Richard Hodges

> could we not just put the websocket headers in separate iovec entries so this could be a zero copy interface in the case of no per-message deflate and no...

I'm not an expert on the origins of Websocket, but my impression is that it was invented as a hack to allow web browsers to defeat corporate firewalls when creating...

Sure, this is possible with latest asio as asio ssl streams became moveabke in boost 1.68

I can knock something up when I get home

My reading of 3.3.2 https://www.rfc-editor.org/rfc/rfc7230#section-3.3.2 suggests that the expectation is that there will be only one content-length field and that it will either contain a list of numbers that are...

However: https://datatracker.ietf.org/doc/html/rfc2616#section-4.2 > Each header field consists of a name followed by a colon (":") and the field value. Field names are **case-insensitive**.

Are you sure your mock stream stream is correct? Have you tried it with a simple asio composed write?

This is great, thank you

Reproduced.

reproduced here for investigation: https://github.com/beast-issues/2221