Nightwatch web scraper

Task

1. First:
   • Go to http://google.com.
   • Search for "pizza".
   • Set the preferences to get 100 results (feature removed).
   • Save the HTML.
2. Create a Go program that does the following:
   • Based on CSS selector definitions read from a local JSON file, extract the following information:
     • URL rank type (organic, local, carousel, knowledge panel, featured snippet),
     • Website title, URL, description, and rank position within its own rank type group.
   • Format the results in a standard format (JSON or YAML, or CSV).
   • The use of well-maintained and mature external libraries is encouraged.
   • Ensure optimal memory and CPU usage.
   • Results should be consistent, even when one or more result types are missing.
   • Bonus points for high performance.

Usage

go run .

• Input HTML is read from pizza.html.
• Selector groups are read from the group-selectors.json.
• Output is saved to result.csv.

Selector groups

Expected JSON format:

{
   "<group name>": {
      "base": "<selector for group item>",
      "title": "<from item, selected by 'base' selector",
      "url": "<from item, selected by 'base' selector",
      "description": "<from item, selected by 'base' selector"
   }
}

Example:

<items-group1>
   <item>
      <title></title>
      <url></url>
      <description></description>
   </item>
   <item>
      ...
   </item>
</items-group1>

{
   "organic": {
      "base": "items-group1 > item",
      "title": "title",
      "url": "url",
      "description": "description"
   }
}