feat: add support to accept Dataframe as input to split text, and added relevant tests (#6302)

* update to support dataframe

* [autofix.ci] apply automated fixes

* [autofix.ci] apply automated fixes (attempt 2/3)

* Update split_text.py

* [autofix.ci] apply automated fixes

* [autofix.ci] apply automated fixes

* update names

* Update src/backend/base/langflow/schema/dataframe.py

Co-authored-by: Gabriel Luiz Freitas Almeida <gabriel@langflow.org>

* [autofix.ci] apply automated fixes

* update to template

* update review changes

* Update Vector Store RAG.json

* fix lint errors

* fix tests

* 📝 (freeze.spec.ts): update test description to match the actual element being tested for better clarity and accuracy

* ✨ (stop-button-playground.spec.ts): improve test reliability by specifying target position for drag action to prevent flakiness

* ✅ (logs.spec.ts): increase timeout from 1000ms to 3000ms for better test reliability
✅ (stop-building.spec.ts): update test selector from "handle-splittext-shownode-data inputs-left" to "handle-splittext-shownode-input documents-left" for accurate testing
✅ (starter-projects.spec.ts): add a 1000ms timeout before asserting visibility of an element for better test stability

---------

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Gabriel Luiz Freitas Almeida <gabriel@langflow.org>
Co-authored-by: cristhianzl <cristhian.lousa@gmail.com>

src/backend/tests/unit/components/processing/test_split_text_component.py

@@ -1,4 +1,5 @@
 import pytest
+from langflow.components.data import URLComponent
 from langflow.components.processing import SplitTextComponent
 from langflow.schema import Data, DataFrame
 
@@ -44,6 +45,7 @@ class TestSplitTextComponent(ComponentTestBaseWithoutClient):
                 "chunk_overlap": 0,
                 "chunk_size": 15,
                 "separator": "\n",
+                "text_key": "text",
                 "session_id": "test_session",
                 "sender": "test_sender",
                 "sender_name": "test_sender_name",
@@ -220,3 +222,53 @@ class TestSplitTextComponent(ComponentTestBaseWithoutClient):
         assert "Second line" in results[1].text, f"Expected 'Second line', got '{results[1].text}'"
         assert "Another text" in results[2].text, f"Expected 'Another text', got '{results[2].text}'"
         assert "Another line" in results[3].text, f"Expected 'Another line', got '{results[3].text}'"
+
+    def test_split_text_with_dataframe_input(self):
+        """Test splitting text with DataFrame input."""
+        component = SplitTextComponent()
+        test_texts = ["First text\nSecond line", "Another text\nAnother line"]
+        data_frame = DataFrame([Data(text=text) for text in test_texts])
+        component.set_attributes(
+            {
+                "data_inputs": data_frame,
+                "chunk_overlap": 0,
+                "chunk_size": 10,
+                "separator": "\n",
+                "session_id": "test_session",
+                "sender": "test_sender",
+                "sender_name": "test_sender_name",
+            }
+        )
+
+        results = component.split_text()
+        assert len(results) == 4, f"Expected 4 chunks (2 from each text), got {len(results)}"
+        assert "First text" in results[0].text, f"Expected 'First text', got '{results[0].text}'"
+        assert "Second line" in results[1].text, f"Expected 'Second line', got '{results[1].text}'"
+        assert "Another text" in results[2].text, f"Expected 'Another text', got '{results[2].text}'"
+        assert "Another line" in results[3].text, f"Expected 'Another line', got '{results[3].text}'"
+
+    def test_with_url_loader(self):
+        """Test splitting text with URL loader."""
+        component = SplitTextComponent()
+        url = ["https://en.wikipedia.org/wiki/London", "https://en.wikipedia.org/wiki/Paris"]
+        data_frame = URLComponent(urls=url, format="Text").as_dataframe()
+        assert isinstance(data_frame, DataFrame), "Expected DataFrame instance"
+        assert len(data_frame) == 2, f"Expected DataFrame with 2 rows, got {len(data_frame)}"
+        component.set_attributes(
+            {
+                "data_inputs": data_frame,
+                "chunk_overlap": 0,
+                "chunk_size": 10,
+                "separator": "\n",
+                "session_id": "test_session",
+                "sender": "test_sender",
+                "sender_name": "test_sender_name",
+            }
+        )
+        results = component.as_dataframe()
+        assert isinstance(results, DataFrame), "Expected DataFrame instance"
+        assert len(results) > 2, f"Expected DataFrame with more than 2 rows, got {len(results)}"
+
+        results = component.split_text()
+        assert isinstance(results, list), "Expected list instance"
+        assert len(results) > 2, f"Expected DataFrame with more than 2 rows, got {len(results)}"

src/backend/tests/unit/schema/test_schema_dataframe.py

@@ -0,0 +1,66 @@
+import pandas as pd
+import pytest
+from langchain_core.documents import Document
+from langflow.schema.data import Data
+from langflow.schema.dataframe import DataFrame
+
+
+@pytest.fixture
+def sample_dataframe():
+    """Create a sample DataFrame for testing."""
+    return pd.DataFrame({"name": ["John", "Jane"], "text": ["name is John", "name is Jane"]})
+
+
+class TestDataFrameSchema:
+    def test_to_data_list(self, sample_dataframe):
+        """Test conversion of DataFrame to list of Data objects."""
+        data_frame = DataFrame(sample_dataframe)
+        data_list = data_frame.to_data_list()
+        assert isinstance(data_list, list)
+        assert all(isinstance(item, Data) for item in data_list)
+        assert len(data_list) == len(sample_dataframe)
+        assert data_list[0].data["name"] == "John"
+        assert data_list[0].data["text"] == "name is John"
+
+    def test_add_row(self, sample_dataframe):
+        """Test adding a single row to DataFrame."""
+        data_frame = DataFrame(sample_dataframe)
+        # Test adding dict
+        new_df = data_frame.add_row({"name": "Bob", "text": "name is Bob"})
+        assert len(new_df) == len(sample_dataframe) + 1
+        assert new_df.iloc[-1]["name"] == "Bob"
+        assert new_df.iloc[-1]["text"] == "name is Bob"
+
+        # Test adding Data object
+        data_obj = Data(data={"name": "Alice", "text": "name is Alice"})
+        new_df = data_frame.add_row(data_obj)
+        assert len(new_df) == len(sample_dataframe) + 1
+        assert new_df.iloc[-1]["name"] == "Alice"
+        assert new_df.iloc[-1]["text"] == "name is Alice"
+
+    def test_add_rows(self, sample_dataframe):
+        """Test adding multiple rows to DataFrame."""
+        data_frame = DataFrame(sample_dataframe)
+        new_rows = [{"name": "Bob", "text": "name is Bob"}, Data(data={"name": "Alice", "text": "name is Alice"})]
+        new_df = data_frame.add_rows(new_rows)
+        assert len(new_df) == len(sample_dataframe) + 2
+        assert new_df.iloc[-2:]["name"].tolist() == ["Bob", "Alice"]
+        assert new_df.iloc[-2:]["text"].tolist() == ["name is Bob", "name is Alice"]
+
+    def test_to_lc_documents(self, sample_dataframe):
+        """Test conversion to LangChain documents."""
+        data_frame = DataFrame(sample_dataframe)
+        documents = data_frame.to_lc_documents()
+        assert isinstance(documents, list)
+        assert all(isinstance(doc, Document) for doc in documents)
+        assert len(documents) == 2
+        assert documents[0].page_content == "name is John"
+        assert documents[0].metadata == {"name": "John"}
+
+    def test_bool_operator(self):
+        """Test boolean operator behavior."""
+        empty_df = DataFrame()
+        assert not bool(empty_df)
+
+        non_empty_df = DataFrame({"name": ["John"], "text": ["name is John"]})
+        assert bool(non_empty_df)