PDF table extractor

How to use it

1. Drop one PDF on the card, paste it with ⌘V / Ctrl+V, or press “Choose PDF”. The first table appears as soon as its page is analysed — there is no Extract button. Copy and download unlock when the analysis finishes.
2. Auto detection looks for ruled tables first, then for columns lined up by text. Choose “Borders” or “Text alignment” to force one method. Switching the mode or the page range reuses the pages already read and updates the preview immediately.
3. Type pages such as “1-3, 5” to analyse only those pages. Commas, Chinese enumeration marks and Arabic commas are accepted, as are Arabic-Indic, Eastern Arabic and full-width digits. Leave the box empty for all pages.
4. Click a table chip to switch the preview. “First row is a header” only changes how the first row is shown and how the XLSX header is styled; the values themselves do not change.
5. If the PDF needs a password, type it in the row that appears and press Enter. The password is only used to open the file in this tab.

Example

Detection: Auto
Pages: (empty = all)
First row is a header: on

Item,Qty,Unit price,Total
Notebook,12,3.50,42.00
Desk lamp,2,24.00,48.00
USB cable,5,4.20,21.00

The preview shows the same values. Copy TSV joins cells with tabs, Download CSV adds a UTF-8 BOM so Excel opens it correctly, and Download XLSX writes one sheet per table. Numbers keep their original spelling — 3.50 stays 3.50.

Rules, limits and known issues

• Text-based PDFs only. Scanned pages, screenshots and tables inside images have no text layer; the tool reports those pages instead of guessing. There is no OCR.
• Detection is a heuristic. Missing rules, narrow gaps, two-column layouts, nested tables and tables that continue across pages can be missed or merged. Switch the detection mode or narrow the page range and try again.
• Merged cells are not rebuilt: text lands in the cell under its centre, and the neighbouring empty cells stay empty.
• Multi-line cells are joined with a single space; line breaks inside a cell are not kept.
• Tables that continue on the next page are not merged; each page is numbered on its own.
• Two-column pages can join rows of the left and right columns into one table.
• Right-to-left is decided by the majority of letters. A table with similar amounts of Arabic and Latin text can be flipped the wrong way; the note above the preview always shows which direction is in use.
• Values are copied as they appear: 89.90, 00721 and 03/04/2026 are not parsed, rounded or reordered.
• Limits: one PDF, 50 MB, 200 pages, 200 tables and 100,000 cells; a single table over 200 columns or 50,000 rows is skipped and named. XLSX holds at most 30 tables.
• The preview renders the first 100 rows; copy and download always use every row.
• If the author restricted copying, the tool still extracts the tables and shows a notice. Make sure you have the right to use them.
• Passwords: only the open (user) password is supported. It is used to open the document in this tab, is never saved, and the PDF’s encryption is not removed.
• Files over 20 MB or 50 pages show a “may take a while” hint; files over the hard limits are rejected with the actual value.

Related tools

Keep working with the extracted table in tools that also run entirely in this tab.