What’s new? v4.83 — A thousand pages open at the page you’re looking at

A thousand-page case file opens at the page you’re looking at From the network folder, opening a large PDF meant downloading it in full before seeing the first page. And every page jump could repeat the full download.

v4.83.0 — A thousand-page case file opens at the page you’re looking at

From the network folder, opening a large PDF meant downloading it in full before seeing the first page. And every page jump could repeat the full download.

The reason lies in how a PDF works: the index that says at which byte each page starts lives at the end of the file. The viewer reads it first and then jumps to the page being viewed. That requires the server to be able to deliver a chunk — and it couldn’t: it delivered the whole case file, over and over.

Measured on a real case file: flipping through thirty pages goes from 64 seconds and 600 MB to 2.7 seconds and 1.9 MB. The same goes for meeting audio and video, where an interrupted download started over from scratch.

Why it hadn’t been done before, which is the interesting part: it wasn’t a network folder problem but an encryption at rest one. Documents were encrypted in a single operation over the whole file, so delivering a chunk meant decrypting everything — and a browsing session is dozens of requests. Now they’re encrypted in blocks, each with its own seal, so reading a page touches one block and not the entire case file. The disk overhead is 0.003%.

Nothing gets converted on its own and nothing stops working. Documents that were already stored are still read exactly the same; the only thing they lack is chunked reading, so an old, large PDF keeps downloading in full until it’s converted. New ones benefit from the improvement from day one, and for old ones there’s an optional converter the administrator can run whenever they want —or never—, which checks the content byte by byte before touching anything.

Updating doesn’t touch any existing file.

large page release

← Back to the index