feat: add Prometheus metrics support and update Dockerfile for exposure

Author: milsman2

Dockerfile

@@ -22,4 +22,5 @@ COPY --from=builder --chown=nonroot:nonroot /app/ /app
 ENV PATH="/app/.venv/bin:$PATH"
 USER nonroot
 WORKDIR /app/src
-CMD ["python", "-m", "python_response_time.main"]
+EXPOSE 8000
+CMD ["python", "-m", "python_response_time.main"]

pyproject.toml

@@ -9,6 +9,7 @@ dependencies = [
     "coverage>=7.13.5",
     "isort>=8.0.1",
     "loguru>=0.7.3",
+    "prometheus-client>=0.24.1",
     "pydantic>=2.12.5",
     "pydantic-settings>=2.13.1",
     "pytest>=9.0.2",

src/python_response_time/core/__init__.py

@@ -2,5 +2,16 @@
 
 from python_response_time.core.config import app_settings
 from python_response_time.core.logging import setup_logger
+from python_response_time.core.metrics import (
+    REQUEST_COUNT,
+    REQUEST_LATENCY,
+    start_metrics_server,
+)
 
-__all__ = ["app_settings", "setup_logger"]
+__all__ = [
+    "app_settings",
+    "setup_logger",
+    "start_metrics_server",
+    "REQUEST_COUNT",
+    "REQUEST_LATENCY",
+]

src/python_response_time/core/metrics.py

@@ -0,0 +1,17 @@
+"""Core metrics for the application."""
+
+from prometheus_client import Counter, Histogram, start_http_server
+
+REQUEST_COUNT = Counter("http_requests_total", "Total HTTP requests", ["status"])
+
+REQUEST_LATENCY = Histogram(
+    "http_request_duration_seconds",
+    "HTTP request latency in seconds",
+    ["status"],
+    buckets=(0.01, 0.02, 0.03, 0.04, 0.05, 0.1, 0.25, 0.5, 1, 2, 5),
+)
+
+
+def start_metrics_server(port: int = 8000):
+    """Start the Prometheus metrics server."""
+    start_http_server(port)

src/python_response_time/main.py

@@ -8,7 +8,13 @@
 from loguru import logger
 from rich.console import Console
 
-from python_response_time.core import app_settings, setup_logger
+from python_response_time.core import (
+    REQUEST_COUNT,
+    REQUEST_LATENCY,
+    app_settings,
+    setup_logger,
+    start_metrics_server,
+)
 
 setup_logger(app_settings.LOG_LEVEL)
 console = Console()
@@ -67,18 +73,20 @@ def run_app():
                     timeout=app_settings.TIMEOUT,
                     verify=app_settings.VERIFY_SSL,
                 )
-                elapsed_ms = (time.perf_counter() - start_time) * 1000
+                elapsed = time.perf_counter() - start_time
                 console.print(
-                    f"{i + 1:>4} | {response.status_code} | {elapsed_ms:.2f} ms"
+                    f"{i + 1:>4} | {response.status_code} | {elapsed * 1000:.2f} ms"
                 )
                 logger.info(
                     {
                         "request": i + 1,
                         "status": response.status_code,
-                        "response_time_ms": elapsed_ms,
+                        "response_time_ms": elapsed * 1000,
                         "url": str(app_settings.TARGET_URL),
                     }
                 )
+                REQUEST_COUNT.labels(status=str(response.status_code)).inc()
+                REQUEST_LATENCY.observe(elapsed)
             except requests.exceptions.SSLError as e:
                 console.print(f"{i + 1:>4} | SSL ERROR")
                 logger.error(
@@ -89,6 +97,7 @@ def run_app():
                     }
                 )
             except requests.RequestException as e:
+                REQUEST_COUNT.labels(status="error").inc()
                 console.print(f"{i + 1:>4} | ERROR")
                 logger.error(
                     {
@@ -107,6 +116,7 @@ def run_app():
 if __name__ == "__main__":
     """Run the application."""
     register_signals()
+    start_metrics_server(port=8000)
     try:
         run_app()
     except KeyboardInterrupt: