@0biwank/screen-capture 2.0.1 → 2.0.2

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@0biwank/screen-capture",
-  "version": "2.0.1",
+  "version": "2.0.2",
   "description": "Native screen + audio capture for macOS with file-backed HLS output.",
   "main": "index.js",
   "gypfile": false,
@@ -41,11 +41,6 @@
     "index.d.ts",
     "prebuilds/**/*.node",
     "prebuilds/SHA256SUMS",
-    "binding.gyp",
-    "include/**/*",
-    "src/**/*",
-    "scripts/**/*",
-    "vendor/ffmpeg/README.md",
     "README.md",
     "SOURCE.md",
     "THIRD_PARTY_NOTICES.md",

package/prebuilds/SHA256SUMS

@@ -1,2 +1,2 @@
-92899877a4d6a464d7742998221b410ea2a404fe113743c8243c1ed3917e82b9  darwin-arm64/native_capture.node
-e6b6659d0e18c782a1cd95964fc7c263f771500ded12b4a1bcc996af448dac59  darwin-x64/native_capture.node
+58b47e3851865c139191fcbe940b2cad6247511b99094874971ba238e63c4aa3  darwin-arm64/native_capture.node
+c5f205c3297f81269d74f2d6a6bd97bbd7f96a7c47c130b19c2f5395ccec2fb7  darwin-x64/native_capture.node

package/binding.gyp

@@ -1,58 +0,0 @@
-{
-  "targets": [
-    {
-      "target_name": "native_capture",
-      "sources": [
-        "src/addon.cpp",
-        "src/SourceHelper.mm",
-        "src/SourceHelperWrapper.mm",
-        "src/DesktopCapturer.mm",
-        "src/CameraCapturer.mm",
-        "src/MediaPipeline.cpp",
-        "src/HLSMuxer/VideoEncoder.cpp",
-        "src/HLSMuxer/AudioEncoder.cpp",
-        "src/HLSMuxer/FileHLSMuxer.cpp"
-      ],
-      "include_dirs": [
-        "<!@(node -p \"require('node-addon-api').include\")",
-        "./vendor/ffmpeg/darwin-<(target_arch)/include",
-        "./include"
-      ],
-      "libraries": [
-        "<(module_root_dir)/vendor/ffmpeg/darwin-<(target_arch)/lib/libavformat.a",
-        "<(module_root_dir)/vendor/ffmpeg/darwin-<(target_arch)/lib/libavcodec.a",
-        "<(module_root_dir)/vendor/ffmpeg/darwin-<(target_arch)/lib/libswscale.a",
-        "<(module_root_dir)/vendor/ffmpeg/darwin-<(target_arch)/lib/libswresample.a",
-        "<(module_root_dir)/vendor/ffmpeg/darwin-<(target_arch)/lib/libavutil.a",
-        "<(module_root_dir)/vendor/ffmpeg/darwin-<(target_arch)/lib/libx264.a",
-        "-lm"
-      ],
-      "cflags!":    ["-fno-exceptions", "-fno-rtti"],
-      "cflags_cc!": ["-fno-exceptions", "-fno-rtti"],
-      "defines": ["NAPI_DISABLE_CPP_EXCEPTIONS"],
-      "conditions": [
-        ["OS==\"mac\"", {
-          "xcode_settings": {
-            "CLANG_ENABLE_OBJC_ARC": "YES",
-            "GCC_ENABLE_CPP_EXCEPTIONS": "YES",
-            "GCC_ENABLE_CPP_RTTI": "YES",
-            "MACOSX_DEPLOYMENT_TARGET": "14.0",
-            "OTHER_LDFLAGS": [
-              "-framework", "Foundation",
-              "-framework", "AVFoundation",
-              "-framework", "CoreAudio",
-              "-framework", "AudioUnit",
-              "-framework", "AudioToolbox",
-              "-framework", "CoreMedia",
-              "-framework", "CoreVideo",
-              "-framework", "ApplicationServices",
-              "-framework", "CoreGraphics",
-              "-framework", "CoreFoundation",
-              "-framework", "ScreenCaptureKit"
-            ]
-          }
-        }]
-      ]
-    }
-  ]
-}

package/include/CameraCapturer.h

@@ -1,54 +0,0 @@
-#pragma once
-#include <string>
-#include <vector>
-#include <functional>
-
-/**
- * CameraDevice
- * ------------
- * Represents a physical camera found by AVCaptureDevice.
- */
-struct CameraDevice {
-    std::string uid;
-    std::string name;
-};
-
-/**
- * CameraFrameCallback
- * -------------------
- * Returns a CMSampleBufferRef (as void*) containing the raw camera frame.
- */
-using CameraFrameCallback = std::function<void(void* cm_sample_buffer_ref)>;
-
-/**
- * CameraCapturer
- * --------------
- * A standalone macOS camera capturer using AVCaptureSession.
- * Isolated from ScreenCaptureKit to ensure a clean architecture.
- */
-class CameraCapturer {
-public:
-    CameraCapturer();
-    ~CameraCapturer();
-
-    /** List all available video input devices (cameras) on this Mac. */
-    static std::vector<CameraDevice> listDevices();
-    
-    /** 
-     * Start capturing from a specific camera.
-     * @param device_uid The unique ID of the camera (empty for default).
-     * @param width Requested width (closest match used).
-     * @param height Requested height (closest match used).
-     * @param fps Requested framerate.
-     * @param cb Callback triggered for every raw frame.
-     */
-    bool start(const std::string& device_uid, uint32_t& width, uint32_t& height, int fps, CameraFrameCallback cb);
-    
-    /** Stop the capture session and release hardware resources. */
-    void stop();
-
-private:
-    // ObjC objects (void* to avoid ObjC headers in C++ includes)
-    void* m_session = nullptr;  // AVCaptureSession
-    void* m_delegate = nullptr; // AVCaptureVideoDataOutputSampleBufferDelegate
-};

package/include/DesktopCapturer.h

@@ -1,83 +0,0 @@
-#pragma once
-
-#include "Types.h"
-#include <atomic>
-#include <functional>
-#include <memory>
-#include <string>
-
-using SegmentReadyCallback = std::function<void(const std::string&)>;
-
-struct CaptureOptions
-{
-    CGDirectDisplayID display_id = 0;
-    bool capture_window = false;
-    CGWindowID window_id = 0;
-    bool no_screen = false;
-
-    bool camera = false;
-    std::string camera_device_uid;
-
-    uint32_t width = 1920;
-    uint32_t height = 1080;
-    uint32_t fps = 30;
-    uint32_t gop_size = 30;
-    uint32_t video_bitrate = 9'000'000;
-    bool crop_enabled = false;
-    uint32_t crop_x = 0;
-    uint32_t crop_y = 0;
-    uint32_t crop_width = 0;
-    uint32_t crop_height = 0;
-
-    bool system_audio = true;
-    bool microphone = false;
-    std::string mic_device_uid;
-    uint32_t sample_rate = 48000;
-    uint32_t audio_bitrate = 128'000;
-    uint32_t channels = 2;
-    bool exclude_app_audio = false;
-
-    std::string output_dir;
-    std::string recording_id;
-    uint32_t segment_time_sec = 4;
-};
-
-class DesktopCapturer
-{
-public:
-    DesktopCapturer();
-    ~DesktopCapturer();
-
-    bool selectSource(CaptureSource source);
-    bool startCapture(const CaptureOptions &options, SegmentReadyCallback segment_ready_callback = nullptr);
-    bool stopCapture();
-    bool pauseCapture();
-    bool resumeCapture();
-    const std::string &lastError() const { return m_last_error; }
-
-    void on_video_sample(void *cm_sample_buffer_ref);
-    void on_camera_sample(void *cm_sample_buffer_ref);
-    void on_audio_sample(void *cm_sample_buffer_ref);
-    void on_mic_pcm(const float *data, uint32_t frames, uint32_t sample_rate, uint32_t channels,
-                    double host_time_sec);
-
-private:
-    void *m_stream = nullptr;
-    void *m_config = nullptr;
-    void *m_filter = nullptr;
-    void *m_delegate = nullptr;
-    void *m_audio_engine = nullptr;
-    void *m_camera_capturer = nullptr;
-
-    CaptureSource m_source;
-    CaptureOptions m_options;
-    std::string m_last_error;
-
-    bool m_capturing = false;
-
-    struct Impl;
-    std::unique_ptr<Impl> m_impl;
-
-    bool setup_mic_capture();
-    void stop_mic_capture();
-};

package/include/HLSMuxer/AudioEncoder.h

@@ -1,75 +0,0 @@
-#pragma once
-
-#include <iostream>
-#include <queue>
-#include <string>
-#include "Types.h"
-#include <vector>
-#include <cstdint>
-
-extern "C"
-{
-    struct AVCodec;
-    struct AVStream;
-    struct AVFormatContext;
-    struct AVCodecContext;
-    struct SwrContext;
-    struct AVAudioFifo;
-    struct AVFrame;
-    struct AVPacket;
-}
-
-class AudioEncoder
-{
-private:
-    // state variables
-    bool m_initialized;
-
-    // conversion essentials
-    int m_sampleRate;
-    int m_channels;
-    int m_bitrate;
-    int64_t m_pts;
-    int64_t m_sampleCount;
-    int64_t m_nextInputPts;
-    std::queue<AVPacket *> m_packetQueue;
-    std::vector<std::string> m_chunkFileNames;
-
-    // FFMpeg
-    const AVCodec *m_codec;
-    OutputMode mode;
-    AVCodecContext *m_codecContext;
-    SwrContext *m_swrContext;
-    AVAudioFifo *m_audioFifo;
-    AVFrame *m_frame;
-
-    // standalone
-    AVStream *m_stream;
-    AVFormatContext *m_formatContext;
-    int currentChunkIndex;
-    double targetChunkDuration;
-    int64_t currentSegmentStartPts;
-    double currentSegmentDuration;
-
-private:
-    bool shouldStartNewChunk() const;
-    void forceNewChunk();
-    // Drain encoder and handle remaining packets based on mode
-    void flushEncoder();
-
-public:
-    AudioEncoder();
-    ~AudioEncoder();
-    bool initialize(int sampleRate, int channels, int bitrate, OutputMode mode);
-    bool encodeSamples(const uint8_t *samples, int numSamples, int64_t pts = -1);
-    AVPacket *getNextPacket();
-    bool hasPackets() const;
-    std::string generatePlaylist() const;
-    void cleanup();
-
-    bool isInitialized() { return m_initialized; }
-    int getSampleRate() { return m_sampleRate; }
-    int getChannels() { return m_channels; }
-    const AVCodecContext *getCodecContext() const { return m_codecContext; }
-    OutputMode getMode() const { return mode; }
-};

package/include/HLSMuxer/FileHLSMuxer.h

@@ -1,63 +0,0 @@
-#pragma once
-
-#include <cstdint>
-#include <functional>
-#include <set>
-#include <string>
-
-extern "C" {
-struct AVBSFContext;
-struct AVCodecContext;
-struct AVFormatContext;
-struct AVPacket;
-struct AVStream;
-}
-
-class FileHLSMuxer {
-public:
-    FileHLSMuxer();
-    ~FileHLSMuxer();
-
-    bool init(const std::string& output_dir,
-              const std::string& recording_id,
-              int segment_time_sec,
-              const AVCodecContext* video_codec_ctx,
-              const AVCodecContext* audio_codec_ctx);
-
-    void set_expected_segment_duration_ms(double duration_ms);
-    void set_segment_ready_callback(std::function<void(const std::string&)> callback);
-
-    void push_video(AVPacket* pkt);
-    void push_audio(AVPacket* pkt);
-
-    void shutdown();
-    bool is_initialized() const { return m_initialized; }
-
-private:
-    int write_packet(AVPacket* pkt, bool is_video);
-    bool init_h264_bsf_if_needed();
-    void log_overlong_segment_if_needed(AVPacket* pkt);
-    void notify_ready_segments();
-
-    AVFormatContext* m_fmt_ctx = nullptr;
-    AVStream* m_video_stream = nullptr;
-    AVStream* m_audio_stream = nullptr;
-    AVBSFContext* m_h264_bsf_ctx = nullptr;
-
-    std::string m_output_dir;
-    std::string m_recording_id;
-    std::string m_playlist_path;
-    std::set<std::string> m_notified_segments;
-    std::function<void(const std::string&)> m_segment_ready_callback;
-    bool m_initialized = false;
-
-    int m_fps = 30;
-    int m_sample_rate = 48000;
-    int m_audio_frame_samples = 1024;
-    int m_video_tb_num = 1;
-    int m_video_tb_den = 30;
-    int m_audio_tb_num = 1;
-    int m_audio_tb_den = 48000;
-    double m_expected_segment_duration_ms = 0.0;
-    int64_t m_last_segment_key_pts = -1;
-};

package/include/HLSMuxer/HLSMuxer.h

@@ -1,13 +0,0 @@
-#include <Foundation/Foundation.h>
-
-extern "C" {
-#include <libavcodec/avcodec.h>
-#include <libavformat/avformat.h>
-#include <libswresample/swresample.h>
-}
-
-class Encoder {
-private:
-  bool m_Initialized;
-  
-};

package/include/HLSMuxer/VideoEncoder.h

@@ -1,90 +0,0 @@
-#pragma once
-/*
-This Encoder in the default format needs to produce H264 packets
-and when in the seperate stream mode it needs to produces TS chunks.
-
-- Input: CVPixelBufferRef, BGRA
-- Output:
-    - Default: H264 packets for the TS Muxer to process
-    - Seperate Stream: TS chunks with playlist.
-
-I need to track frame count, timestamps, keyframes
-*/
-
-#include "Types.h"
-#include <cstdint>
-#include <iostream>
-#include <queue>
-#include <vector>
-
-extern "C"
-{
-  struct AVCodecContext;
-  struct AVPacket;
-  struct AVCodec;
-  struct AVFrame;
-  struct SwsContext;
-  struct AVStream;
-  struct AVFormatContext;
-}
-
-class VideoEncoder
-{
-  // all variables
-private:
-  int frameHeight;
-  int frameWidth;
-  int frameRate;
-  bool isPaused;
-  bool m_initialized;
-
-  // FFMPEG
-  AVCodecContext *m_codecContext;
-  const AVCodec *m_codec;
-  AVStream *m_stream;
-  AVFormatContext *m_formatContext;
-  AVFrame *m_frame;
-  SwsContext *m_swsContext;
-
-  // Timing & state
-  int64_t m_frameCount;
-  int64_t m_pts;
-  int64_t m_dts;
-  int64_t m_lastInputPts;
-  int64_t currentSegmentStartPts;
-  double currentSegmentDuration;
-
-  // Chunking
-  int currentChunkIndex;
-  double targetChunkDuration;
-  OutputMode mode;
-  std::vector<std::string> m_chunkFileNames;
-
-  // output buffer
-  std::queue<AVPacket *> m_packetQueue;
-
-  // all functions
-public:
-  VideoEncoder();
-  ~VideoEncoder();
-  bool initialize(int width, int height, int fps, int bitrate, OutputMode mode, int gopSize = 0);
-  bool encodeFrame(uint8_t *pixelData, int stride, int64_t pts = -1);
-  void cleanup();
-
-  // Chunk control
-  void forceKeyFrame();
-  bool shouldStartNewChunk() const;
-
-  // Muxed mode when we want just single stream
-  AVPacket *getNextPacket();
-  bool hasPackets() const;
-  std::string generatePlaylist() const;
-
-  // getters to get the state of the VideoEncoder
-  bool isInitialized() const { return m_initialized; }
-  int getWidth() const { return frameWidth; }
-  int getHeigth() const { return frameHeight; }
-  int getFPS() const { return frameRate; }
-  int64_t getFrameCount() const { return m_frameCount; }
-  const AVCodecContext *getCodecContext() const { return m_codecContext; }
-};

package/include/MediaPipeline.h

@@ -1,39 +0,0 @@
-#pragma once
-#include <cstdint>
-#include <functional>
-#include <memory>
-#include <string>
-
-struct PipelineConfig {
-    int      width          = 1920;
-    int      height         = 1080;
-    int      fps            = 30;
-    int      gop_size       = 30;
-    int      video_bitrate  = 9'000'000;
-    int      sample_rate    = 48000;
-    int      channels       = 2;
-    int      audio_bitrate  = 128'000;
-    std::string output_dir;
-    std::string recording_id;
-    int      segment_time_sec = 4;
-    std::function<void(const std::string&)> segment_ready_callback;
-};
-
-class MediaPipeline {
-public:
-    MediaPipeline();
-    ~MediaPipeline();
-
-    bool init(const PipelineConfig& cfg);
-
-    void push_video_frame(const uint8_t* data, int stride, int64_t pts_ticks,
-                          bool is_keyframe_hint = false);
-
-    void push_audio_s16(const int16_t* data, int num_frames, int64_t pts_samples);
-
-    void flush_and_shutdown();
-
-private:
-    struct Impl;
-    std::unique_ptr<Impl> m_impl;
-};

package/include/SourceHelper.h

@@ -1,41 +0,0 @@
-#pragma once
-
-#include <Foundation/Foundation.h>
-#include <ScreenCaptureKit/ScreenCaptureKit.h>
-#include <cstdint>
-#include <iostream>
-#include <vector>
-#include <ApplicationServices/ApplicationServices.h>
-/*
-food for thought: should I keep a shared
-list of sources that get refreshed every
-time, or should I just provide the functions
-to get the list?
-
-Both should be easy to do but ig we should
-go with the idea where I just return the list
-right!? Shouldn't complicate it too much
-*/
-
-namespace SourceHelper
-{
-    struct Displays
-    {
-        CGDirectDisplayID id;
-        uint32_t width;
-        uint32_t height;
-        std::string title;
-        bool isMain;
-    };
-    struct Windows
-    {
-        CGWindowID id;
-        uint32_t width;
-        uint32_t height;
-        std::string title;
-        // CGWindow
-    };
-    std::vector<Displays> getDisplaySources();
-    std::vector<Windows> getWindowSources();
-    bool isCaptureableWindow(const NSDictionary* windowInfo);
-}

package/include/SourceHelperWrapper.h

@@ -1,29 +0,0 @@
-#pragma once
-
-#include <vector>
-#include <cstdint>
-#include <string>
-
-namespace SourceHelperWrapper
-{
-    struct Display
-    {
-        uint32_t id;
-        uint32_t width;
-        uint32_t height;
-        std::string title;
-        bool isMain;
-    };
-
-    struct Window
-    {
-        uint32_t id;
-        uint32_t width;
-        uint32_t height;
-        std::string title;
-    };
-
-    std::vector<Display> getDisplaySources();
-    std::vector<Window> getWindowSources();
-    bool isCaptureableWindow(uint32_t windowID);
-}

package/include/Types.h

@@ -1,58 +0,0 @@
-#pragma once
-
-// only use for universal type definitions
-#include <ApplicationServices/ApplicationServices.h>
-#include <CoreFoundation/CFCGTypes.h>
-#include <CoreGraphics/CGDirectDisplay.h>
-#include <CoreGraphics/CGGeometry.h>
-#include <CoreGraphics/CGWindow.h>
-#include <cstdint>
-#include <string>
-
-/*
-    I think I wrote this very good ;)
-    PS Claude helped too! Constructor was it's idea
-
-
-    EDIT: modified to store the scContent
-*/
-struct CaptureSource {
-  bool isWindow;
-  union {
-    CGWindowID windowID;
-    CGDirectDisplayID displayID;
-  };
-  void *scContent;
-  uint32_t width;
-  uint32_t height;
-  std::string title;
-  CGRect captureRect;
-
-  static CaptureSource createWindow(CGWindowID id, uint32_t w, uint32_t h,
-                                    std::string t) {
-    CaptureSource s;
-    s.isWindow = true;
-    s.windowID = id;
-    s.width = w;
-    s.height = h;
-    s.title = t;
-    s.captureRect = CGRectMake(0, 0, w, h);
-    s.scContent = nullptr;
-    return s;
-  }
-
-  static CaptureSource createDisplay(CGDirectDisplayID id, uint32_t w,
-                                     uint32_t h, std::string t) {
-    CaptureSource s;
-    s.isWindow = false;
-    s.displayID = id;
-    s.width = w;
-    s.height = h;
-    s.title = t;
-    s.captureRect = CGRectMake(0, 0, w, h);
-    s.scContent = nullptr;
-    return s;
-  }
-};
-
-enum class OutputMode { STANDALONE, MUXED };

package/include/UploadManager.h

@@ -1,9 +0,0 @@
-#include <cstdint>
-#include <iostream>
-#include <queue>
-#include <boost/asio.hpp>
-
-class UploadManager {
-private:
-  std::queue<uint8_t> uploadQueue;
-};