eclabsF23_data_chicago.py

# IMPORTS
from bs4 import BeautifulSoup
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.common.exceptions import NoSuchElementException
import pandas as pd
import time
import csv

def extract_data1(driver):
    soup = BeautifulSoup(driver.page_source, 'lxml')
    try:
        Store_Name = soup.find("h1").get_text() 
    except AttributeError: #getting an error here means the page didn't load correctly, so we just refresh, wait, and retry
        driver.refresh()
        time.sleep(5)
        soup = BeautifulSoup(driver.page_source, 'lxml')
        Store_Name = soup.find("h1").get_text()
    rating_wrapper = soup.find("div", class_ = "RatingWrapper-sc-6cf8fc1f-0 kxYFiI")
    Review_Counter = rating_wrapper.find("button")

    # If reviews exist, the button will have 5 elements under contents, and the review count will be the third. In this case, the average rating will be under a div
    # Otherwise, reviews don't exist and so we can just set Num_Reviews to 0 amd Avg_Rating to None
    try: 
        Num_Reviews = Review_Counter.contents[2]
        Avg_Rating = rating_wrapper.find("div").get_text()
    except (AttributeError, IndexError):
        Num_Reviews = 0
        Avg_Rating = None
        

    #banner includes stuff like is it recreational or medical, delivery or pickup, etc.
    banner_items = soup.find("div", class_ = "ChipWrapper-sc-b686745d-0 dcyYkr").find_all("div")
    descriptors = " | ".join(map(lambda e : e.get_text(), banner_items))

    phone = soup.find("a", {"aria-label" : "phone"})
    if phone:
        phone_number = phone.contents[1]
    else:
        phone_number = ""

    final = [Store_Name, Num_Reviews, Avg_Rating, descriptors, phone_number]
    return final

def extract_data3(driver): #this should scrape the store details, including address, phone number, license info, etc.
    soup = BeautifulSoup(driver.page_source, 'lxml')
    licenses = " | ".join(map(lambda e : e.get_text(), soup.find_all("div", class_ = 'Text-sc-51fcf911-0 blMBuQ')))
    address = soup.find("div", {"data-testid" : "listing-address"}).get_text()
    return [licenses, address]

c = webdriver.ChromeOptions()
c.add_argument("--incognito")
driver = webdriver.Chrome(options = c)
driver.maximize_window()

fields = ["Name", "Number Reviews", "Avg Rating", "Descriptors", "Phone Number", "Licenses", "Address", "Url"]
file = open(f'./stores_data.csv', 'w', newline = '')
file_writer = csv.writer(file, delimiter = ',', quoting = csv.QUOTE_MINIMAL)
file_writer.writerow(fields)

driver.get("https://weedmaps.com/dispensaries/sunnyside-dispensary-river-north")
driver.find_element(By.XPATH, '/html/body/div[1]/div[2]/div/div/div[2]/button[1]').click() #click the "I am above 21" button

page_urls = pd.read_csv("./all_names_urls.csv", header = None)[0]
chicago_data = pd.read_csv("./chicago_stores_data.csv")
seen_dispensaries=  set(chicago_data["Url"])
base_url = "https://weedmaps.com"

# Loop through urls of cities
for url in page_urls:
    i = 1
    driver.get(base_url + url)
    time.sleep(1)
    url = driver.current_url #sometimes the url listed on the page is not the same url, so we reget it
    while True:
        driver.get(url + f"?page={i}")
        
        time.sleep(1.75)
        try:
            list = driver.find_element(By.XPATH, '//*[@data-testid="map-listings-list-wrapper"]')
        except NoSuchElementException:
            break       
        stores_list = list.find_elements(By.XPATH, './li')
        if len(stores_list) == 0: #if there's no stores listed on the page then we've gone past all the pages of data and can finish the loop
            break
        stores_links = [x.find_element(By.XPATH, "./div[1]/div[1]/div[1]/div/div[2]/a").get_attribute("href") for x in stores_list]
        for link in stores_links:
            if link not in seen_dispensaries:
                seen_dispensaries.add(link)
                driver.get(link)
                data = extract_data1(driver)

                driver.find_element(By.XPATH, '//button[@aria-label="store-details"]').click() #click store details button
                time.sleep(1.75) #wait for everything to load properly
                try:
                    driver.find_element(By.XPATH, '//*[@id="details"]/div/div[1]/div/span/div/button').click() #click license info button
                except NoSuchElementException:
                    pass
                data = data + extract_data3(driver) + [link]
                file_writer.writerow(data)
                file.flush()
        i += 1
pd.concat([chicago_data, pd.read_csv("./stores_data.csv")]).to_csv("stores_data.csv", index = False) #merge collected data with previously collected chicago data