Skip to main content

Pages behind a login

To convert a page behind a login or a paywall that you can access, save it from your browser and send the HTML to POST /api/html. PullMD can't sign in to sites itself.

  1. Save the page

    In your browser, select File → Save Page As. PullMD accepts any HTML file up to 10 MB.

  2. Send it to PullMD
    curl -X POST "http://localhost:3000/api/html?url=https://example.com/release-checklist" \
      -H "Content-Type: text/html" \
      -H "X-Filename: release-checklist.html" \
      --data-binary @release-checklist.html

    url is optional. To make relative links in the result point to the right place, pass the page's original address. With url set, site recipes for that address apply.

# Release checklist

Run through these steps before every release so nothing ships half-finished.

## Before tagging

- Update the changelog with every user-facing change.
- Run the full test suite on a clean checkout.

In the web app, drag the .html file onto the page.

Differences from a URL conversion​

  • Never cached: the conversion creates no history entry or share link, and the stats don't record the file name.
  • No JavaScript: PullMD never runs scripts from an uploaded file, so the headless browser is off. extractor accepts readability or trafilatura.
  • App shells: a file with under 200 characters of text after extraction returns 422. That usually means the page builds its content with JavaScript. If the page is public, convert its URL, because PullMD renders JavaScript pages in a headless browser.

Send the file as Content-Type: text/html or application/xhtml+xml. Any other type returns 400.