Defuddle icon

Defuddle

Extract main webpage content and metadata from HTML using Defuddle (fetch pages with the HTTP Request node)

Defuddle

AI-generated

Summary

Extract main webpage content and metadata by parsing raw HTML input using the Defuddle library, optionally improving processing with the document's absolute URL and various content extraction options.

Inputs

  • HTML (required) — Raw HTML content to parse, typically the body from an HTTP Request node or another source, provided as a string.
  • Document URL — Optional absolute URL of the HTML document to aid relative link resolution and metadata extraction.
  • Options — Collection of optional settings to customize extraction, including whether to convert content to Markdown, how to handle replies, removal of ads or hidden elements, image removal, content standardization, debug info inclusion, preferred language, content CSS selector override, and async extractor usage.

Output shape

A list of JSON items representing the extracted main content and metadata per input item.

One output item is returned for each input item, preserving pairing. If parsing fails and 'Continue on Fail' is enabled, output contains an error object for that item; otherwise, the node execution throws an error. The result includes extracted content in HTML and optionally Markdown and debug info based on options.

Links

Discussion