replaceBase64toimage.py

import pandas as pd
import requests
import json
import sched, time
import base64
from bs4 import BeautifulSoup as BSHTML


# print(df.to_json(orient = 'index', indent = 2))
# print(df.to_json(orient='records', indent = 2))

# url = './admissionerpap.csv'

# df = pd.read_csv(url, sep=',')
# data = df.to_json(orient='records', indent = 2)

def purifyString(item, key):
    soup = BSHTML(item[key], "html.parser")
    images = soup.findAll('img')

    i = 0
    for img in images:
        # print(img['src'][22:])
        try:
            with open("./img/image_{0}_{1}.png".format(i, item['question_id']), "wb") as fh:
                fh.write(base64.b64decode(img['src'][22:]))
        except:
            # print(img)
            img['src'] = ""
            continue
        
        img['src'] = "/admission/questions/image_{0}_{1}.png".format(i, item['question_id'])
        # print(images)
        i += 1
    return format(soup)


with open("admissionerpap.json", "r") as read_file:
   data = json.load(read_file)
# data = json.loads('list.json')
# print(data)
newData = []
# for item in json.loads(data):
for item in data:
    # print(item['type'])
    # try:
    string = purifyString(item, 'question')
    # itemObj = {
    #     "question_id": '{}'.format(item['question_id']),
    #     "quiz_id": '{}'.format(item['quiz_id']),
    #     "order_idx": '{}'.format(item['order_idx']),
    #     "question": format(soup),
    #     # "question": newsrc,
    #     "type": '{}'.format(item['type']),
    #     # "explanation": '{}'.format(item['explanation']),
    # }

    with open("sql.txt", "a") as myfile:
        myfile.write("UPDATE `content_questions` SET `question`='{}' WHERE `question_id`={};\n".format(string.replace("'","\\\'"), item['question_id']))

        # newData.append(itemObj)
    # except:
    #     continue




# with open('./list.json', 'w') as outfile:
#     json.dump(newData, outfile, ensure_ascii=False)
# rows = json.dumps(newData)
# print(rows)
# df2 = pd.DataFrame(newData, columns=["colummn"])
# df2.to_csv('list.csv', index=False)
# print(newData)