easy-csv

is an annotation based CSV utility with support for column names, column positions and nested objects

Usage

Let's say you have a List<UserDetail>, which needs to be converted into a CSV format data or written into a CSV file. Add field level annotations on fields which needs to be considered for the operation in below fashion in your UserDetail object

public class UserDetail {
  
  @CSVHeader(value = "First name")
  @CSVHeaderPosition(value = 1)
  private String fName;
  
  @CSVHeader(value = "Last name")
  @CSVHeaderPosition(value = 2)
  private String lName;
  
  
  @CSVHeaderPosition(value = 4)
  private Address address;
  
  @CSVHeaderPosition(value = 3)
  private List<Vehicle> vehicles;
  .
  .
  .

If you wish to ignore a field for CSV conversion, just don't add the annotations.
If you wish to change column order, just change the value of @CSVHeaderPosition annotated over the field.
Simple! Ain't it? :)
In-case of nested object (in this case, Address), provide @CSVHeaderPosition on the field and @CSVHeaderPosition and @CSVHeader on the fields inside Address.
In-case of List<Object> (in this case List<Vehicle>), same thing as above, but only the first object inside the collection will be considered for CSV (the index to be considered won't be taken from user to avoid ArrayIndexOutOfBoundException).

Now, let's consider a scenario where-in your object has hundreds of fields and you need to add a new field with header position 2, but that position is already taken.
Here it's cumbersome to update the position values of hundreds of fields just because you added a new field. So, you can just add the new field with @CSVHeaderPosition as 2 just above the exisiting field with position 2.

Please note: I used @CSVHeaderPosition to sort fields as JAVA doesn't guarantee field order when using reflection. However, in scenarios where we have same positions for mulitple fields, the sort logic will sort all fields as per the value of @CSVHeaderPosition, which implies fields with same value of position will be sorted in the order in which they come from Field[].

For ex.

public class UserDetail {
  
  @CSVHeader(value = "First name")
  @CSVHeaderPosition(value = 1)
  private String fName;
  
  @CSVHeader(value = "Contact number")
  @CSVHeaderPosition(value = 3)
  private String mobile;
  
  @CSVHeader(value = "Last name")
  @CSVHeaderPosition(value = 2)
  private String lName;
  
  @CSVHeader(value = "Age")
  @CSVHeaderPosition(value = 2)
  private int age;

Headers for above will be printed as First name,Last name,Age,Contact number

Supported data-types:

String
Object
boolean, float, long, double, int, char, short
List<Object>
Object[], boolean[], float[], long[], double[], int[], char[], short[]

V2.0.0.0 - Multithreaded data processing and file writes

Added an implementation for parallel processing and file writes. Here's how this feature works:

1. Utility divides List<Object> into n buckets (specified by client) with capacity list.size() / n each
2. All buckets are then sent for processing (file creation) asynchronously
3. The files are then zipped for compression. The path is provided to the user
4. Once zip creation is done, all files created in above process are deleted

Usage

List<Object> objects  = getThisObjFromNoMansLand();
String dir            = "/Users/abc/xyz/testcsv";
ICSVFileWriter z      = new ParallelZippedFileWriter(13, objects, dir, true);
Result result         = z.writeAndZip();
System.out.println(result)

Result:

Result{taskMetas=[TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-1.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-2.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-3.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-4.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-5.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-6.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-7.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-8.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-9.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-10.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-11.csv'}, TaskMeta{records=939, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-12.csv'}, TaskMeta{records=945, status=200, message='null', filePath='/Users/praveenkamath/Documents/testcsv/WDUBX/bucket-13.csv'}], status=200, message='null', zipPath='/Users/praveenkamath/Documents/testcsv/WDUBX.zip'}

Benefits of using this

1. Files are compressed, so less space consumption (as per tests conducted by me, 13 files of size 15kb each where compressed into single zip file of size 25 kb). Also, in requirements where-in client downloads large data, zipped file can be made downloadable to client rather than a huge file, which is time consuming
2. Files are created asynchronously, so time consumption is less for large data.

Also, added an extra annotation @CSVProperties with defaultValue method. Use this annotation at class level to denote the default value to be used when field value is null.

V1.0.0.1

This version comes with a function to convert collection of data into compressed, Base64 encoded CSV data in bytes format.

Usage

CSVByteUtils t = new CSVByteUtils(",");
List<UserDetail> userDetails = new ArraysList<>();
byte[] b = t.write(userDetails, true); //pass true in 2nd arg if you want to apply header in CSV
String data = new String(Base64.getDecoder().decode(GZip.decompress(b).toString()));
//Do anything with this data, maybe write to a file or send it in bytes somewhere.

As of now, it's upto the caller to manage memory by sending data in chunks. Memory management and file write will be supported in next version.