Exception in thread “main” java.lang.ArrayIndexOutOfBoundsException: Index 18 out of bounds for length 18 [closed]

by 
class CSVFileParser {
    ArrayList<String[]> convertCSVToArrayList(String filePath) {
        String line = "";
        ArrayList<String[]> rows = new ArrayList<>();
        try {
            BufferedReader br = new BufferedReader(new FileReader(filePath));
            while ((line = br.readLine()) != null) {
                String[] row = line.split(",");
                rows.add(row);
            }
            br.close();
        } catch (IOException e) {
            e.printStackTrace();
        }
        return rows;
    }
}

Please help me why its not working. When i try to add 18th index this error come, till 17th its working fine, anyone fix this for me, where I am doing mistake

delivery.csv file
matches.csv file

  • the indexing starts from 0 and for collection that have 18 items in it the length will be 18 but that index of last item will be 17, starting from 0

    – 

  • 1

    Which of the two files is giving you an error? Please post a minimal reproducible example, not a de-contextualized method with two different input files. Chances are the problem is not even in the piece of code you posted.

    – 




  • that delivery file gives me error when i am trying to access it

    – 

  • 1

    are you sure the error is in posted code? I see no indexed access

    – 

  • 1

    line.split(",") is a completely inappropriate way to parse a CSV file. CSV files can have embedded commas inside double-quoted strings, escaped double quotes inside them, even line breaks inside them. Use a proper CSV library like Apache Commons CSV. Parsing CSV files yourself is one of those things programmers always think they can just throw together and it always ends in tears.

    – 




The comments have the answer, but they’re not exactly coherent.

String.split(",", limit) will return an array with length equal to the number of fields found up to limit (corrected based on a comment below), so if there are 17 commas, it will be an array with index [0] to [17], length 18. If limit is 0 or omitted, any empty fields at the end will be omitted. The file delivery.csv has headers with 21 fields, but not all lines are the same length. That’s the first problem, you should specify the expected limit, and all lines should have 20 commas (any short lines should have empty fields at the end like “abc,def,qrs,tuv,,,”).

On the other hand, your software should also be able to deal with errors like this. If you use a package that’s been developed for reading CSV files, that’s probably already been done. If you want to use split() in this way, you have to check the length of the returned array before trying to access it.

Also you might need to validate the CSV file in other ways, it might omit fields in the middle (e.g. instead of “abc,,ghi” it might have “abc,ghi”). If that’s the case, try to get a file that’s formatted correctly, or you might need to encode special cases to correct these problems.

Also, make sure the file doesn’t have cases like “abc,"d,f",ghi” or “abc,d\,f,ghi” which have commas in the data. CSV parsing packages can deal with those cases, it will be harder for you to add code to handle that always correctly.

Leave a Comment