Catch different bucket case.

apps/dc_tools/odc/apps/dc_tools/_stac.py

@@ -186,16 +186,16 @@ def _find_self_href(item: Document) -> str:
 def _get_relative_path(asset_href, self_link):
     if self_link is None:
         return asset_href
-    # elif urlparse(self_link).netloc != urlparse(asset_href).netloc:
-    #     # files are not stored in same domain (e.g. diferent buckets)
-    #     # therefore use the absolute path
-    #     return asset_href
+    self_parts = urlparse(self_link)
+    href_parts = urlparse(asset_href)
 
-    self_path = urlparse(self_link).path
-    href_path = urlparse(asset_href).path
+    if self_parts.netloc.split('.')[0] != href_parts.netloc.split('.')[0]:
+        # files are not stored with same hostname (e.g. different buckets)
+        # therefore use the absolute path
+        return asset_href
 
     try:
-        return str(Path(href_path).relative_to(Path(self_path).parent))
+        return str(Path(href_parts.path).relative_to(Path(self_parts.path).parent))
     except ValueError:
         # if it's not relative, keep as an absolute link
         return asset_href