fix(listing): pick actually the smallest one to update

src/datachain/lib/dc.py

@@ -430,7 +430,11 @@ def parse_uri(
         if listings:
             if update:
                 # choosing the smallest possible one to minimize update time
-                listing = sorted(listings, key=lambda ls: len(ls.name))[0]
+                listing = sorted(listings, key=lambda ls: len(ls.name), reverse=True)[0]
+                if listing.name != ds_name:
+                    # better to create a separate new listing rather then updating
+                    # potentially a very big one that we don't need now
+                    listing = None
             else:
                 # no need to update, choosing the most recent one
                 listing = sorted(listings, key=lambda ls: ls.created_at)[-1]

tests/func/test_catalog.py

@@ -840,13 +840,17 @@ def test_listing_stats(cloud_test_catalog):
 
     catalog.enlist_source(f"{src_uri}/dogs/", update=True)
     stats = listing_stats(src_uri, catalog)
+    assert stats.num_objects == 7
+    assert stats.size == 36
+
+    stats = listing_stats(f"{src_uri}/dogs/", catalog)
     assert stats.num_objects == 4
     assert stats.size == 15
 
     catalog.enlist_source(f"{src_uri}/dogs/")
     stats = listing_stats(src_uri, catalog)
-    assert stats.num_objects == 4
-    assert stats.size == 15
+    assert stats.num_objects == 7
+    assert stats.size == 36
 
 
 @pytest.mark.parametrize("cloud_type", ["s3", "azure", "gs"], indirect=True)