- A compression pointer is when the two higher bits are set, the code was
considering only 0xC0 as a pointer, where in reality anything from 0xC0-0xFF is
a pointer, probably went unnoticed since you need big packets to have long pointers.
- Make sure we can access the lower byte of the pointer by checking len, the
code was careful to not access past the first byte, but ignored the second.
- As per RFC9267 make sure a pointer only points backwards, this one is not so
bad, as the code had a iteration_max that ended up guarding against infinite jumps.
Lightly tested, some eyes are welcome, but these are remote DOSable.
The main motivation for this is to have sinergy with flags parsing, currently
flags for a sockaddr returns a net.Host_Or_Endpoint, but we can't just dial
from it since there isn't a variant.
Consider the following:
```
Options :: struct {
target: net.Host_Or_Endpoint `args:"pos=0,required" usage:"host:port"`,
}
before :: proc() -> (sock: net.TCP_Socket, err: net.Network_Error) {
opt: Options
flags.parse_or_exit(&opt, os.args)
switch t in opt.target {
case net.Host:
sock, err = net.dial_tcp(t.hostname, t.port)
case net.Endpoint:
sock, err = net.dial_tcp(t)
}
return
}
after :: proc() -> (sock: net.TCP_Socket, err: net.Network_Error) {
opt: Options
flags.parse_or_exit(&opt, os.args)
sock, err = net.dial_tcp(opt.target)
return
}
```
For completion, add dial_tcp_from_host() and define the upper functions in terms
of the newly added ones, cuts one repeated block, now:
from_hostname_and_port_string is parse + from_host_or_endpoint
from_hostname_with_port_override is parse + override + from_host_or_endpoint
from_host is to_endpoint + from_endpoint
from_host_or_endpoint is from_endpoint or from_host
The documentation for `setsockopt(2)` mentioned accept filters for
`EINVAL`, but I've found that it can arise for any manner of invalid
values for setting socket options.
We'll just have to leave this as a generic error.
This is a better default than not having it, since it turns errors that
would be signals into error values instead. We could take these as
options but given that we currently don't I think this at the very least
improves on the status quo.
