Assume we are part of a complete and proper installation and encode the
full expected path in to the vncserver script.

It keeps much better track of the X server and startup files than
we can do.

We need to be started as a system service for things to work correctly
anyway, so delegate the work of starting and stopping things to the
system service manager (e.g. systemd).

This is how display managers (e.g. gdm or lightdm) start sessions
and is necessary to get the proper set of environment variables.

We don't want to create files in the users home directory unless we
have to. Users can read about how this file works in our man pages.

We now have config files for more fine grained control of these
things, so avoid duplicating the functionality.

These might contain values we want if we are start from a proper
environment.

It is the most common init system these days so it should not be
hidden in the contrib/ directory.

This also removes all old SysV files from the contrib packages.

Install some example files to make things more easily discoverable.

These are EOL (or soon to be) and we want to start relying on things
present in current distributions.

There are multiple "okay" return values, not just Z_OK. Make sure we
don't bail out needlessly.

We adapted to the wrong new callback in a previous commit.

The CConnection base class wants to be able to write things these days,
so we need to provide it a place to do so.

Read username and passwrod from env.

This allows Xvnc to build with xorg-server 1.20.7, which requires OS
layers to implement a ddxInputThreadInit function when configured with
--enable-input-thread (the default).

relevant xorg-server commit: e3f26605d8

As gettext needs some stuff from Carbon and we don't want to rely on it
being pulled in as a side effect.

Modern MinGW seems to provide this, so simplify things a bit. This also
side steps some of the issue of the windows.h/winsock2.h include
ordering.

Not every mouse has three buttons e.g. laptops. Some OS might not
have support for middle mouse button emulation.

This commit adds emulation for middle mouse button when pressing both
left and right mouse button simultaneously.

This was not necessary since you could use ESC key or simply just
clicking outside of the context menu to close the menu.

Otherwise such clients cannot use Scroll Lock at all, and that is
probably worse than any effects we might get from getting out of sync.

We had an unintentional conflict with PixelBuffer::setSize() here.
But we can simplify this further as this initialization is only used
by the subclass DeviceFrameBuffer, and only once.

Since 53f913a we initialize the underlying PixelBuffer with 0x0
dimensions, which means we need to keep more explicit track of what
we are trying to allocate in the setup methods.

new throws an exception on allocation errors rather than return NULL.

This check is completely backwards and it is currently unknown how
this ever worked.

If the server doesn't support local cursors and want to render them
itself then we need to make sure the local cursor is invisible.

This also makes sure we always have some cursor allocated, so we can
remove the checks in some places.

Commit 4e61f8dbc5 fixed the GUI but forgot
to change the other places in the code.

There was even some confusion in the RFB protocol regarding this, but
the zlib implementation confirms that accepted values for compression
level is 0-9.

The GUI incorrectly noted the max level to be 6 while it in fact is 9.

Even if this note is true it just adds confusion.

The number of colors used isn't something the end-users should have to
concern themselves with. I intentionally left the information in the
man-pages.

The input stream might no longer be the raw socket, so we need to
query what's currently active. That wrapping stream might have its
own buffering and may have more data even if the socket is drained.

There might be more bytes left in the current TLS record, even if
there is nothing on the underlying stream. Make sure we properly
return this when we aren't being requested to block.

The copied rects have already been merged in to the changed rects
at this point if the client doesn't support the CopyRect encoding.