# TeslaCam View

A Flask web app for browsing and playing TeslaCam (Tesla Dashcam/Sentry) recordings from a local drive. It scans a TeslaCam folder for video files, categorizes them by camera view, and presents them in a dark, Tesla-styled web interface with synchronized multi-angle playback.

## Features

- **Recursive scan** of the TeslaCam directory tree (`RecentClips`, `SavedClips`, `SentryClips`, etc.) for `.mp4` files matching the TeslaCam naming convention `YYYY-MM-DD_HH-MM-SS-${camera_view}.mp4`.
- **Categorization by camera view** — front, left repeater, right repeater, rear, left pillar, right pillar — with friendly display names.
- **Multi-angle synchronized playback** — clips sharing the same timestamp are grouped so you can switch between camera angles for the same moment.
- **Search & sort** — filter clips by filename, sort oldest-first or A–Z.
- **Safe streaming** — videos are served via Flask with path resolution relative to the configured TeslaCam root.
- **Dark Tesla-themed UI** built with responsive HTML/CSS (no build step).

## Project Structure

```txt
tesla-cam-view/
├── videoplayer.py          # Flask app: scanning, routing, video streaming
├── requirements.txt        # Python dependencies
├── .env                    # Local configuration (git-ignored)
├── .env.example            # Configuration template (committed)
├── .gitignore
├── templates/
│   ├── index.html          # Main page: camera tabs, player, clip list
│   └── camera_view.html    # Single-camera view page
└── README.md
```

## Requirements

- Python 3.x
- See `requirements.txt`: Flask, python-dotenv, gunicorn

## Setup

1. Create and activate a virtual environment (optional but recommended):

   ```bash
   python -m venv venv
   # Windows
   venv\Scripts\activate
   # macOS/Linux
   source venv/bin/activate
   ```

2. Install dependencies:

   ```bash
   pip install -r requirements.txt
   ```

3. Configure the app via environment variables. Copy the example file and edit it:

   ```bash
   cp .env.example .env
   ```

   `TESLACAM_DIRECTORY` should point to the root folder containing your
   `RecentClips`, `SavedClips`, and `SentryClips` subfolders. Variables set in
   the real environment take precedence over `.env`.

## Configuration

All settings are read from environment variables, with sensible defaults. You
can set them in a `.env` file, export them in the shell, or pass them to the
process manager.

| Variable             | Default         | Description                                     |
| -------------------- | --------------- | ----------------------------------------------- |
| `TESLACAM_DIRECTORY` | `/mnt/TeslaCam` | Root directory of the TeslaCam USB drive        |
| `FLASK_RUN_HOST`     | `0.0.0.0`       | Bind address (local dev server only)            |
| `FLASK_RUN_PORT`     | `5000`          | Port (local dev server only)                    |
| `FLASK_DEBUG`        | `0`             | Enable Flask debug mode (local dev server only) |
| `SCAN_CACHE_TTL`     | `3600`          | Seconds to cache the directory scan. Hit        |
|                      |                 | `/refresh` to force a re-scan.                  |

## Usage

### Local development server

```bash
python videoplayer.py
```

Then open [http://localhost:5000](http://localhost:5000) in your browser.

### Production (WSGI server on Linux)

For production, run the app behind a WSGI server such as **gunicorn**. The
`app` object in `videoplayer.py` is the WSGI entry point:

```bash
# Two workers are sufficient for this app
gunicorn -w 2 -b 0.0.0.0:5000 'videoplayer:app'
```

Make sure `TESLACAM_DIRECTORY` is set for the gunicorn process, e.g.:

```bash
export TESLACAM_DIRECTORY=/mnt/TeslaCam
gunicorn -w 2 -b 0.0.0.0:5000 'videoplayer:app'
```

For a systemd service or other process manager, set the variable in the unit
file / environment rather than relying on `.env`.

- The **main page** lists all clips grouped by camera view with a player and searchable clip list.
- Click a clip to play it; use the camera switch buttons to jump between angles for the same timestamp.
- The `/camera/<view>` route shows all clips for a single camera.

## Routes

| Route                    | Description                                          |
| ------------------------ | ---------------------------------------------------- |
| `/`                      | Main page with all videos categorized by camera view |
| `/camera/<camera_view>`  | All videos for a single camera view                  |
| `/video/<path:filepath>` | Streams a video file from the TeslaCam directory     |
| `/refresh`               | Forces a re-scan of the TeslaCam directory           |

## Camera Views

| Key              | Display Name   |
| ---------------- | -------------- |
| `front`          | Front          |
| `left_repeater`  | Left Repeater  |
| `right_repeater` | Right Repeater |
| `rear`           | Rear           |
| `left_pillar`    | Left Pillar    |
| `right_pillar`   | Right Pillar   |

## Notes

- The directory scan is cached in memory for `SCAN_CACHE_TTL` seconds
  (default 3600) to avoid re-walking the tree on every request. Hit
  `/refresh` to force a re-scan after copying new clips onto the drive.
- Under gunicorn with multiple workers, each worker keeps its own cache; a
  `/refresh` only refreshes the worker that handles the request.
- Debug mode is off by default (`FLASK_DEBUG=0`) and is only used by the local
  dev server. It has no effect when running under a WSGI server like gunicorn.
- The TeslaCam directory must be accessible from the machine running the app
  (e.g., a mounted USB drive or network share).
- Filename parsing assumes the standard TeslaCam naming format; files not
  matching the regex are ignored.