import os
import requests
import time
from gpiozero import Motor, DistanceSensor, AngularServo
from picamera2 import Picamera2
from gtts import gTTS
import pygame
import speech_recognition as sr

# --- GPIO Setup ---
motor_left = Motor(forward=5, backward=6)
motor_right = Motor(forward=13, backward=19)
ultrasonic = DistanceSensor(echo=27, trigger=17)
servo = AngularServo(18, min_angle=0, max_angle=180)

# --- Camera ---
picam = Picamera2()
picam.configure(picam.create_still_configuration())

# --- AI Backend ---
VOICE_API = "http://192.168.215.29:8080/api/ai/voice"
IMAGE_API = "http://192.168.215.29:8080/api/ai/image"

# --- Global State ---
automatic_mode = False

# --- Movement Functions ---
def move_forward(t=1): 
    motor_left.forward()
    motor_right.forward()
    time.sleep(t)
    stop()

def move_backward(t=1): 
    motor_left.backward()
    motor_right.backward()
    time.sleep(t)
    stop()

def turn_left(): 
    motor_left.backward()
    motor_right.forward()
    time.sleep(0.5)
    stop()

def turn_right(): 
    motor_left.forward()
    motor_right.backward()
    time.sleep(0.5)
    stop()

def stop(): 
    motor_left.stop()
    motor_right.stop()

# --- Image Functions ---
def capture_image():
    filename = "/tmp/obstacle.jpg"
    picam.start()
    time.sleep(1)
    picam.capture_file(filename)
    picam.stop()
    return filename

def send_image_to_laptop(filepath):
    with open(filepath, "rb") as f:
        files = {"image": f}
        res = requests.post(IMAGE_API, files=files)
        return res.json().get("caption", "No description found")

# --- Speech Functions ---
def speak(text, lang="en"):
    tts = gTTS(text, lang=lang)
    filename = "/tmp/speak.mp3"
    tts.save(filename)
    pygame.mixer.init()
    pygame.mixer.music.load(filename)
    pygame.mixer.music.play()
    while pygame.mixer.music.get_busy():
        continue

# --- Sensor Functions ---
def scan_surroundings():
    angles = [0, 45, 90, 135, 180]
    for angle in angles:
        servo.angle = angle
        time.sleep(0.3)
        dist = ultrasonic.distance * 100
        print(f"Angle {angle}°: {dist:.2f} cm")
    servo.angle = 90

# --- Autonomous Mode Loop ---
def auto_mode_loop():
    global automatic_mode
    servo.angle = 90
    while automatic_mode:
        dist = ultrasonic.distance * 100
        print(f"Distance: {dist:.2f} cm")
        if dist < 25:
            stop()
            image = capture_image()
            caption = send_image_to_laptop(image)
            speak(f"Stop! Obstacle ahead: {caption}")
            turn_left()
        else:
            move_forward(0.5)
        time.sleep(0.1)

# --- Voice Recording & AI Interaction ---
def record_and_send_voice():
    os.system("arecord -D plughw:1,0 -f cd -t wav -d 4 -r 16000 /tmp/voice.wav")
    rec = sr.Recognizer()
    with sr.AudioFile("/tmp/voice.wav") as source:
        audio = rec.record(source)
    try:
        text = rec.recognize_google(audio, language="en-US")
        print(f"Recognized: {text}")
        res = requests.post(VOICE_API, json={"text": text})
        reply = res.json().get("reply", "")
        print(f"AI says: {reply}")
        speak(reply, lang="en")
        handle_ai_command(reply)
    except Exception as e:
        print("Error:", e)
        speak("Sorry, I didn't understand.")

# --- Command Handling ---
def handle_ai_command(command):
    global automatic_mode
    if "forward" in command:
        move_forward()
    elif "backward" in command:
        move_backward()
    elif "left" in command:
        turn_left()
    elif "right" in command:
        turn_right()
    elif "stop" in command:
        stop()
    elif "scan" in command:
        scan_surroundings()
    elif "auto" in command:
        automatic_mode = True
        auto_mode_loop()
    elif "manual" in command:
        automatic_mode = False
        stop()

# --- Extras ---
def robo_dance():
    for _ in range(2):
        turn_left()
        turn_right()
    speak("I'm dancing!", lang="en")

def play_rock_paper_scissors():
    speak("Rock, paper, or scissors! I choose: paper", lang="en")

from flask import Flask, request, jsonify
from voice_processor import handle_voice_command
from image_processor import handle_image_caption

app = Flask(__name__)

@app.route("/api/ai/voice", methods=["POST"])
def voice():
    text = request.json.get("text")
    result = handle_voice_command(text)
    return jsonify({"reply": result})

@app.route("/api/ai/image", methods=["POST"])
def image():
    file = request.files.get("image")
    if not file:
        return jsonify({"error": "No image uploaded"}), 400
    caption = handle_image_caption(file)
    return jsonify({"caption": caption})

if __name__ == "__main__":
    app.run(host="0.0.0.0", port=8080)

import requests

OLLAMA_URL = "http://localhost:11434/api/generate"

def handle_voice_command(text):
    prompt = f"You are a robot controller. Respond to this command: '{text}'"
    payload = {
        "model": "borch/llama3_speed_chat",
        "prompt": prompt,
        "stream": False
    }
    res = requests.post(OLLAMA_URL, json=payload)
    return res.json().get("response", "").strip()

import requests
import base64
from gtts import gTTS
import os

OLLAMA_URL = "http://localhost:11434/api/generate"

def handle_image_caption(file):
    image_data = base64.b64encode(file.read()).decode("utf-8")
    prompt = "Describe this image."

    payload = {
        "model": "llava",
        "prompt": prompt,
        "images": [image_data],
        "stream": False
    }

    res = requests.post(OLLAMA_URL, json=payload)
    hindi_caption = res.json().get("response", "").strip()

    # Speak it
    tts = gTTS(hindi_caption, lang="hi")
    filename = "/tmp/output.mp3"
    tts.save(filename)
    os.system(f"mpg123 {filename} &")  # Or another player

    return caption