Why a PDF does not reflow, and how ReflowPDF edits one anyway
A PDF stores each line at a fixed place on a fixed page, so editors rewrap text only inside its box. ReflowPDF edits the document behind the pages.
, updated , a 3-minute read
You can change a word in a PDF, but you cannot add a row to its table and have the rest of the page make room. The reason is in the file format.
What a PDF contains
Open a PDF in a text editor and, past the binary header, you will find streams of drawing operators like this one:
BT /F2 14 Tf 108 742 Td (Service Agreement) Tj ETBegin text, use font F2 at 14 points, move to x 108 and y 742 (in points from the bottom left of the page), draw the string, end text. Every line on every page is a command like that. The PDF specification spends 756 pages on how to draw text and shapes exactly, and none on how text should flow. Its tags, where a file has them, name paragraphs and tables for screen readers; the drawing still puts every line at fixed coordinates, and nothing says what moves when a line gets longer.
What a PDF editor can do with that
An editor that works on the PDF itself has to infer structure from positions. Acrobat’s Edit PDF, for example, groups nearby lines into a box and rewraps text inside that box when you type. That is as far as the file lets it go. The heading under the box sits at its own coordinates and does not move down; a box that fills does not continue on the next page; a table is text placed between drawn lines, so it has no row to add. The usual workaround is to drag the content below by hand, or to type new text on top of the page, which is what overlay editors do.
Editing the document instead of the page
Word processors do not have this problem because they keep the document: paragraph A comes before paragraph B, a table follows them, and the pages are laid out from that. Change one thing and the layout is computed again.
ReflowPDF keeps a document of the same kind, in HTML and CSS, and makes the PDF from it. When a paragraph grows, everything after it moves, across pages:
The pages you see while editing are laid out by the ReflowPDF Engine, compiled to WebAssembly and running in your browser. When you download, the same engine makes the PDF on our server, so its pages break where the editor showed them.
Getting a PDF you already have into that form
A document made in the editor is already a document. A PDF from somewhere else has to be rebuilt. Import does that with AI: it reads the pages and writes them back as paragraphs, headings, tables and pictures. It takes a free account and uses one AI page for each page of the PDF; the free plan has 10 a month. Fonts and layout can come out differently from the original, so compare the two before you send it.
A PDF that came out of ReflowPDF can skip the AI. With Keep the PDF editable switched on when you download, the document travels inside the PDF, compressed and encrypted with AES-256-GCM. Imported again, it comes back as it was, without AI. Other readers see an ordinary PDF. The switch is off by default.
Pages are part of the layout
Breaking a document into pages is not a step after layout. Where a table splits decides whether its header repeats, which changes how much fits on the page, which moves the next break; a box that may not be split pushes everything after it. The ReflowPDF Engine decides page breaks while it lays the document out, following the CSS Fragmentation and Paged Media specifications, and does it again after every edit.
That is what lets a table grow by a row in the middle of a long document:
How the engine does against the specifications is published test by test on the conformance pages, including the page and column break suites.
What changes for you
In an editor that works on the page, an edit is followed by repairs: move what is below, fix the column, check the next pages. In ReflowPDF the layout is computed again after every edit: add a line and the content moves; add a row and the table grows across the page break, with its header repeated; and the PDF you download is tagged for screen readers by default.
Questions
- Why can’t PDF editors reflow content?
- A PDF stores text as runs of glyphs at fixed coordinates on fixed pages. It has no paragraphs, tables or page flow to recompute, so when a line gets longer, nothing records what should move after it.
- Does Acrobat reflow text?
- Inside a paragraph box, yes: Acrobat’s Edit PDF groups lines into boxes and rewraps text within them. The content below the box does not move, and text does not continue on the next page.
- How does ReflowPDF edit a PDF I already have?
- Import takes a free account and rebuilds the PDF with AI into paragraphs, headings, tables and pictures, using one AI page per page of the PDF (10 a month on the free plan). Fonts and layout can differ from the original, so compare the two before you send it.
- Who lays out the pages in the editor?
- The ReflowPDF Engine, compiled to WebAssembly and running in your browser. The PDF you download is made by the same engine on our server, so its pages break where the editor showed them.
The editor opens without an account. Type into a document and the pages after it are laid out again; download it as a tagged PDF. Importing a PDF of your own takes a free account.
More from the blog
How to edit a PDF on an iPhone: what is built in, and what takes an editor
An iPhone signs, annotates and fills PDFs out of the box, but cannot change the text already in them. What each built-in tool does, and how to change the text in the browser., a 2-minute readHow to edit a PDF on a Mac: what Preview does, and where it stops
Preview signs, annotates, fills forms and rearranges pages, free and on the Mac. It cannot change the text already in a PDF. What it does, step by step, and how to change the text., a 2-minute readHow to edit a PDF on Android without paying for an app
Android has no built-in PDF editor. Google Drive and Acrobat Reader fill, sign and comment for free; changing the text takes an editor, which runs in Chrome., a 2-minute read







