FLTK only allows 256 different box types, but it doesn't actually check
this when registering new ones.

Move our custom types to a valid range, and add an assert for good
measure to make sure we don't overflow FLTK's internal structures.

Since the contents of the LINGUAS file affects which targets are
created, we must tell CMake to reconfigure things if this file changes.

These aren't supported in C2x, and clang will already now complain.

Update copyright year to 2023

Follow the colors that modern desktops use when it comes to background,
text and selections.

There is something broken with these FLTK draw routines on Windows. They
leave gaps at the start and end of the arc/pie rather than filling the
whole specified span. So we need to nudge the numbers a bit to work
around this.

Inspired by modern Windows appearance, and to some extent macOS. They
have flat boxes and use white, or very light, colors for interactive
elements. Unfortunately we can't directly control the colors of
widgets, so instead we just lighten everything that uses this box type.

GNOME uses a different design, both their older and newer style. But UI
look is less consistent on Linux, so hopefully our new look is decent
enough there as well.

Work around a bug in Fl_Input_Choice where it forgets to set the proper
"input background" on some parts.

We should have a consistent color set over all widgets.

gcc doesn't support -Wformat for the wide format versions of printf()
and friends yet:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=38308

Do what glibc does and have some commented out tags to show future
intent.

gettext replaces all *printf() functions on platforms that don't fully
conform to the POSIX behaviour. Unfortunately, gettext fails to tag
these replacement functions properly so that -Wformat can still do its
thing.

Resolve this by adding a redudant declaration of the relevant functions,
with the attribute tagging in place.

The size of size_t depends on the architecture, so we need to have
different conversion to and from strings. But we don't really need that
range, so avoid the issue by using a standard integer size.

Let's log a bit more details when we need to deal with certificate
exceptions to make it easier to debug things.

The browsers let you add an exception for this case, so we should as
well.

GnuTLS can help use translate certificate issues in to user presentable
strings, so let's clean up that reporting.

The browsers let you add an exception for this case, so we should as
well.

This should have been done in 5f46d55.

The browsers let you add an exception for this case, so we should as
well.

We don't want to proceed unless we've made sure the user has approved
the issues with the certificate. So add an extra check that all status
flags have been dealt with.

GnuTLS should hopefully never set just this flag, but let's be fully
prepared for all scenarios.

We've required GnuTLS 3.x for a long time, so this code has been dead
for a while.

The WM_CLASS we set on all windows is just "vncviewer", so that it
matches the name of the .desktop file, which is what GNOME expects.

These are not valid outside of UTF-16 so seeing them in a UTF-8 sequence
means that something is wrong with that sequence. Best to filter them
out rather than letting them propagate and have unknown effects.

We should handle this in the low-level protocol code as much as possible
to avoid mistakes. This way the rest of the code can assume that strings
are always UTF-8 with \n line endings.

If we can't rely on the OS to handle corking for us, then we need to
enable our own handling of it.

The previous method isn't compatible with CMake's try_compile() as it
will respect CMAKE_EXE_LINKER_FLAGS, but not CMAKE_C_LINK_EXECUTABLE and
friends. This results in the default libraries being completely missing,
and the compile test failing.

They've patched their Xorg sources to no longer require this package.

Make sure we can actually build without XRandR libraries.

Make sure we can actually build without XFixes libraries.

Make sure we can actually build without XTest libraries.

The need for these must have got lost somewhere in the type cleanup.

They are not fully supported and break with some backends, like Ninja.

These are expected to abort if they fail to find the relevant software
and "REQUIRED" is specified.

It is more specific, and it properly sets up propagation when include
directories also need to be used further down a dependency chain.

This is important in case there are build flags that need to propagate
between libraries for things to build correctly.

Get in sync with what's recommended these days.

This is fully automated, so we can't let anything wait for user input or
the job will just hang.

The newlines aren't stripped, so they are quite sufficient separators
for the commands.

We don't want a broken job to consume our entire quota, so make sure
things are killed if the stray too far from a normal runtime.