Warcs have useful content too

requirements.txt

@@ -12,4 +12,5 @@ requests ~=2.32.3
 urllib3 ~=2.3.0
 toolz ~=1.0.0
 tqdm ~=4.67.1
+warcio[all] ~=1.7.5
 wayback ~=0.4.5

scripts/warc_import

@@ -0,0 +1,6 @@
+#!/usr/bin/env python
+from web_monitoring.cli.warc_import import main
+
+
+if __name__ == '__main__':
+    main()

web_monitoring/cli/cli.py

@@ -58,7 +58,6 @@
 import re
 import requests
 import sentry_sdk
-import signal
 import threading
 import time
 from tqdm import tqdm
@@ -667,7 +666,7 @@ def import_ia_urls(urls, *, from_date=None, to_date=None,
     worker_count = worker_count if worker_count > 0 else PARALLEL_REQUESTS
     unplaybackable = load_unplaybackable_mementos(unplaybackable_path)
 
-    with utils.QuitSignal((signal.SIGINT, signal.SIGTERM)) as stop_event:
+    with utils.QuitSignal() as stop_event:
         cdx_records = utils.FiniteQueue()
         cdx_thread = threading.Thread(target=lambda: utils.iterate_into_queue(
             cdx_records,