Output formats¶

>>> import etrafilatura as trafilatura
>>> html = """<html><head><title>Small tools win</title></head><body><article><h1>Small tools win</h1>
... <p>A tool that does one job is cheap to audit, cheap to replace, and composes with the rest through plain text.</p>
... <p>Platforms accumulate switching costs until leaving them costs more than staying; small tools never do.</p>
... <blockquote>Write programs that do one thing and do it well.</blockquote>
... <ul><li>grep searches</li><li>sort orders</li></ul></article></body></html>"""
>>> print(trafilatura.extract(html, output_format="xml"))  # doctest: +ELLIPSIS
<doc fingerprint="...">
  <main>
    <head rend="h1">Small tools win</head>
    <p>A tool that does one job is cheap to audit, cheap to replace, and composes with the rest through plain text.</p>
    <p>Platforms accumulate switching costs until leaving them costs more than staying; small tools never do.</p>
    <quote>Write programs that do one thing and do it well.</quote>
    <list rend="ul">
      <item>grep searches</item>
      <item>sort orders</item>
    </list>
  </main>
  <comments/>
</doc>
>>> print(trafilatura.extract(html, output_format="html"))
<html>
  <body>
    <h1>Small tools win</h1>
    <p>A tool that does one job is cheap to audit, cheap to replace, and composes with the rest through plain text.</p>
    <p>Platforms accumulate switching costs until leaving them costs more than staying; small tools never do.</p>
    <blockquote>Write programs that do one thing and do it well.</blockquote>
    <ul>
      <li>grep searches</li>
      <li>sort orders</li>
    </ul>
  </body>
</html>
>>> import json
>>> sorted(json.loads(trafilatura.extract(html, output_format="json")))
['comments', 'text']