## create Walter the Wikipedia Web Scraper in a method

def walter(year):
    ## create html and open for bs 
    html = 'https://en.wikipedia.org/wiki/Category:' + str(year) + '_initial_public_offerings'
    html_open = urlopen(html)
    
    ## initialize webdriver and go to html site
    ## reference driver
    #global driver
    driver = webdriver.Chrome(ChromeDriverManager().install())
    driver.get(html)

    ## get the html source of the current page (will be the page listing all the IPOs for that year)
    soup_year = bs(html_open)

    ## get container of all the companies. Under html "div"
    container = soup_year.find("div", {"class": "mw-category"})

    ## find all the companies ('c') on page
    c1 = container.find_all("li")

    ## loop through c1 and use 'get_text' function to put all companies into a list ('c2')
    c2 = []
    
    ## create a list of dictionaries to keep the data in
    lst_dict = []
    
    ## create a value to keep track of companies passed in cy. Create value to track exceptions. Reference global passed value
    passed_cy = []
    exceptions = 0
    global passed

    for i in range(0, len(c1)):
        name = c1[i].get_text()
        c2.append(name)

    ## loop through company list (c2) and click on each company in list to visit company's wikipedia page
    for x in c2:
        ## initialize variables of interest with np.NaN. If variable is not available it 
        ## will be still be included in df.
        industry = np.NaN
        founded = np.NaN
        headquarters = np.NaN
        ticker = np.NaN
        
        try:
            ## click on company link
            driver.find_element_by_link_text(x).click()
            ## slow down crawler by 5 seconds
            sleep(5)
            try:
                ## get html source of page
                soup_company = bs(driver.page_source, 'html.parser')  
                ## get company's info table
                table = soup_company.find("table", {"class": "infobox vcard"})
                ## get all rows from the table
                rows = table.find_all("tr")

                ##create values list to keep track of points of interest
                vals = [x, year]

                ## loop through all rows and grab 'industry', 'founded', 'hq'
                ## in info unavailable, variable will be np.NaN
                for i in range(0, len(rows)):
                    ## use try for exception handling
                    try:
                        ## get header of variable and match to point of interest
                        header = rows[i].find("th").get_text()

                        if header == 'Industry':
                            industry = rows[i].find("td").get_text()
                        elif header == 'Founded':
                            founded = rows[i].find("td").get_text()
                        elif header == 'Headquarters':
                            headquarters = rows[i].find("td").get_text()
                        elif header == 'Traded as':
                            ticker = rows[i].find("td").get_text()
                        else:
                            ## if row does not match then skip
                            pass

                    except:
                        pass
            except:
                passed.append(x)
                exceptions += 1
                pass
        except:
            passed.append(x)
            exceptions += 1
            pass

        ## add variables to vals list and append to running list dictionary
        vals.append(ticker)
        vals.append(industry)
        vals.append(founded)
        vals.append(headquarters)
        lst_dict.append(dict(zip(cols, vals))) 

        ## go back one page to listing of companies
        driver.execute_script("window.history.go(-1)")
        
        ## add in delay before going to next company
        sleep(5)
        
    ## add in message that year is complete and how many companies were passed
    print("Walter complete for year " + str(year) + ". Number of companies passed was: " + str(exceptions))
    
    ## close web browser
    #driver.close()
    
    ## quit web browser
    driver.quit()
    
    return lst_dict