This project uses three main patterns: Converters, Parsers and the Conversion object.
The converter objects are the main object in charge of making the conversion and writing back to the output file. The converter takes ruby objects supplied by the parser and converts them into the :to format.
Parsers are a well known concept in CS, in this case I extracted parsing logic on a per data format basis, so that any additional logic per parsing format can be handled in seperate objects. The parsers must always return ruby objects back to the conversion instance. The responsibility is to go from a data format to that data format represented in ruby:
e.g.
JSON -> Hash/Array, CSV -> Array
The conversion object executes checks on the requested conversion to see if the converter exists and if the supplied input and output files exist. After validations it calls the correct parser object and passes its result to the converter.
In this folder the extension of existing classes is handled, some of them have methods added to assist the conversion, only add methods here specifically that can be reused outside of the conversion context. If they are truely conversion specific, the method belongs in a converter or converter assisting object.
To add a new conversion, you need to create a new converter: The convention is: FromToTOConverter
for instance: JSONToCSVConverter. The code base works on naming convention, so follow this closely.
The converter object takes a hash as an argument with the following attributes: the ruby objects supplied by the parser and the output file path. If a parser is not yet present for your data format, add the parser in the parser folder that follows the correct convention: DATAFORMATParser, e.g.: JSONParser
It should have a class method 'parse' which returns the ruby objects needed for the conversion.