This PR suggest some exploratory work implementing the same functionality (the readers in this case) using classes and subclasses.
The reason for this change is:
Reduces some redundancy of the code, in particular when reading headers.
Has a slightly better separation of concerns.
Enables using the header to validate or do something else with the tabular data more easily in the future, implementing dedicated methods within each subclass.
No need to do a thorough review, just gathering some feedback on the approach, pros, cons and things to consider.
I think this approach helps with keeping things consistent and not repeating code. There could be a more more functional way to do this, but I think this is maybe clearer than that would be.
This PR suggest some exploratory work implementing the same functionality (the readers in this case) using classes and subclasses.
The reason for this change is:
No need to do a thorough review, just gathering some feedback on the approach, pros, cons and things to consider.