fix: problem with saving files from APIRequestComponent (#5627)

* fix problem with saving files from APIRequestComponent

* moving api_request component test to appropriate subdir

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

@@ -3,11 +3,10 @@ import json
 import mimetypes
 import re
 import tempfile
-from datetime import datetime
+from datetime import datetime, timezone
 from pathlib import Path
 from typing import Any
 from urllib.parse import parse_qsl, urlencode, urlparse, urlunparse
-from zoneinfo import ZoneInfo
 
 import httpx
 import validators
@@ -177,12 +176,13 @@ class APIRequestComponent(Component):
             )
 
             redirection_history = [
-                {"url": str(redirect.url), "status_code": redirect.status_code} for redirect in response.history
+                {
+                    "url": redirect.headers.get("Location", str(redirect.url)),
+                    "status_code": redirect.status_code,
+                }
+                for redirect in response.history
             ]
 
-            if response.is_redirect:
-                redirection_history.append({"url": str(response.url), "status_code": response.status_code})
-
             is_binary, file_path = self._response_info(response, with_file_path=save_to_file)
             response_headers = self._headers_to_dict(response.headers)
 
@@ -196,11 +196,12 @@ class APIRequestComponent(Component):
                 if file_path:
                     async with async_open(file_path, mode, encoding=encoding) as f:
                         await f.write(response.content if is_binary else response.text)
+                        await f.flush()  # Ensure the file is flushed to disk
+                    metadata["file_path"] = str(file_path)
 
                 if include_httpx_metadata:
                     metadata.update(
                         {
-                            "file_path": str(file_path),
                             "headers": headers,
                             "status_code": response.status_code,
                             "response_headers": response_headers,
@@ -347,7 +348,7 @@ class APIRequestComponent(Component):
                 extracted_filename = filename_match.group(1)
                 # Ensure the filename is unique
                 if (component_temp_dir / extracted_filename).exists():
-                    timestamp = datetime.now(ZoneInfo("UTC")).strftime("%Y%m%d%H%M%S%f")
+                    timestamp = datetime.now(timezone.utc).strftime("%Y%m%d%H%M%S%f")
                     filename = f"{timestamp}-{extracted_filename}"
                 else:
                     filename = extracted_filename
@@ -355,7 +356,7 @@ class APIRequestComponent(Component):
         # Step 3: Infer file extension or use part of the request URL if no filename
         if not filename:
             # Extract the last segment of the URL path
-            url_path = urlparse(str(response.request.url)).path
+            url_path = urlparse(str(response.request.url) if response.request else "").path
             base_name = Path(url_path).name  # Get the last segment of the path
             if not base_name:  # If the path ends with a slash or is empty
                 base_name = "response"
@@ -366,7 +367,7 @@ class APIRequestComponent(Component):
                 extension = ".bin" if is_binary else ".txt"  # Default extensions
 
             # Combine the base name with timestamp and extension
-            timestamp = datetime.now(ZoneInfo("UTC")).strftime("%Y%m%d%H%M%S%f")
+            timestamp = datetime.now(timezone.utc).strftime("%Y%m%d%H%M%S%f")
             filename = f"{timestamp}-{base_name}{extension}"
 
         # Step 4: Define the full file path

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

@@ -0,0 +1,177 @@
+import tempfile
+from pathlib import Path
+from unittest.mock import Mock
+
+import aiofiles.os
+import httpx
+import pytest
+import respx
+from aiofile import async_open
+from httpx import Response
+from langflow.components import data
+
+
+@pytest.fixture
+def api_request():
+    # This fixture provides an instance of APIRequest for each test case
+    return data.APIRequestComponent()
+
+
+def test_parse_curl(api_request):
+    # Arrange
+    field_value = (
+        "curl -X GET https://example.com/api/test -H 'Content-Type: application/json' -d '{\"key\": \"value\"}'"
+    )
+    build_config = {
+        "method": {"value": ""},
+        "urls": {"value": []},
+        "headers": {},
+        "body": {},
+    }
+    # Act
+    new_build_config = api_request.parse_curl(field_value, build_config.copy())
+
+    # Assert
+    assert new_build_config["method"]["value"] == "GET"
+    assert new_build_config["urls"]["value"] == ["https://example.com/api/test"]
+    assert new_build_config["headers"]["value"] == {"Content-Type": "application/json"}
+    assert new_build_config["body"]["value"] == {"key": "value"}
+
+
+# HTTPx Metadata testing
+@pytest.mark.parametrize(
+    ("include_metadata", "expected_properties"),
+    [
+        (False, {"source", "result"}),
+        (True, {"source", "result", "headers", "status_code", "response_headers", "redirection_history"}),
+    ],
+)
+@respx.mock
+async def test_httpx_metadata_behavior(api_request, include_metadata, expected_properties):
+    # Mocking a successful GET request with headers and a redirection
+    url = "https://example.com/api/test"
+    redirected_url = "https://example.com/api/redirect"
+    response_content = {"key": "value"}
+    respx.get(url).mock(return_value=Response(303, headers={"Location": redirected_url}))
+    respx.get(redirected_url).mock(
+        return_value=Response(200, json=response_content, headers={"Custom-Header": "HeaderValue"})
+    )
+
+    # Make the request
+    result = await api_request.make_request(
+        client=httpx.AsyncClient(),
+        method="GET",
+        url=url,
+        save_to_file=False,
+        include_httpx_metadata=include_metadata,
+    )
+
+    # Check returned metadata
+    metadata = result.data
+    assert set(metadata.keys()) == expected_properties, f"Unexpected properties: {set(metadata.keys())}"
+
+    if include_metadata:
+        # Validate individual fields
+        assert metadata["source"] == url
+        assert metadata["headers"] is None
+        assert metadata["status_code"] == 200
+        assert metadata["response_headers"]["custom-header"] == "HeaderValue"
+
+        # Validate redirection history
+        assert metadata["redirection_history"] == [
+            {"url": redirected_url, "status_code": 303}
+        ], "Redirection history is incorrect"
+
+        # Validate result
+        assert metadata["result"] == response_content, "Response content mismatch"
+
+
+# Save to File testing
+@pytest.mark.parametrize(
+    ("save_to_file", "expected_properties"),
+    [
+        (False, {"source", "result"}),
+        (True, {"source", "file_path"}),
+    ],
+)
+@respx.mock
+async def test_save_to_file_behavior(api_request, save_to_file, expected_properties):
+    # Mocking a successful GET request with a response body
+    url = "https://example.com/api/test"
+    response_content = "Test response content"
+    respx.get(url).mock(return_value=Response(200, content=response_content))
+
+    # Make the request
+    result = await api_request.make_request(
+        client=httpx.AsyncClient(),
+        method="GET",
+        url=url,
+        save_to_file=save_to_file,
+    )
+
+    # Check returned metadata
+    metadata = result.data
+    assert set(metadata.keys()) == expected_properties, f"Unexpected properties: {set(metadata.keys())}"
+
+    if save_to_file:
+        # Validate that file_path exists in metadata
+        assert "file_path" in metadata, "file_path is missing in metadata"
+        file_path = metadata["file_path"]
+
+        # Validate that the file exists and its content matches the response
+        assert await aiofiles.os.path.exists(file_path), "Saved file does not exist"
+        async with async_open(file_path, "r") as f:
+            file_content = await f.read()
+        assert file_content == response_content, "File content does not match response content"
+
+        # Cleanup the file
+        await aiofiles.os.remove(file_path)
+    else:
+        # Validate that result exists in metadata
+        assert "result" in metadata, "result is missing in metadata"
+        assert metadata["result"] == response_content.encode("utf-8"), "Response content mismatch in metadata"
+
+
+def test_response_info_binary_content(api_request):
+    response = Mock()
+    response.headers = {"Content-Type": "application/octet-stream"}
+    is_binary, file_path = api_request._response_info(response, with_file_path=False)
+    assert is_binary is True
+    assert file_path is None
+
+
+def test_response_info_non_binary_content(api_request):
+    response = Mock()
+    response.headers = {"Content-Type": "text/plain"}
+    is_binary, file_path = api_request._response_info(response, with_file_path=False)
+    assert is_binary is False
+    assert file_path is None
+
+
+def test_response_info_filename_from_content_disposition(api_request):
+    response = Mock()
+    response.headers = {
+        "Content-Disposition": 'attachment; filename="thisfile.txt"',
+        "Content-Type": "text/plain",
+    }
+    response.request = Mock()
+    response.request.url = "https://example.com/testfile"
+
+    is_binary, file_path = api_request._response_info(response, with_file_path=True)
+
+    assert is_binary is False
+    assert file_path.parent == Path(tempfile.gettempdir()) / "APIRequestComponent"
+    assert file_path.name.endswith("thisfile.txt")
+
+
+def test_response_info_default_filename(api_request):
+    response = Mock()
+    response.headers = {"Content-Type": "text/plain"}
+    response.request = Mock()
+    response.request.url = "https://example.com/testfile"
+
+    is_binary, file_path = api_request._response_info(response, with_file_path=True)
+
+    assert is_binary is False
+    assert file_path.parent == Path(tempfile.gettempdir()) / "APIRequestComponent"
+    assert file_path.name.endswith("testfile.txt")

src/backend/tests/unit/test_data_components.py

@@ -1,29 +0,0 @@
-import pytest
-from langflow.components import data
-
-
-@pytest.fixture
-def api_request():
-    # This fixture provides an instance of APIRequest for each test case
-    return data.APIRequestComponent()
-
-
-def test_parse_curl(api_request):
-    # Arrange
-    field_value = (
-        "curl -X GET https://example.com/api/test -H 'Content-Type: application/json' -d '{\"key\": \"value\"}'"
-    )
-    build_config = {
-        "method": {"value": ""},
-        "urls": {"value": []},
-        "headers": {},
-        "body": {},
-    }
-    # Act
-    new_build_config = api_request.parse_curl(field_value, build_config.copy())
-
-    # Assert
-    assert new_build_config["method"]["value"] == "GET"
-    assert new_build_config["urls"]["value"] == ["https://example.com/api/test"]
-    assert new_build_config["headers"]["value"] == {"Content-Type": "application/json"}
-    assert new_build_config["body"]["value"] == {"key": "value"}