Co-Authored-By: Félix Baylac-Jacqué <felix@alternativebit.fr>

Co-Authored-By: Félix Baylac-Jacqué <felix@alternativebit.fr>

Also add some error codes used in the response header.

Co-Authored-By: Félix Baylac-Jacqué <felix@alternativebit.fr>

Co-Authored-By: Félix Baylac-Jacqué <felix@alternativebit.fr>

Co-Authored-By: Félix Baylac-Jacqué <felix@alternativebit.fr>

Co-Authored-By: Florian Klink <flokli@flokli.de>

We implement the GETAI (aka. get address info) operation. Most of the
glibc applications will rely on this operation to perform the hostname
-> addrs resolutions.

We add some unit tests checking that we're correctly serializing the
address info header and its payload. The expected data used in these
tests has been extracted from various Nscd socket dumps. We also add
an integration test checking that getaddrinfo is able to resolve
localhost.

Co-Authored-By: Florian Klink <flokli@flokli.de>

Previously IPv6-only results would return an empty successful result
message. This doesn't seem to be a valid response from the perspective
of Glibc. In my case it caused Firefox to crash.

We don't have to use a lambda to return a value from a
sub-expression. Rust takes the last value in a scope as return value
even if it isn't a function.

The function belongs to the data format and not to the actual
processing logic. This isn't a clean cut just yet all over the
codebase but moving this very fragile piece into its own function will
enable us to test it in isolation.

This makes the result filtering similar and thus should be less confusing.

This is as efficient as doing the manual pre-allocation as the
`to_vec` method will use the length of the slice for the initial size
allocation of the vector.

Previously this would produce a malformed packet as far as glibc is
concerned.

If there are no addresses the "found" field should not be one and the
other lookup fields can also be left blank.

This moves the "well-known" wire format structs to constants that can
be reused without having to duplicate their definition all the time.

We internally used getai to restpond to the
gethostbyname/gethostbyaddr operations. Sadly, it does not behave as
expected and breaks some tools (like hostname --fqdn, see
#4.

We FFI the right Glibc gethostbyname_2r/gethostbyaddr_2r (now
deprecated) functions and use it to back the GETHOSTBYNAME,
GETHOSTBYNAME6, GETHOSTBYADDR, and GETHOSTBYADDR6 Nscd interfaces.

Using sockburp, we realized the hostent serialization function was
bogus: we totally forgot to serialize the aliases. This commit fixes
this and makes sure we're producing bit-to-bit identical results with
Nscd for gethostbyname/getaddrinfo.

Took me three try to get this right. This is actually the third
full rewrite.

The Nscd behaviour for these two legacy functions is *really*
confusing. We're supposed to ignore the herrno (herrno != errno!!) and
set it to 0 if gethostbyaddr/name returns a non-null hostent. If we
end up with a null hostent, we return the herrno together with a dummy
hostent header.

I tried to keep things as safe as possible by extracting the glibc
hostent to a proper rust structure. This structure mirrors the libc
hostent in a Rust idiomatic way. We should probably try to upstream
the FFI part of this commit to the Nix crate at some point.

Fixes #4

Gethostbyname: use proper glibc function