Link Resolution

link_resolution defines how the engine follows and unpacks redirect/obfuscated links before downloading. It's a sequential step system for resolving a single URL through a chain of fetches and extractions.

For complex multi-step navigation (branching, loops, conditional logic), use Navigation Paths instead.

Structure

link_resolution:
  enabled: true
  resolvers:
    resolver_name:
      source_field: link
      steps:
        - action: fetch
          ...
        - action: extract
          ...

Steps use action: as a tag — this is different from the - fetch: {...} format used in navigation paths.

Steps

`fetch`

- action: fetch
  follow_redirects: true
  headers:
    Referer: "https://example.com"
  timeout_ms: 10000

`extract`

- action: extract
  method: regex
  pattern: "window\\.location = ['\"]([^'\"]+)['\"]"
  group: 1

- action: extract
  method: selector
  selector: "a#download-link"
  attribute: href

- action: extract
  method: json_path
  pattern: "data.download_url"

Field	Notes
`method`	`regex`, `selector`, `json_path`, `xpath`, `text`
`pattern`	For `regex` / `json_path`
`group`	Regex capture group (default: `0`)
`selector`	CSS selector for `selector` method
`attribute`	HTML attribute to extract
`fallback`	Value to use if extraction fails

`transform`

- action: transform
  transformations:
    - type: trim
    - type: url_normalize
      base_url: https://example.com

`wait`

- action: wait
  duration_ms: 2000

Examples

JavaScript redirect page

link_resolution:
  resolvers:
    redirect_page:
      source_field: link
      steps:
        - action: fetch
          follow_redirects: true
        - action: extract
          method: regex
          pattern: "location\\.href=['\"]([^'\"]+)['\"]"
          group: 1

HTML page with a download button

link_resolution:
  resolvers:
    download_page:
      source_field: link
      steps:
        - action: fetch
          follow_redirects: true
        - action: extract
          method: selector
          selector: "a.download-button"
          attribute: href
        - action: transform
          transformations:
            - type: url_normalize
              base_url: https://example.com

Build URL from extracted ID

link_resolution:
  resolvers:
    id_to_url:
      source_field: link
      steps:
        - action: extract
          method: regex
          pattern: "/file/([A-Za-z0-9]+)"
          group: 1
        - action: transform
          transformations:
            - type: template
              template: "https://cdn.example.com/download/{value}"

Use case	Tool
Follow a redirect or extract a URL from a page	`link_resolution`
Multiple steps, branching, loops, conditions	`paths`
Hosters page with many mirrors	`paths`
Per-host programmatic download	`hosts.resolver`

Structure​

Steps​

fetch​

extract​

transform​

wait​

Examples​

JavaScript redirect page​

HTML page with a download button​

Build URL from extracted ID​

When to use Link Resolution vs Navigation Paths​