Skip to content

[Feature]: Extract complete X threads with configurable Markdown output #86

Description

@miguelcorderocollar

Feature Category

Settings / Configuration

Problem or Use Case

When extracting an X/Twitter post that belongs to a thread, ExtractMD currently captures only the selected post. This loses the surrounding context and requires users to extract each post manually, then assemble the thread themselves.

Users also need control over how a multi-post thread is represented in Markdown: one continuous document or clearly separated posts, plus an optional list of original post links.

Proposed Solution

Extend X/Twitter extraction to detect and collect the complete authored thread in chronological order, while excluding unrelated replies and quote-post conversation branches.

Provide settings for:

  • Thread scope: extract only the selected post, or the complete thread.
  • Text layout: combine all thread text into one continuous Markdown document, or separate posts into distinct, consistently numbered sections.
  • Source links: append every extracted post URL in a links section at the bottom, with an option to disable it.
  • Default behavior: use safe, backwards-compatible defaults that preserve the current single-post output unless a user opts in.

The extractor should support X's lazy-loaded content and fail gracefully if only part of a thread can be retrieved.

Alternatives Considered

  • Keep extracting only the selected post and require manual collection of the rest of the thread. This is slower and makes preserving thread context error-prone.
  • Always extract the whole thread with a fixed layout. This would remove useful control for users who only need a single post or prefer a particular Markdown structure.

Impact

High (significant improvement for most users)

Additional Context

Acceptance criteria:

  • Starting extraction from any post in an X thread can produce the complete authored thread in chronological order.
  • Users can select combined text or separate-per-post Markdown output.
  • Users can include or omit a bottom section listing every extracted post URL.
  • Single-post extraction continues to work unchanged.
  • Settings are available in the extension's existing settings UI and persist correctly.
  • Tests use synthetic X fixtures and cover thread detection, ordering, both layouts, link inclusion/exclusion, and unrelated replies.

Metadata

Metadata

Assignees

No one assigned

    Labels

    area: extractionContent detection and Markdown extractionarea: outputExport formats and generated Markdownarea: settingsSettings, preferences, and configurationenhancementNew feature or requestintegration: XX/Twitter content extraction

    Projects

    Status
    Backlog

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions