Refaire support

correction.py

@@ -16,6 +16,8 @@ parser = argparse.ArgumentParser()
 parser.add_argument("--overwrite", action="store_true",
                     help="Force redo requests even if output exists")
 parser.add_argument("--limit", type=int, help="limit calls to gemini rpo integer")
+parser.add_argument("--refaire", action="store_true",
+                    help="Redo specific copies/labels defined in refaire.json")
 args, _ = parser.parse_known_args()
 
 
@@ -675,6 +677,67 @@ def process_single_task(task_tuple):
         flush_thread_log()
 
 if __name__ == "__main__":
+    if args.refaire:
+        refaire_path = Path(INPUT_DIR) / "refaire.json"
+        overwritten_path = Path(INPUT_DIR) / "overwritten_correction.json"
+
+        if refaire_path.exists():
+            with open(refaire_path, "r", encoding="utf-8") as f:
+                refaire_list = json.load(f)
+
+            overwritten_data = []
+            if overwritten_path.exists():
+                with open(overwritten_path, "r", encoding="utf-8") as f:
+                    overwritten_data = json.load(f)
+
+            dirty_results = False
+
+            for copie_name, labels in refaire_list:
+                pid = copie_name.replace("Copie", "")
+                copie_dir = Path(INPUT_DIR) / copie_name
+
+                # If list is empty, redo all labels available for this Copie
+                if not labels:
+                    labels = [p.stem for p in copie_dir.glob("*.pdf")]
+
+                for label in labels:
+                    # 1. Extract and backup old corrections
+                    if label in results:
+                        for batch in results[label]:
+                            to_remove = None
+                            for item in batch:
+                                if item.get("id") == pid:
+                                    to_remove = item
+                                    break
+                            if to_remove:
+                                batch.remove(to_remove)
+                                overwritten_data.append({
+                                    "pid": pid,
+                                    "label": label,
+                                    "data": to_remove,
+                                    "timestamp": time.time()
+                                })
+                                dirty_results = True
+                        # Clean up empty batches
+                        results[label] = [b for b in results[label] if b]
+
+                    # 2. Make new group and add to tasks
+                    pdf_path = copie_dir / f"{label}.pdf"
+                    if pdf_path.exists():
+                        idx = get_next_group_idx(INPUT_DIR, label)
+                        height = grouping.get_pdf_height(str(pdf_path))
+                        grouping.create_jpg(label, idx, [(pid, str(pdf_path), height)], INPUT_DIR)
+                        new_group_path = str(Path(INPUT_DIR) / label / f"Group_{idx+1}.jpg")
+                        tasks_to_process.append((new_group_path, label))
+
+            if dirty_results:
+                with open(output_path, "w", encoding="utf-8") as f:
+                    json.dump(results, f, indent=2)
+                with open(overwritten_path, "w", encoding="utf-8") as f:
+                    json.dump(overwritten_data, f, indent=2)
+        else:
+            print(f"Warning: --refaire flag used, but {refaire_path} not found.", file=sys.stderr)
+
     print(f"Starting processing on {len(tasks_to_process)} tasks with {NB_THREADS} threads...")
 
     with concurrent.futures.ThreadPoolExecutor(max_workers=NB_THREADS) as executor: