Output is correct

Question about the task

Resolved

So I made this algorithm where I add all Strings (filenames) from the console to an ArrayList, then I add the file part numbers to a TreeSet (lines 25 - 28) then to an ArrayList (line 29), then I iterate (lines 30 - 46) through the filename ArrayList to find the one which part number comes first by comparing it to the part-number Arraylist, then create the streams to copy the content, the output is correct on my pc with my files, but still doesn't validate :. - (

Let's build a file from various pieces.
Read file names from the console.
Each file has a name: <someName>.partN.

For example, Lion.avi.part1, Lion.avi.part2, ..., Lion.avi.part37.

The file names are supplied in random order. The word "end" is used to stop reading in file names.
In the folder where all the files are located, create a file without the "part" suffix, i.e. without ".<partN>".

For example, Lion.avi.

Use a buffer to copy all the bytes from the partial files to the created file.
Copy the first in the proper order, first the first part, then the second, ..., finally - the last part.
Close the streams.

Requirements:

The program must read file names from the console until the word "end" is entered.
Create a stream to write to the file without the "part" suffix (".<partN>") in the folder with all the "part" files.
Copy all the bytes from the *.partN files to the new file.
You should use a buffer for the reading and writing.
The file streams must be closed.
Don't use static variables.

package com.codegym.task.task18.task1825; import java.io.*; import java.util.ArrayList; import java.util.Comparator; import java.util.Set; import java.util.TreeSet; /* Building a file */ public class Solution { public static void main(String[] args) throws IOException { BufferedReader reader = new BufferedReader(new InputStreamReader(System.in)); String s; ArrayList<String> strings = new ArrayList<>(); while (!(s = reader.readLine()).equals("end")){ strings.add(s); } File file = new File(strings.get(0).substring(0,strings.get(0).indexOf("part"))); file.createNewFile(); TreeSet < Integer > integers = new TreeSet<>(); for (String s1: strings) { integers.add(Integer.parseInt(s1.substring(s1.indexOf("part") + 4,s1.length()))); } ArrayList<Integer> arrayList = new ArrayList<>(integers); for (int a = 0; a < integers.size(); a++) { for (String s1 : strings ) { if (Integer.parseInt(s1.substring(s1.indexOf("part") + 4,s1.indexOf('.',s1.indexOf("part") + 4))) == arrayList.get(a)) { FileInputStream inputStream = new FileInputStream(s1); FileOutputStream outputStream = new FileOutputStream(file, true); byte[] buffer = new byte[inputStream.available()]; inputStream.read(buffer); for (byte b : buffer ) { outputStream.write(b); } inputStream.close(); outputStream.close(); } } } /*for (Integer i : integers ) { System.out.println(i); }*/ } }

Comments (4)

Popular
New
Old

You must be signed in to leave a comment

Guadalupe Gagnon

Level 37 , Tampa, United States

23 August 2020, 18:52

solution

#1 The first problem that I see is that "part" is too loose of a place to split a string for the filename to get the part number. What if the file was "apartment.avi.part#", or "partner.txt.part#", or "party.jpg.part#". Do you see how this could be a problem? You will need to come up with a more complete way to only get the last number(s) of the string. #2 line 33 is going to throw an Exception. The code gets the Integer correctly on line 27 but on line 33 does something complete different. I would suggest using line 27 to parse properly on this line. *tips* You can use a TreeMap to instead of a TreeSet to order all the files. The key would be the integer part number and the value would be the full file name. Then you wouldn't have to refigure this out in the loop at line 30. At line 37 the code reads the whole file in one shot, but on lines 38-41 it outputs it 1 byte at a time. I would recommend writing in the same way that was read, in one shot. Just pass the byte array to the write method to do this. Also I would recommend only making one OutputStream and not a new one for every single input. Opening and closing streams is costly for the computer. If you open one, every new write just follows where the last left off, you do not need to append.

yehuda b

Level 23 , Beersheba, Israel

23 August 2020, 19:29

First of all, I can not thank you enough not only for this reply, but for all the other ones throughout the help section which I came across countless times throughout my journey here @ codegym, honestly my main mentorship so far in programming (as codegym is my only exposure so far) is from your answers. re: the task, I did end up validating it with a TreeMap as you already recommended other people to do, but I do think the reason it didn't validate is because of your last tip which was not making a new OutputStream, because the #1 that you wrote ended up being validated without being changed, #2 was changed b4 validating as I only coded it like that because my personal files end with ".txt" (was a mistake to post question with that, though I'm almost sure I fixed it b4 trying to validate) and the first and second tips are not reasons not to validate (which is prob. why you call them "tips") but the third tip maybe they have a log for the OutputStreams which only validates one of them?? I wonder. Thanks again you're the best!!

Guadalupe Gagnon

Level 37 , Tampa, United States

24 August 2020, 14:17

You're welcome and thanks for the feed back. I did all of these tasks and many of them I initially failed. I would then start modifying pieces here and there until something passed. So, I don't always know why the validator kicks back some code while passing others, so I try to give as many hints and tips as possible so that the poster has plenty of things to try out until something works. As far as the OutputStream and appending, I see that this fails many tasks. I do have a theory: In java you can use something called reflection to intercept methods and streams and modify them. I bet the validator is programmed to do this, especially with streams. It would make sense because it would be highly insecure to let random coders off the internet have access to files on CodeGyms actual servers. It can instead point these streams to hard coded data for testing purposes. I bet the outputstream is tested to see if the data being output matches what they expect if you were to output everything that was read in with the addition of the added data.

yehuda b

Level 23 , Beersheba, Israel

24 August 2020, 17:24

Oh Wow! makes so much sense, which makes me want to elaborate, I wonder if with the $50 plan an actual human validates it, cause somehow I feel that although the amount of tasks offered and the convenience of having everything validated right away, is something to appreciate and prob. isn't available even in universities, the fact that the implementation of the solutions is validated by a machine prob. withholds many important points about your code which either are genius or the opposite, that's why (as per reading one of your answers), I check other's solutions in the help section even after successfully validating mine. I remember in the syntax quest @ this task, my solution was perfectly validated with this: githubLink I was so proud at the time of my 179 line solution!