ADV_3.py

import newspaper
from newspaper import Article
import concurrent.futures

URLs = ['http://www.foxnews.com/',
        'http://www.cnn.com/',
        'http://www.derspiegel.de/',
        'http://www.bbc.co.uk/',
        'https://theguardian.com',]


def get_headlines():
    '''
    Function for printing 5 firtst headlines from news sites specified in URLs list
    Input: none
    Output: none, print statements instead
    '''
    for url in URLs:                                                                #for each element in URLs list
        result = newspaper.build(url, memoize_articles=False)                       #create an instance of an article and set the url attribute value to thread parameter and memoize_articles attribute value to False
        print('\n''The headlines from %s are' % url, '\n')                          #print string with url from URLs as inserted value
        for i in range(1,6):                                                        #loop 5 times
            art = result.articles[i]                                                #set art variable to a refrence to an instruction for getting i article from the website
            art.download()                                                          #download article specified in art variable
            art.parse()                                                             #parse article specified in art variable
            print(art.title)                                                        #print headline of the article
        
def get_headlines_thread(thread):
        '''
        Function for accesing a website from url provided and getting first 5 headlines
        Input: string. url of the website
        Output: string. first 5 headlines of the website
        '''
        art_content = ""                                                            #create string in which headlines will be stored
        result = newspaper.build(thread, memoize_articles=False)                    #create an instance of an article and set the url attribute value to thread parameter and memoize_articles attribute value to False
        art_content += '\n''The headlines from %s are\n\n' % thread                 #add to art_content string another string with website url as inserted value
        for i in range(1,6):                                                        #loop 5 times
            art = result.articles[i]                                                #set art variable to a refrence to an instruction for getting i article from the website
            art.download()                                                          #download article specified in art variable
            art.parse()                                                             #parse article specified in art variable
            art_content += art.title + "\n"                                         #join art content string, headline of article specified in art variable, and an enter key
        return art_content,thread                                                   #return art_content string

def concurent_get_headlines():
    '''
    Function for councurrently printing url of a news website and 5 of its headlines
    Input: no Input
    Output: no output, the functions ends with a print statement instead
    '''
    threaded_message = [None] * len(URLs)                                           #initialise list of None values with the same length as URLs list
    with concurrent.futures.ThreadPoolExecutor(max_workers=5) as executor:          #create Executor class from concurrent.futures module and ThreadPoolExecutor subclass with max_workers argument set to 5
        threads = {executor.submit(get_headlines_thread,url): url for url in URLs}  #create threads dictionary with keys being submit method from Executor class and values being ulrs from URLs list
        for thread in concurrent.futures.as_completed(threads):                     #for each thread in pool of threads
            headlines = thread.result()                                             #initialise healines tuple and place in it two return values of thread function
            threaded_message[URLs.index(headlines[1])]= headlines[0]                #get the index of a value inside the URLs list that is equal to second element of headlines tuple, 
                                                                                    #and under the same index inside threaded_message list, place first element of headline tuple
        print("\n".join(threaded_message))                                          #print strings in threaded message list joined by enter keys

if __name__ == '__main__':
    import timeit
    print("--------------------NON-CONCURENT--------------------")
    elapsed_time = timeit.timeit("get_headlines()", setup="from __main__ import get_headlines", number=2)/2                 #create a timer with passed statement and setup code being the get_headlines() function,
                                                                                                                            #set the number of executions to two, divide the whole thing by two, and save the results in elapsed_time variable           
    print("--------------------CONCURENT--------------------")
    elapsed_time2 = timeit.timeit("concurent_get_headlines()", setup="from __main__ import concurent_get_headlines", number=2)/2 #create a timer with passed statement and setup code being the concurent_get_headlines() function,
                                                                                                                                 #set the number of executions to two, divide the whole thing by two, and save the results in elapsed_time2 variable               
    print("non_concurent: ",elapsed_time," concurent: ",elapsed_time2)                                                           #print execution time of non_concurent and concurent function