Can you process HTML you already have?

Trafilatura can take an existing HTML copy and pull out main text and available metadata. Use the local path for an approved file when you only need conversion. Inspect the output for missing headings, dates or text next to a picture.

What should you save with the text?

Keep the original file or source reference, date and conversion settings. The project lists plain text and structured formats such as JSON. Decide which fields to retain and check whether you may reuse the text or only cite it.

What costs money if you run it locally?

The software is Apache 2.0. Budget for the machine and any storage. Keep a small set of approved samples because a change in the source layout or library version can change what appears in the output.

Did the text copy keep the closing deadline?

Sketch a fictional community notice where the deadline appears in the main text and the footer has a different date.

Select the evidence
Mark the deadline sentence, its heading and the date next to the footer label.
Check the extracted result
Keep the sentence with the deadline if the converter found it. If it did not, preserve the original reference and review the source.