Merge branch 'main' into min-max-size-in-api

mediar-ai · Sep 16, 2024 · 6974342 · 6974342
2 parents 20b5b93 + ad01105
commit 6974342
Show file tree

Hide file tree

Showing 18 changed files with 329 additions and 80 deletions.
diff --git a/.cargo/config.toml b/.cargo/config.toml
@@ -1,9 +1,17 @@
 [target.x86_64-pc-windows-msvc]
 rustflags = [
-    "-C", "target-feature=+crt-static",
-    "-C", "link-args=/DEFAULTLIB:ucrt.lib /DEFAULTLIB:libvcruntime.lib libcmt.lib",
-	"-C", "link-args=/NODEFAULTLIB:libvcruntimed.lib /NODEFAULTLIB:vcruntime.lib /NODEFAULTLIB:vcruntimed.lib",
-	"-C", "link-args=/NODEFAULTLIB:libcmtd.lib /NODEFAULTLIB:msvcrt.lib /NODEFAULTLIB:msvcrtd.lib",
-	"-C", "link-args=/NODEFAULTLIB:libucrt.lib /NODEFAULTLIB:libucrtd.lib /NODEFAULTLIB:ucrtd.lib",
+  "-C",
+  "target-feature=+crt-static",
+  "-C",
+  "link-args=/DEFAULTLIB:ucrt.lib /DEFAULTLIB:libvcruntime.lib libcmt.lib",
+  "-C",
+  "link-args=/NODEFAULTLIB:libvcruntimed.lib /NODEFAULTLIB:vcruntime.lib /NODEFAULTLIB:vcruntimed.lib",
+  "-C",
+  "link-args=/NODEFAULTLIB:libcmtd.lib /NODEFAULTLIB:msvcrt.lib /NODEFAULTLIB:msvcrtd.lib",
+  "-C",
+  "link-args=/NODEFAULTLIB:libucrt.lib /NODEFAULTLIB:libucrtd.lib /NODEFAULTLIB:ucrtd.lib",
 
 ]
+
+[target.x86_64-unknown-linux-gnu]
+rustflags = ["-C", "link-arg=-Wl,--allow-multiple-definition"]
diff --git a/.github/workflows/release-app.yml b/.github/workflows/release-app.yml
@@ -7,9 +7,9 @@
 name: Release App
 
 on:
-  # push:
-  #   tags:
-  #     - "v*"
+  push:
+    tags:
+      - "v*"
   workflow_dispatch:
 
 jobs:

diff --git a/Cargo.toml b/Cargo.toml
@@ -13,7 +13,7 @@ resolver = "2"
 
 
 [workspace.package]
-version = "0.1.82"
+version = "0.1.83"
 authors = ["louis030195 <[email protected]>"]
 description = ""
 repository = "https://github.com/mediar-ai/screenpipe"

diff --git a/Formula/screenpipe.rb b/Formula/screenpipe.rb
@@ -1,16 +1,16 @@
 class Screenpipe < Formula
   desc "Library to build personalized AI powered by what you've seen, said, or heard."
   homepage "https://github.com/mediar-ai/screenpipe"
-  url "https://github.com/mediar-ai/screenpipe/releases/download/v0.1.82/screenpipe-0.1.82-aarch64-apple-darwin.tar.gz"
-  version "0.1.82"
+  url "https://github.com/mediar-ai/screenpipe/releases/download/v0.1.84/screenpipe-0.1.84-aarch64-apple-darwin.tar.gz"
+  version "0.1.84"
 
   on_macos do
     if Hardware::CPU.arm?
       url "https://github.com/mediar-ai/screenpipe/releases/download/v#{version}/screenpipe-#{version}-aarch64-apple-darwin.tar.gz"
-      sha256 "90e9b699b54c3b38840acb5f6c90b12f9923df575099d3e9a3574a7ed46f0db3" # arm64
+      sha256 "fbfb6f20dc2a0ea9cab8dddc37261f6803fa55b013c4378cfe6c3975edddb3bb" # arm64
     else
       url "https://github.com/mediar-ai/screenpipe/releases/download/v#{version}/screenpipe-#{version}-x86_64-apple-darwin.tar.gz"
-      sha256 "36bca4157bc6655d3e4f0f186f6da272ecae2128f0c128ec3f7dc3d174fdef16" # x86_64
+      sha256 "c1071dd2ba6f7e26e318d75ae2c97ef8f2959d2640a836d09242e5abd88aac9c" # x86_64
     end
   end
 

diff --git a/README.md b/README.md
@@ -111,11 +111,8 @@ brew install screenpipe
 ```bash
 screenpipe 
 ```
-we just released experimental apple native OCR, to use it:
-```bash
-screenpipe --ocr-engine apple-native
-```
-or if you don't want audio to be recorded
+
+if you don't want audio to be recorded
 ```bash
 screenpipe --disable-audio
 ```
@@ -135,10 +132,7 @@ by default screenpipe is using whisper-large that runs LOCALLY to get better qua
 ```bash
 screenpipe --audio-transcription-engine deepgram
 ```
-by default screenpipe is using a local model for screen capture OCR processing to use the cloud (through unstructured.io) for better performance use this flag:
-```bash
-screenpipe --ocr-engine unstructured
-```
+
 friend wearable integration, in order to link your wearable you need to pass user ID from friend app:
 ```bash
 screenpipe --friend-wearable-uid AC...........................F3

diff --git a/screenpipe-app-tauri/components/screenpipe-status.tsx b/screenpipe-app-tauri/components/screenpipe-status.tsx
@@ -207,6 +207,9 @@ const DevModeSettings = () => {
                 </div>
                 <p className="text-sm text-muted-foreground mb-4">
                   on = use CLI for more control
+                  <br />
+                  in dev mode, backend won&apos;t <br />auto start when starting the
+                  app
                 </p>
               </div>
             </CardContent>

diff --git a/screenpipe-app-tauri/src-tauri/Cargo.toml b/screenpipe-app-tauri/src-tauri/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "screenpipe-app"
-version = "0.2.37"
+version = "0.2.39"
 description = ""
 authors = ["you"]
 license = ""
@@ -12,11 +12,11 @@ rust-version = "1.75"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [build-dependencies]
-tauri-build = { version = "=2.0.0-rc.9", features = [] }
+tauri-build = { version = "=2.0.0-rc.11", features = [] }
 
 [dependencies]
 serde_json = "1.0"
-tauri = { version = "2.0.0-rc.10", features = [ "macos-private-api", "tray-icon", "devtools", "image-png", "image-ico"] }
+tauri = { version = "2.0.0-rc.14", features = [ "macos-private-api", "tray-icon", "devtools", "image-png", "image-ico"] }
 tower-http = { version = "0.4.0", features = ["cors", "trace"] }
 http = "0.2"
 

diff --git a/screenpipe-audio/src/core.rs b/screenpipe-audio/src/core.rs
@@ -1,3 +1,4 @@
+use crate::AudioInput;
 use anyhow::{anyhow, Result};
 use chrono::Utc;
 use cpal::traits::{DeviceTrait, HostTrait, StreamTrait};
@@ -11,8 +12,6 @@ use std::{fmt, thread};
 use tokio::sync::mpsc::UnboundedSender;
 use tokio::sync::Mutex;
 
-use crate::AudioInput;
-
 #[derive(Clone, Debug, PartialEq)]
 pub enum AudioTranscriptionEngine {
     Deepgram,
@@ -213,7 +212,7 @@ pub async fn record_and_transcribe(
                         .map_or(false, |arc| arc.load(Ordering::Relaxed))
                     {
                         let mut audio_data = audio_data_clone.blocking_lock();
-                        audio_data.extend_from_slice(bytemuck::cast_slice(data));
+                        audio_data.extend_from_slice(bytemuck::cast_slice::<i16, f32>(data));
                     }
                 },
                 error_callback,
@@ -227,7 +226,7 @@ pub async fn record_and_transcribe(
                         .map_or(false, |arc| arc.load(Ordering::Relaxed))
                     {
                         let mut audio_data = audio_data_clone.blocking_lock();
-                        audio_data.extend_from_slice(bytemuck::cast_slice(data));
+                        audio_data.extend_from_slice(bytemuck::cast_slice::<i32, f32>(data));
                     }
                 },
                 error_callback,
@@ -241,7 +240,7 @@ pub async fn record_and_transcribe(
                         .map_or(false, |arc| arc.load(Ordering::Relaxed))
                     {
                         let mut audio_data = audio_data_clone.blocking_lock();
-                        audio_data.extend_from_slice(bytemuck::cast_slice(data));
+                        audio_data.extend_from_slice(data);
                     }
                 },
                 error_callback,
@@ -356,7 +355,9 @@ pub async fn list_audio_devices() -> Result<Vec<AudioDevice>> {
 
 pub fn default_input_device() -> Result<AudioDevice> {
     let host = cpal::default_host();
-    let device = host.default_input_device().unwrap();
+    let device = host
+        .default_input_device()
+        .ok_or(anyhow!("No default input device detected"))?;
     Ok(AudioDevice::new(device.name()?, DeviceType::Input))
 }
 // this should be optional ?