grab_others.py

from bs4 import BeautifulSoup as bs
from urllib.request import urlopen

###Project Start: 6/14/21
###Project Last Updated: 6/28/21

####### Functions #######

def grab_others():

    ####### Initialization #######

    #initialization of the project, gaining info we need
    url = "https://coinmarketcap.com/"
    page = urlopen(url)
    html = page.read().decode("utf-8")
    soup = bs(html, "html.parser") #lxml or html.parser

    ######## TOTAL DATA GRAB ########

    #grabs html tag <tr> (table row) data from coinmarketcap.com and saves to variable named data
    data = soup.find_all("tr")

    #initialization of two lists to store the price and name of each cryptocurrency
    names = []
    prices = []

    #for loop iterating through all elements of data
    for element in data:

        #grab name, which is inside of the class cmc-link in our url
        name = element.find("a", class_="cmc-link")

        # append name to names
        names.append(name)

        #converting element to string-form of html, as it's easier to parse for the price
        element = element.prettify()

        #this comment starts the price information in HTML 
        price_start = element.find("<!-- -->")

        #cut down on the string such that it only contains information after the found start index
        element = element[price_start:]
        
        #find the immediate next decimal place, representative of the numbers decimal place
        price_end = element.find(".")

        #add the index of the decimal point (plus 3) to account for cents value
        price_end = price_end + 3

        #the price is the remaining string between the initial index of the html comment '<!-- -->' 
        price = element[:price_end]

        #remove excess information (html comments and excess spaces)
        price = price.replace("<!-- -->", "").strip()

        #append price to list prices
        prices.append(price)

    #stores all other (90) prices in prices list. Index of 11, as the 0 index of prices is void
    prices = prices[11:]

    #remainder prices: I am keeping this for the time being in case it's needed or desired to be parsed later, but most likely this is a...
    #TODO discard this extra information after determining usefulness
    #print(topten)

    #convert list of prices to floats
    for i in range(len(prices)):
        prices[i] = float(prices[i])

    #initialization of two lists to store links (which contains name of currency)
    currencies = []
    symbols = []

    #remove first element of list, which is void. Remaining list contains only elements of class bs4.element.Tag
    names.pop(0)

    #parse throught the list of names to get the coin name and the symbol
    for i in range(len(names)):
        
        #find the span with class crypto symbol, save it to symbol, and append the list symbols with its' value
        symbol = names[i].find("span", class_="crypto-symbol")
        symbols.append(symbol)

        #find all span tags in this particular element of the list
        spans = names[i].find_all("span")

        #if spans is an empty list (in this case, follows divergent website logic from other core groups), continue to next iteration and save the indices where there is an empty list
        if spans == []:
            continue
        else:
            #otherwise, append the list of currences with the middle element of spans, which happens to contain the name
            currencies.append(spans[1])

    #split symbols list into list containing the other 90 (see top ten data grab for top ten)
    symbols = symbols[10:]

    #adjusts names of cryptocurrencies
    coin_names = []
    for coin in currencies:
        
        #convert coin to type string
        coin = str(coin)

        #find string indeces of span tags
        span_start = coin.find("<span>")
        span_end = coin.find("</span>")

        #cut string to be just key information
        coin_name = coin[span_start:span_end]
        
        #append to coin_names the coin name excluding "<span>"
        coin_names.append(coin_name[len("<span>"):])

    #initialization of a list called coin symbols to store just thes coin symbol
    coin_symbols = []

    #for every index of symbols
    #this method can be done without being a numeric index of the list symbols by just iterating
    for coin_symbol in symbols:

        #cast to string
        coin_symbol = str(coin_symbol)

        #find the indices of the start and end of the symbol
        span_start = coin_symbol.find("<span class=\"crypto-symbol\">")
        span_end = coin_symbol.find("</span>")
        
        #remove all unnesscary info
        coin_symbol = coin_symbol[span_start:span_end]
        coin_symbol = coin_symbol[len("<span class=\"crypto-symbol\">"):]

        #append the element to coin_symbols
        coin_symbols.append(coin_symbol)

    #initialize dict of coins
    coin_dict = {}

    #coin_names should be len 90, same as symbols and prices
    for i in range(len(coin_names)):

        #insert into coin dictionary the name of each coin, it's rank (index i) and the associated price and symbol
        coin_dict[coin_names[i]] = {"rank": i + 11, "price" : prices[i], "symbol": coin_symbols[i]}

    return coin_dict