reporter

.env

@@ -16,8 +16,8 @@ DB_PASSWORD=Vlado9674+
 # =========================
 # INDEXER
 # =========================
-ROOT_PATH=u:\Dropbox\Ordinace\
+ROOT_PATH=z:\Dropbox\Ordinace\
 ROOT_NAME=DropboxOrdinace
 BATCH_SIZE=1000
-BACKUP_PATH=u:\OnedriveOrdinace\OneDrive\DropBoxBackupClaude\
+BACKUP_PATH=w:\Onedrive\DropBoxBackupClaude\
 BACKUP_PASSWORD=Vlado7309208104++

indexer/hasher.py

@@ -1,10 +1,24 @@
+import ctypes
+
 from blake3 import blake3
 
+# Windows atributy pro cloud/placeholder soubory
+_FILE_ATTRIBUTE_OFFLINE              = 0x00001000
+_FILE_ATTRIBUTE_RECALL_ON_OPEN       = 0x00040000
+_FILE_ATTRIBUTE_RECALL_ON_DATA_ACCESS = 0x00400000
+_CLOUD_MASK = _FILE_ATTRIBUTE_OFFLINE | _FILE_ATTRIBUTE_RECALL_ON_OPEN | _FILE_ATTRIBUTE_RECALL_ON_DATA_ACCESS
+
+
+def is_cloud_placeholder(path: str) -> bool:
+    """Vrátí True pokud soubor není lokálně stažený (Dropbox/OneDrive placeholder)."""
+    attrs = ctypes.windll.kernel32.GetFileAttributesW(path)
+    if attrs == 0xFFFFFFFF:  # INVALID_FILE_ATTRIBUTES
+        return False
+    return bool(attrs & _CLOUD_MASK)
+
 
 def blake3_file(path, chunk_size=1024 * 1024):
-    """
-    Spočítá BLAKE3 hash souboru po blocích (bez načtení do paměti)
-    """
+    """Spočítá BLAKE3 hash souboru po blocích (bez načtení do paměti)."""
     h = blake3()
     with open(path, "rb") as f:
         for chunk in iter(lambda: f.read(chunk_size), b""):

main.py

@@ -11,6 +11,7 @@ from indexer.db import (
 )
 from indexer.events import batch_log_events
 from indexer.backup import ensure_backed_up
+from indexer.hasher import is_cloud_placeholder
 
 
 def main():
@@ -72,15 +73,22 @@ def main():
         files_to_backup = []
 
         # 5a) NEW files — compute BLAKE3, batch INSERT
+        skipped_files = []
         if new_paths:
             print(f"  Hashing {len(new_paths)} new files...")
             new_files = []
             for p in new_paths:
                 f = fs[p]
+                if is_cloud_placeholder(f["full_path"]):
+                    reason = "not synced (cloud placeholder)"
+                    print(f"  WARN: skip {p}: {reason}")
+                    skipped_files.append((p, reason))
+                    continue
                 try:
                     content_hash = blake3_file(f["full_path"])
                 except (FileNotFoundError, PermissionError, OSError) as e:
                     print(f"  WARN: skip {p}: {e}")
+                    skipped_files.append((p, str(e)))
                     continue
                 new_files.append({
                     "relative_path": p,
@@ -168,7 +176,7 @@ def main():
         # ── 7. Finalize ──
         stats = {
             "total": len(fs),
-            "new": len(new_paths),
+            "new": len(new_files) if new_paths else 0,
             "modified": len(modified_paths),
             "deleted": len(deleted_paths),
             "unchanged": len(unchanged_paths),
@@ -196,6 +204,12 @@ def main():
     print(f"Modified : {stats['modified']}")
     print(f"Deleted  : {stats['deleted']}")
     print(f"Unchanged: {stats['unchanged']}")
+    if skipped_files:
+        print(f"Skipped  : {len(skipped_files)} (hash failed)")
+        print("-" * 60)
+        for path, reason in skipped_files:
+            print(f"  SKIP: {path}")
+            print(f"        {reason}")
     print("=" * 60)
 
     # ── 8. Generate Excel report ──

recovery.py

@@ -17,7 +17,7 @@ from indexer.config import BACKUP_PATH, BACKUP_PASSWORD
 from indexer.db import get_connection
 from indexer.backup import blob_path
 
-DEFAULT_OUTPUT_DIR = r"U:\recovery"
+DEFAULT_OUTPUT_DIR = r"\\tower\Pomoc\Recovery"
 
 
 def show_last_runs(n: int = 10):