import cv2
from moviepy.editor import VideoFileClip
from gemini_vision import process_frame_with_gemini, summarize_with_gemini, extract_code_with_gemini

def frame_generator(video_path):
    video = cv2.VideoCapture(video_path)
    while True:
        ret, frame = video.read()
        if not ret:
            break
        yield frame
    video.release()

def process_video(video_path, summary_words):
    # Process frames with Gemini Vision Pro
    extracted_text = ""
    for frame in frame_generator(video_path):
        extracted_text += process_frame_with_gemini(frame) + " "

    # Extract audio and transcribe
    video = VideoFileClip(video_path)
    audio = video.audio
    audio_path = "temp_audio.wav"
    audio.write_audiofile(audio_path)

    # Use Gemini Vision Pro for transcription (assuming it can handle audio)
    transcription = process_frame_with_gemini(audio_path, mode="audio")

    # Summarize content using Gemini
    summary = summarize_with_gemini(transcription, max_words=summary_words)

    # Extract code using Gemini
    extracted_code = extract_code_with_gemini(transcription)

    return {
        'summary': summary,
        'extracted_text': extracted_text,
        'transcription': transcription,
        'extracted_code': extracted_code
    }