feat: add rss component (#8134)

* add RSS component

* add tests

* [autofix.ci] apply automated fixes

* Update rss.py

---------

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

src/backend/base/langflow/components/data/__init__.py

@@ -3,6 +3,7 @@ from .csv_to_data import CSVToDataComponent
 from .directory import DirectoryComponent
 from .file import FileComponent
 from .json_to_data import JSONToDataComponent
+from .rss import RSSReaderComponent
 from .sql_executor import SQLComponent
 from .url import URLComponent
 from .webhook import WebhookComponent
@@ -13,6 +14,7 @@ __all__ = [
     "DirectoryComponent",
     "FileComponent",
     "JSONToDataComponent",
+    "RSSReaderComponent",
     "SQLComponent",
     "URLComponent",
     "WebhookComponent",

src/backend/base/langflow/components/data/rss.py

@@ -0,0 +1,69 @@
+import pandas as pd
+import requests
+from bs4 import BeautifulSoup
+
+from langflow.custom import Component
+from langflow.io import IntInput, MessageTextInput, Output
+from langflow.logging import logger
+from langflow.schema import DataFrame
+
+
+class RSSReaderComponent(Component):
+    display_name = "RSS Reader"
+    description = "Fetches and parses an RSS feed."
+    icon = "rss"
+    name = "RSSReaderSimple"
+
+    inputs = [
+        MessageTextInput(
+            name="rss_url",
+            display_name="RSS Feed URL",
+            info="URL of the RSS feed to parse.",
+            tool_mode=True,
+            input_types=[],
+            required=True,
+        ),
+        IntInput(
+            name="timeout",
+            display_name="Timeout",
+            info="Timeout for the RSS feed request.",
+            value=5,
+            advanced=True,
+        ),
+    ]
+
+    outputs = [Output(name="articles", display_name="Articles", method="read_rss")]
+
+    def read_rss(self) -> DataFrame:
+        try:
+            response = requests.get(self.rss_url, timeout=self.timeout)
+            response.raise_for_status()
+            if not response.content.strip():
+                msg = "Empty response received"
+                raise ValueError(msg)
+            # Check if the response is valid XML
+            try:
+                BeautifulSoup(response.content, "xml")
+            except Exception as e:
+                msg = f"Invalid XML response: {e}"
+                raise ValueError(msg) from e
+            soup = BeautifulSoup(response.content, "xml")
+            items = soup.find_all("item")
+        except (requests.RequestException, ValueError) as e:
+            self.status = f"Failed to fetch RSS: {e}"
+            return DataFrame(pd.DataFrame([{"title": "Error", "link": "", "published": "", "summary": str(e)}]))
+
+        articles = [
+            {
+                "title": item.title.text if item.title else "",
+                "link": item.link.text if item.link else "",
+                "published": item.pubDate.text if item.pubDate else "",
+                "summary": item.description.text if item.description else "",
+            }
+            for item in items
+        ]
+
+        # Ensure the DataFrame has the correct columns even if empty
+        df_articles = pd.DataFrame(articles, columns=["title", "link", "published", "summary"])
+        logger.info(f"Fetched {len(df_articles)} articles.")
+        return DataFrame(df_articles)

src/backend/tests/unit/components/data/test_rss.py

@@ -0,0 +1,129 @@
+from unittest.mock import Mock, patch
+
+import pytest
+import requests
+from langflow.components.data.rss import RSSReaderComponent
+from langflow.schema import DataFrame
+
+from tests.base import ComponentTestBaseWithoutClient
+
+
+class TestRSSReaderComponent(ComponentTestBaseWithoutClient):
+    @pytest.fixture
+    def component_class(self):
+        """Return the component class to test."""
+        return RSSReaderComponent
+
+    @pytest.fixture
+    def default_kwargs(self):
+        """Return the default kwargs for the component."""
+        return {
+            "rss_url": "https://example.com/feed.xml",
+        }
+
+    @pytest.fixture
+    def file_names_mapping(self):
+        """Return an empty list since this component doesn't have version-specific files."""
+        return []
+
+    def test_successful_rss_fetch(self):
+        # Mock RSS feed content
+        mock_rss_content = """
+        <?xml version="1.0" encoding="UTF-8"?>
+        <rss version="2.0">
+            <channel>
+                <item>
+                    <title>Test Article 1</title>
+                    <link>https://example.com/1</link>
+                    <pubDate>2024-03-20</pubDate>
+                    <description>Test summary 1</description>
+                </item>
+                <item>
+                    <title>Test Article 2</title>
+                    <link>https://example.com/2</link>
+                    <pubDate>2024-03-21</pubDate>
+                    <description>Test summary 2</description>
+                </item>
+            </channel>
+        </rss>
+        """
+
+        # Mock the requests.get response
+        mock_response = Mock()
+        mock_response.content = mock_rss_content.encode("utf-8")
+        mock_response.raise_for_status = Mock()
+
+        with patch("requests.get", return_value=mock_response):
+            component = RSSReaderComponent(rss_url="https://example.com/feed.xml")
+            result = component.read_rss()
+
+            assert isinstance(result, DataFrame)
+            assert len(result) == 2
+            assert list(result.columns) == ["title", "link", "published", "summary"]
+            assert result.iloc[0]["title"] == "Test Article 1"
+            assert result.iloc[1]["title"] == "Test Article 2"
+
+    def test_rss_fetch_with_missing_fields(self):
+        # Mock RSS feed content with missing fields
+        mock_rss_content = """
+        <?xml version="1.0" encoding="UTF-8"?>
+        <rss version="2.0">
+            <channel>
+                <item>
+                    <title>Test Article</title>
+                    <!-- Missing link -->
+                    <pubDate>2024-03-20</pubDate>
+                    <!-- Missing description -->
+                </item>
+            </channel>
+        </rss>
+        """
+
+        mock_response = Mock()
+        mock_response.content = mock_rss_content.encode("utf-8")
+        mock_response.raise_for_status = Mock()
+
+        with patch("requests.get", return_value=mock_response):
+            component = RSSReaderComponent(rss_url="https://example.com/feed.xml")
+            result = component.read_rss()
+
+            assert isinstance(result, DataFrame)
+            assert len(result) == 1
+            assert result.iloc[0]["title"] == "Test Article"
+            assert result.iloc[0]["link"] == ""
+            assert result.iloc[0]["summary"] == ""
+
+    def test_rss_fetch_error(self):
+        # Mock a failed request
+        with patch("requests.get", side_effect=requests.RequestException("Network error")):
+            component = RSSReaderComponent(rss_url="https://example.com/feed.xml")
+            result = component.read_rss()
+
+            assert isinstance(result, DataFrame)
+            assert len(result) == 1
+            assert result.iloc[0]["title"] == "Error"
+            assert result.iloc[0]["link"] == ""
+            assert result.iloc[0]["published"] == ""
+            assert "Network error" in result.iloc[0]["summary"]
+
+    def test_empty_rss_feed(self):
+        # Mock empty RSS feed
+        mock_rss_content = """
+        <?xml version="1.0" encoding="UTF-8"?>
+        <rss version="2.0">
+            <channel>
+            </channel>
+        </rss>
+        """
+
+        mock_response = Mock()
+        mock_response.content = mock_rss_content.encode("utf-8")
+        mock_response.raise_for_status = Mock()
+
+        with patch("requests.get", return_value=mock_response):
+            component = RSSReaderComponent(rss_url="https://example.com/feed.xml")
+            result = component.read_rss()
+
+            assert isinstance(result, DataFrame)
+            assert len(result) == 0
+            assert list(result.columns) == ["title", "link", "published", "summary"]