esearch.pipeline

{
  "description": "Customizable ingest pipeline for the 'RAG en Acción' index",
  "version": 1,
  "processors": [
    {
      "set": {
        "field": "text",
        "copy_from": "body_content",
        "override": false
      }
    },
    {
      "set": {
        "field": "metadata.url",
        "copy_from": "url"
      }
    },
    {
      "set": {
        "field": "metadata.title",
        "copy_from": "title"
      }
    },
    {
      "set": {
        "field": "metadata.addedToDB",
        "copy_from": "last_crawled_at"
      }
    },
    {
      "set": {
        "field": "metadata.meta_description",
        "copy_from": "meta_description"
      }
    },
    {
      "remove": {
        "field": [
          "additional_urls",
          "domains",
          "links",
          "url_path_dir2",
          "url_path",
          "url_path_dir1",
          "url_path_dir3",
          "body_content",
          "url_port",
          "url_host",
          "headings",
          "url_scheme",
          "title",
          "last_crawled_at",
          "is_truncated",
          "meta_description",
          "main",
          "url"
        ],
        "ignore_missing": true
      }
    },
    {
      "inference": {
        "model_id": ".multilingual-e5-small_linux-x86_64",
        "input_output": [
          {
            "input_field": "text",
            "output_field": "vector"
          }
        ]
      }
    }
  ]
}