Dataframe

While breadroll 🥟 uses the nomenclature Dataframe for the data structure it works with, it is only similar to a dataframe in the API that is provided to the user, in truth it is an array of records (objects, key-value pair) given by Array<Record<string, unknown>> - the key of string is the the header values from the parsed file, and the value of unknown basically holds the values, ie.every Array<Record<string, unknown>> is a "row" in the dataframe.

While the data in the dataframe are multivariate, they are "homogenous", ie. all values of a column has the same type of data at any given point in time unless the data is modified or transformed using the convenience function provided. To explain better, for all { key: value } pair in the array of object, key will have the same value type across all { key : value } pair in the array.

How?

Well as mentioned above; a Dataframe has a default type of Array<Record<string, unknown>>, to use custom types and open up developer experience features like auto-completion; you can pass the type to any of the four Breadroll.open methods, ie. https, local, supabaseStorage, and json eg. let's say you have a type or an interface called Cities, you can use like so;

Breadroll.open.local<Cities>("../path/to/file", ","); // Promise<Dataframe<Cities>>

Now that's one way to do it; now, let's say you have a huge payload from an API request ie. Array<Record<string, unknown>>, then you can use the Dataframe constructor directly like so;

new Dataframe<Cities>([{}, {}, ...]) // Dataframe<Cities>