HTML to text converter

What an HTML to text converter has to preserve

Removing markup is not the same as removing every substring between angle brackets. HTML carries boundaries as well as words. A paragraph must not run into the next paragraph, a line break must remain a line break, and two neighboring inline elements must not gain a space that was never in the source. Lists and tables need a plain-text shape of their own. This converter uses list markers, tabs between table cells, and newlines between rows so the result remains readable after the tags are gone.

Character references are part of that job too. A source string such as &, ©, or a numeric Unicode reference represents a character, not literal entity syntax. The extractor decodes the full HTML entity set before writing the result. Whitespace inside ordinary elements is collapsed as HTML collapses it; whitespace inside <pre> remains intact.

“Visible text” is a source contract, not a screenshot claim

The result is based on the HTML supplied above. Script, style, template, fallback noscript, and form subtrees are omitted. Elements carrying hidden or aria-hidden="true" are omitted as an explicit source decision. A meaningful image alt value appears as a named image replacement because it is the image's textual alternative, even though a browser does not normally paint it beside a successfully loaded image.

The word “visible” stops there. An external stylesheet can hide a class, move content off screen, or add words through a pseudo-element. Viewport size, fonts, user settings, and JavaScript can change what a browser finally paints. This page does not fetch those resources, run a layout engine, or claim pixel equivalence. The receipt keeps that boundary beside every result instead of burying it in an FAQ.

Paste source, paste a page selection, or open a file

If you paste tags, the tags are the input. If you copy a selection from a rendered page, browsers commonly place both plain text and HTML on the clipboard; this tool reads the HTML flavor first so headings, lists, table rows, and image alternatives can still be interpreted. A local .html or .htm file reaches the same engine. All three paths stay in the tab, and the complete output remains available to copy or download even when a very large preview is clipped for display.

A URL is intentionally not an input. Fetching arbitrary sites would introduce login state, crawler rules, cross-origin failures, remote tracking, and a misleading promise about pages whose final content exists only after scripts run. For a URL, first save or copy the source you are allowed to use. For structured data rather than readable prose, use the HTML table to CSV converter. For content that should keep headings, links, emphasis, and code syntax, use HTML to Markdown instead.