Hi all,
I would have thought that calling `close()` once in the server and once in the client should have been enough, but it is not.
Is it really expected that I have to use `destroy()` and not only `close()`?
If it's expected that the process hangs, it's not obvious that `close()` does not really close and that you have to use `destroy()` (This behaviour is only described on the dartdoc of
Socker#destroy). It would be great to override the inherited documentation of Socket#close to warn about that and to point to `destroy()`.
Cheers,
Alexandre