I have a data structure from SQL.query(Repo...) as such:

IO.inspect results.columns:

["tour_id", "name", "year", "num_shows", "num_cities", "num_festivals", "num_countries"]

and the actual rows, IO.inspect results.rows:

[
  [5, "tour", 2022, 13, 10, 4, 2],
  [1, "asia", 2018, 4, 4, 3, 3],
  [2, "north america", 2018, 39, 17, 2, 2],
  [3, "europe", 2018, 13, 12, 9, 10],
  [4, "las vegas shows", 2018, 3, 1, 0, 1]
]

I am struggling to do something like Enum.zip where the result would be:

[
 [ tour_id: 5, name: "tour", year: 2022, num_shows: 13, ....],
 [ tour_id: 1, name: "asia", year: 2018, num_shows: 4, ....],
]

Ultimately, I want to do a group_by based on year:

2018: [
        [ tour_id: 1, name: "asia", year: 2018, num_shows: 4, ....],
        [ tour_id: 3, name: "europe", year: 2018, num_shows: 13, ....],
      ],
2022: [ ... ]

Or, however else one may suggest it should be formatted. I thought converting the rows and columns to keyword lists would do the trick, and then I can do the group_by.

Thank you

>Solution :

If you want a keyword list of column name -> row value at the end, the first step would be to convert the column names to atoms:

columns =
  ["tour_id", "name", "year", "num_shows", "num_cities", "num_festivals", "num_countries"]
  |> Enum.map(&String.to_atom/1)

Then you can map over the rows and zip each one with the columns:

rows = [
  [5, "tour", 2022, 13, 10, 4, 2],
  [1, "asia", 2018, 4, 4, 3, 3],
  [2, "north america", 2018, 39, 17, 2, 2],
  [3, "europe", 2018, 13, 12, 9, 10],
  [4, "las vegas shows", 2018, 3, 1, 0, 1]
]

zipped = for row <- rows, do: Enum.zip(columns, row)

IO.inspect(zipped)

[
  [
    tour_id: 5,
    name: "tour",
    year: 2022,
    num_shows: 13,
    num_cities: 10,
    num_festivals: 4,
    num_countries: 2
  ],
  [
    tour_id: 1,
    name: "asia",
    year: 2018,
    num_shows: 4,
    num_cities: 4,
    num_festivals: 3,
    num_countries: 3
  ],
  [
    tour_id: 2,
    name: "north america",
    year: 2018,
    num_shows: 39,
    num_cities: 17,
    num_festivals: 2,
    num_countries: 2
  ],
  [
    tour_id: 3,
    name: "europe",
    year: 2018,
    num_shows: 13,
    num_cities: 12,
    num_festivals: 9,
    num_countries: 10
  ],
  [
    tour_id: 4,
    name: "las vegas shows",
    year: 2018,
    num_shows: 3,
    num_cities: 1,
    num_festivals: 0,
    num_countries: 1
  ]
]

And finally group by the year:

IO.inspect(Enum.group_by(zipped, fn row -> row[:year] end))

%{
  2018 => [
    [
      tour_id: 1,
      name: "asia",
      year: 2018,
      num_shows: 4,
      num_cities: 4,
      num_festivals: 3,
      num_countries: 3
    ],
    [
      tour_id: 2,
      name: "north america",
      year: 2018,
      num_shows: 39,
      num_cities: 17,
      num_festivals: 2,
      num_countries: 2
    ],
    [
      tour_id: 3,
      name: "europe",
      year: 2018,
      num_shows: 13,
      num_cities: 12,
      num_festivals: 9,
      num_countries: 10
    ],
    [
      tour_id: 4,
      name: "las vegas shows",
      year: 2018,
      num_shows: 3,
      num_cities: 1,
      num_festivals: 0,
      num_countries: 1
    ]
  ],
  2022 => [
    [
      tour_id: 5,
      name: "tour",
      year: 2022,
      num_shows: 13,
      num_cities: 10,
      num_festivals: 4,
      num_countries: 2
    ]
  ]
}