Choose a file or paste the markup. The headings, paragraphs, lists, tables and links are read out of it and typeset onto the page size you choose — as text you can select and search, in a file of a few kilobytes.
Result
pages
Because it is the document rather than a photograph of it. Your fonts, colours and layout are not reproduced; the headings, paragraphs, lists, tables and links are. The trade is deliberate: you get selectable, searchable text in a file of a few kilobytes instead of a heavy picture of a page.
Yes. Either works, and if you do both, the file wins. Pasting is often the easier path on a phone, where saving a page as a file is more work than copying it.
That is the reason this tool works the way it does. The markup is parsed as data and never inserted into a page, so scripts in it cannot execute in your browser, no remote resource is fetched while it is read, and javascript: links are removed rather than carried into the PDF.
No. Fetching them would mean reaching out to somebody else's server from your browser, which is exactly what this tool is built to avoid. The number of images left out is reported with the result rather than passed over in silence.
Yes. Every link with a real web address becomes a clickable area in the finished document, and the count is shown with the result. Relative links — the ones that only mean something on the original site — are dropped, because they would lead nowhere.