TablesDemo

This package demonstrates a minimalist tabular data type, Table, in Julia. As in AbstractTables.jl, our objective is to encourage modularity and extensibility in the development tabular data facilities. To this end, the present package itself provides as few user-facing components as possible; rather, the Table API is primarily designed to satisfy other tabular data interfaces that provide familiar functionality such as IO, manipulation, modeling, and visualization. In particular, this package demonstrates how a tabular data type such as Table can take advantage of the StructuredQueries query framework as extended by AbstractTables.

Interfaces

The Table data type satisfies each of the three AbstractTable interfaces. As such, it inherits all of the provided methods of these interfaces -- in particular those of the StructuredQueries-provided querying framework:

julia> using TablesDemo

julia> tbl = TablesDemo.Table(
               name = ["Niamh", "Roger", "Genevieve", "Aiden"],
               age = [27, 63, 26, 17],
               eye_color = ["green", "brown", "brown", "blue"]
           )
TablesDemo.Table
│ Row │ name        │ age │ eye_color │
├─────┼─────────────┼─────┼───────────┤
│ 1   │ "Niamh"     │ 27  │ "green"   │
│ 2   │ "Roger"     │ 63  │ "brown"   │
│ 3   │ "Genevieve" │ 26  │ "brown"   │
│ 4   │ "Aiden"     │ 17  │ "blue"    │


julia> qry = @query tbl |>
           groupby(age > 20, eye_color) |>
           summarize(avg_age = mean(age))
A Query.


julia> typeof(ans)
StructuredQueries.Query{TablesDemo.Table}

julia> collect(qry)
TablesDemo.Table
With the following grouping predicate aliases:
    age > 20 => predicate_1

│ Row │ predicate_1 │ eye_color │ avg_age │
├─────┼─────────────┼───────────┼─────────┤
│ 1   │ true        │ "brown"   │ 44.5    │
│ 2   │ false       │ "blue"    │ 17.0    │
│ 3   │ true        │ "green"   │ 27.0    │