# from ultralytics import YOLO
# import cv2
# import matplotlib.pyplot as plt
# import matplotlib.patches as patches
# import numpy as np
# import requests

# model = YOLO('best (5).pt')
# img_url = 'https://www.greendna.in/cdn/shop/products/1296x728_Holy_Basil_1155x.jpg?v=1591462900'
# response = requests.get(img_url, stream=True)
# img_array = np.asarray(bytearray(response.content), dtype=np.uint8)
# img = cv2.imdecode(img_array, cv2.IMREAD_COLOR)

# classes_ = {0: 'anthurium', 1: 'clivia', 2: 'dieffenbachia', 3: 'dracaena', 4: 'gloxinia', 5: 'kalanchoe', 6: 'orchid', 7: 'sansevieria', 8: 'violet', 9: 'zamioculcas'}

# results = model.predict(source=img, conf = 0.4)

# # results = model.predict('api/default_1280-720-screenshot.webp', confidence=40, overlap=30).json()
# boxes = results[0].boxes.xyxy.tolist()
# classes = results[0].boxes.cls.tolist()
# names = results[0].names
# confidences = results[0].boxes.conf.tolist()

# print(boxes)
# print(classes)
# print(names)
# print(confidences)

# # Iterate through the results
# for box, cls, conf in zip(boxes, classes, confidences):
#     x1, y1, x2, y2 = box
#     confidence = conf
#     detected_class = cls
#     name = names[int(cls)]

# def plot_img_bbox(img, target):
#     fig, a = plt.subplots(1,1)
#     fig.set_size_inches(10, 10)
#     a.imshow(img)
#     for i, box in enumerate(target):
#         #print(target['boxes'])
#         x, y, width, height  = box[0], box[1], box[2]-box[0], box[3]-box[1]
# #         if arr[target['labels'][i]] == 'ad':
#         rect = patches.Rectangle((x, y),
#                                      width, height,
#                                      linewidth = 2,
#                                      edgecolor = 'r',
#                                      facecolor = 'none')
#         a.text(x, y-20, classes_[classes[i]], color='b', verticalalignment='top')

#         a.add_patch(rect)
#     plt.show()

# # if length of boxes is zero that means no deceptive popups were found
# plot_img_bbox(img, boxes)

import requests
from ultralytics import YOLO
import cv2
import matplotlib.pyplot as plt
import matplotlib.patches as patches
import numpy as np
import gradio as gr

model = YOLO('best (5).pt')

def plot_img_bbox(img, target, save_path, classes):
    fig, a = plt.subplots(1, 1)
    fig.set_size_inches(10, 10)
    classes_ = {0: 'anthurium', 1: 'clivia', 2: 'dieffenbachia', 3: 'dracaena', 4: 'gloxinia', 5: 'kalanchoe', 6: 'orchid', 7: 'sansevieria', 8: 'violet', 9: 'zamioculcas'}
    a.imshow(img)
    for i, box in enumerate(target):
        x, y, width, height = box[0], box[1], box[2] - box[0], box[3] - box[1]
        rect = patches.Rectangle((x, y), width, height, linewidth=2, edgecolor='r', facecolor='none')
        a.text(x, y - 20, classes_[classes[i]], color='b', verticalalignment='top')
        a.add_patch(rect)
    plt.savefig(save_path)
    plt.close()  

    upload_url = upload_to_cloudinary(save_path)

    return upload_url

def upload_to_cloudinary(local_file_path):
    upload_url = 'https://api.cloudinary.com/v1_1/ddvajyjou/image/upload'
    files = {'file': open(local_file_path, 'rb')}
    params = {'upload_preset': 'nb6tvi1b'}  

    response = requests.post(upload_url, files=files, params=params)

    if response.status_code == 200:
        return response.json()['secure_url']
    else:
        print(f"Error uploading to Cloudinary: {response.status_code}")
        return None

def index(img_url):
    response = requests.get(img_url, stream=True)
    img_array = np.asarray(bytearray(response.content), dtype=np.uint8)
    img = cv2.imdecode(img_array, cv2.IMREAD_COLOR)
    
    print(img_url)

    results = model.predict(source=img, conf = 0.4)

    boxes = results[0].boxes.xyxy.tolist()
    classes = results[0].boxes.cls.tolist()
    names = results[0].names
    confidences = results[0].boxes.conf.tolist()

    print(boxes)
    print(classes)
    print(names)
    print(confidences)

    final_url = plot_img_bbox(img, boxes, 'image.png', classes)
    return final_url

inputs_image_url = [
    gr.Textbox(type="text", label="Image URL"),
]

outputs_result_dict = [
    gr.Textbox(type="text", label="Result Dictionary"),
]

interface_image_url = gr.Interface(
    fn=index,
    inputs=inputs_image_url,
    outputs=outputs_result_dict,
    title="Popup detection",
    cache_examples=False,
)

gr.TabbedInterface(
    [interface_image_url],
    tab_names=['Image inference']
).queue().launch()