diff --git a/Dockerfile b/Dockerfile index 5d6d29b..e880d32 100644 --- a/Dockerfile +++ b/Dockerfile @@ -15,12 +15,14 @@ RUN apt-get update && apt-get install -y --no-install-recommends \ git \ curl \ g++ \ + tini \ && rm -rf /var/lib/apt/lists/* \ && ln -sf /usr/bin/python3.12 /usr/bin/python \ && ln -sf /usr/bin/python3.12 /usr/bin/python3 -RUN curl -LsSf https://astral.sh/uv/install.sh | sh -ENV PATH="/root/.local/bin:${PATH}" +RUN curl -LsSf https://astral.sh/uv/install.sh | sh \ + && cp /root/.local/bin/uv /usr/local/bin/uv \ + && chmod 755 /usr/local/bin/uv WORKDIR /app @@ -42,5 +44,8 @@ ENV FORCE_CPU=false \ HF_HOME=/models/huggingface \ INSIGHTFACE_HOME=/models/insightface -ENTRYPOINT ["/app/entrypoint.sh"] +HEALTHCHECK --interval=1h --timeout=10s --start-period=60s --retries=3 \ + CMD test -f /app/entrypoint.sh || exit 1 + +ENTRYPOINT ["tini", "--", "/app/entrypoint.sh"] diff --git a/entrypoint.sh b/entrypoint.sh index 2e2d25d..5af0308 100644 --- a/entrypoint.sh +++ b/entrypoint.sh @@ -1,17 +1,48 @@ #!/bin/bash set -e +export PYTHONUNBUFFERED=1 + +# Function to log model status before each run +check_models() { + echo "šŸ“¦ Checking models..." + + if [ -d "/models/insightface/models/buffalo_l" ]; then + echo " āœ… InsightFace Buffalo_L: downloaded" + else + echo " ā¬‡ļø InsightFace Buffalo_L: downloading (~300MB)..." + fi + + if [ -d "/models/huggingface/hub" ] && [ "$(find /models/huggingface/hub -maxdepth 1 -type d 2>/dev/null | wc -l)" -gt 1 ]; then + echo " āœ… HuggingFace models: downloaded" + else + echo " ā¬‡ļø HuggingFace models: downloading (SigLIP ~1GB, YOLOv9c ~500MB)..." + fi + + echo "šŸš€ Starting if-curator..." +} + SCHEDULE="${CRON_SCHEDULE:-}" +AUTO="${AUTO_MODE:-false}" + +# Check if interactive mode is viable +if [ "$AUTO" != "true" ] && [ ! -t 0 ]; then + echo "āŒ AUTO_MODE is not enabled and no TTY is attached." + echo "āŒ Set AUTO_MODE=true for headless/automated runs." + exit 1 +fi + +# Always run once on startup +echo "ā–¶ Running on startup..." +check_models +uv run if-curator if [ -z "$SCHEDULE" ]; then - echo "ā–¶ No CRON_SCHEDULE set ā€” running once" - uv run if-curator + echo "ā–¶ No CRON_SCHEDULE set ā€” exiting" exit 0 fi echo "ā–¶ CRON_SCHEDULE set to: $SCHEDULE" -echo "ā–¶ Running if-curator on schedule..." - -# Hand off to the Python scheduler +echo "ā–¶ Switching to scheduled mode..." exec uv run python3 /app/scheduler.py diff --git a/if_curator/cli.py b/if_curator/cli.py index f8d28cd..d64ee6c 100644 --- a/if_curator/cli.py +++ b/if_curator/cli.py @@ -210,41 +210,131 @@ def _resolve_strategy(strategy: str, has_embedding: bool) -> tuple[int | str, st def upload_to_frigate(jobs: list[dict]) -> None: - """Upload processed face crops to Frigate via API.""" + """Upload processed face crops to Frigate via API with detailed logging.""" frigate_url = os.environ.get("FRIGATE_URL", "") if not frigate_url: - rprint("[yellow]FRIGATE_URL not set, skipping upload.[/yellow]") + rprint("[yellow]āš ļø FRIGATE_URL not set, skipping upload.[/yellow]") return - uploaded, failed = 0, 0 + rprint("\n[bold cyan]šŸ“¤ Uploading to Frigate[/bold cyan]") + rprint(f" Target: [dim]{frigate_url}[/dim]") + + # Count total files to upload + total_files = 0 for job in jobs: name = job["person"]["name"] person_dir = os.path.join(Config.OUTPUT_DIR, name) if not os.path.isdir(person_dir): continue + total_files += sum( + 1 for f in os.listdir(person_dir) + if f.lower().endswith((".jpg", ".jpeg", ".png", ".webp")) + ) - for fname in sorted(os.listdir(person_dir)): - fpath = os.path.join(person_dir, fname) - if not fname.lower().endswith((".jpg", ".jpeg", ".png", ".webp")): + if total_files == 0: + rprint(" [yellow]No images found to upload.[/yellow]") + return + + rprint(f" People: [bold]{len(jobs)}[/bold], Total images: [bold]{total_files}[/bold]") + + uploaded, failed = 0, 0 + + with Progress( + SpinnerColumn(), + TextColumn("[progress.description]{task.description}"), + BarColumn(), + TaskProgressColumn(), + console=console, + ) as progress: + upload_task = progress.add_task("[green]Uploading to Frigate", total=total_files) + + for job in jobs: + name = job["person"]["name"] + person_dir = os.path.join(Config.OUTPUT_DIR, name) + if not os.path.isdir(person_dir): + progress.console.print(f" [dim]ā­ļø {name}: no output directory, skipping[/dim]") continue - try: - with open(fpath, "rb") as f: - resp = requests.post( - f"{frigate_url}/api/faces/train/{name}/classify", - files={"file": (fname, f, "image/jpeg")}, - timeout=30, - ) - if resp.status_code == 200: - uploaded += 1 - else: + + person_files = sorted( + f for f in os.listdir(person_dir) + if f.lower().endswith((".jpg", ".jpeg", ".png", ".webp")) + ) + + if not person_files: + progress.console.print(f" [dim]ā­ļø {name}: no images found[/dim]") + continue + + progress.console.print(f" šŸ“ {name}: uploading {len(person_files)} image(s)...") + person_uploaded = 0 + person_failed = 0 + + for fname in person_files: + fpath = os.path.join(person_dir, fname) + try: + with open(fpath, "rb") as f: + resp = requests.post( + f"{frigate_url}/api/faces/train/{name}/classify", + files={"file": (fname, f, "image/jpeg")}, + timeout=30, + ) + if resp.status_code == 200: + uploaded += 1 + person_uploaded += 1 + else: + failed += 1 + person_failed += 1 + progress.console.print( + f" [red]āœ— {fname}: HTTP {resp.status_code}[/red]" + ) + try: + error_detail = resp.json().get("message", resp.text[:100]) + progress.console.print(f" [dim]{error_detail}[/dim]") + except Exception: + progress.console.print(f" [dim]{resp.text[:100]}[/dim]") + except requests.exceptions.ConnectionError: failed += 1 - logger.warning(f"Frigate upload failed for {name}/{fname}: {resp.status_code}") - except Exception as e: - failed += 1 - logger.warning(f"Frigate upload error for {name}/{fname}: {e}") + person_failed += 1 + progress.console.print( + f" [red]āœ— {fname}: Connection refused[/red]" + ) + except requests.exceptions.Timeout: + failed += 1 + person_failed += 1 + progress.console.print( + f" [red]āœ— {fname}: Request timed out (30s)[/red]" + ) + except Exception as e: + failed += 1 + person_failed += 1 + progress.console.print( + f" [red]āœ— {fname}: {type(e).__name__} - {e}[/red]" + ) - rprint(f" [green]Frigate upload: {uploaded} succeeded, {failed} failed[/green]") + progress.advance(upload_task) + # Per-person summary + if person_failed == 0: + progress.console.print( + f" āœ… {name}: {person_uploaded}/{person_uploaded} uploaded" + ) + else: + progress.console.print( + f" āš ļø {name}: {person_uploaded} succeeded, {person_failed} failed" + ) + + # Grand summary + rprint("\n [bold]Frigate Upload Summary:[/bold]") + rprint(f" āœ… Succeeded: [green]{uploaded}[/green]") + if failed: + rprint(f" āŒ Failed: [red]{failed}[/red]") + else: + rprint(f" āŒ Failed: 0") + + if failed > 0: + rprint(" [yellow]Check logs above for per-file error details.[/yellow]") + + if failed == total_files and total_files > 0: + rprint(" [bold red]All uploads failed. Verify FRIGATE_URL is reachable and API is enabled.[/bold red]") def _perform_selection(assets: list, limit: int | str, name: str, selection_mode: str, entity_type: str) -> list: @@ -420,3 +510,4 @@ def main() -> None: if __name__ == "__main__": main() + diff --git a/scheduler.py b/scheduler.py index a6ec20a..7d12a00 100644 --- a/scheduler.py +++ b/scheduler.py @@ -8,6 +8,7 @@ import os import sys import subprocess import time +from pathlib import Path try: from croniter import croniter @@ -16,28 +17,66 @@ except ImportError: sys.exit(1) SCHEDULE = os.environ["CRON_SCHEDULE"] -NOW = time.time() +MODELS_DIR = os.environ.get("HF_HOME", "/models/huggingface") +INSIGHTFACE_DIR = os.environ.get("INSIGHTFACE_HOME", "/models/insightface") +# Ensure all subprocess output appears in docker logs +RUN_ENV = {**os.environ, "PYTHONUNBUFFERED": "1"} + + +def check_models(): + """Log model download status before each run.""" + print("šŸ“¦ Checking models...", flush=True) + + buffalo = Path(INSIGHTFACE_DIR) / "models" / "buffalo_l" + if buffalo.exists(): + print(" āœ… InsightFace Buffalo_L: downloaded", flush=True) + else: + print(" ā¬‡ļø InsightFace Buffalo_L: downloading (~300MB)...", flush=True) + + hf_hub = Path(MODELS_DIR) / "hub" + if hf_hub.exists() and any(hf_hub.iterdir()): + print(" āœ… HuggingFace models: downloaded", flush=True) + else: + print(" ā¬‡ļø HuggingFace models: downloading (SigLIP ~1GB, YOLOv9c ~500MB)...", flush=True) + + print("šŸš€ Starting if-curator...", flush=True) + + +def calc_time_display(seconds): + """Convert seconds to human-readable time.""" + hours = int(seconds // 3600) + minutes = int((seconds % 3600) // 60) + if hours > 0: + return f"{hours}h {minutes}m" + return f"{minutes}m" + + +NOW = time.time() cron = croniter(SCHEDULE, NOW) next_run = cron.get_next(float) while True: now = time.time() if now >= next_run: - print(f"\nā–¶ [{time.strftime('%Y-%m-%d %H:%M:%S')}] Starting if-curator...") - result = subprocess.run(["uv", "run", "if-curator"]) + print(f"\nā–¶ [{time.strftime('%Y-%m-%d %H:%M:%S')}] Starting if-curator...", flush=True) + check_models() + result = subprocess.run( + ["uv", "run", "if-curator"], + env=RUN_ENV, + ) if result.returncode == 0: - print(f"āœ… Run complete.") + print("āœ… Run complete.", flush=True) else: - print(f"āš ļø Run exited with code {result.returncode}") + print(f"āš ļø Run exited with code {result.returncode}", flush=True) next_run = cron.get_next(float) wait = next_run - time.time() if wait > 0: - print(f"ā–¶ Next run: {time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(next_run))} (in {int(wait // 3600)}h {int((wait % 3600) // 60)}m)") + print(f"ā–¶ Next run: {time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(next_run))} (in {calc_time_display(wait)})", flush=True) continue wait = next_run - now - print(f"ā–¶ Next run: {time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(next_run))} (in {int(wait // 3600)}h {int((wait % 3600) // 60)}m)") - time.sleep(min(wait, 3600)) # Check every hour at most, or sooner if needed + print(f"ā–¶ Next run: {time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(next_run))} (in {calc_time_display(wait)})", flush=True) + time.sleep(min(wait, 3600))