plugin: extract-seo

[cldellow]

      // Extract information like title, metadesc, author, publish date,
      // preview image.
      "extract-seo": {
        // optional; absent implies .*
        "url-regex": ".*",

        // optional
        "database": "dbname",

        // optional; defaults to dss_seo
        "table": "dss_seo",

        // optional; if present, will try to extract body of article
        "extract-article": true
      },

Needs https://github.com/cldellow/datasette-scraper#extract_from_responsescraper-config-url-response