How to reflow a PDF on Android

Short answer: a PDF describes a fixed page, so “reflow” means rebuilding that page’s text into paragraphs that wrap to your screen, the way an EPUB does. In Lectern’s PDF reader, open the book’s detail sheet and tap Reflow text. It only works on PDFs that actually contain text; a scanned page has none to work with.

What reflow actually changes

A PDF stores glyphs at fixed coordinates on a fixed page, which is why zooming in on a phone means panning back and forth across a line. Reflow throws that page layout away and rebuilds the words into ordinary paragraphs: it clusters the text into lines, joins lines into paragraphs, splits multi-column pages into a left-to-right reading order, de-hyphenates words that were broken across a line ending, and reconnects a sentence that was cut off by a page break.

Once that’s done, the result behaves like any other book Lectern opens: your normal reader typography (font, text size, line spacing, margins) applies, and it stays sized to your screen instead of the original page. The reflowed text is cached the first time, keyed to that book, so reopening it later is instant and the original PDF file on disk is never touched.

Step by step: reflow a PDF in Lectern

1. Open your library and tap the PDF you want to read (not long-press, a normal tap opens the book’s detail sheet).
2. Look for two buttons under the main Open as PDF button: Reflow text and Listen.
3. Tap Reflow text to read it as flowing paragraphs, or tap Listen to open the same reflowed text and start read-aloud straight away.
4. Wait through the one-time build on a long PDF: a “Preparing reflow” progress bar appears while Lectern extracts and reconstructs the text. Short files finish almost instantly; a long, image-heavy book can take a little longer, once.
5. Adjust text size, line spacing and margins from the reader’s own settings, exactly as you would for an EPUB.
6. Switch back any time by reopening the book’s detail sheet and choosing Open as PDF for the original fixed page; you come back to the same page.

What stays available once you reflow

Reflowed text runs through the same reading screen as every other book in Lectern, not a stripped-down PDF-only view, so the usual tools come with it:

How columns, footnotes and headers behave

This is the part that makes reflow worth using instead of a plain text dump, and it’s also where the honest limits are:

• Multi-column pages (papers, newsletters) are detected and split into left-to-right reading order, so the columns don’t interleave into nonsense. A page with no clear column gap is treated as one column, same as before.
• Footnotes at the bottom of a page, in visibly smaller type, are pulled out and placed after that page’s body text instead of breaking into the middle of a paragraph. They’re still included, just moved to the end of the page’s flow.
• Running headers, footers and bare page numbers that repeat across most pages are detected and dropped automatically, so you don’t hear or read “Chapter 3 · 42” between every paragraph. This detection needs a document of at least 10 pages to tell a repeating header from a coincidence, so a short handout or a single chapter export may still carry its header text into the flow.
• Tables aren’t rebuilt into rows and columns. Reflow reconstructs reading order, not a spreadsheet, so a table’s cells come out as running text in roughly the order they sit on the page. That’s fine for a table that’s mostly a labelled list; a dense data table reads better in the original fixed layout.
• A sentence split by a page break is joined back together across the two pages, so you don’t hear it cut off mid-thought at the old page boundary.

When reflow does not help

• Scanned or photographed pages. If a PDF is an image of a page rather than real text, there is nothing for reflow to work with, and Lectern doesn’t run OCR (optical character recognition) to create one. Try selecting a word in the PDF first: if nothing gets highlighted, it’s a scan. Lectern will tell you directly with “This PDF has no text layer to reflow (it looks like scanned images). Use page or scroll mode instead.” Run the file through a separate OCR tool first (a free desktop option like OCRmyPDF works well), then reopen it in Lectern and reflow will work normally.
• Forms, sheet music, comics and infographics. Anything where the visual position of elements on the page is the content, not just its container, loses meaning once turned into a paragraph stream. Read these as a fixed page.
• Dense data tables and heavily annotated diagrams. As above: the numbers survive, the grid doesn’t. This is really the same page-vs-text trade-off covered in EPUB vs PDF: reflow can make a PDF behave more like an EPUB, but it can’t give it an EPUB’s layout freedom retroactively.
• When you need the original page number. Reflow re-paginates around your text size instead of the printed page, so if you’re citing a specific page for a reference or a class, open the PDF in its original fixed layout to read off the real page number.

A quick way to check first

Before reflowing a long document, open it as a PDF and try to select a word on one page. If a text selection appears, the file has a text layer and reflow will work; if nothing selects, it’s an image and reflow has nothing to build from. This is the same test worth running before trying to read a PDF aloud on any app, not just Lectern.

Frequently asked