Klaus Post

What is the problem you are trying to solve? Is this a 1.14 change? I'm not sure I understand why it is a problem to supply the dictionary when resetting?

Ah, ok, so like a [Stateless compression function](https://godoc.org/github.com/klauspost/compress/flate#StatelessDeflate) that accepts a dictionary or a [Stateless Writer](https://godoc.org/github.com/klauspost/compress/flate#NewStatelessWriter), where it keeps the history between calls, but discards everything else.

I don't have plans to submit it to the stdlib unless someone expresses a need for it. The threshold for adding stuff is of course a bit higher for the...

@nhooyr You can test https://github.com/klauspost/compress/pull/216 Slightly clunky API to remain backwards compatible.

It should store `Hex(SHA256("sts-users/cn=\d0\9c...."))`. But it would require a significant rewrite and testing, which I don't see happening soon.

@harshavardhana The problem is the length, and this seems to fail on linux. Using SHA256 for these gives a fixed length and should be strong enough for it to never...

We will not be addressing this in the Community Edition.

@felixhandte I don't disagree with anything you say, but I think you should be more proactive in your approach. Experience shows that if there is a problem an encoding is...

@felixhandte Option 1. Anything else is just overengineering and will just cause confusion down the line. If client cannot handle that they shouldn't request it - simple. But that is...

> It's just a recommendation, decoders are free to support higher or lower limits, depending on local constraints. No. Decoders should *always* support 8MB when requesting zstd encoding in a...