fix of the json upload format

Web/app.py

@@ -11874,18 +11874,14 @@ csrf = CSRFProtect(app)
 def upload_csv_batch():
     """
     Route for batch adding new items to the inventory via CSV.
-    Handles CSV parsing, bulk image upload (conversion to WebP), GridFS storage,
-    and groups identical items based on their Name.
+    Handles CSV parsing, bulk image upload with deduplication (SHA-256 hash matching),
+    GridFS storage, code generation, location syncing, and grouped item creation.
     """
     import pandas as pd
     import ast
-    #if 'username' not in session:
-    #    return jsonify({'success': False, 'message': 'Nicht angemeldet'}), 401
+    import hashlib
 
-    username = session['username']
-    # permissions = _get_current_user_permissions() ... (anpassen wie in Original)
-    # if not _action_access_allowed(permissions, 'can_insert'):
-    #     return jsonify({'success': False, 'message': 'Einfüge-Rechte erforderlich'}), 403
+    username = session.get('username', 'System')
 
     fs = get_gridfs()
     upload_session_id = str(uuid.uuid4())[:8]
@@ -11908,10 +11904,11 @@ def upload_csv_batch():
     if 'Name' not in df.columns:
         return jsonify({"success": False, "message": "Die CSV muss zwingend eine 'Name' Spalte enthalten."}), 400
 
-    # 3. Bilder verarbeiten, nach WebP konvertieren und in GridFS speichern
-    # Mapping: Original-Dateiname (ohne Pfad/Erweiterung) -> GridFS Filename (.webp)
-    image_mapping = {}
+    # 3. Bilder verarbeiten & Duplikate im selben Durchlauf filtern (Hash-Matching)
+    image_mapping = {}      # Original-Dateiname (ohne Ext) -> GridFS Filename (.webp)
+    processed_hashes = {}   # SHA-256 Hash -> GridFS Filename (.webp)
     processed_count = 0
+    dedup_count = 0
     error_count = 0
 
     for index, image in enumerate(uploaded_images):
@@ -11929,6 +11926,18 @@ def upload_csv_batch():
                 error_count += 1
                 continue
 
+            # SHA-256 Hash des Bildinhalts zur Erkennung identischer Bilder
+            img_hash = hashlib.sha256(image_bytes).hexdigest()
+
+            if img_hash in processed_hashes:
+                # Bild ist identisch zu einem bereits verarbeiteten Bild im selben Batch
+                existing_filename = processed_hashes[img_hash]
+                image_mapping[base_name_no_ext] = existing_filename
+                dedup_count += 1
+                app.logger.info(f"{image_log_prefix} Duplikat erkannt ({original_secure_name}). Wiederverwendung von: {existing_filename}")
+                continue
+
+            # Neues Bild verarbeiten und nach WebP konvertieren
             optimized_io = io.BytesIO()
             with Image.open(io.BytesIO(image_bytes)) as img:
                 if img.mode not in ('RGB', 'RGBA'):
@@ -11945,7 +11954,7 @@ def upload_csv_batch():
             optimized_io.seek(0)
             new_filename = f"{uuid.uuid4().hex}_{int(time.time())}.webp"
 
-            # Speichern in GridFS analog zu upload_item
+            # In GridFS speichern
             file_id = fs.put(
                 optimized_io,
                 filename=new_filename,
@@ -11957,7 +11966,8 @@ def upload_csv_batch():
                 }
             )
 
-            # Im Mapping speichern (damit wir sie später der CSV zuordnen können)
+            # In Hash-Tabelle und Mapping sichern
+            processed_hashes[img_hash] = new_filename
             image_mapping[base_name_no_ext] = new_filename
             processed_count += 1
 
@@ -11965,10 +11975,14 @@ def upload_csv_batch():
             app.logger.error(f"{image_log_prefix} Processing failed: {str(e)}")
             error_count += 1
 
-    # 4. Items gruppieren (Analog zu series_group_id aus upload_item)
-    df['Name'] = df['Name'].fillna('Unbenannt').astype(str)
+    # 4. Predefined Locations laden
+    try:
+        predefined_locations = it.get_predefined_locations()
+    except Exception:
+        predefined_locations = []
 
-    # Optional: Fülle NaN Werte in der CSV mit sinnvollen Defaults für die Datenbank
+    # 5. Dataframe bereinigen & gruppieren
+    df['Name'] = df['Name'].fillna('Unbenannt').astype(str)
     df = df.fillna({
         'Ort': 'Unbekannt',
         'Beschreibung': '',
@@ -11985,9 +11999,22 @@ def upload_csv_batch():
         series_group_id = str(uuid.uuid4()) if item_count > 1 else None
         parent_item_id = None
 
+        # Basis-Code für automatisierte Seriencodes ermitteln
+        first_row_code = str(group.iloc[0].get('Code_4', '')).strip()
+        base_code = first_row_code if first_row_code else None
+
         for position, (index, row) in enumerate(group.iterrows(), start=1):
 
-            # Bilder aus der CSV-Zeile extrahieren und über das image_mapping mappen
+            # Ort automatisch zu predefined_locations hinzufügen, falls neu
+            ort_val = str(row['Ort']).strip()
+            if ort_val and ort_val not in predefined_locations:
+                try:
+                    it.add_predefined_location(ort_val)
+                    predefined_locations.append(ort_val)
+                except Exception as e:
+                    app.logger.warning(f"Ort {ort_val} konnte nicht hinzugefügt werden: {e}")
+
+            # Bilder für diesen Artikel zuordnen
             item_image_filenames = []
             if 'Images' in row and pd.notna(row['Images']):
                 try:
@@ -12006,7 +12033,7 @@ def upload_csv_batch():
                 try:
                     res = ast.literal_eval(str(col_data))
                     return res if isinstance(res, list) else []
-                except:
+                except Exception:
                     return []
 
             filter_upload = parse_filter_col(row.get('Filter', '[]'))
@@ -12015,28 +12042,37 @@ def upload_csv_batch():
 
             reservierbar = bool(row.get('Reservierbar', False))
 
-            # DB Insert Funktion aufrufen (mit korrigierten, positionellen Parametern)
+            # Code_4 Behandlung: Falls in CSV definiert nutzen, sonst Batch-Code erzeugen
+            row_code = str(row.get('Code_4', '')).strip()
+            if row_code:
+                unique_code = row_code
+            elif 'generate_unique_batch_code' in globals():
+                unique_code = generate_unique_batch_code(base_code, position)
+            else:
+                unique_code = None
+
+            # DB Insert (exakt abgestimmt auf die 10 positionellen Argumente)
             item_id = it.add_item(
-                str(row['Name']),
-                str(row['Ort']),
-                str(row['Beschreibung']),
-                item_image_filenames,
-                filter_upload,
-                filter_upload2,
-                filter_upload3,
-                str(row['Anschaffungsjahr']) if row['Anschaffungsjahr'] else None,
-                str(row['Anschaffungskosten']) if row['Anschaffungskosten'] else None,
-                str(row['Code_4']) if row['Code_4'] else None,
+                str(row['Name']),                             # 1. Name
+                ort_val,                                      # 2. Ort
+                str(row['Beschreibung']),                     # 3. Beschreibung
+                item_image_filenames,                         # 4. Image Filenames (GridFS)
+                filter_upload,                                # 5. Filter 1
+                filter_upload2,                               # 6. Filter 2
+                filter_upload3,                               # 7. Filter 3
+                str(row['Anschaffungsjahr']) if row['Anschaffungsjahr'] else None,   # 8. Jahr
+                str(row['Anschaffungskosten']) if row['Anschaffungskosten'] else None, # 9. Kosten
+                unique_code,                                  # 10. Unique Code / Code_4
                 reservierbar=reservierbar,
                 series_group_id=series_group_id,
                 series_count=item_count,
                 series_position=position,
                 is_grouped_sub_item=(position > 1),
                 parent_item_id=parent_item_id,
-                isbn='',
-                item_type='other',
-                library_category='',
-                is_library=False
+                isbn=str(row.get('ISBN', '')),
+                item_type=str(row.get('Item_Type', 'other')),
+                library_category=str(row.get('Library_Category', '')),
+                is_library=bool(row.get('Is_Library', False))
             )
 
             if item_id:
@@ -12047,12 +12083,15 @@ def upload_csv_batch():
                 app.logger.error(f"Fehler beim Erstellen von Item: {row['Name']} (Index {index})")
 
     app.logger.info(
-        f"Batch Upload abgeschlossen: {len(created_item_ids)} Items erstellt. {processed_count} Bilder verarbeitet.")
+        f"Batch Upload abgeschlossen: {len(created_item_ids)} Items erstellt. "
+        f"{processed_count} neue Bilder hochgeladen, {dedup_count} Bild-Duplikate zusammengeführt."
+    )
 
     return jsonify({
         "success": True,
-        "message": f"Upload erfolgreich. {len(created_item_ids)} Items importiert und {processed_count} Bilder konvertiert.",
+        "message": f"Upload erfolgreich. {len(created_item_ids)} Items importiert. {processed_count} neue Bilder gespeichert ({dedup_count} Duplikate zusammengeführt).",
         "created_count": len(created_item_ids),
         "images_processed": processed_count,
+        "images_deduplicated": dedup_count,
         "images_failed": error_count
     }), 200