def url_to_sentiment(url):
"""Takes a URL from the user, """
from newspaper import Article
a = Article(url)
a.download()
a.parse()
article = a.text[:]
r = str(article)
r = r.splitlines()
analyzer = SentimentIntensityAnalyzer()
a = [
] #initialize the empty list 'a', where we will store the polarity scores of the individual lines
for i in range(0, len(r)):
a.append(str(analyzer.polarity_scores(r[i])))
letter_list = [
] #initialize the list where we will store all the letters of the list of polarity scores
#loop through the list of polarity scores and turn the whole thing into one long string called 'my_string'
for j in range(0, len(a)):
for k in range(0, len(a[j])):
letter_list.append((a[j][k]))
my_string = ''.join(map(str, letter_list))
#remove some punctuation from 'my_string', leaving } to be used to split into a list later
my_string = my_string.replace("'", '')
my_string = my_string.replace("{", '')
my_string = my_string.replace(",", '')
my_string = my_string.replace(' ', ' ')
my_string = my_string.replace(': ', ':')
#split back into a list of strings with punctuation removed
url_list_inp = my_string.split("}")
#initialize my lists of values for the four sentiments, neg, neu, pos, and comp
neg = []
neu = []
pos = []
comp = []
#scrapes 'my_list' for the values that correspond to each of the sentiments
#and sorts them into their respective lists.
for g in range(0, len(url_list_inp)):
for h in range(0, len(url_list_inp[g])):
if (url_list_inp[g][h] == ".") and (url_list_inp[g][h - 5:h - 1]
== "neg:"):
neg.append(url_list_inp[g][h - 1:h + 3])
if (url_list_inp[g][h] == ".") and (url_list_inp[g][h - 5:h - 1]
== "neu:"):
neu.append(url_list_inp[g][h - 1:h + 3])
if (url_list_inp[g][h] == ".") and (url_list_inp[g][h - 5:h - 1]
== "pos:"):
pos.append(url_list_inp[g][h - 1:h + 3])
if (url_list_inp[g][h] == ".") and (url_list_inp[g][h - 5:h - 1]
== "und:"):
comp.append(url_list_inp[g][h - 1:h + 3])
if (url_list_inp[g][h - 2] == '-'):
comp.append(url_list_inp[g][h - 2:h + 3])
#initialize a new group of lists, which will store the values of neg, neu, pos,
#after their values are tranformed to floats
neg_float = []
neu_float = []
pos_float = []
comp_float = []
index = []
#creates an index
for i in range(0, 7211):
index.append(i + 1)
#scrapes the respective lists, converts them to floats, deposits them
#into their respective _float lists.
for eins in range(0, len(neg)):
neg_float.append(float(neg[eins]))
for zwei in range(0, len(neu)):
neu_float.append(float(neu[zwei]))
for drei in range(0, len(pos)):
pos_float.append(float(pos[drei]))
for vier in range(0, len(comp)):
comp_float.append(float(comp[vier]))
#initialzes a new list which will only include from instances where
#comp_float i != 0.0
neg_float_new = []
neu_float_new = []
pos_float_new = []
comp_float_new = []
index_new = []
#create an index
for i in range(0, 7211):
index_new.append(i + 1)
#scrape comp_float looking for 0.0 values. if this index value has no
#corresponding comp_float value, remove corresponding neg,neu,float vals
for i in range(0, len(comp_float)):
if (comp_float[i] == 0.0):
pass
else:
neg_float_new.append(neg_float[i])
neu_float_new.append(neu_float[i])
pos_float_new.append(pos_float[i])
comp_float_new.append(comp_float[i])
#calculates the mean of each list, rounding the results to 3 decimal places
neg = stat.mean(neg_float_new)
neu = stat.mean(neu_float_new)
pos = stat.mean(pos_float_new)
comp = stat.mean(comp_float_new)
x = pos
y = neu
z = neg
url_list_inp = [x, y, z, comp]
#print (str(url_list_inp))
return url_list_inp
