Csv

Csv encapsulates a tabular data structure, as found in a CSV file or spreadsheet. This class is a thin wrapper around an Iterator[Row] and a header Row. It allows client code to:

• read CSV data from a tabular data-source (String, File, InputStream, or anything else)
• Change the column structure (add, remove, rename and reorder columns)
• Map data values according to a function (e.g. a lookup, a data-conversion, a default value)
• Export the results

This class may be useful for:

• ETL of externally-generated data
• Pre-formatting data before Diffing it during a reconciliation. @see Diff

Csv is designed to be lazy. All materialisation of the datasource is done OUTSIDE of the Csv, i.e. as late as possible. Thus, there should be no appreciable time cost to constructing a Csv, merging Csvs, restructuring columns, and so forth. The time cost is only paid once, when the client code exports or otherwise materialises then Csv.

TODO - CAS - 07/08/2014 - A Y-shaped pipeline (spits out two CSVs) TODO - CAS - 07/08/2014 - Aggregator 1 - combine multiple columns TODO - CAS - 07/08/2014 - Aggregator 2 - combine multiple rows - provide a predicate for row grouping/inclusion/exclusion