About this site

A small collection of PDF tools that need nothing but a browser.

What it is

J27J offers fourteen free PDF operations: merge, split, extract pages, delete pages, reorder pages, rotate, insert blank pages, compress, add page numbers, change page size, strip metadata, extract text, convert pages to images and images to PDF. All of them run inside your browser. No account, no installation, no upload.

Why local-only

Because uploading was never necessary. Browsers have been capable of this for years: they can parse a PDF's object structure and render a page to a bitmap. Merging two files means moving page objects from one document tree to another — doing that in the browser is identical to doing it on a server, except that the file never leaves your machine.

Contracts, medical records, scans of identity documents, unreleased proposals: none of these should cross someone else's server just to change format. So this site has had no upload endpoint since the first line of code.

What it deliberately does not do

  • PDF to Word. The hard part is not the text, it is reconstructing the layout — which line was a heading, which block was a table, where the images sat. That needs a layout model running server-side. A browser-only version produces a document that takes longer to repair than to retype.
  • Unlocking encrypted PDFs. Most of that demand sits on the wrong side of copyright and compliance lines. Not here.
  • Editing existing text. A PDF's content stream is coupled to its font subsets, so changing one word can break the rest of the paragraph. Safe editing needs professional software, not a web form.

The alternatives for those three: open the PDF in Word or LibreOffice and save as .docx; ask the sender for the password; and edit the original document the PDF was exported from, then export again.

How to verify any of this

You do not have to take the above on trust. Open any tool page, then either disconnect from the internet or open your browser's developer tools on the Network tab, and drop in a PDF. The tool keeps working, and no upload request appears.

Open-source components

PDF parsing and rendering use pdf.js (Mozilla, Apache-2.0). Page manipulation uses pdf-lib (MIT). Multi-file downloads are zipped with fflate (MIT). pdf.js also ships external resources: Adobe's character maps, Liberation and Foxit substitute fonts, and a few WebAssembly decoders. Copyright notices and full licence texts are on the licences page.

How the site is funded

Advertising. There are no paid tiers, no watermarks and no file-size upsell; display ads pay for the bandwidth. Ads are labelled and kept away from the controls, and blocking them does not affect any tool. Details are in the privacy policy.

Contact

Found a bug or want to suggest a tool: guaidaowei@gmail.com.