the assignment page called "The numerics" because it will aid you greatly. The formulas for force are in the diagram under "Pairwise force." All of the formulas you need are listed in the assignment. I promise. I did this assignment last semester without any prior physics knowledge. Just follow the directions, don't overthink things. PS If you look carefully, t isn't even in any of the formulas!! Just dt (Δt) = 25000.

How do I get the console to open?

User 347 said: I'm attempting to run a simple program I found in the text book so I can practice using StdIn and StdOut. I got it to run fine in the command prompt, but I can't seem to get Eclipse to cooperate with me. To the best of my understanding, I need to work through the console window. However, I don't seem to have one, or maybe I just can't find it. Any help would be greatly appreciated.

Professor Humphreys said: You can use StdIn and StdOut within eclipse; you just type into the little window at the bottom of your run. The command prompt on Windows is usually located in the start menu under "Accessories". You can also get there by going to Start->Run... and typing "cmd" in the dialog box that comes up.

Professor Sullivan said: Window>Show View>Console

If you've explicitly closed the console window this will "get it back" for you.

KS

Command Line with Std Libraries

User 222 said: I've been reading up on everyone else's similar problems with being able to compile in Command Prompt. I've changed the CLASSPATH and everything, and I'm still having problems compiling...Any other suggestions?

User 14 said: It's hard to know what the problem is without seeing what the compiler is telling you and what commands you're using. The easiest way to fix this is to see a TA at office hours, but that will have to wait until Sunday. The only other suggestion I can give is to make sure you refer classpath to all sources of libraries you're using. So if you are referencing both stdlib.jar and files in the current directory, you would need to say "javac -cp .;j:\workspace\stdlib.jar something.java" In this case, the ".;" at the beginning is important.

Professor Sullivan said: Be sure that the CLASSPATH environment variable includes the full path to the stdlib.jar file, including the name of the file, stdlib.jar. If your path name has any embedded spaces, you need to enclose the whole thing in double quotes. Come see me if you can't get it to work.

USA.txt

User 250 said: I am having trouble getting the PlotFilter program to run. I do not know how to get the USA.txt folder into the arguments so that I can run the program. How can I do this?

Professor Humphreys said: On the slides, USA.txt is redirected into standard input; it's not an argument. So run: java PlotFilter < USA.txt If you're using Eclipse, you will need to copy-and-paste the contents of the file into the input window and press control-Z to terminate the input, as always.

Professor Sullivan said: I think Prof. H. meant to say, at the command line do this:

java PlotFilter < USA.txt

Notice the "< USA.txt", which connects the standard input of the program to the file, USA.txt

If you want it to work in Eclipse, open USA.txt in an editor and copy and paste the contents of the file into the Eclipse console window

random stdlib error

User 89 said: I've had my stdlib working fine until I closed the cmd prompt window I had been working in. When I started a new cmd prompt window, I received the following error:

symbol: variable StdIn
location: Class Gravity2
double mass1 = StdIn.redDouble();
^

This appears to treat StdIn like a variable, but I have Eclipse and CLASSPATH set up correctly, as they worked 15 minutes ago.


User 14 said: Changing classpath while the command prompt is open won't always immediately update it. Try closing all command prompt windows, then changing the classpath to the correct value. Then when you open a new command prompt, everything should work, ideally. But If that doesn't work, try just setting the classpath after the javac or java command. i.e. "javac -cp ..."

Lesley said: StdIn.redDouble, hrm?

Gravity2 and Robograde

User 124 said: I've repeatedly submitted Gravity.java, and every time at the very end, Robograde reads just one last sample that has output values that don't match mine... I have no idea what to take from that. Help?

Professor Humphreys said: Looking at your robograde output (and your grade!) it appears that you have gotten all the points for Gravity.java, so I'm not sure what your concerned about. Your program is correct.

BluePlanet and Eclipse

User 124 said: This program would not run in my Terminal (I have a Mac), yet I received full credit for it in Robograde. In my Terminal it says i have 3 errors, all having to do with not naming "Std." I'm pretty sure I downloaded all the appropriate files into my library, but what's going wrong?

Professor Sullivan said: You have to set up your CLASSPATH environment variable to include the path to the stdlib.jar file. You do this for any command line compilation independent of your Eclipse settings. There are several other postings on this forum on this issue

USA.txt

User 250 said:

Where can I download the file USA.txt so that I can upload it?

Professor Sullivan said: Booksite, appropriate chapter and section (as stated in class)

BluePlanet

User 124 said: I've done everything I know to try and add the right files to this program, yet this keeps popping up in my terminal window... Help? d-172-26-31-196:~/desktop/CS101/BluePlanet/src michaelincerto$ javac BluePlanet.java BluePlanet.java:6: cannot find symbol symbol : variable StdDraw location: class BluePlanet StdDraw.setXscale(-R, +R); ^ BluePlanet.java:7: cannot find symbol symbol : variable StdDraw location: class BluePlanet StdDraw.setYscale(-R, +R); ^ BluePlanet.java:8: cannot find symbol symbol : variable StdDraw location: class BluePlanet StdDraw.picture(0.0, 0.0, "earth.gif"); ^ 3 errors d-172-26-31-196:~/desktop/CS101/BluePlanet/src michaelincerto$

NBody.zip?

User 196 said:

I am trying to get the things i need from the NBody file downloaded off the site but whenever i download it it will only save as a .zip file. Im was not sure what to do becaus I know this wouldnt work so I tried to extract it to make it a .jar but it will never let me get it. So i cannot get all the audio/picture files for the individaul assignment. What should I do?

Nathan Wethington

Professor Humphreys said: NBody.zip is, well, a zip file. It's just a bunch of files that you will find useful. They're not Java programs or anything, just unzip the file somewhere and you will have a directory full of universe descriptions.

Lesley said:

If you have Windows XP or higher, do right click - explore on nbody.zip. Then select all (click in the folder window it pops up and press 'control' and 'a' or edit - select all) and copy the files (right click - copy, edit - copy, or ctrl+c) and then copy them into where you need them (the bin folder of your project most likely) by doing edit-paste or ctrl+v.

If you don't have this "explore" option, you could consider downloading shareware like WinZip which will extract the files for you.

Planets leaving the solar system

User 380 said:

For the individual assignment, I plug in the sample values given on the assignment page:

5 2.50e11

1.496e11 0.000e00 0.000e00 2.980e04 5.974e24 earth.gif

2.279e11 0.000e00 0.000e00 2.410e04 6.419e23 mars.gif

5.790e10 0.000e00 0.000e00 4.790e04 3.302e23 mercury.gif

0.000e00 0.000e00 0.000e00 0.000e00 1.989e30 sun.gif

1.082e11 0.000e00 0.000e00 3.500e04 4.869e24 venus.gif

and the planets start to orbit the sun, but when they get about 60% of the way around the orbit, they shoot straight off into space. Is this supposed to happen or is there some problem with my leapfrog approximation?

Professor Sullivan said: That doesn't seem like it's consistent with the behavior of astronomical bodies under the force of gravity! My guess is that the model is wrong, not the laws of gravity

using planets.txt

User 53 said: The help page for the individual assignment recommends that we read in planets.txt using StdIn. and print it before starting to write the program. How exactly do we do this?

Professor Sullivan said: Figuring that out is really the essence of the exercise. Apply some of the strategies we've talked about before. Break the problem into parts and then solve the parts separately. For example, ...

• Read in the data and store it in appropriate variables (including arrays)
• Then print out those variables (including arrays -- using for loops of course)
  

selection does not contain a main type

User 473 said: When I try to run Triangle, it shows that selection does not contain a main type. How to solve this problem? Thanks.

Professor Sullivan said: Be sure that Triangle.java is selected in Eclipse when you issue the run command, and be sure that you typed public static void main and all that stuff exactly right.

Readings planets.txt

User 200 said: I'm having trouble getting NBody to work by reading the planets text file as input. When I manually type in all of the arguments from the text file into the command prompt(which took a LOOONG time), the program works fine, but when I actually type % java NBody < planets.txt, I get an ArrayOutofBounds Exception: 0. I went into the text file and made sure everything was spaced appropriately, and I figure that the ArrayOutofBounds Exception means that my computer can at least find the text file, but it's not inputting any of the arguments into the program. Any ideas?

Lesley said: Make sure you're using StdIn to read the values, not args[].

Professor Sullivan said: First, you can cut and paste the contents of the file into the Eclipse console pane. Just open the file in any word processor, e.g., notepad, wordpad, etc (which are both available under Windows Start>Applications). Then select all the text (control-A), copy to your clipboard (control-C), and paste into the Eclipse window (control-V). I'd start by doing this and see if it runs okay in Eclipse, and take it from there.

Confusion about Assignment 2

User 347 said: I'm a bit stuck on Assignment 2. My gravity 1, 2, and 3 programs all worked fine, but my NBody program doesn't. As soon as my planets cross the 90 degree mark, they fly off into space. I've seen several posts about this but I haven't been able to figure out why my planets are doing this. I've checked all of my signs and everything seems to be ok, but my planets keep flying away! It seems like if all of my other programs worked, my planets should stay in the solar system.

Also, I read that we should not include steps 1-3 in the same loop, but it seems like we kind of have to in order to make sure that the x and y components of the gravitational force are constantly updating. Otherwise, the system is using values for which components have been changed but unaccounted for.


Professor Sullivan said: I recommend using the debugger to see what's going on with the values in your program. It's clearly doing something wrong at some point. Set a breakpoint at the beginning of the main animation loop. You can use the debugger's resume command to start the program running again from a breakpoint. Execution will stop when control gets back to the breakpoint. You might have to resume quite a few times before your planets get themselves into the position where the bug takes hold.

Assignment 2 being weird...

User 347 said: My assignment 2 is being really weird. It seems like whenever I stop using the arrays for fx and fy and have the program constantly calculate the acceleration, things start working. However, the program only works when the sun is the last element on the list. If I read in the information as given, only Venus orbits the sun while the rest of the planets fly away vertically.

When I do use arrays for fx and fy in my calculations to find the acceleration, the planets just sling shot around the sun. What's going on here? Nothing I do works.


User 14 said: Make sure your loops are structured properly. You want object to calculate its net force based on the forces between every other body in the system. It might help to go through an example for one of the planets at the beginning, using the debugger. Go through your code step by step and see if the forces are being correctly calculated for every other object.

Having problems with StdDraw.picture

User 270 said: I tried edit the Bouncing Ball program so that it can show the earth picture instead of a black ball (like what Professor did in the class). However, eclipse keep underline the word "earth" with a red line, which preventing from running the program. I already copy and paste the earth.gif file from the NBody zip file the bin and src folder of the project. I also tried to drag the picture file to project on the left side of the eclipse (like what Professor did in the class). I even tried to save the whole NBody zip file in the project folder and add the NBody through libery > property > add external jar. None of these methods worked, eclipse still underline the word earth with a red line saying variable is unreasolved. I don't understand why.

Professor Sullivan said: Without seeing your code it's hard to guess what's wrong, but it sounds like you are using earth as a variable name (undefined), rather than specifying the file name (of type String), "earth.gif" (with the double quotes) as a parameter to the StdDraw.line function.

starfield.jpg

User 53 said:

I've been following the steps in the Help page for Assignment 2. Step 3 says to use StdDraw.picture to plot the background starfield.jpg . I've written the code exactly how prompted to in the book and by eclipse, but I keep getting the same message:

Exception in thread "main"

at StdDraw.getImage(

at StdDraw.picture(

at NBody.main(

I've downloaded the stdlib and NBody zip files into the workspace. The last line of the error message refers to a line of code in my program that sets the x-scale for -R to R, as indicated by the instructions. Do you know what's wrong? Who can I see about this problem before Sunday?

java.lang.RuntimeException: image starfield.jpg not foundStdDraw.java:481)StdDraw.java:496)NBody.java:28)

Lesley said: The picture starfield.jpg is not in the directory where you are running your program (I duplicated your error by doing this at least). You need to UNZIP NBody.zip and put the images and text files in the src and/or bin directory/ies of your project. If you don't know how to unzip a file, there are instructions for how to do this at #420

Is it possible?

User 52 said: How are the planets suppose to follow the laws of gravitation when we are only calculating the net forces between 4 planets when there are actually 9 planets in the solar system.

Please elaborate on these extremely difficult programs that even 4th years cannot comprehend. (Maybe....)


Lesley said:

Please see the "enrichment" section of the checklist for physics details of the NBody prodedure. But here are some more things for you to think about.

The Sun's mass is about 333,000 times Earth's mass.

The mass of Jupiter is 318 times that of Earth.

(On top of that: Sun to Earth: 1AU; Earth to Jupiter: about 4AU)

You should also be bringing up that the orbits of the planets are elliptical, not circular due to the fact that the sun is not at (0,0) like the text files claim but rather at one focus. Though if you've taken any astronomy class you'll know that the orbits are elliptical with very little eccentricity. (Wikipedia says Earth's Eccentricity is 0.016710219)

If you prefer, there's a text file called elliptical-nine.txt... Though I think that's probably a worse example because it over exaggerates the eccentricity (see http://freeutah.net/images/blogspot/planet_orbits1.jpg)

You should also consider in your question that the sun is moving, so how can we calculate anything without including the whole galaxy, universe, etc? We don't take into consideration how the planets are rotating... The model blatantly says it uses constant acceleration... It is just an approximation that works well given certain circumstances.

Program with Home Dir. and Stdlib

User 331 said: I am having trouble loading my home dir to eclipse. My home dir. loads but when eclipse asks for the workspace and i use my home dir, it comes out with an error "j:\workspace\.metadata\.log".

Also when I still had eclipse working, i tried to get stdlib to work and it wouldn't. Now i cant delete it off my desktop because it says its being used in a program but nothing is running. I dont know if this is related.

Dan



User 14 said: I think restarting your computer will get rid of that second problem. And possibly the first too.

Planets really slow

User 270 said:

By using dt = 25000 and StdDraw.shoe(dt), the planets are extremely slow. Is it supposed to be like that?

Lesley said: If you use StdDraw.show(25000) that's quite a delay. You only need to do StdDraw.show(20) or StdDraw.show(30) at most. (StdDraw.show() is unrelated to dt)

README Template?

User 52 said: Why do the directions say to upload the README file on Eclipse? When the file to be uploaded must be a txt file? If done on Eclipse, it would be a java file? yes? Just confused...

User 14 said: You can edit txt files in Eclipse. If you want to create a .txt file while in Eclipse, go to New --> File and then type in "readme.txt"

Galaxy Simulation extra credit deadline

User 459 said: Hey, I was wondering if the galaxy sim. extra credit is also due on Wednesday. It would be nice if it were due a couple of days later.

Step 2:Planet not moving on given velocity

User 254 said: I set scale x(-2.5E11,2.5e11) y(-2.5e11,2.5e11). In step 2, my planet does not move on given velocity, but move as discribed when I changed the velocity for 10^4 to 10^9. the FAQ does not help,cause i have not moved to step3, there is no t value to change. and my Std show time is 20.Thanks

User 14 said: I'm not sure what you mean by "there is no t value to change." You need a time variable in order to calculate new positions for each planet. If this time value is larger, the planets will appear to move faster. You shouldn't need to change the velocities.


Professor Sullivan said: It's true that you might not need an explicit representation of time. To do the animation, all you really need to know is the difference in time between two frames: delta-T.

Deluxe Bouncing Ball

User 489 said:

I'm trying to set up for the bouncing ball test and by earth starts in the upper right going to the upper right corner. When it hits the wall I get the following errors

Exception in thread "main"

at StdAudio.play(StdAudio.java:147)DeluxeBouncingBall.java:15)

at DeluxeBouncingBall.main(

I'm assuming that the audio files it is calling are not there. And I can't find them in the Nbody Zip. So where are they, and where do I put them in the DeluxeBouncingBall file to get this thing running????

java.lang.RuntimeException: audio pop.wav not found

Professor Sullivan said: Download the .wav files from the book site. Read the chapter/section (online) in which the topics are discussed. Links are there. Put the files in the bin not the src directory of the project in your workspace.

Student.java Stdin

User 489 said:

I tryed to complie the students.java file and got 8 errors about

"Cannot find symbol

Sysmo: varriable StdIn"

The location is every StdIn command given. ??????

Professor Sullivan said: We've talked about this quite a few times by now. You need to have stdlib.jar as part of your project.

HELP

User 124 said:

I cannot find a way to run a program using StdIn or StdOut, and I have tried and tried to make it work with the help of others but I can't. I'm using a Mac and I would love it if I could get specific instructions on what to do with downloading the right files and such so I know if I've done everything right or if there's some other problem.

Professor Humphreys said: the procedure for getting stdlib.jar to work in eclipse is IDENTICAL on the Mac to the procedure on windows. 1) Create the java project 2) Project menu --> Properties 3) Choose "Java Build Path" on the left 4) Choose the "Libraries" tab 5) click the "Add External Jars" button 6) navigate to the location of stdlib.jar That's it.

Another StdIn problem

User 489 said:

I'm trying to get the planets.txt as a String planets = StdIn.readString(); And then print them out using a System.out.println(planets);

when I compile I get the error of

Javac :file not found: PlanetsReadIn.java

Usage:Javac <optional><Source Files>

use -help for a list of pissible options

I'm assuming that the external jar file is not doing it's job cause I can see the StdIn.class in my directory....????

Professor Sullivan said: You're not meant to use one statement to read in the names of all the planets. Rather, each line of input includes on planet name as a string. See a TA for more help. Someone will have to look at your code.

Another StdIn problem update

User 489 said:

After a Google I seteled on the below code.

public

String planets = StdIn.readString(

System.

}

}

But the compiler still dosnt like my planets.txt as a string read in

unclosed character literal

?????? P.S could it be made so students can add replies to the tech forum so we can update our own postings or possible help others???? THXS

class PlanetsReadIn {public static void main(String[] args){'planets.txt');out.print(planets);

User 14 said: This post is very difficult to read, but you need to use double (") quotes around planets.txt rather than single(') quotes.

Clearing image trails....

User 427 said: When I run my nBody program, the planets leave behind at trail of images everywhere they do. How do I prevent this?

Lesley said:

Quoting the assignment page:

After each time step (i) draw the background image starfield.jpg, (ii) redraw all the bodies in their new positions, and (iii) control the animation speed using StdDraw.show().

StdDraw.show()

User 505 said: I am trying to use the StdDraw.show() to control the animation. However, when I type that line of code at the end of my program, i get an error that read "Unreachable code". I have the stdlib jar file isntalled correctly, so i dont know what the problem is.


Thanks.


Professor Sullivan said: The term unreachable code refers to code that can never be executed. For example, consider this code fragment:

if (false) {
StdOut.println("Hi");
}

The code inside the curly braces is unreachable. There's no way that it will ever be executed.

In your case, there's no problem with your stdlib.jar. Rather, the problem is in the logic of your program. Read through your code very carefully and figure out precisely how it executes. You'll very likely see that there's a problem with the logic.

Kevin Sullivan

I cannot get StdDraw.picture to work

User 320 said:

Perhaps it went over my head during lecture, but I cannot get images from NBody. I copied and pasted the NBody folder (unzipped) into my src and bin and wrote the code StdDraw.picture(0.0, 0.0, "earth.gif"); for BluePlanet.java, but I always get an error message.

Is there a StdDraw.getImage involved? If so, where is it, because such a command is not in the standard library.

Could you please go over the steps to getting pictures such as earth.gif and starfield.jpg to work again?

Professor Sullivan said: These files (one way or another) need to be placed into the bin subdirectory of your project, if you want them to be found when running the program from in Eclipse.

Unsure how to compile and run Assignment 2

User 296 said: I can only compile programs that use StdIn/StdOut in eclipse (I installed Stdlib.jar already). When attempting to compile in command prompt, I get error messages for each usage of StdIn/StdOut. This is not a problem for programs that take command line arguments, or input from the console, but I don't know how to run NBody.java to take input from planets.txt. The only way I know how to do this is through the console, but this does not allow me to use Stdlib. The alternative is compiling in eclipse, but then I don't know how to receive input from a .txt file.


Professor Sullivan said: Best to see the TA's in the Stacks.

Stdlib.jar problems and problem reading files

User 560 said: I have modified the class path on my desktop to include the stdlib.jar file, but i still get error msgs when compiling in the command prompt for every use of stdout/in.

I am also having problems reading input from files in eclipse. I am just trying to read data from planets.txt to test if it works, but when i open the console and run the program, it gives me error msgs when i type < planets.txt. Does the file have to be somewhere specific?


Professor Humphreys said: This has been posted about MANY times on the forums. You need to copy/paste the input into Eclipse. You cannot redirect input like you can at the command line. Open the input file in a text editor (like eclipse, notepad, etc), copy the contents, and paste them into the console window.

Blinking Planets

User 531 said: On my N-body program, the planets follow the desired path, but the further their N-value gets from 0, the more the start blinking when they are being drawn. Body 0 is always present, body 1 blinks a little, body 2 more, and so on and so forth until the planets at the outer edge are hard to follow. Im using a for loop to print bodies 0 to N-1, and reprinting the starfield every time to clear my trails. There is sometimes a visible delay when printing the bodies, you can tell what order they are being printed in. Any advice to make it run smoother?

Lesley said:

Blinking is usually caused by incorrect use of StdDraw.show() but without seeing your code, it's hard to know exactly what is wrong. There are office hours in the stacks right now that you should consider coming to.

Planets velocity

User 531 said: In my Nbody program, planets start out doing what they are supposed to but it seems like their velocities just go up and up and up. using planets.txt, the orbits stay cirular but increase in radius as the planets pick up speed until they are offscreen. My force IS clearing before every frame. I used the formulas a=f*m, V=a*(t-(t/2), and position = v*(t+t/2) like the assignment help said to. can anyone spot my mistake?

User 14 said: I don't think those are the right formulas for velocity and position. The right formulas are on the assignment page. Velocity should be (v_x + Δt a_x, v_y + Δt a_y), and there is a similar looking formula for position below that.

how to play 2001.mid

User 543 said: StdAudio.play() only plays .wav files, right? But the file we want to play is a .mid file. Do we have to convert this .mid file to .wav somehow? And if so, any hints on how we should go about doing that?

Lesley said: Misleadingly the documentation would lead you to believe its capabilities are limited to .wav files. However, it can play .mid files too. Try it. :)

Error message when adding audio

User 223 said:

I tried to add sound to my program, so I moved the 2001.mid file to the src folder, and I also added the code you had metioned in the helpsheet at the beginning of my program StdAudio.play("2001.mid"), but for some reason whenever I run the program an error message comes:

Exception in thread "main"

at StdAudio.play(

at NBody.main(

java.lang.RuntimeException: audio 2001.mid not foundStdAudio.java:147)NBody.java:14)I'm not sure what I am doing wrong. A TA in office hours got it to work on the command line, but I still can't run it in eclipse. The other files I moved into the source directory worked just find so what is different about audio? Thanks for the help.

Professor Sullivan said: You should drag and drop the file into your Eclipse project at the top level. What that actually does is to put it in the bin folder.

Lesley said: She had done that. It was in the bin (and src) folder, I checked. It worked at the command line with no problems, but eclipse refused to recognize the file's existence. ...Any ideas?

User 167 said: If you put a file in a project's src directory, then Eclipse will copy it to the bin directory when it compiles the code. However, for some reason StdAudio seems to look for the file in the root of the whole project, so you need to put the midi file there instead. I saw this problem several times last night, and I don't know why StdAudio works like this; it seems like it's probably a bug in StdAudio, especially since StdDraw looks for images in the correct place. -- Dan

Path of the planets

User 323 said: When going around the Sun, are the planets supposed to follow a circular path or an elliptical one?

Lesley said: The data from planets.txt will model planets taking a circular path. The data from planets-elliptical.txt -> elliptical paths.

Path of the planets, again

User 323 said: When I run my NBody program using planets-ellipcital.txt, one of the planets close to the sun actually goes through the sun during its path. Is this a problem?

Lesley said: No, it's not a problem, though that's not actually what's happening. It doesn't go through the sun; its image is improperly scaled which gives the illusion of it going through the sun.

Massive Loop

User 260 said: I've followed steps 1-3 and, based on my successful groupwork assignments, i know i'm calculating the force and acceleration properly. My planets revolve for a short period of time then slingshot off into deep space. I think what is happening is step 1 is accounting for the forces on the originial x and y positions, but not the new, updated positions. I know the help file said you should not use a massive loop to encompass the three steps, but it seems like that is the only way to ensure the forces are updated according to the new positions. How can i make my planets revolve instead of fly off screen?

Lesley said:

How are you plotting these new positions? Consider storing the coordinates in a variable, which you can reference when you need to calculate forces.

Your (infinite) loop needs to recalculate the force of each planet on each other planet before plotting the position. There's a difference between the infinite loop to keep it going, and the loops your program uses to do each step in the physics calculations.

Professor Sullivan said: There will be a pretty big loop, right? Within the simulation loop you need to calculate net forces on each planet, and that itself requires a nested loop.

standard audio

User 440 said: my stdaudio.play("2001.mid") messes up my program and it gets stuck. what is wrong? it runs without the music perfectly fine.



Lesley said:

How exactly is it "messing up your program"?

Make sure you're not trying to play the file every 20 milliseconds. Using Professor Sullivan's outline:

• ....
• Set up graphical output
  • Set up X scale
  • Set up Y scale
  • PLAY MUSIC HERE
• Run Simulation/Animation Loop -- a while loop
• Update planet positions and velocities -- from group work
• Redraw the universe
• Draw background
• Draw each of N planets -- a for loop
• DON'T PLAY MUSIC HERE
• Wait reasonable number of milliseconds

Professor Sullivan said: That's for sure: you don't want to try to play the tune each time through the simulation loop!

Did Robograde die?

User 146 said: As soon as I upload a new version of a program, it gives me the Compiling (file name here) ... then promptly does nothing. No additional information besides a lovely grade of 1.0. Is this just me?

Professor Humphreys said: This was fixed, robograde is all better now! Sorry!

Audio for NBody wont play

User 66 said:

My NBody program works, but when i try to add the audio i get an error....

audio 2001.mid not found

I have downloaded and redownloaded the file. Right now it is in my src with the rest of the images and everything and they dont error, why is the audio file not working?....it plays on my computer when i open it, so its not corrupt.

Lesley said:

#465

ie. The conclusion is to put it in the folder one level above bin or src. ie. j:\workspace\NBody\src -> it should go in the NBody folder

Image Errors

User 213 said: I've put the NBody image files into my bin folder, but every time I try to run my program it comes up with the error message
"Exception in thread "main" java.lang.RuntimeException: image starfield.jpg not found
at StdDraw.getImage(StdDraw.java:481)
at StdDraw.picture(StdDraw.java:496)
at NBody.main(NBody.java:74)"
What am I doing wrong?




User 14 said: Try dragging them into the Eclipse project. This will put things in the correct place.

resubmit quiz 2

User 244 said:

i have consulted the Zhiheng about this and he suggested that i post this question here.

i went to lab and did quiz 2 last thursday, but i got a zero because i seem to have submitted the program skeleton without my code. is there any possible way for me to redo the quiz and submit it again for a grade?

cast String&gt;Double

User 437 said: What is the code to cast from String to Double from an Array of String?

Lesley said:

You'll need to make a new double array with the same length of the String array and then parse and store each element of the String array into an element in the double array.

"Image Fetcher 0" java.lang.OutOfMemoryError: Java heap space

User 27 said: I added StdIn.java, StdDraw.java, and StdAudio.java to my project alongside NBody.java. However, when I ran NBody from the command line, my audio is choppy and the planets don't even get the chance to load before the StdDraw window crashes. Apparently java is running out of memory, but I have no idea how to fix it.

Professor Sullivan said: James, We really cannot diagnose program malfunctions on these forums. You may not post code, and without looking at code it's all but impossible to say what's going on. We provide TA and faculty office hours to deal with such issues.

looping an audio file

User 380 said: What is the best way to go about looping an audio file to play again once it finishes playing. For instance, in the NBody program, 2001.mid only plays once and then there is no sound if you want to watch the animation further. Is there an easy way to make the audio file play again when it is over?

Professor Sullivan said: Joseph, You could write a while loop that would keep playing it again and again, but then the rest of your code would never run! What you really need here is a construct we won't cover this semester, concurrent threads of control. When you run a program in 101, there is one "thread of control" that snakes it's way through your program executing one line after another. Imagine that you can have several threads executing concurrently. Well, in that case, you could have one running the loop that repeats the music, and another running the loop that does the simulation. If you're interested, search the web for Java Threads.

Group work robograde

User 240 said: I just want to clarify:
We can run Robograde on group assignments and then change things if we don't get 100%, right?
Thanks!


User 14 said: Correct

StdIn Planets.txt

User 489 said:

By following the recommended Procedures for the Gravity Lab, I got the planets read in program to compile and run but it only returns the 5.

String P = StdIn.readString();

System. out .println(P);

Do I need to put the string in an array or something to get the rest of it???

Professor Humphreys said: Hi Grant. readString only gets the *next* string, separated by spaces. See slide 14 of the Input and Output lecture for a description of how the StdIn functions work. They're also documented in the book in section 1.5, where there are many examples.

groupwork3

User 71 said: I know how to read a parameter, but I've no idea how to not read any parameter. When I leave blank in the brackets of the signature it always said syntax error. What am I supposed to do?

Professor Humphreys said: public static void some_function_that_takes_no_parameters() { // Do something fascinating }

Professor Sullivan said: public static void some_function_that_takes_no_parameters() { // Do nothing fascinating }

groupwork3

User 71 said: When I try to return an array of double I wrote"return arrayname;", just like what they did in the book, but the debugger kept saying:"can't convert double[] to double". I don't know where I did wrong.
Thank you.


User 14 said: Make sure your return type is an array of doubles instead of just a double. For instance:

public double[] foo()

instead of

public double foo()

Extra Credit files won't load

User 226 said: The website refuses to upload by extra credit file, even though I used the correct name. Is there something else I can do to make it upload?

User 14 said: Donna came to me after lab today with this question. She said she put everything in extracredit.zip but the submission system wouldn't accept it. The submission is closed now, so I couldn't help her out, but does anyone know why this might happen?

Submission closed 9:55 p.m

User 234 said:

When I was submiting my extracredit.zip and nbody_readme.txt, I've got the message about technical problem. In result extracredit.zip was downloaded, but nbody_readme.txt not(time was 9:55 p.m). After that submission was closed. Could you please help me?

Professor Humphreys said: Can you explain what the "technical problem" was? We got 429 successful submissions.

Click Sound in StdAudio

User 486 said:

So I wrote this program that plays the mario theme song tonight that ends up playing a bunch of notes quickly, and the tones and lengths are right, but it doesn't sound very good because it keeps clicking; is this just because because java/eclipse/computers are not supposed to actually play songs this way, or is there a way to make it sound smoother?

-Jamie Johnston

Professor Sullivan said: Jamie,

First, good work for going above and beyond the required work. It's just great when we hear that people are starting to write real code.

My suspicion is that you're calling play many times, once for each tone. If you do this, there will be short gaps between tones, and I wouldn't be surprised to hear them as pops. Try creating a single big long array, and see how that works. You might, for example, define a new function like this:

public static double[] append (double[] a, double[] b) { ... }

This function would return a new array of length a.length + b.length, whose first a.length elements are identical to those of a, and the rest of which are identical to those of b. You could then use this function repeatedly to string together a bunch of separate tone arrays.

KS

sumArrays.java

User 236 said: the goal of this program is to "Write a function sumArrays that takes as parameters two arrays of doubles of equal length and that returns a new array of doubles of that same length equal to the sum of the two given arrays. (Add elements at index i in the given arrays to compute the value of the new array at index i)."

What output should I have if the arrays for two of the tests in Robograder are cases when the two arrays do not have the same dimension?
false?

and I am having the same problem as Jun Wang, and yes, I have what you said to him, Mr. H.


Professor Sullivan said: Your question is a really good one. In general, as a designer of a function, you have to decide whose job it is to make sure the function is used correctly: is it the job of the function to check (and to do something special in case of misuse) or is it the job of the client (who risks a serious malfunction if it uses a function incorrectly -- since the function itself doesn't check)? In this case, we've clearly decided that it's the client's job to use the function correctly. What that means for you is that you don't have to check or to do anything: just assume that you get arrays as parameter values that satisfy the requirement: the arrays must be (and are assumed to be) of equal length. Note: if you want to write really robust (dependable, bug-free) code, then you want functions to check that they are being used correctly. If you want to write code that's simple and fast, then you don't want functions to have to do such checks. As in much of engineering, it's really a matter of trade-offs, and there is no answer that's best in all cases.

KS 

Professor Humphreys said: These are all excellent points! However, the assignment said that they can assume they are equal length; my test unintentionally violated this. I have updated the test to be in line with the spec. Sorry.

Groupwork3 - tone

User 33 said: Can someone explain exactly what this project is? I've been skimming all the functions to write, and this one doesn't make sense. For example, how big is the array that i'm returning? Is it a function of time, or the frequency?

Professor Humphreys said: It was covered in lecture.

Groupwork 3 - Credit Card

User 375 said: Robograder is asking for "validCreditCardNumber.java" and the submission screener is looking for "validateCreditCardNumber.java."

Either I cannot submit the program robograder is looking for, or robograder will not bother to look at "validate..."

thanks,
alan


Professor Humphreys said: Whoops! I really appreciate the bug reports. I hope it underscores the fact that it's incredibly hard to get software right in ever last detail :) (It's fixed)

lab today??

User 98 said: do we have to go, or like is there a file we have to submit at the end?

Professor Sullivan said: There is lab today, as there is every Thursday. Your TA will explain when you arrive.

More Mario Help

User 486 said: So I added

public

{

e[f] = c[f];

e[g+c.

}

to the beginning of my code, and it works for a while in the program, but I get a heap space error after about 75 notes; this is even after I thought I reset the note array. Do you know why it would do this?

static double[] append(double[] c, double[] d) double[] e = new double[c.length+d.length];for (int f=0; f<c.length; f++)for (int g=0; g<d.length;g++)length] = d[g];return e;

Professor Sullivan said: Your posting seems to be malformed. The code as posted is syntactically incorrect and I can't tell from it what you're doing. Feel free to send it to me in a separate e-mail.

Assingment 2 after the fact

User 305 said: i know its after the fact but i was wondering if a TA or someone could help me get assingment 2 to function properly, i have everything but getting the planets to rotate, i just want to know why its not working, my physics and initial image is fine, but for some reason inside my while loop positions aren't being regenerated, just wondering if anyone was available on campus today to go through this real quick. Thank you.

Lesley said:

I'm not sure what the policy on 101E students attending 101 labs is this semester... I say this because the solution was discussed in labs on Thursday.

Professor Humphreys made sure to go over assignments' solutions in class last semester. I'm not sure if he's planning to do this again this semester, though.

Feel free to come by the stacks Sun, Mon, Tues from 6-10 and a TA will be more than willing to help you then. With ~500 students, it's really impossible to offer help outside of these specified times (even just a few minutes before or after in the stacks), especially for the sake of fairness.

Professor Sullivan said: Jose,

We'd also be glad to see you in our office hours, e.g., on Monday. I'm glad you're determined to debug your program. In the meantime, may I suggest that you try two strategies: (1) read your code super-carefully and mentally simulate what it does; (2) use the debugger to watch it operate. Pay attention to small issues: an x here instead of a y, for example.

KS

TA's at Stacks on the 14th???

User 489 said: Just checking to see if the 101 TA's are suposed to be at Stacks today from 18:00 to 22:00????

Professor Sullivan said: Yes: Sunday - Tuesday.

Groupwork 3, Question 4

User 264 said: I've tried to do #4 by following the relevant example in the book, but it says "the method must return a result of type boolean". What am I doing wrong?

Professor Sullivan said: If you declare a method as returning a value of a given type, then the method must actually return a value of that type. In fact, you have to be sure that on any given execution path, it returns a value of that type. The following is wrong, for example, because there's an execution path on which the method fails to return a value of the declared type.

public static void boolean foo(boolean bar) {
if (bar) return bar;
}

What happens if bar is false? The method fails to return any value. That will give you the error you're seeing.

No 101 TA's at Stacks tonight....

User 489 said:

Update on the TA's at Stacks.

I'v been here since 17:55 and have only seen TA's for 201 and 126. No one here has seen any TA's for 101.

I'm giving up.............. Mabe some one can help Tuseday....

User 14 said: TA Office hours for cs101 are Sunday-Tuesday from 6-10 in the stacks. These office hours are made available on the main class page. There are no cs101 TA office hours on Thursdays.

Capitalization in Groupwork #17

User 380 said: Should we discriminate between capital and lowercase letters in this program (i.e. if you search for the char 't' and there is a 'T', should the count include this or no)?

Professor Humphreys said: Just search for the provided character. 't' and 'T' are not the same.

Group Project 3 errors

User 323 said: For the first few programs that I tried for Group Project 3, I get the following errors that extends to more than a 100 lines.

ERROR! COMPILATION FAILED! Compiler error messages below:

identity.java:1: class, interface, or enum expected
{\rtf1\ansi\ansicpg1252\cocoartf949

and

identity.java:14: not a statement
\cf2 int\cf0 a = Integer.parseInt(args[0]);\
^

I'm not sure why I'm getting those errors. Could you help me so I can fix them?


Professor Humphreys said: looks to me like you accidentally saved your java files as "rich text" format. Make sure your files are plain text (e. g. Edit them in eclipse)

Group Project 3 errors (contd)

User 323 said: I checked that but it seems like the file that I submitted is a plain text format. I changed the names to programname.java, and I also checked its property.

One thing that I've noticed is that all of my programs for Group Project 3 gives me an exclamation mark (not an error) and a yellow wavy line under for a particular line, which is the second line where you declare your function. (it's the portion that is underlined)

public static int boo(int a) {

When I click on the exclamation mark, it gives me the option of 1. Change modifiers to final where possible & 2. Change to Constructor.


Professor Humphreys said: *I* checked your files -- they are RTF, not Java. Please use Eclipse to edit your files, and submit only plain text.

Recursions

User 166 said:

I was wondering how recursions are read. Like for program 2.3.4 (Recursive Graphics), draw is called upon 4 times within the function. So when draw is called the first time does it goes back to the top and restarts the draw(. . .) function, or does it read the rest of the draws and then restarts the function?

public static void draw( . . .)

{

if (base case) return; // Sullivan added this line to the original message

. . .

draw(. . .) --> like does this goes back to the draw function

draw(. . .)

draw(. . .)

draw(. . .) --> or the the rest of the draws get read before it goes back to the function?

}

(Edited by Kevin Sullivan - original submission Saturday, 16 February 2008, 03:02 AM)

Professor Sullivan said: Lynna,

Great question. Let's call the occurrences of draw top, call1, call2, call3 and call4.

You start off by calling top. Let's call this the outer call to top. This outer call to top then either returns immediately (in the base case) or goes on to call1, a recursive (nested) call to top.

Suppose that the outer call is not a base case, and therefore does not return immediately. Control flows on to the recursive call1. Top calls itself. Control is now in an inner (recursive, nested) invocation of top. This recursive call now either returns (in the base case) or it goes on to call1 (again), which would be another, more deeply nested, recursive call. Suppose that the base case occurs. The inner call to top returns. Where did that call come from? It came from call1 in the outer invocation of top. This call1 call is completed, and so control in the outer invocation of top now passes on to call2, and the recursive process starts all over again. Once all the nested invocations that result from call2 have completed, control passes to call3, then to call4, and then finally the outer invocation of top is done. Of course, that outer call could actually have been nested inside a surrounding recursive call! Eventually, though, after getting wound up and unwinding and getting wound up again and unwinding again, etc., etc., the recursion does terminate, with the job done.

A really good way to see how this works is to run a recursive program several times in the debugger: the first time with the base case, the second time on a problem one step away from the base case, the third time two steps away.

KS

Standard In- reading chars

User 200 said: After working on the nchar part of the groupwork, I tried testing it by using StdIn to read one String and one char as parameters. However, the program read both input values as Strings, and even though I used StdIn readChar on the second parameter, I still had to use another function to convert that value into a char. Is there a problem with the readChar function, or is there a certain way I have to input the char when I run the program?


User 14 said: It looks like StdIn.readChar will try to read in spaces and new line characters. If you want to read in the next non-blank character, you'll have to use something like StdIn.readString().charAt(0). That will read in the next string and parse the first character into a char.

conversion from int to string

User 323 said: how do you convert a int type array into a string type array?
and another related question:
how do you convert an individual element of an int type array into a string?


Lesley said:

1. similar process to #490 aka you need to do it element by element, there's no magic code you should use to convert a whole array at once. Also keep in mind the easiest way to convert an int into a string (see #2 below).

2. strarray[i] = intarray[i] + "";

Referring to the last character in an string

User 323 said: if you have a string of unknown length and you want to refer to the last character of that string, how would you do that?


Lesley said:

There is a function charAt() for a String. charAt(0) returns the character at position "0" (if you took the String and made in an array of characters, the 0th element) - for instance

String s = "Test";

System.out.println(s.charAt(0) + "," + s.charAt(1)); // This prints "T,e"

There is also a function s.length(), see if you can figure out how to correctly use these two functions together. Try doing things like

1. s.charAt(s.length() + 1)
2. s.charAt(s.length() - 1)
3. s.charAt(s.length())

and see which ones throw exceptions and which one does what you want.

Groupwork #7

User 124 said: Is there some sort of method/formula to find the greatest common divisor between two integers?

Lesley said:

http://en.wikipedia.org/wiki/Greatest_common_divisor#Calculating_the_gcd

There are three different ways I can think to implement it off the top of my head (there are plenty more, I'm sure): the euclidean algorithm (easily done with recursion), a loop where it tests if every number between the minimum of a,b and 1 (in that order) divides a and b (stops upon finding one), and finding the prime power factorization of a and b and then multiplying together common terms. I'm thinking that 3rd one is a bit much though and that one of the first two options would be a better solution to try and implement, but if you are bored and want a challenge...

nchar(#17) problem

User 270 said: I have been thinking about this problem for really long time and still can't figure out a way to attack this problem. I tried to go stacks today to ask TA for help, but i didn't see any TA there. Only operation we can perform on String is concatenate, and we can't perform any operation on char type. Also, we can't even convert them from one to another. After thinking for really long time, I wonder if anyone can give me a clue on how to attack this problem. Thanks.

Lesley said:

I was there. There is a list up by the ITC desk which people signed their names on to get help since I was the only one here. This is the procedure for most nights actually.

The key to nchar is probably the function chatAt. I have already posted how this function is used in a previous thread tonight. Think about this process:

How do we know how many 'L''s there are in the string HELLO?

H - not an L

E - not an L

L - that's 1... We had none, we increment by 1

L - 2; we had one, now we have one more, we increment by 1

O - not an L

So after looping through the string, we arrive at the conclusion that there are 2 Ls. See if you can figure out how to make this into a program.

Unsubmitting Group Work

User 179 said: Just curious as to where the "Un-submit" option on the groupwork went.

Professor Humphreys said: The "unsubmit" button was never needed and only caused confusion. Just submit the work again, it will overwrite the old one.

what is the difference?

User 49 said: what is the difference between carriage return and line feed? Also, what is the difference between Stdout.print(xxx); Stdout.println(); and just Stdout.println(xxxx) thanks

Professor Sullivan said: Carriage return and newline are actually separate characters. New line advances the cursor down the screen one line. Carriage return moves the cursor back to the beginning of the line. When we say "start a new line" that translates into "print a carriage return and a line feed." StdOut.print(xxx) prints xxx without a carriage return or line feed at the end. StdOut.println(xxx) prints xxx with a carriage return and line feed at the end. StdOut.println() prints just a carriage return and line feed.

KS

Using Color

User 166 said: If we wanted to use color in our Art.java program what function would that be? Cause I've tried using StdDraw.setPenColor(CYAN) or something and CYAN doesn't work, but if I take it out it defaults to black. And how would I draw filled shapes with different colors (like StdDraw.filledCircle(...) is a black circle).

Lesley said:

StdDraw.setPenColor(StdDraw.CYAN) will work.

Similarly use StdDraw.ANY_STDDRAW_COLOR_CONSTANT_HERE inside any of the functions that take a Color as a parameter.

If you wanna get really spiffy you can do something like this:

import java.awt.Color;

public class Art {

public static void main(String[] args) {

Color mycolor = new Color(120, 92, 133);

StdDraw.setPenColor(mycolor);

}

}

These are RGB (woahhh remember THAT assignment??) values (0-255). You can google java.awt.Color for more details (including things like alpha support).

factorial GroupWork 3

User 393 said:

14. Write a function factorial that takes as a parameter a single value n of type int (which your program can assume is non-negative) and that returns a value of type int equal to n!

I am not sure what this question is asking for, the n! is confusing.

Here is what I understood it to mean, but the test failed.

Verifying the existence of a proper factorial function in factorial.class
Found one! Now I'll go ahead and test it.
According to your software, factorial(1) = 1
According to your software, factorial(5) = 5
But that is not correct.
According to your software, factorial(8) = 8
But that is not correct.
According to your software, factorial(3) = 3
But that is not correct.
According to your software, factorial(10) = 10
But that is not correct.
TEST FAILED: Factorial didn't seem to work. Please see the above messages.

Professor Humphreys said: This will help: http://en.wikipedia.org/wiki/Factorial

Parameters for sierpinski

User 200 said: I was wondering how it would be possible to get the recursion working for sierpinski by only taking one parameter. I looked at the help page and got it to the point where it works by taking 4 parameters (n, x, y, and size), similar to H-tree, but that seems like the only feasible way to do this.

By only taking n as a parameter, that means that x, y and size have to be functions of n. However, since n could start at any number, yet the initial point and size have to start at the same values no matter what, there's no easy way to do that. Instead, it would make sense to have n start at one, make the other variables functions of n, and count up n to the desired degree of depth, but since we have to recursively call this function that supposedly takes only one parameter, there seems to be no efficient way to keep track of such a counter.

Would we be allowed to write one function that takes n as a parameter, but then recursively call another function that takes 4 variables n, x, y and size as parameters? I keep trying to come up with other ideas, but this seems like the only plausible solution, given what we've learned so far.




Professor Humphreys said: The largest triangle is supposed to be roughly centered and take up the entire window. That *does* make x, y, and size functions of n, in some sense. Your recursive function itself should take four parameters. Your PROGRAM should only take one, on the command line.

Strict comparison in totient

User 449 said: Hey,

When my group ran the totient problem for the first time, it worked except for the case where our program gave:

totient(1) = 0

where the correct answer was:

totient(1) = 1

The program specification is: ...returns the number of integers less than n that are relatively prime to n.

If this comparison is strict, then there are no integers less than 1 relatively prime to it -- however, if integers can be less than or equal to n, this includes 1 in the set, which is relatively prime to itself. I didn't notice any other posts about this -- did I misinterpret something?

-Joel



Professor Humphreys said: Excellent point. The spec is wrong. It should say "less than or equal to". I will fix this.

Appointment with TA

User 71 said:

I need to summit a regrade request this week, so I will appreciate if any of you would have time available tomorrow after noon after 4, or Friday after 3, or at weekends. Thanks a lot. Jun

Lesley said: There are office hours Sun, Mon, Tues. Just bring it by then and it will get taken care of.

Office hour and a maybe not simple question

User 71 said: 1. I'm sorry but I forgot the specific office hours times, and I can't find them anywhere, so could anyone tell me when they are, thanks a lot.
2. I wrote a simple program to randomly pop up a picture of a building to help me prepare for my history exam, but is there a feasible(I mean feasible for somebody like me..) way for me to change the program to an ".exe" or anything like that(or Java runtime or whatever?), so I don't have to use cmd or the Eclipse to run it every time, thank you very much.
Even it's hard, I would love to know if it won't take you too much time to explain it here.
Thank you very much


User 167 said: Office hours: $@RESOURCEVIEWBYID*68@$ This is on the sidebar from the main moodle page (labeled "TA Office Hours"). As for executables, you probably don't want to make your code a .exe. Remember that Java runs inside an interpreter - a standalone .exe would need to contain the entire interpreter to run. However, you can package your code into an executable .jar file, and I believe you can do this through eclipse. When trying to figure out how to do something like this, google is an invaluable resource. Here are a few sites I found that might be useful (I googled "executable jar eclipse") : http://www.cs.uta.fi/pgui/2005/local-uta/tutorials.html Discusses how to use Eclipse to build Java Applets and executable Jars. http://java.sun.com/docs/books/tutorial/deployment/jar/ Describes how to build and manipulate jars directly, without needing Eclipse

Professor Sullivan said: Thanks, Dan, that's exactly right: Export as executable jar. --KS

Canvas Color

User 236 said: Is there a way to set the canvas (what we draw on) color for out stddraw window with out declaring a JFrame and such? I want to have a background color that will make the colors of my drawn recursions more visible.

Thank You.


Professor Humphreys said: The method you want is: public static void clear(Color color) So StdDraw.clear( StdDraw.BLACK ); will clear the screen to, well, black.

Individual Assignment 3 Question

User 270 said: I wonder if I can use two recursion functions in the part 2 of the Individual Assignment #3. The second recursion function is very simple and has a fixed number of n. I just added it to make it looks more creative. I wonder is that ok.

Professor Humphreys said: absolutely! For fun, see if you can do something that is *mutually* recursive -- function A calls function B, and function B calls function A.

Canvas scale problems for individual assignment

User 188 said:

Hi:

I'm working on my individual assignment, but for some reason when I call the triangle function the triangle doesn't appear on the screen unless I set the X and Y scales. I know we're not supposed to set the size of the canvas, so what's going wrong? Is there any way I can force it to appear in the centre of the canvas?

Thanks,

Becca

Professor Sullivan said: If you don't explicitly set the X and Y scales, then they are (by default) 0.0 to 1.0

Groupwork 3 Robograde

User 71 said: It said there is a compiling problem with my #21, which I didn't encounter when running it on my laptop, and it seemed that what it is looking for ValidateCreditCardNumber, not what is asked in the assignment "Valid"CreditCardNumber.
my error messgae: class validCreditCardNumber is public, should be declared in a file named validCreditCardNumber.java


User 167 said: Ok, here's what's going on behind the scenes: When you start a file with something like "public class validCreditCardNumber" you're creating a new class, and all the functions you're writing are actually static members of that class. You'll learn a lot more about what that means later on in the course, when we get to classes, but for the moment just remember that all of the functions you define in that file are part of the class, and the name of the class is "validCreditCardNumber". The thing is, Java requires that every public class (again, you'll understand the term "public" better once we've gone over classes) be defined in its own file, and the file needs to have the same name as the class. Thus, if your file starts with "public class validCreditCardNumber", then the file MUST be named "validCreditCardNumber.java". In your case, it isn't - it's named "ValidateCreditCardNumber.java". Since they don't match, you should expect to get exactly the error above: class validCreditCardNumber is public, should be declared in a file named validCreditCardNumber.java There are two ways to fix this error. One is to rename the file to validCreditCardNumber.java, so that the filename matches the class name. The other is to rename the class to ValidateCreditCardNumber, by changing the appropriate line to read "public class ValidateCreditCardNumber" Of course, one of these changes will let you submit the file through Robograde, and one of them won't. In this case the confusion appears to be our fault - the assignment asks you to use a class named after the validCreditCardNumber function, while Robograde expects you to use the name "ValidateCreditCardNumber". We'll fix this shortly and announce which one we actually want, but for now you can use the name ValidateCreditCardNumber - this will let you get feedback from Robograde immediately, and we'll let you know when we get this set what name you should ultimately use. By the way, a useful Eclipse shortcut: If you want to change a filename AND the name of the class in it at the same time, right-click on the file in Eclipse, and select "Refactor -> Rename" from the menu. This will even automatically change the name if you used it in other java files (again, you'll understand why you'd want to use the name in other files once we do more with classes).

Professor Humphreys said: The submission system will ONLY accept files named "validateCreditCardNumber.java", so the function should have that name. From the beginning of the assignment: In each instance, you should include your function inside a class of the same name as the function. Contrary to Dan's post, none of the functions, class names, or file names should be capitalized.

Assignment 3 - Colors

User 177 said: I'm working on Art.java and I think it's kind of boring now. I was wondering if there is a way to get java to make different color circles.
Thanks - Macie


Lesley said: #550

uninverted triangle outline on assignment page Assign 3

User 305 said: The assignment says "First, make sure that your program draws a single black triangle when N is set to 1." but i see an outline of an uninverted triangle on the assignment page, do we need to include that outline triangle in our assignment and then draw the solid inverted black ones inside it?

Lesley said: The grey outline triangle was optional last semester. I'm assuming that still holds.

Professor Sullivan said: Folks, I'll step in on this one: No surrounding triangle, please. Just follow the written instructions.

Exam 1

User 305 said:

I might have missed the announcement, but when will we get our Exam1 back in 101E?

Professor Humphreys said: tomorrow

Regrades

User 305 said:

Regrades are supposed to be written, can they be typed and turned in, in person, or actually hand written?

Professor Humphreys said: Typing is okay. Just so long as there is a paper trail.

Art.java

User 86 said: I added a picture to my Art.java design. Is this going to affect the grading process since there is no way to attach the .jpg file? Thanks!

Professor Humphreys said: Put the picture on the web somewhere, and put a comment at the top of Art.java indicating where we can download the picture.

Sierpinski Triangle

User 166 said:

1. So we can't just draw the Sierpinki Triangle with an imput of 4 parameters? (n, x, y, size), it has to be just 1 parameter?

2. And when writing Art.java does that have the same idea (an imput of just 1 parameter)?

3. And when you grade the program, will the main function be tested too (for both programs)?

Lesley said:

Your API will look like this:

class: Sierpinski

• triangle(x, y, size) //draws a triangle
• sierpinski(n, x, y, size) //recursive caller of triangle
• main(args) // N is a command line argument; calls sierpinski

All of those things need to be present (you need a "main"). I'm assuming your Art.java will look very similar in structure. If you look at HTree, it has three methods (drawH, draw, and main) which are comparable to the ones outlined above.

Professor Sullivan said: Lesley is right. If you're lost on Sierpinski, go back and study HTree (which we went over in class in great detail). We're confident that if you download and run and study HTree, and then think about the Sierpinski requirements, you will see that the same basic scheme works for both programs.

Art.java

User 166 said:

I didn't read the assignment correctly so I created a sierpinksi function that takes in 4 parameters and calls itself 3 times (n, x, y, size). So I don't have a triangle function, but it still works. But I can't do that right? I have to create the triangle function to be graded and I have to make Sierpinski function take only 1 parameter, but you don't need a triangle function, you could do it in 1 function. Also is that true for the Art.java assignment? I've created a bunch of functions that doens't call on another function, it just calls upon itself a couple of times. So do I need to call on another function or do I not have to? Or does Art.java not matter?

Does that mean the x, y, and size is up to our choosing? (cause that makes life a lot easier).

And then we have to set the scale to fit our triangle?

Professor Sullivan said: Lynna,

> I don't have a triangle function, but it still works

You probably have code that draws a filled triangle embedded in the sierpinski function . All you have to do is to restructure your program, putting that code into a separate function. You should do this in order to get full credit. Prof. H posted a message a day or two ago restating precisely the functions that your program must include.

I don't yet fully understand your other questions.

KS

question

User 440 said: Mr. Sullivan
I did my sierpinski triangle starting from the middle of the first triangle. It made it more difficult but it works the same way....Is that ok?

Color

User 177 said: Where can I get java.awt.Color?

User 14 said: add the line:

import java.awt.Color;

to the top of your file.

Outline of Triangle

User 218 said: The following is on the assignment directions:

"Also, when writing your program, exercise good design: include a (non-recursive) function triangle() that draws an equilateral triangle of a specified size at a specified location."

However, on a previous forum topic Professor Sullivan said that you don't need to include it.....so which is it?

Professor Humphreys said: Those statements are not contradictory! The triangle people were asking about was just the faint outline around everything else. *THAT* is not necessary.

Edge lengths of triangle

User 347 said: Is the size of the overall triangle in Sierpinski.java supposed to be a function of n or is it supposed to have edge lengths of 1 as suggested on the help page?

Professor Humphreys said: Do whatever you need to do so that the overall triangle fills up most of the window.

Art.java

User 220 said: I made an Art.java program that looks a lot cooler when N is greater than 7. May we indicate somewhere that we would like this done and where, if we can?

Professor Humphreys said: Probably in the readme. Alternately, just have your program print an informative message when it starts: (in main somewhere) System.out.println( "PLEASE RUN THIS WITH AN ARGUMENT OF 7 OR MORE FOR MAXIMUM EFFECT" );

How many methods?

User 347 said: How many methods should our final Sierpinski.java have?
Should it be:
Class Sierpinski
triangle
Sierpinski
main

or should it just be:
Class Sierpinski
Sierpinski
main

or does it matter?


Professor Humphreys said: The assignment says to write a triangle function

Professor Humphreys said: The assignment says to write a triangle function

Random colours

User 188 said:

Is there any way to alternate colours or randomly select colours as part of the recursion? My Art.java looks rather dull in black and white outlines, but if I do filled all in one colour you can't see anything.

I've tried assigning different StdDraw.COLOUR values into an array, then geenrating random numbers to get them out, but that doesn't work.

Anyone know how to do this?

Thanks,

Becca

Professor Sullivan said: For each of the red, green and blue channels you might try "some function of n" mod 256. The mod 256 guarantees that the result is between 0 and 255.

question

User 440 said: I did my program differently than everyone else. Instead of starting at (.5, 0) I started at the middle of the triangle......this made it way more difficult but it works the right way.....is this ok?

Professor Humphreys said: As long as you have a recursive function and a "triangle" function, you're fine.

What does this mean?

User 198 said: What does it mean whenever Eclipse says "This method has a constructor name." with the yellow squiggly line on the method header? hm.

Lesley said:

public class MyClass {

public static void MyClass() {

}

}

will give this "error." When you learn about Object Oriented programming later in the course, you'll learn what a constructor is and why it's giving this error. For now, it's not something you need to worry about, but you can discern that it happens when a class has a method with its same name (case sensitive).

art.java

User 266 said: Is there anyway we know if our Art.java is "creative" enough???

Professor Sullivan said: Do your best in the available time to create something interesting, moving, colorful, fun, funny, deep, artsy, whimsical, inspired, aesthetic, or whatever. What more can you do? Oh, and have fun with it! --KS

art.java additions

User 211 said: If we want to add music to our art.java program to accompany our creative recursive program how can we upload the music file?

Professor Humphreys said: I've added an optional upload.

zip

User 266 said: how do you make a zip file to submit the artwork with our Art.java program

Lesley said: Download some shareware called Winzip and then you can right click files to zip them. Dunno if there's a built in function in Windows. No clue about Macs.

#21 -vCCN array length

User 531 said: are we to assume, from the example, that the input data will always be 9 digits long? If not, are we given an input value with which to initialize the aray?

Lesley said: Well you're given an array as your parameter, so you can just use array.length like you've been using all along. Though I think Robograde only tests against 9 digit inputs.

StdDraw.setCanvasSize

User 326 said: I know the help for the assingment says not to use set canvas size, but are we allowed to use StdDraw.setXscale and StdDraw.setYscale? If I don't, I can't get my drawings to show up where they are supposed to.

Lesley said:

Those functions are all equivalent. You should be able to write a program which is not dependant upon using any of them.

If you hadn't waited until the last minute, I would have suggested going to office hours so a TA could help you make your code work without them.

I don't know if points are being taken off for using these functions. But your sierpinski and triangle functions shouldn't be dependant upon window size at ALL so you should just be able to modify the original size and x, y you pass to sierpinski in your main routine to make it draw correctly.

Professor Sullivan said: The default X and Y scales are 0.0 to 1.0. The coordinates of your points will (should) all be in this square.

midpoint of a line

User 221 said: is there any java language in the math library that calculates the midpoint of a line when given two endpoints? i was thinking of using this in my sierpinski triangle code...

Professor Sullivan said: You can just calculate it yourself.

If the coordinates of the two end-points are (x0,y0) and (x1,y1), then what are the coordinates of the mid-point? What is the x coordinate of the mid-point? What is the y coordinate? How do you calculate the number halfway between two given numbers?

name of function for Art.java

User 323 said: what should the name of the function be for Art.java?

Lesley said: Whatever you want it to be!

Professor Humphreys said: public static void whatever_you_want_it_to_be( ... )

uploading Art_extra.zip

User 376 said: When I try to upload my Art_extra.zip it only says "Preparing to upload Art_extra.zip..." and then has the continue button. It never says upload complete. I don't know if it is my computer or what, but it never uploads.


Professor Humphreys said: Not sure why that would be happening, but your file has uploaded successfully.

Saving StdDraw as other file type

User 380 said: When I save my StdDraw images as a jpeg, they always come out blurry and nowhere near as crisp as in the stddraw window. Is there a way to save these files in another format, like a bmp?

Lesley said:

You can save as .png which might make it clearer.

You can also press the "print screen" button on your keyboard (up near the top usually) and then open a program like Paint and press ctrl+v aka paste and magically your screen will come up. To get just the current window press alt+prntscrn. Then you can save as .bmp

Art.java

User 31 said: Can we implement iterations within our functions to create fractals?

I guess the crux of my question lies in how much of the program must be done in recursion.


Professor Sullivan said: You can certainly include iteration in addition to recursion. But you have to use recursion. --KS

jpg

User 120 said: Did we have to have the 2 additional lines that turned our art into a jpg file when we submitted our art to moodle? I submitted my project earlier this week and didn't have the 2 save lines in it.

Professor Sullivan said: NO

redirect standard input from file to terminal

User 305 said: Is there any way to redirect standard input from file to terminal within the same program? That is, get data from a file then make it interactive with the terminal?

Professor Humphreys said: Not easily. If you want to both read from a file *and* the user, you're better off just using the "In" class (see the book website) and explicitly opening / reading from the file in your program. You can just pass the name of the file to read as a command line argument.

Eclipse

User 166 said: Is there a reason why Eclipse won't open up any of my programs anymore? Like the src file no longer exists under the java project, but if I open up my J drive, it's all there and I can open up the program in notepad or something. Now all of my projects have that red error x on the project like something's wrong, but the program runs fine out of the command prompt.

User 167 said: That probably means Eclipse is confused about your workspace. This usually means one of three things: 1) Eclipse just got confused. Usually this means something interrupted it while it was loading, with the result that Eclipse has forgotten about some of your folders. To fix this, select all your projects, right-click on one of them, and select 'Refresh'. 2) You changed your workspace. In this case Eclipse will show whatever's in the new workspace. To fix this, go to 'File->Choose Workspace' and reselect your usual workspace. 3) The .metadata folder, or one of the .project files, was moved or corrupted. This is extra information that Eclipse stores in the workspace to help it keep track of all the extra information it has about your projects. To fix this, reimport - In the list of files on the right (where it normally lists your projects, java files, etc.) right-click somewhere empty. In the resulting menu, select "Import..." IN that dialog, expand "General" to find "Existing projects into workspace". Then select your workspace as the root directory - it should give you a list of projects that are in the workspace folder but not registered as part of the Eclipse workspage. Check all of them and hit "import" to load them. If none of these work, come talk to a TA during office hours.

mergeSort merge function

User 380 said: When I looked at the merge function in the mergeSort groupwork program, it says it takes only one array as a paramter, but if we are merging two arrays together, shouldn't it take 2?

Professor Sullivan said: MergeSort sorts an array of numbers. The function takes one array. Read on: The way that it does is by splitting the given array into two sub-arrays, sorting each of them, and then merging the two sorted halves.

Lesley said:

It's not merging as in array "a" has 5 elements and array "b" has 2 and array "c" is array a's 5 elements followed by array b's 2...

It's merging as in the decks of cards analogy given in the file. And really instead of two piles, it is one pile where it is sorted until a midpoint (ie. 1, 7, 2, 4, 9, 6, 8, 22, 18, 3; 9/6 is the mid point and on either side things are sorted between a lower and upper bound) and merge() "merges" these "two (sorted) arrays" together, from lower to upper (around that midpoint). (So after merge worked on that PART it would go 1, 7, 2, 4, 6, 8, 9, 22, 18, 3)

RE: mergeSort merge function

User 380 said:

right, i understand the part about breaking the array into halves, until there are just a bunch of 1 number arrays. but then we are supposed to call the merge function to merge the left and right halves. Is this still supposed to be done taking just one array as a paramter?

Lesley said: I replied to your original post. I hope it clarified.

bubblsort

User 71 said: it says: "/* * Return the value of the largest element in the "suffix" of" for the indexOfSmallestElementInSuffix function, should that be "return the value of the smallest"? and it seems that (I might be wrong because I just take a brief look at it) it dosen't fulfil its goal(which it should, because the bublesort does anything except sorting the numbers)

Professor Sullivan said: Many teachers teach that code should contain extensive explanatory comments. A problem with this strategy is that comments can be inconsistent with code. You see that here. What I'm trying to say is that I screwed up when I wrote the comment. Yes, it should say "smallest"! Thanks for pointing out the error. Just go ahead and write the code we've asked you to write. The incorrect comment is confusing, but if you ignore it, you'll be fine.

MergeSort

User 71 said: I now came up with an algorithm that only works if you call the MergeSort passing it the whole array, so I wonder if we have to create an algorithm that works for any kind of calling (which seems not required by the main program)? Thank you

Lesley said: "passing it the whole array" - That's the correct implementation. :)

Groupwork4

User 71 said: I'm sorry about this, but there seems to be a lot of inconsistency in the comments among which I don't know which specific one to ignore. Are the comments basically the requirement for not only what the functions do but also how they do them , or should I just look at the assignment page and ignore all the comments and write the program in my own way? One more question, could I create my own sub-functions(I mean functions that would be called by the functions that are already required)? Thank you very much.

Professor Humphreys said: You can, of course, add your own functions if you think they are helpful. It's important that you implement the algorithm as specified (on the assignment page); our grading scripts will test that the array is sorted *and* that you used the right number of comparisons.

MergeSort Again

User 380 said: Since the merge function inside the mergeSort class returns void, could we just eliminate the merge function and include the code that would go in that within the mergeSort function?

Professor Humphreys said: You can always write any program in just one function, but why would you want to do that? Good abstractions and functional decomposition are what differentiate well-written working programs from poorly-written working programs.

Professor Sullivan said: Right. The program would still work, but it wouldn't be as easily understood by people who read the code. There are, in general, many different programs that compute the same function, but they're not all equally good with respect to a range of important factors, including ease of understanding and ease of change.

changing output of merge

User 380 said: Can we change the output of merge from void to something else then, such as double[]?

Professor Sullivan said: In general, yes; for this particular exercise, no. --KS

auxilliary arrays

User 380 said: sorry to keep bombarding you with questions, but is there a way to set up an auxilliary array that can be referred two in two separate methods without having to return the array type? i.e. i want mergesort and merge to be able to reference the same array.

Professor Sullivan said: If you declare a variable outside of several functions in a class, then they can all access it. Such a variable is called a "global" variable. Using global variables is a practice that is often best avoided, but sometimes it's just what's needed.

Sorry, but Merge this time...

User 71 said: I might be an idiot, but have we learned how to use private static method? And since all functions seem to be set to static method, I guess we don't have to worry about the instance variables?

Professor Sullivan said: Private simply means "cannot be called by a method in another class." The "API" is the set of *public* methods that a class defines. Additional private methods are "for internal use only."

RoboGrade for Pascal

User 54 said: For the group work will we be able to RoboGrade for the Pascal file? I uploaded my file and there was no Robo grading option for it.

Professor Sullivan said: It will appear "soon" -- sorry for the delay.

RoboGrade for MergeSort, Count??????

User 305 said: Please clarity the RoboGrade parameters for MergeSort, my program puts the elements in order but it doesn't match RoboGrade's count, does my program have to match the count exactly to get all the points? Not clear on what needs to be done to get all the points for this part of the assignment.

Professor Humphreys said: yes, they must match. Please re-read the robograde output.

can we change the function to non-static?

User 71 said: otherwise even if I create varibles outside methods, I couldnt find a way to let Mergesort and merge refer to the same array without any return, because for recurssion I have to constantly change the length of the array

Lesley said:

Think about it this way instead:

You have your array that is being passed, called "arr" and as you saw in one of the groupworks, you can modify this array without returning it, ie saying inside the method that arr[0]=27 without returning arr will still change the value of arr[0] to 27 once the function completes. IE:

public static void whatever(double[] myarr) {

myarr[0] = 27.0;

}

public static void main(String[] args) {

double[] arr = new double[5];

System.out.println(arr[0]); //prints 0.0

whatever(arr);

System.out.println(arr[0]); // prints 27.0

}

You shouldn't have to have any variables outside of your methods. You only need those temporary array(s) inside the merge function, right? (you only need one to my knowledge) So you can just declare them there (anew) every time you need them.

(Edited by Daniel Lepage - original submission Sunday, 16 March 2008, 06:04 PM)

further question for Non-static..

User 71 said: then if I created temp varibale inside mergesort, the only way I can make merge() refer to it is to read it as parameter(since both of the two methods have no returns), but since merge can only read one array, it make me confused how to use merge() combine two arrays with only one array as parameter and without reading any public arrays. Am I getting there? or am I totally in the wrong way? thank you.

Lesley said: you'd want it to be inside merge() not mergeSort() - mergeSort is only 5 lines.

an outline of my algorithm?

User 71 said: btw, an outline of my algorithm: recursive mergesort, each call two mergsort to sort left and right half, and using merge to combine two, untill the length of sub-arrays become one. For doing that, two mergesort will call itself twice with different parameters and using merge to combine them. And to achieve that, I have to make merge refer to the sub-arrays sorted by the result of two mergesort just called, but since they are in the same function, and merge can only read one array, and both of the two methods are void method, I have to create two non-static arrays in the class to be constantly changed and refered to in the recurssion. Am I getting there? or am I totally in the wrong way? Thank you again for being patient.

Lesley said: There aren't actually two subarrays that need to be declared - they're just part of the big array as referenced by the indices lower and upper. Please see #670

Professor Sullivan said: Lesley's right. What you pass to merge is a big *sub* array each half of which is sorted. When merge completes that big sub-array is completely sorted, in that the two halves have been merged. According to our instructions, you may use a temporary array the size of the big subarray into which you insert elements in sorted order. You can then copy that sorted array back into the subarray of the original array.

Thanks so much to the previosu reply

User 71 said: I think I got it. The relationship of the two counts looks perfect and reasonable, though it seems that my captain doesnt want a group meet. Just wanna say thanks.

How could I set an element of an array to be void

User 71 said: and how could I refer to that as an expression of boolean?(say if I didn't set a element of an array to anything, I still don't know how to get a true or false out of that fact with boolean expressions)
Thank you so much.


Professor Sullivan said: If the elements of array A are objects of a reference type, then you can do this:

A[i] = null;

if (A[i] == null) { /* do something */ }

DataCollection

User 77 said: Ok so I've got a working MergeSort program, and I was wondering about using DataCollection.incrementCount(). If we compare values using an if... else statement, should we have DataCollection.incrementCount under the if AND the else, or should we only count it once?

Additionally, I've tried running the AlgorithmComplexity.java file with my working MergeSort program, and it gives me error messages about my MergeSort program. Is this a problem, since the MergeSort Program sorts properly and follows the correct technique?


Professor Sullivan said: Just be sure that you count exactly one comparison each time you do a comparison. If you do an if...else, you might just put a call to increment() immediately before or after.

What error message?

Re: DataCollection

User 77 said: the error message is ArrayIndexOutOfBoundsException.
It doesn't seem to make sense because it seems like I would get the same error if I just ran MergeSort.java on its own, but I don't.


Professor Humphreys said: Sounds like it's time to use the debugger!

Professor Sullivan said: Yup! There's clearly a bug in the code.

Professor Sullivan said: Be sure to call the DataCollection initialization routine properly, otherwise the array for holding values won't be set up properly. That is one thing that could easily result in the error you're seeing (although it's not the only problem that could be causing the error you're getting).

Edit Distance Submission

User 222 said: I went to submit my EditDistance.java into the uploader on the assignment page, but the file requested is EditDistance.Java (captial J in Java). I had to rename my program so the .java was a .Java for it to accept it. You might want to change that for the convenience of everyone else. Also, does my changing of the .java to .Java mess with my program at all? Thanks

Professor Humphreys said: whoopsie! Fixing now.

lab quiz 4

User 77 said: For the averagearray function portion of the quiz, it says to initialize the ith elements of the first row and first column to equal i, however the image provided shows that they are instead initialized to i+1. Which is correct?

Professor Humphreys said: the picture is correct, and the text is meant to start counting at 1, not zero. Sorry for the confusion, I wish everyone started counting from the same base :(

Professor Sullivan said: I had already corrected the assignment online.

Quiz 4 Grading

User 187 said: Will the Robograde feature be available for quiz 4?

Quiz 4; problem 1

User 437 said:

This question is regarding problem 1 on quiz 4.

How can I print each element of the double array as fractions, like the instruction image, instead of decimals?

Lesley said:

public static double[][] averagearray(int n)

Since a fraction is not a double, you don't have to worry about your program doing this. But you can google for some Fraction classes or attempt to write your own should you want to overachieve.

Professor Sullivan said: Right: you don't have to do this. Just use double, and make sure the values are equivalent to those shown in the figure. But if you have time and inclination it would be worth following Lesley's suggestion: write your own Fraction class. A fraction has a numerator and a denominator, each an integer. You should be able to take it from there!

Drawing a rectangle??

User 531 said: Im trying to draw a rectangle for a program I am writing, but none of the StdDraw functions seem to work for me. I have the coordinates of the 4 corners of the rectangles, but when I use StdDraw.FilledPolygon(x[], y[]), it draws little bow-ties instead of rectangles. Is there some other way to draw them??

Lesley said:

You have your coordinates out of order in your arrays.

a	c
b	d

It is going a -> d -> c -> b instead of a -> b -> d -> c

Not able to type in an argument into console window and run program

User 543 said:
I'm not able to put in an argument to my console window and get my program to run for some reason.
Under my console window, it says
StdIn[Java Application]C:\Program Files\Java\jre1.6.0_04\bin\javaw.exe

Whenever I hit the play button to run my program, in the console window it says "Type a string:" and then after I put in an argument it says Your string was:
and then says Type an int:






Professor Sullivan said: It sounds like you're inadvertently running the main routine in a class other than the one you mean to run. Be sure the class containing your main routine is selected in Eclipse.

Arithmatic with complex numbers?

User 188 said:

I was running through quiz 4 this evening when I realised that I have no idea how to do addition and multiplication with complex numbers. The rest I can play by ear, but what do I do for this? Just add and multiply the real and imaginary components separately, or is there some sort of complex ritual?

Thanks!

Yrs,
Panicking CLAS student .

Lesley said: http://www.google.com/search?hl=en&q=add+and+multiply+complex+numbers

Individual Assignment Run Time

User 380 said: Is there a way to get the control panel to work with StdOut or a way to use java -Xmx300m EditDistance < input.txt > output.txt in the console in eclipse. Or do I have to go and change all of my StdOuts to System.outs and use the control panel to find the time it takes for the computation?

Lesley said:

just for knowledge (since so many people keep asking) running from the command prompt:

j:\cs\editdistance\src\>javac -cp j:\cs\stdlib.jar;. EditDistance.java

j:\cs\editdistance\src\>java -cp j:\cs\stdlib.jar;. -Xmx300m EditDistance < input.txt > output.txt

where your source is in src ALONG with the input.txt file(s)

lastxgap8.txt and lastygap8.txt

User 71 said: My programs ran all very all and got the right answer until I got to these two files, it said array out of bounds, and when I opened them I found that they all just had one string and the only difference between them is that lastxgap8 had an extra "a" at the end. Is that a problem with my program? Or is that a prolem with the file? Thank you so much.

Lesley said: Yeah this is a problem with your program. Logically think through why it is doing that and you will figure out what it is doing wrong. I also think that last semester those files were not tested against your program.

Just ignore the previous post

User 71 said: Now I realize what my problem is, but is there a way to enlarge the memory that will be used when Eclipse run the java, so I will be able to debug with inputs as big as that of the "ecoli3000.txt".
Now it works on that two, but doesn't work on this one.
Thank you.

color transparency.

User 531 said:

Say i have a color:

Color sky = new Color(0,0,255);

If i want to make a lighter shade of this color (closer to white), how would I do this? By progressivly adding to the red and green?

Lesley said:

There's a difference between transparency and shade. To play around with shade open up something like MSPaint, double click one of those colors at the bottom and then click define custom colors. You can see how changing rgb affects the color.

Should you want TRANSPARENCY, you should use Color sky = new Color(0, 0, 255, 100) or something, play around with that last value (its "alpha" value) - try putting some other object underneath it to see what exactly it is doing.

Quiz 4, AverageArray

User 276 said:

Problem 1 for Quiz for has a picture demonstrating what averagearray(5) will look like, with is:

{1.0, 2.0, 3.0, 4.0, 5.0}
{2.0, 2.0, 2.5, 3.25, 4.125}
{3.0, 2.5, 2.5, 2.875, 3.5}
{4.0, 3.25, 2.875, 2.875, 3.1875}
{5.0, 4.125, 3.5, 3.1875, 3.1875}

However, when I run the test script, the robograde is looking for a matrix like this:

{0.0, 1.0, 2.0, 3.0, 4.0}
{1.0, 1.0, 1.5, 2.25, 3.125}
{2.0, 1.5, 1.5, 1.875, 2.5}
{3.0, 2.25, 1.875, 1.875, 2.1875}
{4.0, 3.125, 2.5, 2.1875, 2.1875}

This doesn't seem to coincide with the instructions. Maybe the instructions were changed but the program that tests the script was not.


Professor Sullivan said: Oh, man! The instructions are right. The roboscript is stlil wrong. We'll fix it.

ecoli 500000

User 380 said: Is there any way we can actually use the file ecoli500000.txt? When I first tried it, it said it ran out of memory, so i tried to increase the memory to 5000Mb, which was a bad idea in retrospect since I just realized I dont have that much memory, which explains why it says it exceeds the maximum representable size, but when i try 2000Mb or 1500Mb, etc. the command prompt says it could not create the Java Virtual Machine.

Weird!

User 71 said: Before my program won't work with ecoli3000 and 7000, but it runs pretty fast.
Then I changed something that I don't think will really make a different, but it does, only that the program works much slower. And the weird thing happened, I redo the change, the program will now work with ecoli3000 and 7000! But it works slowly as well! What might be a reasonable answer?
I ran the program in Eclipse as well as in cmd, and I am sure I did compile the program in cmd everytime there is a change.
Thank you.

Right Mouse Click

User 380 said:

I know the StdDraw.mousePressed method will tell if the left mouse button is pressed, but is there a way to implement certain fucntions when the right mouse button is pressed instead?

Lesley said: Last semester I edited StdDraw for more mouse abilities (they're all deprecated or something) - you'll have to implement your own routine in it for handling right clicks. See http://forum.java.sun.com/thread.jspa?forumID=57&threadID=571926 (this deals with methods inherited from StdDraw "implementing" MouseListener, something which I do not think is covered in 101).

Professor Sullivan said: Lesley's right: Go read the code in StdDraw.java. Let us know if you can't find it. (It's available on the booksite, and from the Useful Links page on the class web site. If you're willing to put in the effort, we'll help you to understand the code and how to extend it to do what you need. Doing this sort of thing is one of the things that CS101 is meant to enable you to do!

BubbleSort grading funny?

User 151 said: I, in working with my group have written the necessary code for BubbleSort.java, however when uploading the file to be RoboGraded, it is not working properly. In Eclipse, we added code to the Main function to print out the sorted array, and everything works fine. However, in the RoboGrade, the output looks exactly like the input. I have tried all the inputs from RoboGrade in the Eclipse main routine, and they seem to work. What could be going wrong? We have tried uploading several times, though the same result occurs.

Individual Assignment Output

User 270 said: I not sure what is suppose to be the output for the the example given in the individual assignment. It says it suppose to output the "optimal solution." I m not exactly sure what that is, because the I don't exactly understanding the whole biology thing in the assignment. So, can you give me the output we suppose to get for the example in the individual assingment?

Lesley said: Please see the help page, the section "Input, Output, and Testing"

Professor Humphreys said: I agree with these posts. There's no way to make progress without understanding the assignment. If you could be more specific about *what* you don't understand, then we can help. Otherwise you will need to go to office hours. In general it's not possible to complete an assignment without understanding the assignment. The assignment itself requires no knowledge of biology; all the terms and relationships that you need to do the work are defined in the assignment itself. If you don't understand something, please tell us *what* you don't understand, so we can help you.

Individual Assignment - Problem with Opt Array

User 52 said: When I print out the opt array, I get the answer 7, (and the answer 8 when using text file gene57), but some parts of the array are wrong.

The first two rows of the opt array are correct, as it is with the bottom 2 rows and the last column. But somehow there is a mistake in the middle.

I get this as my result.

7 8 10 12 13 15 16 18 20
6 6 8 10 11 13 14 16 18
4 5 6 8 9 11 12 14 16
6 4 4 6 7 9 11 12 14
7 5 3 4 5 7 9 10 12
9 7 5 3 3 5 7 9 10
8 8 6 4 3 3 5 7 8
10 8 7 5 3 2 3 5 6
12 10 8 6 4 2 1 3 4
14 12 10 8 6 4 2 1 2
16 14 12 10 8 6 4 2 0

the bolded parts are where I think the problem starts. But i don't understand why the first row is correct, if some parts are wrong in the beginning.

Am i just doing the wrong coding?


Lesley said: The formula might just be leading to the first row being correct because of the way incorrect values situate themselves (ie: coincidence). I would suggest coming to office hours for help with debugging what is wrong.

Professor Sullivan said: Lesley's right, but even before coming to office hours you should use the *debugger* to carefully step through and watch what your program is doing. If you haven't been brave enough to use the debugger until now, well, the time is now!

Distance - readme.txt

User 546 said: Hi!
I didn't understand what exactly should be written in the "distance" for each file in readme.txt file. Is it the optimal alignment? or the sum of the penalties?

Thanks a lot!


Lesley said:

"The edit-distance is the score of the best possible alignment between the two genetic sequences over all possible alignments"

The edit distance *is* the score of the optimal alignment which *is* the sum of the penalties.

The optimal alignment is an alignment, not a number/distance.

RoboGrade Group Assignment 4 - Sorting

User 151 said: I am not sure, but there could be an issue with the Group Assignment 4 Robograde. My group has submitted the BubbleSort and MergeSort java files, but when we run the RoboGrade routine, the output is totally different from what we get when we run the exact same inputs through Eclipse. What should we do about this?

what does StdIn.readAll().split("\\s+") mean?

User 49 said: In the Program 4.2.4, this is a line String[] a = StdIn.readAll().split("\\s+"). I know StdIn.readAll() means read all the text but what is.split? and what is the String"\\s+"? I did not see it inthe original String"the and was his......him" Thanks.

Professor Humphreys said: http://java.sun.com/j2se/1.4.2/docs/api/java/lang/String.html#split(java.lang.String)

Lesley said:

the string inside of split is a regular expression - see http://java.sun.com/j2se/1.5.0/docs/api/java/util/regex/Pattern.html#sum

It basically means split on a whitespace character (such as a space, tab, newline...) - split makes an array so in your string

"the and was his.....him" after split the array would be

{"the", "and", "was", "his.....him"}

if you had "the and\nthis is crazy\t woahhh" it would be

{"the", "and", "this", "is", "crazy", "woahhhh"}