PDF to CSV

It runs PyMuPDF's table detection on every page, which looks for the ruling lines and aligned text that make up a table. Each table it finds is written to the CSV row by row, and all pages are scanned in one pass.

All of them are written into one .csv file in page order, with the rows of one table following the rows of the previous one and no separator between them. If you need each table in its own file or on its own sheet, use PDF to Excel instead.

Usually the PDF lays out its data with spaces and tabs that only look like a table, without the structure the detector needs, or the page is a scan with no text layer at all. For scans, run OCR PDF first so there is text to detect.

Cells that contain a comma, a double quote, or a newline are wrapped in double quotes, and any quotes inside them are doubled, following RFC 4180. The file is saved as UTF-8, so accented characters and non-Latin scripts survive the trip into a spreadsheet.

No. CSV is plain text, so bold text, colors, merged cells, and column widths are dropped and only the cell values remain. Numbers are exported as the text shown in the PDF, including any currency symbols or thousands separators.

You'll be asked for the password after upload, and extraction continues once it's entered. The password is used only to open the file in your browser.

No. The table detection engine is loaded into your browser as WebAssembly and the CSV is generated locally, so the document never leaves your device. The first run takes a little longer while the engine loads.

Double-clicking the file works in most setups, since the tool uses the standard comma delimiter. If your Excel is set to a locale that expects semicolons, use Data > From Text/CSV and pick comma as the delimiter.