PDF a Excel de balde

Como funciona

A PDF has no real tables — only pieces of text placed at exact positions on the page. This tool reads those positions with pdf.js and rebuilds the rows and columns from the layout, then writes them to an Excel workbook or a CSV file, all inside your browser.

1. 1
   The text layer is read

   pdf.js (the PDF engine Firefox uses) lists every run of text on every page with its x/y position and font size. Pages with no text at all are scans; they are flagged, and the OCR tool is suggested for them.
2. 2
   Rows and columns are rebuilt

   Text sharing a baseline becomes a row. Columns are the vertical gaps that no table row crosses — found across the whole page (or every page, when all pages go on one sheet), so a single heading or footer cannot merge two columns.
3. 3
   Cells are typed

   Columns that are mostly amounts become real numbers: 1,234.56, 1.234,56, (12.00) and -12.00 are all understood. Dates are kept exactly as printed. The bank-statement option also joins wrapped description lines to the transaction above and drops repeated column headers.
4. 4
   Preview, then download

   You see the table before saving. SheetJS writes the .xlsx (one sheet per page, or all pages in one sheet), or a UTF-8 CSV is produced directly.

Ferramentas relacionadas