Implement pre-processing for the embed api

[akolson]

Overview

This task involves implementing a robust pre-processing mechanism for the Embed API that performs validations and transformations on received input. This task is dependent on learningequality/le-utils#117. The implementation should be made off https://github.com/learningequality/curriculum-automation/pull/2

Description and outcomes

• The input expected by the embed API is displayed below;

{
  "topics": [
      {
          "id": "02d6763539bd4dfc998464a72768d0cc",
          "title": "Target topic",
          "description": "Target description",
          "language": "en",
          "ancestors": [
              {
                "id": "02d6763539bd4dfc998464a72768d0ee",
                "title": "Parent topic title",
                "description": "Parent topic description",
                "language": "en",
              }
          ]
      }
  ],
  "resources": [
      {
          "id": "02d6763539bd4dfc998464a72768d0bb",
          "title": "Resource title",
          "description": "Resource description",
          "text": "Resource text",
          "language": "en",
      },
  ],
  "metadata": {
      "channel_id": "738a44c0d5ba4c69900494083e0b059e",
      "channel_title": "Channel title",
      "some_additional_field": "some_random_value",
  },
}

• Ensure the presence of required fields and validate their formats.
• Convert input into a standardized format compatible with the ensemble of models within the Embed API.

Accessibility requirements

Not applicable

Acceptance criteria

• The Embed API only allows input data with all required fields
• The input data is standardized into a format compatible with the ensemble of models within the Embed API.
• Tests to verify the correctness of the input pre-processing functions.

---

Resources