#59 add refresh query param to force refresh of S3 cache

cellxgene_gateway/gateway.py

@@ -22,9 +22,7 @@ from flask import (
     send_from_directory,
     url_for,
 )
-from flask_api import status
 from werkzeug.middleware.proxy_fix import ProxyFix
-from werkzeug.utils import secure_filename
 
 from cellxgene_gateway import env, flask_util
 from cellxgene_gateway.backend_cache import BackendCache

cellxgene_gateway/items/s3/s3item_source.py

@@ -7,10 +7,11 @@
 # OR CONDITIONS OF ANY KIND, either express or implied. See the License for
 # the specific language governing permissions and limitations under the License.
 
-from os.path import basename, dirname, join
+from os.path import basename, dirname
 from typing import List
 
 import s3fs
+from flask import request
 
 from cellxgene_gateway import dir_util
 from cellxgene_gateway.items.item import ItemTree, ItemType
@@ -18,6 +19,10 @@ from cellxgene_gateway.items.item_source import ItemSource, LookupResult
 from cellxgene_gateway.items.s3.s3item import S3Item
 
 
+def truthy(val: str):
+    return val.lower() in ["true", "1"]
+
+
 class S3ItemSource(ItemSource):
     def __init__(
         self,
@@ -76,10 +81,10 @@ class S3ItemSource(ItemSource):
 
         if not self.s3.exists(url):
             raise Exception(f"S3 url '{url}' does not exist.")
-
+        refresh = truthy(request.args.get("refresh", default="false")) or not self.use_listings_cache
         s3key_map = dict(
             (self.remove_bucket(filepath), "s3://" + filepath)
-            for filepath in sorted(self.s3.ls(url, refresh=not self.use_listings_cache))
+            for filepath in sorted(self.s3.ls(url, refresh=refresh))
         )
 
         def is_annotation_dir(dir_s3key):