feat: add --force flag to allow re-computing and overwriting existing snapshots

src/lifecycle_retrieve.py

@@ -286,10 +286,11 @@ def push_lifecycle_to_hf(lifecycle_en: dict, lifecycle_zh: dict,
 # Run one snapshot
 # ---------------------------------------------------------------------------
 def run_snapshot(snapshot_month: str, all_papers: list[dict],
-                 existing_splits: set[str], no_push: bool = False):
+                 existing_splits: set[str], no_push: bool = False,
+                 force: bool = False):
     split_name = _snapshot_to_split(snapshot_month)
 
-    if split_name in existing_splits:
+    if split_name in existing_splits and not force:
         print(f"  {_GRAY}⊘  {snapshot_month} — already on HF, skipping{_RESET}")
         return
 
@@ -332,6 +333,8 @@ def main():
                         help="Compute all missing bimonthly snapshots")
     parser.add_argument("--no-push", action="store_true",
                         help="Skip pushing results to HuggingFace")
+    parser.add_argument("--force", action="store_true",
+                        help="Re-compute and overwrite existing snapshots")
     args = parser.parse_args()
 
     print(f"\n  {_BOLD}📊 Lifecycle Snapshot Retriever{_RESET}\n")
@@ -386,7 +389,8 @@ def main():
     print(f"  {_DIM}Snapshots to process: {len(snapshots)}{_RESET}\n")
 
     for snapshot in snapshots:
-        run_snapshot(snapshot, all_papers, existing_splits, no_push=args.no_push)
+        run_snapshot(snapshot, all_papers, existing_splits,
+                     no_push=args.no_push, force=args.force)
 
     print(f"\n  {_GREEN}{_BOLD}✓{_RESET} Done\n")