#!/usr/bin/env python
# coding: utf-8

# # EXPLORING HACKER NEWS

# #### In the project we are going to Analyse 2 things:
# 1-Do ASK HN or Show HN receive more comments on average?
# 2-Do post created at a certain time receive more comments on average?

# In[36]:


from csv import reader
open_file=open('hacker_news.csv')
read=reader(open_file)
hn=list(read)
headers=hn[0]
hn=hn[1:]


# In[37]:


print(headers)


# In[38]:


print(hn[:5])


# In[39]:


ask_posts=[]
show_posts=[]
other_posts=[]
for i in hn:
    title=i[1]
    if title.lower().startswith('ask hn'):
        ask_posts.append(i)
    elif title.lower().startswith('show hn'):
        show_posts.append(i)
    else:
        other_posts.append(i)


# In[40]:


print('Number of Ask_Posts',len(ask_posts))
print('Number of Show_Posts',len(show_posts))
print('Number of Other_Posts',len(other_posts))


# In[41]:


print(ask_posts[:5])


# ### Calculating Total and Average Ask comments

# In[42]:


total_ask_comments=0
for i in ask_posts:
    num_comments=i[4]
    num_comments=int(num_comments)
    
    total_ask_comments+=num_comments
    
print('Total Ask Comments',total_ask_comments)
avg_ask_comments=total_ask_comments/len(ask_posts)
print('Average Ask Comments',avg_ask_comments)
  

    
    
    


# #### Calculating Total and Average Show Comments

# In[43]:


total_show_comments=0
for i in show_posts:
    num_comments=i[4]
    num_comments=int(num_comments)
    total_show_comments+=num_comments
print('Total Show Comments are:',total_show_comments)
avg_show_comments=total_show_comments/len(show_posts)
print('Average show comments',avg_show_comments)


# #### Apparantly Ask comments receive more comments on topic based on the calculated average which is 14.03

# In[44]:


import datetime as dt


# In[45]:


result_list=[]
counts_by_hour={}
comments_by_hour={}

for i in ask_posts:
    created_at=i[6]
    num_comments=i[4]
    num_comments=int(num_comments)
    result_list.append((created_at,num_comments))
#print(result_list)
for i in result_list:
    date=dt.datetime.strptime(i[0], "%m/%d/%Y %H:%M")
    date=date.strftime("%H")
    comment=i[1]
    if date not in counts_by_hour:
        counts_by_hour[date]=1
        comments_by_hour[date]=comment
    else:
        counts_by_hour[date]+=1
        comments_by_hour[date]+=comment


# In[46]:


print(comments_by_hour)


# In[47]:


print(counts_by_hour)


# In[48]:


avg_by_hour=[]
for i in comments_by_hour:
    
    avg_by_hour.append([i,comments_by_hour[i]/counts_by_hour[i]])


# In[49]:


print(avg_by_hour)


# In[52]:


swap_avg_by_hour=[]
for i in avg_by_hour:
    swap_avg_by_hour.append((i[1],i[0]))


# In[54]:


print(swap_avg_by_hour)


# In[56]:


sorted_swap=sorted(swap_avg_by_hour,reverse=True)


# In[57]:


print(sorted_swap)


# In[62]:


for i,j in sorted_swap[:5]:
    date=dt.datetime.strptime(j,'%H')
    date=date.strftime('%H:%M')
    print('{} {:.2f} average comments per post'.format(date,i))
    


# ### Here we understoon that at 15:00 every day we can expet the highest number of comments per post(ASK) based on our calculations