Does it work on scanned PDFs?

No — scanned PDFs are images with no text layer. You'd need OCR first to add a text layer.

Why are my cells merged or split?

The clustering uses fixed tolerances (3 pt Y threshold). Complex layouts with merged cells, multi-line rows or variable row heights may not reconstruct cleanly.

Are multi-page tables supported?

Yes — text from all pages is combined with a Y offset per page so rows don't merge across pages.

No. Extraction happens entirely in your browser via pdf.js.

PDF Table to CSV

Reconstruct tables from a text-based PDF into CSV — local.

100% private — no uploads PDF ToolsFree & unlimited Read tutorial

How to use PDF Table to CSV

Follow these steps — everything happens instantly in your browser.

1
Step 1. Drop a PDF containing a table onto the zone.
2
Step 2. Click “Extract table to CSV” — text positions are read and clustered into rows/columns.
3
Step 3. Copy or download the CSV result.