Add Sentry SDK integration

requirements.txt

@@ -2,3 +2,4 @@ requests==2.28.2
 python-dateutil==2.8.2
 tqdm==4.65.0
 geojson==3.0.1
+sentry-sdk==1.18.0

src/scraper/main.py

@@ -3,10 +3,12 @@ import logging
 import os
 import pathlib
 import pickle
+import subprocess
 import sys
 import typing as t
 from datetime import date, datetime, timedelta
 
+import sentry_sdk
 from tqdm import tqdm
 
 from src.const import TIMEZONE
@@ -16,6 +18,17 @@ from src.scraper.train import Train
 DATA_DIR = pathlib.Path("data/")
 
 
+def get_git_revision_short_hash() -> str:
+    try:
+        return (
+            subprocess.check_output(["git", "rev-parse", "--short", "HEAD"])
+            .decode("ascii")
+            .strip()
+        )
+    except subprocess.CalledProcessError:
+        return "unknown"
+
+
 def load_dataset(file_path: pathlib.Path) -> dict[t.Any, t.Any]:
     try:
         with open(file_path, "rb") as f:
@@ -38,6 +51,15 @@ def main() -> None:
         )
         sys.exit(1)
 
+    sentry_dsn = os.getenv("SENTRY_DSN")
+    if sentry_dsn is not None:
+        sentry_sdk.init(
+            dsn=sentry_dsn,
+            release=get_git_revision_short_hash(),
+            traces_sample_rate=1.0,
+        )
+        logging.info("Activated sentry error reporting")
+
     # Today + ~3 hours
     today: date = (datetime.now(tz=TIMEZONE) - timedelta(hours=3)).date()
     today_path: pathlib.Path = DATA_DIR / today.strftime("%Y-%m-%d")