CodeGym/Help with Java Tasks/Please advise, on paper logic seems good but conditions 2...

Level 41

13.01.2021
544views
15comments

Please advise, on paper logic seems good but conditions 2 and 3 not sufficient

Question about the task

Resolved

I can't see what I am doing wrong, please help!

Read 3 file names from the console: file1, file2, file3.
Split file1 as follows:
Write the first half of its bytes to file2,and write the second half of its bytes to file3.
If the number of bytes in file1 is odd, then file2 should contain the larger half.
Close the streams.

Requirements:

The program should read a file name three times from the console.
Use FileInputStream to read from a file, and use FileOutputStream to write to files.
The first half of the bytes in the first file must be written to the second file.
The second half of the bytes in the first file must be written to the third file.
The FileInputStream and FileOutputStream must be closed.

package com.codegym.task.task18.task1808; /* Splitting a file */ import java.io.*; public class Solution { public static void main(String[] args) throws IOException { BufferedReader br = new BufferedReader(new InputStreamReader(System.in)); String file1 = br.readLine(); String file2 = br.readLine(); String file3 = br.readLine(); br.close(); FileInputStream fileInputStream = new FileInputStream(file1); int totalBytes = fileInputStream.available(); int file3FilesSize = totalBytes / 2; int file2FilesSize = totalBytes / 2 + totalBytes % 2; int file4FileSizeWhenEqual = totalBytes / 2; byte[] largerHalfSize = new byte[file2FilesSize]; byte[] smallerHalfSize = new byte[file3FilesSize]; byte[] equalHalfSize = new byte[file4FileSizeWhenEqual]; if (totalBytes % 2 != 0) { //here i am reading the first larger half of the file and putting it into the array for (int i = 0; i < fileInputStream.available() - file3FilesSize; i++) { largerHalfSize[i] = (byte) fileInputStream.read(); } //here i am reading the second smaller half of the file and putting it into the array for (int i = fileInputStream.available() - file3FilesSize; i < fileInputStream.available() ; i++) { smallerHalfSize[i] = (byte) fileInputStream.read(); } //this writes the content of largerhalfsize into file2 but largerhalfsize is an array with no values yet. FileOutputStream fileOutputStream = new FileOutputStream(file2); fileOutputStream.write(largerHalfSize); FileOutputStream fileOutputStream2 = new FileOutputStream(file3); fileOutputStream2.write(smallerHalfSize); fileInputStream.close(); fileOutputStream.close(); fileOutputStream2.close(); } else { //here i am reading the first half in case the size is an even number for (int i = 0; i < fileInputStream.available() / 2; i++) { equalHalfSize[i] = (byte) fileInputStream.read(); } //here i am reading the second half in case the size is an even number for (int i = fileInputStream.available() / 2; i < fileInputStream.available() ; i++) { smallerHalfSize[i] = (byte) fileInputStream.read(); } FileOutputStream fileOutputStream3 = new FileOutputStream(file2); fileOutputStream3.write(largerHalfSize); FileOutputStream fileOutputStream4 = new FileOutputStream(file3); fileOutputStream4.write(smallerHalfSize); fileInputStream.close(); fileOutputStream3.close(); fileOutputStream4.close(); } /* write(byte[] b) Writes b.length bytes from the specified byte array to this file output stream. */ /* I need to read from file1 the first half into file 2 (so the larger half if applicable) and the second half (the smaller half) into file 3 */ } }

Comments (15)

Popular
New
Old

You must be signed in to leave a comment

Nouser

Level 36 , Germany

13 January 2021, 11:45

useful

you started pretty good

int totalBytes = fileInputStream.available();
int smallerHalf = totalBytes / 2;
int largerHalf = totalBytes - smallerHalf;

the last line just as a hint the rest is pretty complex. Loops line 32 and 37 could be idential (they do exactly the same). The number of bytes you want to read are equal. read() always starts at the internal pointer of the stream. So you could express it like

for (int i = 0; i < anyOfTheTwoHalfsSize; i++) {

But you also could use int writtenBytes = write(byte[]), start, length) with that construct written two times you can do everything lines 30 - 60 do (afair it was explained in a lesson before) I haven't tested the overall code and logic as it was to complex for me ;)

Andrei

Level 41

13 January 2021, 15:37

Let's take a size of totalBytes (file1 inputstream size) of 11 bytes. The loops in line 32 and 37 they do the same thing, BUT the starting point for each loop differs. In the case of file1 size being odd (see line 30 where the check for this is) For loop in line 32 the starting point is the beginning of the array until i < 6 (11-5) - so the first 6 elements, resulting the larger half. For loop in line 37 the starting point is i = 6 (11-5) and it loops until 10. So the last 5 elements. These 2 loops create 2 byte[] arrays in which I wrote the 2 halves of file1. In file2 the larger file I wrote array largerHalfSize and in file3 I wrote smallerHalfSize. This is why I believe my logic is good but don't understand why it isn't validating I don't know what you mean by "read() always starts at the internal pointer of the stream." - i don't know what that internal pointer of the stream is.

Nouser

Level 36 , Germany

13 January 2021, 16:22

useful

fileInputStream.read(); starts to reade where the streams internal pointer is and not where some loop you have iterates. So the following two streams pointing two the same file and haven't read a byte since declaration would read the same bytes from the file for (int i = 0; i < 100; i++) someFileInputStream.read(); same as for (int i = 100; i < 200; i++) someFileInputStream2.read(); if you do a read() then some internal pointer in the stream advances. The next call to read would read the second byte and so on. That's why while (stream.available() > 0) stream.read(); is working. Or did you give the read method some index what to read? So that internal pointer is pointing to the last byte read. So your loops work somehow but the logic you have put into is wrong. And as said, I haven't checked further cause it is making me head spin ;)

Andrei

Level 41

13 January 2021, 16:35

OK, I understand better now! Let's see what I can put together and I'll come back . :)

Andrei

Level 41

13 January 2021, 17:48

I have redone the program which in theory works, but when I check the program it gives the compiler error below and I don't understand why because the readAllBytes() method is used as it should be...

public class Solution {
    public static void main(String[] args) throws IOException {
        BufferedReader br = new BufferedReader(new InputStreamReader(System.in));

        String file1 = br.readLine();
        String file2 = br.readLine();
        String file3 = br.readLine();
        br.close();

        FileInputStream fileInputStream = new FileInputStream(file1);

        int totalBytes = fileInputStream.available();
        int file3FilesSize = totalBytes / 2;
        int file2FilesSize = totalBytes / 2 - file3FilesSize;

        byte[] totalBytesArray = new byte[totalBytes];
        byte[] largerHalfSize = new byte[file2FilesSize];
        byte[] smallerHalfSize = new byte[file3FilesSize];

        FileOutputStream fileOutputStream = new FileOutputStream(file2);
        FileOutputStream fileOutputStream1 = new FileOutputStream(file3);

totalBytesArray = fileInputStream.readAllBytes();

        for (int i = 0; i < largerHalfSize.length; i++){
            largerHalfSize[i] = totalBytesArray[i];
        }

        for (int i = 0; i < smallerHalfSize.length ; i++){
            for (int j = totalBytes; j > file2FilesSize ; j--){
                smallerHalfSize[i] = totalBytesArray[j];
            }
        }

        fileOutputStream.write(largerHalfSize, 0, file2FilesSize);
        fileOutputStream1.write(smallerHalfSize, 0, file3FilesSize);

        fileInputStream.close();
        fileOutputStream.close();
        fileOutputStream1.close();
}
}

Andrei

Level 41

13 January 2021, 17:49

But I get this error when I check the solution: Error in com/codegym/task/task18/task1808/Solution.java on line 34 cannot find symbol symbol: method readAllBytes() location: variable fileInputStream of type java.io.FileInputStream

Andrei

Level 41

13 January 2021, 17:51

I have tried multiple ways including using your suggested solution " while (stream.available() > 0) stream.read(); " but I don't understand if the pointer remembers where it left off so I can use it for the file3 writing...

Guadalupe Gagnon

Level 37 , Tampa, United States

13 January 2021, 18:08

solution

Two problems: #1

int file2FilesSize = totalBytes / 2 - file3FilesSize

This is the incorrect math. file3FilesSize is half the total. If you take half the total and subtract half the total, you are going to end up with 0. #2 Codegym's validator uses Java 8 while the readAllBytes() method was not available until Java 9. You won't be able to use that. Instead, just replace that line with:

fileIntputStream.read(totalBytesArray);

It will do the same thing but won't cause errors with the validator. You should also set your programming language level in IntelliJ to 8. Do this by going to - File>Project Structure (ctrl+alt+shift+s is the keyboard shortcut) - Next select the tab on the left "Modules" if its not already on that tab - Next the select the modules 1 at a time and click the "Sources" tab in "Sources" and set the Language Level to "8 -Lambdas, type annotatiions etc." See pic:

After that, IntelliJ will error out any class/method that was not in Java 8 and you wont have to deal with that with the validator.

Guadalupe Gagnon

Level 37 , Tampa, United States

13 January 2021, 18:12

useful

Lines 25-33, As a tip try this out:

int i = 0;
for (; i < largerHalfSize.length; i++){
   largerHalfSize[i] = totalBytesArray[i];
}

// use a new counter to start at 0 of the next array
// but continue using i where it left off in previous loop
for (int j = 0; i < totalBytesArray.length ; i++, j++){
   smallerHalfSize[j] = totalBytesArray[i]; // see j and i used here
}

I will do something like this in situations such as this task. Also, for lines 35 and 36, because you are writing the whole array you dont need the second and third arguments of the start and finish. This is sufficient:

fileOutputStream.write(largerHalfSize);
fileOutputStream1.write(smallerHalfSize);

Guadalupe Gagnon

Level 37 , Tampa, United States

13 January 2021, 18:26

useful

As a tip: You should never have more than 1 stream open at one time. Also you focus only on what the stream is for when you open a stream, closing it immediately after the work is done. This is because Streams are 99% of the time going to be the slowest part of your code. So, your code would more appropriately be structured like this:

// ***after reading 3 file names***

/* step 1
open input stream for file 1
read all bytes into totalBytes array
close that input stream*/

/* step 2
do the math to split the larger half and the smaller half
and get the data ready to be written
*/

/*step 3
open stream for file 2
write the data from step 2
close that output
*/

/*step 3
open stream for file 3
write the data from step 2
close that output
*/

These are only a tips. You don't have to change your code or spend hours trying to make your future code "perfect". Just try to keep this in mind while you are leveling up and practice it.

Andrei

Level 41

14 January 2021, 07:34

Check out this bad boy.... noice! This morning (with a fresh mind) I noticed the error you mentioned with the division and subtraction which might have prevented me from validating some of the other ways I had tried and were not working.. oh well. Your solution with outside counter was exactly was I was looking for, I tried to implement this but I couldn't figure out how.

public class Solution {
    public static void main(String[] args) throws IOException {
        BufferedReader br = new BufferedReader(new InputStreamReader(System.in));

        String file1 = br.readLine();
        String file2 = br.readLine();
        String file3 = br.readLine();
        br.close();

        FileInputStream fileInputStream = new FileInputStream(file1);
        int totalBytes = fileInputStream.available();
        byte[] totalBytesArray = new byte[totalBytes];
        fileInputStream.read(totalBytesArray);
        fileInputStream.close();

        int file3FilesSize = totalBytes / 2;
        int file2FilesSize = totalBytes - file3FilesSize;

        byte[] largerHalfSize = new byte[file2FilesSize];
        byte[] smallerHalfSize = new byte[file3FilesSize];

        int i = 0;
        for (; i < largerHalfSize.length; i++){
            largerHalfSize[i] = totalBytesArray[i];
        }

        for (int j = 0; i < totalBytesArray.length; i++, j++){
            smallerHalfSize[j] = totalBytesArray[i];
        }

        FileOutputStream fileOutputStream = new FileOutputStream(file2);
        fileOutputStream.write(largerHalfSize);
        fileOutputStream.close();

        FileOutputStream fileOutputStream1 = new FileOutputStream(file3);
        fileOutputStream1.write(smallerHalfSize);
        fileOutputStream1.close();
    }
}

Andrei

Level 41

14 January 2021, 07:35

Also, I rearranged the code to close the input/output stream as soon as the work was done with them. Thank you for the suggestions, they were really handy!

Guadalupe Gagnon

Level 37 , Tampa, United States

14 January 2021, 14:41

Big thumbs up. Your code looks exactly like what I was trying to describe. I ran some speed tests where I took the code like it was before and compared it to code like you have it above. I used a file that I have on my computer that +/- a million individual bytes. The old code ran in about 16 ms, which is pretty fast. The new version of the code ran in 5 ms, which is 3 times faster. These differences are very trivial for individual tasks, like reading one file, but if you can imagine a server center that is handling 1000's of requests every second; that small difference could start to add up to a couple second delay in processing individual requests. Sort of like when a webpage gets too much traffic at once and starts loading slowly on people's computers. In this particular instance, restructuring your code to keep the Streams open as little as possible gave you some noticeable results. This was a very easy thing to do. I have a few "rules" like this that I use in my code to help keep it clean and fairly fast; BUT creating highly optimized code every time is near impossible. Most of the algorithms that are highly optimized today are the result of years of research and trial/error. I don't recommend you spend a lot of time now trying to write the absolute best code possible, you would spend so much time doing so that you would slow down your learning. However, you are at the point that you should be picking up some tricks. Start creating your own rules to follow in your code, such as keeping streams limited, and create a unique style all your own. It will involve some trial... and a lot of error... and it will evolve over time as you learn new patterns and tricks. I hardly made up any of the stuff I use today all by myself. I learned most of it by seeing it in other people's code and saying "wow, that is way better than what I do". Anyways... You're doing awesome. Noticeably improving week after week. Keep it up!

Andrei

Level 41

14 January 2021, 15:34

Wow, that is so nice of you to say these things, thank you so much, I really appreciate it! 😁 I haven't really thought about creating my own rules and my own style it seems like such a foreign concept to me, but somehow it makes sense, because I have seen so many ways to solve a problem. Unfortunately, most of the times I think that they way I have solved it is not good or efficient enough and probably lose time thinking like this when I should be satisfied that it is done to the best of my current abilities. I have a question: when you were trying to solve some of the programs here did you use mostly pen and paper or you felt a small idea/hunch form in your brain and started following that to see where it leads?

Guadalupe Gagnon

Level 37 , Tampa, United States

14 January 2021, 15:56

useful

I've never used pen and paper, but I do create an outline of my initial plan with comments, which is the same thing really. My process: I start by copying the task conditions into the code as a comment. I use this as a start to my outline then I create the steps just like my above comment:

//step1: do this
//step 2: do this
//etc

After that I just start typing away. Most of the time my initial ideas don't come together and work as intended, but the "failed" ideas give me a better understanding of the whole problem. With a little debugging I can modify the spots that don't work, maybe rework some sections, and finally arrive at the correct solution. The debug, modify, rework step can happen a few times before the correct solution is found. Compartmentalizing your code makes this easier. This is actually what makes OOP languages so successful, how they can break up chunks of code into logical sections based on the duties they perform. You will see this in the big tasks coming up in levels 21+ where they break the code into multiple classes with multiple methods. The Games do this as well.