Skip to content

Delete & Restore

Soft Delete

By default, delete() performs a soft delete — the record is hidden from list() and get(), but all data and metadata are preserved on disk.

import dfstore

dfstore.delete("old_dataset")

The record won't appear in dfstore.list() after this. Attempting dfstore.get("old_dataset") raises DFNotFoundError.

Checking Soft-Deleted Records

Soft-deleted records are still accessible with include_deleted=True:

all_records = dfstore.list(include_deleted=True)

# Filter to deleted only
deleted = all_records[all_records["deleted"] == True]
print(deleted[["name", "updated_at"]])

Restore

Undo a soft delete:

dfstore.restore("old_dataset")

# Now accessible again
df = dfstore.get("old_dataset")

Hard Delete

Permanently removes all data and metadata. This cannot be undone.

dfstore.delete("old_dataset", hard=True)

After a hard delete:

  • The Parquet files are removed from disk.
  • The metadata entry is removed from the index.
  • restore() will raise DFNotFoundError.

Summary

Action Effect Reversible?
delete(name) Hides the record; data preserved Yes — restore()
delete(name, hard=True) Removes all data and metadata No
restore(name) Undoes a soft delete —