address comments

Author: Yuan325

src/langchain_google_spanner/loader.py

@@ -78,7 +78,7 @@ def _load_doc_to_row(
     doc: Document,
     metadata_json_column: str,
     parse_json: bool = True,
-):
+) -> tuple:
     """
     Load document to row.
 
@@ -90,7 +90,7 @@ def _load_doc_to_row(
                     JSON object is used to compare with Spanner reads.
     """
     doc_metadata = doc.metadata.copy()
-    row = [doc.page_content]
+    row = []
     # store metadata
     for col in table_fields:
         if (
@@ -100,6 +100,8 @@ def _load_doc_to_row(
         ):
             row.append(doc_metadata[col])
             del doc_metadata[col]
+        if col == CONTENT_COL_NAME:
+            row.append(doc.page_content)
     if metadata_json_column in table_fields:
         metadata_json = {}
         if metadata_json_column in doc_metadata:
@@ -118,7 +120,7 @@ def _batch(datas: List[Any], size: int = 1) -> Iterator[List[Any]]:
 
 
 class SpannerLoader(BaseLoader):
-    """Loads data from Google CLoud Spanner."""
+    """Loads data from Google Cloud Spanner."""
 
     def __init__(
         self,
@@ -158,7 +160,7 @@ def __init__(
         self.metadata_json_column = metadata_json_column
         formats = ["JSON", "text", "YAML", "CSV"]
         if self.format not in formats:
-            raise Exception("Use on of 'text', 'JSON', 'YAML', 'CSV'")
+            raise Exception("Use one of 'text', 'JSON', 'YAML', 'CSV'")
         self.databoost = databoost
         self.client = client
         self.staleness = staleness

tests/integration/test_spanner_loader.py renamed to tests/test_spanner_loader.py

@@ -19,11 +19,7 @@
 from google.cloud.spanner import Client, KeySet  # type: ignore
 from langchain_core.documents import Document
 
-from langchain_google_spanner.loader import (
-    Column,
-    SpannerDocumentSaver,
-    SpannerLoader,
-)
+from langchain_google_spanner.loader import Column, SpannerDocumentSaver, SpannerLoader
 
 project_id = os.environ["PROJECT_ID"]
 instance_id = os.environ["INSTANCE_ID"]