yasithdev opened a new pull request, #662: URL: https://github.com/apache/airavata/pull/662
## Summary Two experiment-data fixes found while verifying the clean-slate devstack end to end. ### "No files" for SDK-launched experiment output The Python SDK persists an **absolute** `experiment_data_dir` (`/project/experiment`). The staging **write** side (`DataStagingTask.buildDestinationFilePath`) strips the leading `/` and anchors the dir under the storage root, but the **read** side (`storage_resources.resolve_user_storage_path`) honored it as-is, so `list_dir`/`dir_exists` resolved against the SFTP chroot root (outside the storage root) and listed nothing — while UI-launched experiments (bare-relative dir) listed fine. Strip the leading `/` and anchor under `~/`, matching the write side. ### Clone shows creation time as "56 years ago" Clone re-sends the full experiment via `update_experiment` with `creation_time` cleared, but `ExperimentRepository.saveExperiment` only stamped `creation_time` on insert, so the update nulled the column (`longToTimestamp(0)` → NULL → renders as epoch 1970). Make `creation_time` set-once: on update, carry the stored value forward. ## Test plan - `resolve_user_storage_path` unit checks: SDK absolute `/Default_Project/Echo_goldstandard/` → `~/Default_Project/Echo_goldstandard`; UI relative unchanged (no regression). Confirmed the staged output file exists at the location the patched path resolves to. - Real clone via the SDK flow against the live server: cloned experiment's `creation_time` is now a valid timestamp (was `NULL` pre-fix). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
