Pass auth headers to lizard, but not to S3

CHANGES.rst

@@ -7,6 +7,8 @@ Changelog of threedi-scenario-downloader
 - Improved test setup: a pytest fixture automatically sets up the api
   key. Add one to ``test_api_key.txt``.
 
+- When downloading files from amazon S3, we don't send along lizard authentication headers.
+
 
 1.3 (2024-01-17)
 ----------------

threedi_scenario_downloader/downloader.py

@@ -320,7 +320,11 @@ def get_task_download_url(task_uuid):
 def download_file(url, path):
     """download url to specified path"""
     logger.debug(f"Start downloading file: {url}")
-    r = requests.get(url, stream=True)
+    if "amazon" not in url:
+        r = requests.get(url, auth=("__key__", get_api_key()), stream=True)
+    else:
+        # Amazon s3 request: don't send lizard's auth headers.
+        r = requests.get(url, stream=True)
     r.raise_for_status()
     with open(path, "wb") as file:
         for chunk in r.iter_content(1024 * 1024 * 10):