######### Extract ######### .. module:: turbohtml.extract Pull content and data out of HTML. Extraction runs through the node methods, and the records those methods return are re-exported from this namespace for discoverability -- :class:`~turbohtml.Article`, :class:`~turbohtml.Link`, :class:`~turbohtml.StructuredData`, and :class:`~turbohtml.MicrodataItem` -- while staying importable from the package root. :func:`boilerplate` adds the per-paragraph view of :doc:`the main-content scoring `: each paragraph unit of a page classified good or boilerplate, the ``justext`` / ``boilerpy3`` call shape. .. autofunction:: boilerplate .. autoclass:: Paragraph .. autoclass:: Extraction :members: justext :func:`dates` recovers a page's publication or modification date from its ```` tags, JSON-LD, ``