added text processing and tests

src/app.py

@@ -63,7 +63,6 @@ def process_video(video_path: str, max_images: int) -> list[dict]:
     result_content = []
     # TODO: Change max_image to slider
     frames = get_frames(video_path, max_images)
-    # Take frame and attach to result_content with timestamp
     for frame in frames:
         image, timestamp = frame
         with tempfile.NamedTemporaryFile(suffix=".png", delete=False) as temp_file:
@@ -71,4 +70,16 @@ def process_video(video_path: str, max_images: int) -> list[dict]:
             result_content.append({"type": "text", "text": f"Frame {timestamp}:"})
             result_content.append({"type": "image", "url": temp_file.name})
     logger.debug(f"Processed {len(frames)} frames from video {video_path} with frames {result_content}")
-    return result_content
+    return result_content
+
+def process_user_input(message: dict, max_images: int) -> list[dict]:
+    if not message["files"]:
+        return [{"type": "text", "text": message["text"]}]
+
+    if message["files"][0].endswith(".mp4"):
+        return [{"type": "text", "text": message["text"]}, *process_video(message["files"][0], max_images)]
+
+    return [
+        {"type": "text", "text": message["text"]},
+        *[{"type": "image", "url": path} for path in message["files"]],
+    ]

tests/{test_video.py → test_media.py}

@@ -5,7 +5,7 @@ from PIL import Image
 from pathlib import Path
 import tempfile
 
-from src.app import get_frames, process_video
+from src.app import get_frames, process_video, process_user_input
 
 # Get the project root directory
 ROOT_DIR = Path(__file__).parent.parent
@@ -100,3 +100,137 @@ def test_process_video_invalid_path():
 
     with pytest.raises(ValueError):
         process_video("nonexistent_video.mp4", 3)
+        
+def test_process_user_input_text_only():
+    """Test processing user input with text only (no files)."""
+    message = {
+        "text": "This is a test message",
+        "files": []
+    }
+    
+    # Add the max_images parameter
+    result = process_user_input(message, 5)
+    
+    # Should return a single text item
+    assert len(result) == 1
+    assert result[0]["type"] == "text"
+    assert result[0]["text"] == "This is a test message"
+
+
+def test_process_user_input_with_video():
+    """Test processing user input with a video file."""
+    video_path = os.path.join(ROOT_DIR, "assets", "test_video.mp4")
+    assert os.path.exists(video_path), f"Test video not found at {video_path}"
+    
+    message = {
+        "text": "Video analysis",
+        "files": [video_path]
+    }
+    
+    result = process_user_input(message, 4)
+    
+    # Should have at least 3 items (text + at least one frame with text and image)
+    assert len(result) >= 3
+    
+    # First item should be the message text
+    assert result[0]["type"] == "text"
+    assert result[0]["text"] == "Video analysis"
+    
+    # Following items should be frame text and images
+    assert result[1]["type"] == "text"
+    assert result[1]["text"].startswith("Frame ")
+    
+    assert result[2]["type"] == "image"
+    assert "url" in result[2]
+    assert os.path.exists(result[2]["url"])
+
+
+def test_process_user_input_with_images():
+    """Test processing user input with image files."""
+    # Create temporary image files for testing
+    with tempfile.NamedTemporaryFile(suffix=".jpg", delete=False) as img1, \
+         tempfile.NamedTemporaryFile(suffix=".png", delete=False) as img2:
+        
+        image_paths = [img1.name, img2.name]
+        
+        message = {
+            "text": "Image analysis",
+            "files": image_paths
+        }
+        
+        result = process_user_input(message, 5)
+        
+        # Should have 3 items (text + 2 images)
+        assert len(result) == 3
+        
+        # First item should be the message text
+        assert result[0]["type"] == "text"
+        assert result[0]["text"] == "Image analysis"
+        
+        # Following items should be images
+        assert result[1]["type"] == "image"
+        assert result[1]["url"] == image_paths[0]
+        
+        assert result[2]["type"] == "image"
+        assert result[2]["url"] == image_paths[1]
+        
+    # Clean up temp files
+    for path in image_paths:
+        if os.path.exists(path):
+            os.unlink(path)
+
+
+def test_process_user_input_empty_text():
+    """Test processing user input with empty text but with files."""
+    video_path = os.path.join(ROOT_DIR, "assets", "test_video.mp4")
+    
+    message = {
+        "text": "",  # Empty text
+        "files": [video_path]
+    }
+    
+    # Add max_images parameter
+    result = process_user_input(message, 3)
+    
+    # First item should be empty text
+    assert result[0]["type"] == "text"
+    assert result[0]["text"] == ""
+    
+    # Rest should be video frames
+    assert len(result) > 1
+
+
+def test_process_user_input_handles_empty_files_list():
+    """Test that an empty files list is handled correctly."""
+    message = {
+        "text": "No files",
+        "files": []
+    }
+    
+    # Add max_images parameter
+    result = process_user_input(message, 3)
+    assert len(result) == 1
+    assert result[0]["type"] == "text"
+    assert result[0]["text"] == "No files"
+
+
+def test_process_user_input_max_images_effect():
+    """Test that max_images parameter correctly limits the number of frames."""
+    video_path = os.path.join(ROOT_DIR, "assets", "test_video.mp4")
+    
+    message = {
+        "text": "Video with few frames",
+        "files": [video_path]
+    }
+    
+    result_few = process_user_input(message, 2)
+    result_many = process_user_input(message, 5)
+    
+    # Count actual frames (each frame has a text and image entry)
+    frames_few = (len(result_few) - 1) // 2  # -1 for initial text message
+    frames_many = (len(result_many) - 1) // 2
+    
+    # Should respect max_images parameter
+    assert frames_few <= 2
+    assert frames_many <= 5
+    assert frames_few < frames_many