Comment by andix | Hacker News Reader

Comment by andix | Hacker News Reader

Parent Full thread

Actually I'm a bit disappointed that it can't format 128 bit integers or byte arrays. That would allow formatting UUIDs. I'm not a huge fan of public facing integer IDs. There is always the risk of leaking some kind of critical information with ascending IDs.

So I will probably keep Base64URL formatting my UUIDs to make them shorter for URLs, QR Codes and so on.

Quick example:

  20b30b32-d421-4cfb-bdbc-9a4e0475abea
  =>
  MguzICHU-0y9vJpOBHWr6g

It's important to keep in mind that there are different ways to convert an UUID to a byte array (little/big endian, order of segments, ..)

Even if it could format 128 bit integers, I don't see how that would be any better than base64-encoding them. Base64 is already pretty close to the maximum efficiency you can get for shortening UUIDs while still using URL-safe characters.

To add some of the additional features Squids provide. You can also convert any smaller integer to a string with Base64.

I actually used sqids algo for something very different where I had to encode arbitrary sized byte arrays. And with Ruby it was very simple to remove limit - I think it was matter of monkey-patching https://github.com/sqids/sqids-ruby/blob/main/lib/sqids.rb#L... to return infinity.

The reason for the limit is most likely to ensure interoperability with libraries in other languages where working with bignums is much more complicated.

This limit could easily be lifted with byte array encoding support.

The shortuuid library can do that:

https://pypi.org/project/shortuuid/

Reply on news.ycombinator.com