fix item_count double incremented

src/crawlee/memory_storage_client/_creation_management.py

@@ -181,6 +181,7 @@ def create_dataset_from_directory(
     from crawlee.memory_storage_client.dataset_client import DatasetClient
 
     item_count = 0
+    has_seen_metadata_file = False
     created_at = datetime.now(timezone.utc)
     accessed_at = datetime.now(timezone.utc)
     modified_at = datetime.now(timezone.utc)
@@ -189,6 +190,7 @@ def create_dataset_from_directory(
     metadata_filepath = os.path.join(storage_directory, METADATA_FILENAME)
 
     if os.path.exists(metadata_filepath):
+        has_seen_metadata_file = True
         with open(metadata_filepath, encoding='utf-8') as f:
             json_content = json.load(f)
             resource_info = DatasetMetadata(**json_content)
@@ -202,7 +204,6 @@ def create_dataset_from_directory(
 
     # Load dataset entries
     entries: dict[str, dict] = {}
-    has_seen_metadata_file = False
 
     for entry in os.scandir(storage_directory):
         if entry.is_file():

tests/unit/memory_storage_client/test_dataset_client.py

@@ -138,3 +138,13 @@ async def test_iterate_items(dataset_client: DatasetClient) -> None:
     assert len(actual_items) == item_count
     assert actual_items[0]['id'] == 0
     assert actual_items[99]['id'] == 99
+
+
+async def test_reuse_dataset(dataset_client: DatasetClient, memory_storage_client: MemoryStorageClient) -> None:
+    item_count = 10
+    await dataset_client.push_items([{'id': i} for i in range(item_count)])
+
+    memory_storage_client.datasets_handled = []  # purge datasets loaded to test create_dataset_from_directory
+    datasets_client = memory_storage_client.datasets()
+    dataset_info = await datasets_client.get_or_create(name='test')
+    assert dataset_info.item_count == item_count