Hi I've written a web scraper for tracking gold price. Can someone please review my code and suggest improvements. It gets the price of gold from a specific url then stores the price against date in a db file further makes a plot then uploads image to firebase. Further this image and gold price is sent to your whats-app by using the twilio dependency.

Also please suggest methods for deployment on web. Any suggestions on further improvements are also welcome.

#Gold Price Tracker Everyday.
import sqlite3
import random
import numpy as np
from uuid import uuid4
import base64
import PyPDF2
import openpyxl as wb
import urllib.request,urllib.parse,urllib.error
from pdfminer.pdfdocument import PDFDocument
from matplotlib import pyplot as plt
import ssl
import os
import io
from twilio.rest import Client
from twilio.jwt.access_token import AccessToken
from twilio.jwt.access_token.grants import ChatGrant
import sys
import requests
import firebase_admin
from firebase_admin import credentials
from firebase_admin import storage

from firebase_admin import credentials

account_sid='my sid for twilio'
account_auth = "authorization id for twilio"


grams=1032.6/1000 # Denotes the number of grams that is present with you currently.
print("Your Current Grams of Gold is %f"%grams)
global date

def get_price():

    '''This module is to get the price of gold online'''

    html=urllib.request.urlopen(url,context=ctrx).read()# Sending the request to the #designated url
    reader = PyPDF2.PdfFileReader(memoryFile)

    contents = reader.getPage(0).extractText().split('\n')# Parsing the gold price
    gp=float(str(contents[298]).replace(',',''))# Today's gold price.
    return gp  # Returns today's gold price.

def message():# Prints the message
    '''Prints the message that is required'''
    print("Selling at today's price will fetch Rs %f"%(g_p*grams))

def get_date_time():
    '''Function to get present date and time from appspot'''

    date = str(d1[0]).replace('b','').replace('\'','')# Gets Date from appspot
    time = str(d1[1]).replace('b','').replace('\'','')

def populate_database(today_date,today_price):
# Here I am creating a database and populating the entries there
    # Used to update the database with the current price
     conn = sqlite3.connect('.\Price_tracker.db')
     conn.row_factory = lambda cursor, row: row[0]
     cur = conn.cursor()

     cur.execute('''INSERT INTO dp_tracker(Date,Price) VALUES (?,?) ''',(today_date,today_price,))

     dates_list=cur.execute('''SELECT Date FROM dp_tracker''').fetchall()
     price_list=cur.execute('''SELECT Price FROM dp_tracker''').fetchall()



def plot():

    ''' This function is used to plot the gold price scraped online'''
    gp=get_price()# Today's gold Price
    date=d[0]#Todya's date

    dp=populate_database(str(date),float(gp)) #Passing today's date and today's gold price and returns a list of all

    dx=dp[0] #A list of dates obtained from database
    py=dp[1]#A list of prices obtained from excel file
    fig = plt.figure()

    plt.bar(dx_pos, py, align='center', alpha=0.5,figure=fig)
    plt.title("Gold Price Tracker",figure=fig)

    for i, v in enumerate(py):
        plt.text(dx_pos[i] - 0.15, v + 0.01, str(v),figure=fig)

    return fig

def upload2firebase():

    cred = credentials.Certificate(
    firebase_admin.initialize_app(cred, {
        'storageBucket': 'gold-price-tracker-caa9e.appspot.com'

    img_src = "sample_image.png"
    bucket = storage.bucket()
    blob = bucket.blob(img_src)

    # Create new token
    new_token = uuid4()

    # Create new dictionary with the metadata
    metadata = {"firebaseStorageDownloadTokens": new_token}

    # Set metadata to blob
    blob.metadata = metadata

    # Upload file
    blob.upload_from_filename(filename="./Test.png", content_type='image/png')
        # bucket = storage.bucket()
        # image_data = ""
        # with open("./Test.png", "rb") as img_file:
        #     image_data = base64.b64encode(img_file.read())
        # blob = bucket.blob('test.png')
        # blob.upload_from_string(image_data)
        # return blob.public_url

def send2Phone(gram,price,r):

    '''This function is to send the message to the phone'''
    client=Client(account_sid , account_auth)
    to_what_app_number='whatsapp:my number'
    a="Your current grams of gold is "+str(gram)+" g.\n Selling at today's price will fetch Rs "+str(gram*price)

def loop():

    # t=get_date_time()
    # time=t[1].split(':')
    # count=0
    # if time[0] == '08' and time[1] == '57' and time[2]=='00' :
    #     count=1
    # else:
    #     count=0
    # if count==1:
        q = plot()
        q.savefig('Test.png') #Saving image locally and upload to firebase.
        r = upload2firebase() # Getting public url of the image from firebase


while True:


There are so, so many examples of people scraping stock tracker sites. For beginners it's an understandable urge: you can see the data on the web, and you want to be able to translate those data using a script.

The first thing you should reach for is an API, not a scraper. Scrapers are fragile, inefficient, and sometimes immoral - most website creators intend on human consumption, rather than bot consumption, and may be losing out on ad revenue. It's not clear to me what the business model of the MMTC is so I cannot confirm whether that is the case here, but there are many APIs that will give you the price of gold without having to go through the round trip of PDF-render-PDF-parse.

Import order

There are several different ways to do this; I recommend:

  • Built-in libraries first, alphabetical
  • External libraries second, alphabetical

Global constants

Things like grams should be GRAMS since they're global constants.

ctrx, gold_prices and date should not be at the global level and should be state as represented in function arguments and/or class members.

date needs its name changed to avoid shadowing the built-in datetime.date.

account_* variables should not be hard-coded, and should be saved to a secure secrets wallet. There are multiple ways to do this either via Python libraries or the OS that you are using.

Strings like "./gold-price-tracker-caa9e-firebase-adminsdk-9e39d-72694e4d52.json" and 'gold-price-tracker-caa9e.appspot.com' should be moved to global constants, if not made parametric configuration.


Among other elements of the PEP8 standard, this:

gp=float(str(contents[298]).replace(',',''))# Today's gold price.

should have two spaces before the hash. In fact, it's deeply confusing StackExchange's Python highlighting parser.


get_price does not just get the price. It also adds the price to gold_prices. Why? This appending should not be done in this function.


You stream-ify the result of urlopen to a BytesIO. There is a much better way to do this:

  • Use requests, not urlopen
  • Use the streaming option of requests.get
  • Use the raw stream from the response object, which is already a stream, passing this to PdfFileReader
  • Do not make a BytesIO

Getting the date and time

Python has this built-in: datetime.datetime.now. Do not make an HTTP request to get the current time. If you're worried that the client's time is not reliable, that's a different problem that should be solved at the operating system level with NTP.

Context managers

Read the docs. Use a with statement for your SQLite variables. Also, your call to close() should be in a finally, since the context manager does not actually do a close.

No-op loop

Your final

while True:

does not have any effect. Replace it with a single call to loop().

| improve this answer | |
  • \$\begingroup\$ Wow thanks a lot for that! \$\endgroup\$ – vinaysr93 Apr 30 at 3:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.