Hacker News Reader

Comment by dotancohen | Hacker News Reader

I haven't tried it yet, but what advantage does this have over pandoc?

Pandoc is awesome for converting text-based formats and markup files, but from my knowledge it doesn’t cover some of the more specialized PDF operations (like merging, splitting, rotating pages, etc.) or handling complex office formats with embedded elements. My tool aims to provide a quick, browser-based way to handle all those PDF tasks—no installation or command-line usage required.

I see, thank you!

How do you handle complex office formats with embedded elements? Do you reimplement ODT and other standards?

Thanks for asking! I don’t reimplement those formats myself—LibreOffice does the heavy lifting for parsing and converting office documents with embedded elements. That way, I just leverage an existing engine instead of reinventing the wheel, and it helps preserve formatting as accurately as it can.

So the web app actually loads LO components? How about MS Office? Other esoteric formats?

You mention elsewhere that all this can be done offline once the web app has loaded, all these components are pulled in?

I’m not actually loading LO into the browser—those parts run on the server side, so you still need an internet connection for complex document conversions. The offline functionality mostly covers simpler features like merging PDFs or converting images to PDF with WebAssembly libraries in the browser. For MS Office files or other advanced formats, I rely on LO’s server-side engine to handle parsing and conversion

I see, thank you for taking the time to explain.

Reply on news.ycombinator.com