Parsing CSV: Count the number of columns using Univocity with bean processor

Question

Is it possible to count the number of columns in a CSV file using BeanListProcessor? My goal is to verify that the number of headers is equal to the number of values in the CSV file.

I have tried this solution: what is the fastest way to get dimensions of a csv file in java

But this only applies to a RowProcessor, and I'm using the BeanListProcessor;

BeanListProcessor<SomeObject> rowProcessor = new BeanListProcessor<SomeObject>(SomeObject.class);

I tried overriding the method of RowProcessor in BeanProcess but it is declared final.

Is there any other way of validating CSV files with univocity?

Answer 1

You can always wrap row processors to implement any custom logic to manage how the input should be handled:

public class MyRowProcessor<T> implements RowProcessor {

    private final BeanListProcessor<T> wrappedProcessor;

    public MyRowProcessor(Class<T> beanType) {
        wrappedProcessor = new BeanListProcessor<T>(beanType);
    }


    @Override
    public void processStarted(ParsingContext context) {
        wrappedProcessor.processStarted(context);
    }


    @Override
    public void rowProcessed(String[] row, ParsingContext context) {
        //validate the input row here
        if(row.length != context.headers().length){
            return; // Example: I'm skipping the row if the number of columns doesn't match number of headers
        }

        //send the row to the wrapped processor (or skip, or do anything you want)
        wrappedProcessor.rowProcessed(row, context);
    }

    @Override
    public void processEnded(ParsingContext context) {
        wrappedProcessor.processEnded(context);
    }

    public List<T> getBeans() {
        return wrappedProcessor.getBeans();
    }
}

All you have to do now is to use your custom implementation:

MyRowProcessor<SomeObject> processor = new MyRowProcessor<SomeObject>(SomeObject.class);

settings.setRowProcessor(processor);
settings.setHeaderExtractionEnabled(true);

CsvParser parser = new CsvParser(settings);

parser.parse(new File("/path/to/file"));

List<TestBean> beans = processor.getBeans();