Or rpmbuild gets grumpy.

This reverts commit d6e39658ae. Apparently
this is broken on macOS so it's not something we can make use of.

Some of these were incorrectly calculated so the server or client would
wait too long before proceeding with decoding.

Change all of these to be a more explicit calculation to avoid such
issues in the future.

TCP_CORK is a Linux thing and BSD has TCP_NOPUSH instead.

We didn't include the proper headers to get the correct define, so
corking was never enabled.

Otherwise we might end up owners of something we cannot deliver data on,
which can hang applications.

Setting `rfbport` to `-1` disables TCP port listening.

It took way too long to converge on a sane value, so reduce the window
by a factor of ten. This seems to work smoothly.

So that people can easily find them and not file bug reports for
things that are just questions.

We want users to use the templates so we don't miss any relevant
information.

We computed a safe area if a client gave us a bogus one, but we didn't
actually use it. Fix this properly and make sure we don't pass on bad
coordinates further.

So the current clipboard state is properly reflected in the desktop
session.

This was out of sync with the client handling for no good reason.

The peer expects a response, so we should also be able to respond that
there is no clipboard data currently available.

The extended clipboard protocol has the ability for the peer to request
things to be sent automatically, without a request message. Make sure we
honor such settings.

It's a better match for what the key symbolises, even though it is not
as common on Unix systems. But we want to avoid getting in to tailoring
things for any one specific input method implementation.

We don't know the size of the output buffer here and giving the input
string size is just confusing and upsets the compiler in some cases.

This is what display managers do, so it is expected by many
environments.

We terminated the child instead of the parent after fork().

Reported by Jan Grulich from Red Hat.

Windows either completely omits WM_KEYUP for these, or only sends it
just before the next WM_KEYDOWN for the key.

This covers some Alt+key presses as well.

Some of the virtual key codes for Japanese keys are used for other
things on other layouts, so we need to make sure we only use this look
up table when the layout is indeed set to Japanese.

This was already handled for most exit conditions, but not the early
ones.

Mainly fixes for Japanese and Korean keyboard layouts.

Added vkey mappings for Japanese keyboards because the special keys for Japanese input do not work on TigerVNC currently.

Not sure how this has changed, but ImageMagick is now using the first
-size to also determine how to scale the SVGs, which is not what we
want.