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

# # Working on Hacker news articles 

# In[8]:


from csv import reader
file=open('hacker_news.csv')
dataset=reader(file)
hn=list(dataset)

print(hn[:5])
print(len(hn))


# In[9]:


hn_headers=hn[0]
print(hn_headers)
hn=hn[1:]
print(hn[:5])


# In[20]:


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)
        
len(ask_posts) 
print('\n')
len(show_posts) 
 


# In[21]:


print(ask_posts[:2])
print('\n')
print(show_posts[:2])


# In[22]:


total_ask_comments=0
length=0
for row in ask_posts:
    a=row[4]
    total_ask_comments+=int(a)
    length+=1
avg_ask_comments=total_ask_comments/length
print(avg_ask_comments)


# In[23]:


total_show_comments=0
show_length=0
for row in show_posts:
    a=row[4]
    total_show_comments+=int(a)
    show_length+=1
avg_show_comments=total_show_comments/show_length
print(avg_show_comments)


# In[68]:


import datetime as dt
result_list=[]
for row in ask_posts:
    date=row[6]
    comments=int(row[4])
    result_list.append([date,comments])
    
#print(result_list)    
print(len(ask_posts))    
print(len(result_list))    
counts_by_hour={}
comments_by_hour={}
total=0
for row in result_list:
    date=row[0]
    comment=int(row[1])
    #print(date)
    dt_object=dt.datetime.strptime(date,"%m/%d/%Y %H:%M")
    #print(start_date)
    hour=dt_object.strftime("%H")
    #print(hour)
    if hour in counts_by_hour:
        counts_by_hour[hour]+=1
        comments_by_hour[hour]+=comment
    else:
        counts_by_hour[hour]=1
        comments_by_hour[hour]=comment
        
    total+=comment    
        
print(counts_by_hour)
print(comments_by_hour)
print(total)


# In[74]:


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


# In[91]:


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

sorted_swap=sorted(swap_avg_by_hour, reverse=True)

print("Top 5 hours for Ask posts comments")
for i in sorted_swap[:5]:
    print(i)
print('\n')
demo=sorted_swap[0]
#print(demo[1])
date_object=dt.datetime.strptime(demo[1],"%H")
Hour=date_object.strftime("%H:%M")
#print(Hour)
print("{}: {:.2f} average commnets per post".format(Hour,demo[0]))


# # Around 3 pm is the right time to put the post in order to receive maximum comments