Handling Dirty Data is Dirty
For a project I'm currently working on I need to write a Csv Parser. Simple enough, right? Well... each csv file has different parsing criteria, each has a completely different structure and data, and each manages to change its ways every now and then. Not only that, each file needs to be merged into a common schema!
I'm trying to use TDD as a method to tackle this. I'm going to have a IParser interface where the concrete implementation will take a IDataSource and an IParsingStrategy injected into it. Nevermind the code model, developing the UI that will allow a user to change the parsing strategy and persisting those changes is going to be an even bigger headache.
Watching JP Boodhoo's dnrTV episodes for inspiration.
Similar Posts
- Software Estimation is Like Predicting a 9-Ball Game
- Installing Windows Home Server On Old Hardware
- I Survived JP's Nothing but Dot Net Boot camp!
