import cv2
import numpy as np
from io import BytesIO
from PIL import Image
from ppadb.client import Client as AdbClient


def get_current_screen():
    return current_screen


def capture_current_screen():
    global current_screen
    current_screen = device.screencap()

    return current_screen


def find_center(x1, y1, x2, y2):
    centerX = round(x1 + (x2 - x1) / 2)
    centerY = round(y1 + (y2 - y1) / 2)
    return centerX, centerY


def tap(x, y=None):
    # Check if x is an int
    if isinstance(x, int):
        if not isinstance(y, int):
            raise ValueError("y must be an int when x is an int")
        # Construct the location string from both x and y
        location = f"{x} {y}"
    # Check if x is a string
    elif isinstance(x, str):
        location = x
    else:
        raise TypeError("x must be either an int or a string")

    # Assuming 'device' is a previously defined object with a 'shell' method
    action = f"input tap {location}"
    print(action)
    device.shell(action)


def tap_button(template):
    button = find_template(template)
    if len(button) == 0:
        return
    tap(f"{button[0][0]} {button[0][1]}")


def swipe(start, end, duration=1000):
    action = f"input swipe {start} {end} {duration}"
    print(action)
    device.shell(action)


def look_for_templates(templates):
    for name, template in templates.items():
        locations = find_template(template)
        if len(locations) > 0:
            return name, locations

    return None, None


def find_template(template_image):
    target_image = Image.open(BytesIO(get_current_screen()))

    # Convert the image to a NumPy array and then to BGR format (which OpenCV uses)
    target_image = np.array(target_image)
    target_image = cv2.cvtColor(target_image, cv2.COLOR_RGB2BGR)

    h, w = template_image.shape[:-1]

    # Template matching
    result = cv2.matchTemplate(target_image, template_image, cv2.TM_CCOEFF_NORMED)

    # Define a threshold
    threshold = 0.9  # Adjust this threshold based on your requirements

    # Finding all locations where match exceeds threshold
    locations = np.where(result >= threshold)
    locations = list(zip(*locations[::-1]))

    # Create list of rectangles
    rectangles = [(*loc, loc[0] + w, loc[1] + h) for loc in locations]

    # Apply non-maximum suppression to remove overlaps
    rectangles = non_max_suppression(rectangles, 0.3)

    # Initialize an empty list to store coordinates
    coordinates = []

    for startX, startY, endX, endY in rectangles:
        # Append the coordinate pair to the list
        coordinates.append(find_center(startX, startY, endX, endY))

    # Sort the coordinates by y value in ascending order
    return sorted(coordinates, key=lambda x: x[1])


def non_max_suppression(boxes, overlapThresh):
    if len(boxes) == 0:
        return []

    # Convert to float
    boxes = np.array(boxes, dtype="float")

    # Initialize the list of picked indexes
    pick = []

    # Grab the coordinates of the bounding boxes
    x1 = boxes[:, 0]
    y1 = boxes[:, 1]
    x2 = boxes[:, 2]
    y2 = boxes[:, 3]

    # Compute the area of the bounding boxes and sort by bottom-right y-coordinate
    area = (x2 - x1 + 1) * (y2 - y1 + 1)
    idxs = np.argsort(y2)

    # Keep looping while some indexes still remain in the indexes list
    while len(idxs) > 0:
        # Grab the last index in the indexes list and add the index value to the list of picked indexes
        last = len(idxs) - 1
        i = idxs[last]
        pick.append(i)

        # Find the largest (x, y) coordinates for the start of the bounding box and the smallest (x, y)
        # coordinates for the end of the bounding box
        xx1 = np.maximum(x1[i], x1[idxs[:last]])
        yy1 = np.maximum(y1[i], y1[idxs[:last]])
        xx2 = np.minimum(x2[i], x2[idxs[:last]])
        yy2 = np.minimum(y2[i], y2[idxs[:last]])

        # Compute the width and height of the bounding box
        w = np.maximum(0, xx2 - xx1 + 1)
        h = np.maximum(0, yy2 - yy1 + 1)

        # Compute the ratio of overlap
        overlap = (w * h) / area[idxs[:last]]

        # Delete all indexes from the index list that have overlap greater than the threshold
        idxs = np.delete(
            idxs, np.concatenate(([last], np.where(overlap > overlapThresh)[0]))
        )

    # Return only the bounding boxes that were picked
    return boxes[pick].astype("int")


client = AdbClient(host="127.0.0.1", port=5037)
device = client.device("192.168.178.32:5555")

current_screen = capture_current_screen()