tt

desktop/src/main.ts

@@ -146,7 +146,7 @@ const registerPrivilegedSchemes = () => {
  *
  * This window will show the HTML served from {@link rendererURL}.
  */
-const createMainWindow = async () => {
+const createMainWindow = () => {
     // Create the main window. This'll show our web content.
     const window = new BrowserWindow({
         webPreferences: {
@@ -160,7 +160,7 @@ const createMainWindow = async () => {
         show: false,
     });
 
-    const wasAutoLaunched = await autoLauncher.wasAutoLaunched();
+    const wasAutoLaunched = autoLauncher.wasAutoLaunched();
     if (wasAutoLaunched) {
         // Don't automatically show the app's window if we were auto-launched.
         // On macOS, also hide the dock icon on macOS.
@@ -367,7 +367,7 @@ const main = () => {
     // Note that some Electron APIs can only be used after this event occurs.
     app.on("ready", async () => {
         // Create window and prepare for the renderer.
-        mainWindow = await createMainWindow();
+        mainWindow = createMainWindow();
         attachIPCHandlers();
         attachFSWatchIPCHandlers(createWatcher(mainWindow));
         registerStreamProtocol();

desktop/src/main/services/auto-launcher.ts

@@ -38,7 +38,7 @@ class AutoLauncher {
         }
     }
 
-    async wasAutoLaunched() {
+    wasAutoLaunched() {
         if (this.autoLaunch) {
             return app.commandLine.hasSwitch("hidden");
         } else {

desktop/src/main/services/ml-clip.ts

@@ -49,7 +49,7 @@ const clipImageEmbedding_ = async (jpegFilePath: string) => {
     return normalizeEmbedding(imageEmbedding);
 };
 
-const getRGBData = async (jpegFilePath: string) => {
+const getRGBData = async (jpegFilePath: string): Promise<number[]> => {
     const jpegData = await fs.readFile(jpegFilePath);
     const rawImageData = jpeg.decode(jpegData, {
         useTArray: true,
@@ -64,7 +64,7 @@ const getRGBData = async (jpegFilePath: string) => {
     const ny2 = 224;
     const totalSize = 3 * nx2 * ny2;
 
-    const result: number[] = Array(totalSize).fill(0);
+    const result = Array(totalSize).fill(0);
     const scale = Math.max(nx, ny) / 224;
 
     const nx3 = Math.round(nx / scale);

desktop/src/main/services/ml-face.ts

@@ -47,5 +47,7 @@ export const faceEmbedding = async (input: Float32Array) => {
     const results = await session.run(feeds);
     log.debug(() => `onnx/yolo face embedding took ${Date.now() - t} ms`);
     /* Need these model specific casts to extract and type the result */
-    return (results.embeddings as unknown as any)["cpuData"] as Float32Array;
+    return (results.embeddings as unknown as Record<string, unknown>)[
+        "cpuData"
+    ] as Float32Array;
 };