01 / Read
The file is read here
The delimiter and the character set are found by looking at the file. Quoted cells with commas, quotes and line breaks are read as one cell.
Free tool
Drop a CSV file and see it as a table on real pages before you save. Commas, semicolons, tabs and pipes are found for you, quoted cells with commas and line breaks are read correctly, and the header row is repeated on every page. Wide tables are fitted to the page, in landscape if that helps. Cells are shown exactly as they are in the file. The text in the PDF can be selected and searched. Nothing is uploaded.
Page
Table
The file
Press Cmd S to download. Your page settings are remembered in this browser, never the file.
Your CSV is read and laid out in your browser. Nothing is uploaded.
How it works
01 / Read
The delimiter and the character set are found by looking at the file. Quoted cells with commas, quotes and line breaks are read as one cell.
02 / Lay out
Each column gets the width its cells need. Long cells wrap, the text shrinks a little if that helps, and very wide tables continue on more pages.
03 / Save
The preview is a page of the file itself. Download PDF writes every page, with the header repeated and text you can select.
Guide
To convert a CSV to PDF is to draw the rows and columns of a comma-separated file as a table on pages. This tool does it in your browser. It repeats the header row on every page, gives each column the width it needs, and saves a PDF with real text, so you can select, copy and search it. The cells are printed exactly as they are in the file: no number is rounded, no date is reformatted, and a code such as 007 keeps its zeros.
Commas, semicolons, tabs and pipes are all found by looking at the first rows, and the box named Cells split by shows which one was used. Change it if the guess is wrong. A cell in double quotes can hold the delimiter, a line break or a doubled quote (""). A file that starts with a byte order mark is read correctly. Files saved by older programs in Windows-1252, Latin-1 or another character set can be read by choosing it under Characters. If the file has rows with fewer cells than the widest row, the empty cells are left blank and the page says so.
The default setting, Fit to the page width, works in steps. First the text may shrink to as little as 80% of the size you picked, never below 5 pt, so every row stays on one line. If that is not enough, long cells wrap onto more lines inside their column, but a word is only cut when it is longer than about 14 characters. If even the narrowest columns do not fit, the text shrinks further, down to 5 pt. Past that, the extra columns continue on more pages, with the first column repeated so each row can still be recognised. A small line at the top of those pages says which columns are shown.
The other choice, Keep size, add column pages, never shrinks the text. It wraps long cells, and when the columns still do not fit it puts the rest on more pages the same way.
How many rows fit on a page. At 9 pt a one-line row is 9 × 1.3 = 11.7 pt of text plus 2 × 2.88 pt of padding, 17.46 pt in all. An A4 landscape page is 595.28 pt tall. With Normal margins (36 pt) and page numbers on, 595.28 − 36 − 50 = 509.28 pt is left for the table, and 17.46 pt of that is the header. The remaining 491.82 pt holds 28 rows (28 × 17.46 = 488.9 pt). A file with 28 data rows is one page, and 29 rows is two.
A long file. A file with 20,000 one-line data rows at these settings becomes 20,000 ÷ 28 = 714.3, so 715 pages, each with the header on top. The preview shows one page at a time, so it stays quick, and Download PDF writes the pages a few at a time and shows its progress.
Numbers lined up. A column where nine in ten cells look like numbers, such as 1,249.50 or 18 400.00 or 89,90, is lined up on its right edge. Only the position changes. The text of each cell is the same as in the file.
To go the other way, use CSV to Excel or CSV to JSON. For plain text that is not a table, Text to PDF keeps every line as typed.
Questions
No. The file is read in this tab, laid out here, and the PDF is written here. Nothing is sent to a server, and the file is not stored. Only your page settings are remembered in this browser.
No. Every cell is printed as the text in the file. 007 stays 007, 1E5 stays 1E5 and 03/04/2026 stays as it is. Numbers can be lined up on the right, which only moves them.
Yes. It is not a picture of a table, so you can select it, copy it and search it in any PDF reader.
Yes, when First row is the header is on, which it is by default. Turn it off if your file has no header, and the first row is then printed like any other.
The PDF font has Latin, Greek, Cyrillic and Vietnamese letters only. A letter outside that is drawn as a question mark, and the line under the tool says how many there were.
The page sets no row limit. A file of 20,000 one-line rows becomes 715 pages. A very long file makes a very long PDF and takes longer, and above 150,000 rows the page warns you first. Your browser's memory is the real limit.
Not on this page. Remove the columns in a spreadsheet and save the file as CSV again, then drop it here.
No. The table uses one clean font and a light grey header, which print well in black and white. You can change the size, the margins, the shading and the lines.