extractor.ex

defmodule ExtractorInvoice do
  @schema %{
    type: "object",
    properties: %{
      invoice_number: %{
        type: "string",
        description: "The unique identifier of the invoice"
      },
      invoice_date: %{
        type: "string",
        format: "date",
        description: "The date the invoice was issued (YYYY-MM-DD)"
      },
      total_with_vat: %{
        type: "number",
        description: "Total amount including VAT"
      },
      total_without_vat: %{
        type: "number",
        description: "Total amount excluding VAT"
      },
      items: %{
        type: "array",
        description: "List of line items in the invoice",
        items: %{
          type: "object",
          properties: %{
            label: %{
              type: "string",
              description: "Description of item or service"
            },
            quantity: %{
              type: "number",
              description: "Quantity of the item"
            },
            price: %{
              type: "number",
              description: "Unit price of the item (without VAT)"
            }
          },
          required: ["label", "quantity", "price"],
          additionalProperties: false
        }
      }
    },
    required: [
      "invoice_number",
      "invoice_date",
      "total_with_vat",
      "total_without_vat",
      "items"
    ],
    additionalProperties: false
  }

  def extract(content) do
    {:ok,
     %{
       choices: [
         %{
           "message" => %{
             "content" => raw
           }
         }
         | _
       ]
     }} =
      OpenAI.chat_completion(
        [
          model: "gpt-5",
          messages: [
            %{
              role: "system",
              content: ~s"""
              You are the best accountant assistant.
              The user is providing the text of the invoice he wants to analyse. Your goal is to extract only the useful data and get the data of the invoice. Concentrate only on extracting the data from the user input.
              """
            },
            %{
              role: "user",
              content: content
            }
          ],
          response_format: %{
            type: "json_schema",
            json_schema: %{name: "extracted_data", strict: true, schema: @schema}
          },
          stream: false
        ],
        %OpenAI.Config{
          http_options: [recv_timeout: :timer.seconds(240)]
        }
      )

    Jason.decode(raw)
  end
end