Improve effeciency of TextSplitter.split_documents, iterate once (lan…

…gchain-ai#5111)

# Improve TextSplitter.split_documents, collect page_content and
metadata in one iteration

## Who can review?

Community members can review the PR once tests pass. Tag
maintainers/contributors who might be interested:

@eyurtsev In the case where documents is a generator that can only be
iterated once making this change is a huge help. Otherwise a silent
issue happens where metadata is empty for all documents when documents
is a generator. So we expand the argument from `List[Document]` to
`Union[Iterable[Document], Sequence[Document]]`

---------

Co-authored-by: Steven Tartakovsky <[email protected]>

langchain/text_splitter.py

@@ -64,10 +64,12 @@ def create_documents(
                 documents.append(new_doc)
         return documents
 
-    def split_documents(self, documents: List[Document]) -> List[Document]:
+    def split_documents(self, documents: Iterable[Document]) -> List[Document]:
         """Split documents."""
-        texts = [doc.page_content for doc in documents]
-        metadatas = [doc.metadata for doc in documents]
+        texts, metadatas = [], []
+        for doc in documents:
+            texts.append(doc.page_content)
+            metadatas.append(doc.metadata)
         return self.create_documents(texts, metadatas=metadatas)
 
     def _join_docs(self, docs: List[str], separator: str) -> Optional[str]:

tests/unit_tests/test_text_splitter.py

@@ -146,3 +146,25 @@ def test_iterative_text_splitter() -> None:
         "Bye!\n\n-H.",
     ]
     assert output == expected_output
+
+
+def test_split_documents() -> None:
+    """Test split_documents."""
+    splitter = CharacterTextSplitter(separator="", chunk_size=1, chunk_overlap=0)
+    docs = [
+        Document(page_content="foo", metadata={"source": "1"}),
+        Document(page_content="bar", metadata={"source": "2"}),
+        Document(page_content="baz", metadata={"source": "1"}),
+    ]
+    expected_output = [
+        Document(page_content="f", metadata={"source": "1"}),
+        Document(page_content="o", metadata={"source": "1"}),
+        Document(page_content="o", metadata={"source": "1"}),
+        Document(page_content="b", metadata={"source": "2"}),
+        Document(page_content="a", metadata={"source": "2"}),
+        Document(page_content="r", metadata={"source": "2"}),
+        Document(page_content="b", metadata={"source": "1"}),
+        Document(page_content="a", metadata={"source": "1"}),
+        Document(page_content="z", metadata={"source": "1"}),
+    ]
+    assert splitter.split_documents(docs) == expected_output