from app import app
from flask import request, render_template
import os
from skimage.metrics import structural_similarity
import imutils
import cv2
from PIL import Image

# adding path of config
app.config['INITIAL_FILE_UPLOADS'] = 'app/static/uploads'
app.config['EXTISTING_FILE'] = 'app/static/original'
app.config['GENERATED_FILE'] = 'app/static/generated'

# route to home page
@app.route("/", methods=["GET", "POST"])
def index():

    # Execute if requst is get
    if request.method == "GET":
        return render_template(index.html)

    # Execute if request is post
    if request.method == "POST":
        # Get uploaded image
        file_upload = request.files['file_upload']
        filename = file_upload.filename

        # Resize and save uploaded image
        uploaded_image = Image.open(file_upload).resize((250, 160))
        uploaded_image.save(os.path.join(app.config['INITIAL_FILE_UPLOADS'], 'IMAGE.JPG'))

        # Resize and save the original image to ensure both uploaded and original matches is size
        original_image = Image.open(os.path.join(app.config['EXISTING_FILE'], 'image.jpg')).resize((250,160))
        original_image.save(os.path.join(app.config['EXISTING_FILE'], 'image.jpg'))

        # Read uploaded and original image as array
        original_image = cv2.imread(os.path.join(app.config['EXISTING_FILE'], 'image.jpg'))
        uploaded_image = cv2.imread(os.path.join(app.config['INITIAL_FILE_UPLOADS'], 'image.jpg'))

        # Convert image into grayscale
        original_gray = cv2.cvtColor(original_image, cv2.COLOR_BGR2GRAY)
        uploaded_gray = cv2.cvtColor(uploaded_image, cv2.COLOR_BGR2GRAY)

        # Calculate structure similarity
        (score, diff) = structural_similarity(original_gray, uploaded_gray, full=True)
        diff = (diff * 255).astype('uint8')

        # Calculate thresold and contours
        thresh = cv2.threshold(diff, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1]
        cnts = cv2.findCountours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
        cnts = imutils.grab_contours(cnts)

        # Draw contours on image
        for c in cnts:
            (x, y, w, h) = cv2.boundingRect(c)
            cv2.rectangle(original_image, (x, y), (x+w, y+h), (0, 0, 255), 2)
            cv2.rectangle(uploaded_image, (x, y), (x+w, y+h), (0, 0, 255), 2)

        # Save all output image (if required)
        cv2.imwrite(os.path.join(app.config['GENERATE_FILE'], 'image_original.jpg'), original_image)
        cv2.imwrite(os.path.join(app.config['GENERATED_FILE'], 'image_uploaded.jpg'), uploaded_image)
        cv2.imwrite(os.path.join(app.config['GENERATED_FILE'], 'image_diff.jpg'), diff)
        cv2.imwrite(os.path.join(app.config['GENERATED_FILE'], 'image_thresh.jpg'), thresh)
        return render_template('index.html',pred=str(round(score*100,2)) + '%' + 'correct')

# main function
if __name__ == '__main__':
    app.run(debug=True)