Untitled notebook

Section

import Ecto.Query

# Maybe drop existing?
# from(m in Sanbase.ProjectEcosystemMapping) |> Sanbase.Repo.delete_all()
# from(e in Ecosystem) |> Sanbase.Repo.delete_all()w

alias Sanbase.Project
alias Sanbase.Repo
alias Sanbase.Ecosystem

import Ecto.Query

csv_file = "put_csv_file"
data = File.read!(csv_file) |> NimbleCSV.RFC4180.parse_string()

slug_ecosystems =
  data
  |> Enum.map(fn row -> [Enum.at(row, 0), Enum.at(row, 10)] end)
  |> Enum.take(500)
  |> Enum.map(fn [slug, ecosystems_str] ->
    ecosystems = String.split(ecosystems_str, ",", trim: true) |> Enum.map(&String.trim/1)
    [slug, ecosystems]
  end)

slug_to_id = from(p in Project, select: {p.slug, p.id}) |> Repo.all() |> Map.new()

ecosystems =
  Enum.map(slug_ecosystems, &Enum.at(&1, 1))
  |> List.flatten()
  |> MapSet.new()

for e <- ecosystems do
  Ecosystem.create_ecosystem(e)
end

ecosystem_to_id =
  from(p in Ecosystem, select: {p.ecosystem, p.id}) |> Repo.all() |> Map.new()

for [slug, ecosystems] <- slug_ecosystems, e <- ecosystems do
  project_id = Map.get(slug_to_id, slug)
  ecosystem_id = Map.get(ecosystem_to_id, e)

  create(project_id, ecosystem_id)
end