{
  "$id": "https://opendatafabric.org/schemas/sources/v1alpha1/PrepStep",
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "description": "Defines the steps to prepare raw data for ingestion.",
  "$defs": {
    "Decompress": {
      "description": "Pulls data from one of the supported sources by its URL.",
      "type": "object",
      "required": [
        "format"
      ],
      "properties": {
        "format": {
          "$ref": "https://opendatafabric.org/schemas/sources/v1alpha1/CompressionFormat",
          "description": "Name of a compression algorithm used on data.",
          "unevaluatedProperties": false
        },
        "subPath": {
          "type": "string",
          "description": "Path to a data file within a multi-file archive. Can contain glob patterns."
        }
      }
    },
    "Pipe": {
      "description": "Executes external command to process the data using piped input/output.",
      "type": "object",
      "required": [
        "command"
      ],
      "properties": {
        "command": {
          "type": "array",
          "items": {
            "type": "string"
          },
          "description": "Command to execute and its arguments."
        }
      }
    }
  },
  "oneOf": [
    {
      "allOf": [
        {
          "properties": {
            "kind": {
              "type": "string",
              "const": "Decompress"
            }
          },
          "required": [
            "kind"
          ]
        },
        {
          "$ref": "#/$defs/Decompress"
        }
      ]
    },
    {
      "allOf": [
        {
          "properties": {
            "kind": {
              "type": "string",
              "const": "Pipe"
            }
          },
          "required": [
            "kind"
          ]
        },
        {
          "$ref": "#/$defs/Pipe"
        }
      ]
    }
  ]
}