Turn a PDF table into CSV

or drop one here
Choose a PDF first.

Turn native-text PDF tables into CSV without uploading the file. Select one table for a single CSV or several for a ZIP, and compare a source sample with its cells. Spreadsheet-safe mode protects leading formula characters, while exact-text mode keeps them unchanged. Every generated CSV is parsed again and compared cell by cell before the download appears.

How to turn a PDF table into CSV

  1. Choose a PDF

    Press Choose a PDF, or drop one on the page. The file stays in this browser tab.

  2. Choose detected tables

    The tool groups aligned selectable text into rows and columns. Select one table for a direct CSV or several tables for a ZIP.

  3. Choose cell safety

    Spreadsheet-safe mode protects leading formula characters. Exact-text mode keeps those characters unchanged.

  4. Check and download

    Every generated CSV is parsed again and compared cell by cell. ZIP downloads are reopened and checked before the download appears.

What table detection can and cannot prove

A PDF records words at page coordinates, not cells in a spreadsheet. This tool looks for repeated horizontal and vertical alignment in the native text layer. It keeps the source boxes behind every detected cell and outlines their combined area in the receipt.

The exact CSV check starts after that inference. It proves that the chosen cell matrix was serialized and parsed without a changed value. It cannot prove that a merged heading belonged to column two instead of columns two and three. The side-by-side preview is part of the result, not decoration. To convert a PDF table to CSV the page must carry a real text layer, because a scanned image gives the detector nothing to align.

Questions

Does PDF to CSV upload my file?
No. The PDF, detected cells, preview and CSV stay in this browser tab.
Does this work on a scanned table?
Not yet. This version needs native selectable text. A scan needs OCR, and this tool refuses it instead of making an empty CSV.
Which tables can it detect?
It looks for text aligned into repeated rows and columns. Simple native-text tables work best. Merged cells, rotated tables, forms and multi-column page layouts may need correction outside this tool.
What if the PDF has several tables?
Select any tables you need. One selection downloads as a CSV. Multiple selections download as a ZIP with one CSV per table, so unrelated tables are never combined.
What does spreadsheet-safe mean?
A cell beginning with a character a spreadsheet may run as a formula gets a leading apostrophe. Other cell text is unchanged.
What does the check prove?
It proves the CSV can be parsed into the same rows, columns and cell strings that the detector produced. It does not prove that every visual table boundary was inferred correctly, which is why the source and cell previews are shown together.

The other pdfpix tools

All 30 of them run in a browser tab like this one. Each reads its own output back and checks it before offering the download, and none of them needs an account.

Where to go next