Skripta

#Import of required libraries for this project
import requests
import json
import time
import re
import pandas
import openpyxl

#Defining the start and the end of the first day that we want to analyze
start = 1501545600
end = 1501631999

#Initialization of every list
dates = []
comments = []
scores = []
positive_sentiment = []
negative_sentiment = []
positive_slash_negative = []
sentiment_score = []

#Import of positive and negative pools of words
p = open("positive.txt",'r')
n = open("negative.txt",'r')

x = p.read().split("\n")
y = n.read().split("\n")

i = 1
number_of_days = 365
while i < number_of_days:
    #URL creation
    url = 'https://api.pushshift.io/reddit/submission/search/?after='+str(start)+'&before='+str(end)+'&sort_type=score&sort=desc&subreddit=Bitcoin&category=best&size=30'
    r = requests.get(url)

    #Data export from the Response
    responseData = r.json()
    data = responseData['data']

    #Reseting all the counters and scores to 0 for the new posts
    counter = 0
    positive_score = 0
    negative_score = 0
    total_score = 0
    total_comments = 0
    for post in data:
        #Data export from the post
        post_id = post['id']
        title = post['title']
        comment = post['num_comments']
        score = post['score']
        if 'selftext' not in post:
            text = ""
        else:
            text = post['selftext']

        total_score = total_score + score
        total_comments = total_comments + comment

        positive_counter=0
        negative_counter=0
        #TITLE ANALYSYS
        title = re.sub('[,.!?;:]',' ',title)
        words_title = re.findall(r"[\w']+", title)
        for word in words_title:
            for positive_word in x:
                if positive_word == word.lower():
                    positive_counter = positive_counter + 1
                    break
            for negative_word in y:
                if negative_word == word.lower():
                    negative_counter = negative_counter + 1
                    break

        #IF TEXT AVAILABLE TEXT ANALYSYS
        if text != "":
            text = re.sub('[,.!?;:]',' ',text)
            words_text = re.findall(r"[\w']+", text)

            positive_counter=0
            negative_counter=0
            for word in words_text:
                for positive_word in x:
                    if positive_word == word.lower():
                        positive_counter = positive_counter + 1
                        break
                for negative_word in y:
                    if negative_word == word.lower():
                        negative_counter = negative_counter + 1
                        break
        if text == "":
            counter = counter + 1
        positive_score = positive_score + positive_counter
        negative_score = negative_score + negative_counter

    #Date transformation from timestamp to datetime
    date = time.strftime('%Y-%m-%d', time.localtime(start))

    #Adding all the elements to their lists
    dates.append(date)
    comments.append(str(total_comments))
    scores.append(str(total_score))
    positive_sentiment.append(str(positive_score))
    negative_sentiment.append(str(negative_score))
    positive_slash_negative.append(str(positive_score/negative_score))
    sentiment_score.append(str(positive_score-negative_score))

    #Writing out to the console that the data for this date have been finished
    day = {"date":date,"positive score":positive_score,"negative score:":negative_score}
    print(day)

    #Changing the iterator for the next day
    i=i+1
    start+=86399
    end+=86399

#Data export to datasheet
new_dataframe = pandas.DataFrame(
    {
                                 "Date": dates,
                                 "Num of comments on 30 posts": comments,
                                 "Score on 30 posts": scores,
                                 "Positive Sentiment": positive_sentiment,
                                 "Negative Sentiment": negative_sentiment,
                                 "P/N": positive_slash_negative,
                                 "Sentiment Score": sentiment_score
    }
)
writer = pandas.ExcelWriter('output.xlsx')
new_dataframe.to_excel(writer,'Sheet1')
writer.save()