How this tool works
Wikipedia import uses the MediaWiki plain-text extract API. The preparation step normalizes line endings and repeated blank lines and handles section-heading markers without changing ordinary words. Copying or downloading an imported extract includes source and attribution information.
Plain text → normalized line endings and spacing → selected heading handlingA worked example
The sample “Hello world.” contains 12 Unicode code points and two whitespace-separated words.
Good to know
An extract may omit tables, formulas, references and other article elements. It is not a complete scholarly citation or an assurance of article accuracy. Imported material keeps its applicable reuse terms.
Reference: MediaWiki: TextExtracts API (opens in a new tab) ↗
A clean extract is not the same as the entire page
A Wikipedia page contains article text alongside navigation, images, tables, citations and other elements. The import asks MediaWiki for a plain-text extract rather than downloading that entire presentation and pretending every element can be preserved as ordinary sentences.
Some information can therefore be missing, especially where an article relies on a formula, table or illustration. Read the source article when that context matters. A convenient extract is a starting point for notes, not a replacement for checking the underlying material.
The Wikipedia image and file extractor reports a different view of an article’s referenced media with separate source and rights notices. The Wikipedia category explorer instead lists direct category members; neither result is part of a plain-text article extraction.
An invisible character can survive ordinary formatting removal. Use the character inspector to identify a suspicious pasted code point rather than assuming blank-looking text is empty.
Read more: MediaWiki: TextExtracts API and caveats (opens in a new tab) ↗
Source information should travel with the words
Removing visual formatting does not remove an author’s contribution or the applicable reuse terms. That is why the copy and download actions include the imported article link, contributor-history link and a licence pointer. They remain separate from the prepared body in the preview so both are easy to inspect.
If you edit the imported passage locally, the copied attribution notes that change. Check the original page for additional notices and follow the relevant terms when reusing text. The tool does not certify that a particular reuse meets every requirement.
Read more: Wikimedia Foundation: licensing of content (opens in a new tab) ↗
Preparation changes layout, not facts
The local step standardizes line endings, removes surplus blank lines and handles wiki-style heading markers. It does not summarize, fact-check or rewrite the sentences. Keeping the words unchanged makes it easier to compare your result with the available extract.
A word count based on whitespace is a simple token count, not a universal linguistic analysis. Likewise, Unicode code points are not always the same as the number of visible symbols: an emoji or accented sequence can contain several underlying code points.
Import is optional and explicit
Paste an HTTPS Wikipedia article address and choose Import article to make one direct request from your browser. No article request is made while you type. Use Lead only if the complete available extract is longer than you need.
If the service is unavailable or a request is cancelled, you can still paste plain text and prepare it locally. Editing the input during a pending request cancels that import so a late response does not overwrite your new text.
A few questions, answered
Does this send pasted text to Wikipedia?
No. Import sends the chosen article title to Wikipedia. Pasted article text is prepared locally and is not sent to our server.
Can I import an old revision?
This importer reads the current available article extract. Historical revision URLs are rejected rather than silently returning another version.
Can I paste raw HTML?
This field expects plain text. HTML is not rendered, executed or converted into a complete article layout.
Why is the output shorter than the page?
The extract API can omit some page elements. Check the original article for tables, formulas, figures, references and other context.