Ignore unwanted fields for parsing

[magoarcano]

When headers=true sometimes we don't want al the fields in the returned parsed items.

Now if headers line is: id, name, date, description I get the parsed items: [{id: "23", name: "John", date: "2021-01-01", description: "very long text I don't care"}, ...]

It would be nice to return just the fields we need, specially when there are many unwanted fields, the file is very big, or field values are very long.

Better to get: [{id: "23", name: "John", date: "2021-01-01"}, ...]

We could define the un/wanted fields in the transformHeader function, being ignored if it returns null. Or adding other parameters to the config: ignore (providing a list of the fields to ignore), and only (providing a list of fields to include only them)

[pokoli]

Thanks for your suggestion but I do not think we should implement that.

The objetive of PapaParse is just to parse the data and return all the parsed file. If you just need a subset of the information for each row is just a mather of filtering this data after the file is parsed.

If you can are about performance you can use streaming to filter the data.