Print only the part of the article you want
Untick the sections you did not come for, drop the references, the images or the tables, cut a single paragraph or a single table row — then generate. What you get is a PDF of the part you actually wanted, in your browser, with no account.
Most of a Wikipedia article is not what you came for
On a well-cited article the reference list runs to pages, and after it come Notes, Further reading, External links and a stack of navigation boxes. If you are printing to read, or to hand out, or to keep, that is a lot of paper spent on apparatus.
The usual answer is to export the whole thing and delete pages afterwards in a PDF editor, which leaves the page numbers, the contents page and the citation markers all describing a document that no longer exists. Choosing before you generate is a different thing: everything is typeset around what you kept.
Turning it on
In the app’s header there is a Content Selection button with an On/Off pill. Switch it on, then search and open an article as usual. Instead of going straight to a PDF, the article opens in a panel called Article Contents: its sections down the left, a preview on the right, and Generate PDF in the corner when you are done.
Untick anything you do not want. Select All and Clear All are there for when it is quicker to start from nothing and add three sections back.
Dropping the references, properly
Untick References — and Notes, Bibliography, Further reading, whatever this article calls them — and they are gone from the document, the contents page and the bookmarks.
The part that is easy to miss: the body text is now full of
[1], [2], [3] markers pointing at
a list that is not there any more. Those are stripped too, one by one,
and only the ones whose target actually went — a marker whose reference
survived is left exactly where it is. You end up with clean prose rather
than prose peppered with dead numbers.
Or drop a whole kind of thing
Above the section list is a row of four filters with counts — Text, Figures, Images and Tables. Switching one off removes that category from the whole document at once, which is the quick way to a text-only copy that prints in a fraction of the pages.
Three things are deliberately never swept up by this. Headings are not offered at all — removing one would orphan its section and break the contents page and the bookmarks. The infobox has its own control. And a displayed equation, which is technically an image, is exempt from Images: somebody turning off photographs almost never means “and delete the mathematics”.
Finer than a section
Click a section to preview it, and everything in it becomes editable in place. This is where a whole-section tick box is too blunt.
- Show first N paragraphs At the top of the preview. Keeps a section’s opening and drops the rest — useful for a long history section you want the gist of.
- ✕ Remove from PDF On any block: a paragraph, a list, a quotation, a table, a picture.
- Hide, ⇄ Replace, and a width slider On pictures. Swap an image for another from the article, or set how wide it sits on the page.
- ✕ on a table row or column A wide comparison table often carries three columns you care about and nine you do not.
- Click the text to edit it Paragraphs, headings and captions. Click, type, click away.
- ⠿ Drag to reorder Move a block up or down the section.
- + Add note Adds a paragraph of your own at the end of the section — a heading for a handout, a line of context for whoever you are giving it to.
The document says what you changed
Wikipedia’s licence asks you to indicate when you have changed the text, and a document that has been cut is a changed document. So the closing credit says which kinds of change were made — added, altered, removed — rather than a blanket disclaimer, and the margin carries a thin rule beside anything you added or edited. This is on by default and costs a verbatim export nothing, because a document with no changes has nothing to mark.
A pure deletion gets the sentence and no rule, which is right: there is nothing left on the page to draw a line beside.
Changing your mind afterwards
The panel does not close when you generate. There is a ✎ Resume Editing button in the viewer bar, so you can look at the PDF, decide the fourth section should go after all, and pick up exactly where you were.
Your choices also survive the session. A document you come back to through Recent is fetched from Wikipedia again rather than stored, so the selection has to be re-applied to a freshly parsed article — and a deleted paragraph is found again by its content, never by its position. If somebody has edited that paragraph on Wikipedia in the meantime, or there are two the app cannot tell apart, the block simply comes back. Every way it can fail leaves you with more of the article, never with the wrong paragraph deleted.
Three kinds of edit are not remembered by a recall: text you edited in place, table rows and columns you deleted, and reordering. Removed sections, removed blocks, trims and the type filters all come back. If you have done heavy in-place editing, keep the PDF rather than planning to regenerate it.
Next
Section selection works the same way on a book of several articles, where each article appears as a chapter with its sections nested underneath — so you can take one section from one article and all of another. If the result is going on paper, the printing settings are worth a minute first. There is a fuller account of what does and does not work well, and of exactly what leaves your browser, on the about page.
Free, and nothing to sign up for.
Last updated 3 September 2026.