[bind10-dev] Resolver skeleton, location of logic, and plans/discussionstarter for that

[Jelte Jansen]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 01/21/2011 12:06 PM, Stephen Morris wrote:
> 
> The first thing should be to step back, look at the components, and define their responsibilities.  E.g. if asiolink is the interface to ASIO - handling the sending and receiving of packets asynchronously - then that is all it should do.  Resolver objects or recursive logic has no place in it.
> 

that's exactly why i started out describing the things there are now and where
they are located, as we are still taking this as the base :)

so we might need to reintroduce the 'interface to asio' task that has been on
the board for a while last week, generalize that as much as possible, then put
the actual resolving code (and related things) in its own library, and make
bin/resolver not do much more than parse option and call run() (I think we
should use that pattern for all module binaries anyway, for various reasons)

I can think of several specific things to do here, but they might all be so
incredibly related that splitting them up into separate tasks might be more
trouble than worth

Jelte
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk09QHcACgkQ4nZCKsdOncVCxACfc/98KO9rF9RXlANZj/QzJ6Go
eycAoMd2tJPMU2erAw3OtWyczEP8OZ6f
=gZXn
-----END PGP SIGNATURE-----