This is a demo project for CSV tagging, that serves as support for the series of posts on ui and ux for small programs started on Aliaume's personal webpage.
Given a CSV file with columns containing date, strings, and floats
the goal is to add a new column (or edit an existing column)
whose content is a string belonging to a finite set.
Typically, the input CSV will be a list of bank transactions,
and the goal is to assign a category to every transaction. The archetypical
example of category being solo or shared expanse.
The above program is almost one hundred percent about human interface.
Indeed, the categories solo or shared cannot be inferred from
the input data, and every tag has to be asked to the user. Therefore,
the algorithmic content of the program is reduced to its bare minimum:
parsing CSV files, writing CSV files, and managing a list of records.
This allows us to focus on the human interface, and compare approaches
in different languages, without having to fundamentally change
the underlying data model.
The repository is organised in terms of sub-folders, one sub-folder
corresponds to one way to solve the project's requirements.
The different implementations do not need to be code, and their
development stage may differ. Every sub-folder contains a README.md
file explaining the approach and listing its pros and cons.
The evaluation, comparison, and list of the approaches remains a work in progress. In an ideal world, you will find below a list of key indicators that discriminates the maturity of the approaches (non-linear edition, state restoration, etc.), together with informal indicators of their usability (easy to install, easy to use, easy to modify, etc.).