import cv2
import mediapipe as mp
import numpy as np
import gradio as gr

# Initialize MediaPipe pose and drawing utilities
mp_pose = mp.solutions.pose
pose = mp_pose.Pose(min_detection_confidence=0.5, min_tracking_confidence=0.5)
mp_drawing = mp.solutions.drawing_utils

# Common angle calculation function
def calculate_angle(a, b, c):
    a, b, c = np.array(a), np.array(b), np.array(c)
    radians = np.arctan2(c[1]-b[1], c[0]-b[0]) - np.arctan2(a[1]-b[1], a[0]-b[0])
    angle = np.abs(radians * 180.0 / np.pi)
    return angle

def check_leg_raise_feedback(landmarks):
    # Using left leg landmarks as reference
    hip = [landmarks[mp_pose.PoseLandmark.LEFT_HIP.value].x,
           landmarks[mp_pose.PoseLandmark.LEFT_HIP.value].y]
    knee = [landmarks[mp_pose.PoseLandmark.LEFT_KNEE.value].x,
            landmarks[mp_pose.PoseLandmark.LEFT_KNEE.value].y]
    ankle = [landmarks[mp_pose.PoseLandmark.LEFT_ANKLE.value].x,
             landmarks[mp_pose.PoseLandmark.LEFT_ANKLE.value].y]
    
    angle = calculate_angle(hip, knee, ankle)
    leg_lift = 1 - knee[1]  # approximate vertical lift (adjust as needed)
    accuracy = max(0, min(100, (1 - abs(angle - 180) / 50) * 100))
    
    feedback = "Correct Leg Raise" if angle > 160 and leg_lift > 0.4 else "Incorrect Leg Raise"
    if angle < 160:
        feedback += " - Keep Legs Straight"
    if leg_lift < 0.4:
        feedback += " - Raise Legs Higher"
    return feedback, int(accuracy)

def draw_accuracy_bar(image, accuracy):
    bar_x, bar_y = 50, image.shape[0] - 50
    bar_width, bar_height = 200, 20
    fill_width = int((accuracy / 100) * bar_width)
    color = (0, 255, 0) if accuracy >= 80 else (0, 0, 255) if accuracy < 50 else (0, 255, 255)
    cv2.rectangle(image, (bar_x, bar_y), (bar_x + bar_width, bar_y + bar_height), (200, 200, 200), 2)
    cv2.rectangle(image, (bar_x, bar_y), (bar_x + fill_width, bar_y + bar_height), color, -1)
    cv2.putText(image, f"Accuracy: {accuracy}%", (bar_x, bar_y - 10),
                cv2.FONT_HERSHEY_DUPLEX, 0.6, (255, 255, 255), 2)

def analyze_leg_raises(video_path):
    cap = cv2.VideoCapture(video_path)
    frame_width, frame_height = int(cap.get(3)), int(cap.get(4))
    fps = cap.get(cv2.CAP_PROP_FPS) if cap.get(cv2.CAP_PROP_FPS) > 0 else 30

    output_video = "output_leg_raises.mp4"
    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
    out = cv2.VideoWriter(output_video, fourcc, fps, (frame_width, frame_height))

    while cap.isOpened():
        ret, frame = cap.read()
        if not ret:
            break
        image = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
        results = pose.process(image)
        image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
        if results.pose_landmarks:
            mp_drawing.draw_landmarks(image, results.pose_landmarks, mp_pose.POSE_CONNECTIONS)
            landmarks = results.pose_landmarks.landmark
            feedback, accuracy = check_leg_raise_feedback(landmarks)
            draw_accuracy_bar(image, accuracy)
            color = (0, 255, 0) if "Correct" in feedback else (0, 0, 255)
            cv2.putText(image, feedback, (50, 50),
                        cv2.FONT_HERSHEY_COMPLEX, 1, color, 3)
        out.write(image)
    cap.release()
    out.release()
    return output_video

gr.Interface(
    fn=analyze_leg_raises,
    inputs=gr.Video(),
    outputs=gr.Video(),
    title="Leg Raises Form Analyzer",
    description="Upload a video of your leg raises and receive form feedback!"
).launch()