very, very simple WP with midi CC out

include/libremidi/backends/winuwp/helpers.hpp

@@ -0,0 +1,63 @@
+#pragma once
+#define NOMINMAX 1
+#define WIN32_LEAN_AND_MEAN 1
+#include <libremidi/detail/midi_api.hpp>
+
+#include <mutex>
+#include <string>
+#include <thread>
+#include <vector>
+
+#include <winrt/Windows.Devices.Enumeration.h>
+#include <winrt/Windows.Devices.Midi.h>
+#include <winrt/Windows.Foundation.h>
+#include <winrt/Windows.Storage.Streams.h>
+
+namespace libremidi
+{
+inline void winrt_init()
+{
+  // init_apartment should only be called on the threads we own.
+  // Since we're the library we don't own the threads we are called from,
+  // so we should not perform this initialization ourselves.
+  // winrt::init_apartment();
+}
+
+namespace
+{
+using namespace winrt;
+using namespace winrt::Windows::Foundation;
+using namespace winrt::Windows::Devices::Midi;
+using namespace winrt::Windows::Devices::Enumeration;
+using namespace winrt::Windows::Storage::Streams;
+
+// Helper function to allow waiting for aynchronous operation completion
+// from the thread in STA. The only benefit from it compared to the
+// get() function from winrt is that we avoid an assertion if waiting
+// from the STA thread.
+template <typename T>
+auto get(T const& async)
+{
+  if (async.Status() != AsyncStatus::Completed)
+  {
+    slim_mutex m;
+    slim_condition_variable cv;
+    bool completed = false;
+
+    async.Completed([&](auto&&, auto&&) {
+      {
+        slim_lock_guard const guard(m);
+        completed = true;
+      }
+
+      cv.notify_one();
+    });
+
+    slim_lock_guard guard(m);
+    cv.wait(m, [&] { return completed; });
+  }
+
+  return async.GetResults();
+}
+}
+}