Write Bean containing a list

[enR!Gi]

Hi there,

I never found this in the forum or the documentation.
So my question is, how can I write a bean which contains a list, and for each element in the list a new csv line is being written.
I recognized that with ICsvDozerBeanWriter it is possible to access specific indexes within the list, but what I want is that every element in the list is being written.

public class Bean {
    private List<Recipient> recipients
}

So a mapping for a specific index would be

private static final String[] FIELD_MAPPING = new String[] { "recipients[0].mail"};

and my thought was, it might be possible to define something like

private static final String[] FIELD_MAPPING = new String[] { "recipients[].mail"};

which would then, instead of using only one index, write the whole list.

Is this possible, or is there another way to solve this?

Thanks and regards,
René

[James Bassett]

Hi René,

Typically,
1 bean instance == 1 call to write() == 1 row of CSV

It looks like you're trying to write each recipient's mail field on a new line, so why not iterate through each Recipient and write them instead (of Bean).

Then your field mapping would simply be "mail". Think of the field mapping as a way of mapping from one bean to one row of CSV.

Does that help?

[enR!Gi]

Hi,

ok - thank you.
The proposal you suggested also came to my mind but I never accomplished an implementation.

But according to your answer, I guess my approach was wrong anyway. This is what I've tried so far:

for (Bean bean : beans) {
    for (Recipient recipient : bean.getRecipients()) {
        beanWriter.write(bean, processors); // <-- And this will not work, of course
    }
}

So the problem is obvious - I would have to pass over an instance of Recipient, but then the mapping I used to initialize my BeanWriter would fail. And that's where I am stuck right now.
And if I understood you correctly, this cannot work anyway, as the write-operation is always attached to a single bean.

What I want to have is a CSV File that looks like this (Assuming I have an additional attribute called 'date' within 'Bean'):

Recipient	Date
mail1@example.com	2012-12-06
mail2@example.com	2012-12-06

Thanx in advance and best regards, René

[James Bassett]

Ok I see now! In order for that to work, you'll need a reference to the Bean in Recipient, e.g.

public class Recipient {
    private String mail;
    private Bean bean;
    // getters/setters here
}

Then you can have the following as your field mapping:

private static final String[] FIELD_MAPPING = 
    new String[] { "mail", "bean.date"};

[James Bassett]

Of course, for a simple scenario such as yours it's probably a lot easier to keep things simple and use a CsvListWriter instead of a CsvDozerBeanWriter. Yes it's not as cool, but it gets the job done :)

for (Bean bean : beans) {
    for (Recipient recipient : bean.getRecipients()) {
        Object[] row = {recipient.getMail(), bean.getDate()};
        listWriter.write(Arrays.asList(row), processors);
    }
}

Obviously there are limits to using CsvDozerBeanWriter - but for complex nested object graphs it is a lot easier to use.

[enR!Gi]

Ok, thank you - will play a littlebit wth it and see what I'll use.