Reputation: 119
I have the following code that works great. It get my IP address out of a file and counts how many times they appear in the logfile.
def count_ips():
fp=open('logfile','r')
store=[]
while 1:
line=fp.readline()
if not line:
break
if line[-1:] == '\n':
line=line[:-1]
data1=line.split('"')
data2=data1[0].split(' ')
store.append({'IP':data2[0],'Date':data2[3]+' '+data2[4],'Action':' '.join(data1[1:-2]),'Browser':' '.join(data1[-2:])})
fp.close()
count={}
for i in store:
if i['IP'] in count:
count[i['IP']] +=1
else:
count[i['IP']] =1
avg=0
cnt=0
for i in count:
avg+=count[i]
cnt+=1
avg=avg/cnt
print 'average hit is: %i' % avg
for i in count:
if count[i] > 10:
print i +' %i' % count[i]
count_ips()
I dont really know how I got to this point but in this section. I would like to sort by the count before I print it out. Biggest number on the bottom.
for i in count:
if count[i] > 10:
print i +' %i' % count[i]
I feel at this point im just looking at things wrong and dont see the easy fix for my little dilemma.
Thank you for you help! Jason
Upvotes: 1
Views: 5995
Reputation: 3828
Whenever I have to treat dictionaries as data I use pandas.
import pandas as pd
pd.DataFrame(list(dict.items()), columns= ['IP','count']).sort_values('count')
Notice that the items from the dictionary are called with dict.items()
and then passed to a list. If using python 2.X then you should omit the list()
call.
Upvotes: 0
Reputation: 494
so assume that you have a dictionary d which contain keys that are IPs and values are the counts.
>>> d = {'1.1.1.1':5, '2.2.2.2':4}
Here is what I would do in a one liner:
>>> sorted((d[ip], ip) for ip in d)
[(4, '2.2.2.2'), (5, '1.1.1.1')]
You can also use parameter reverse=True to sorted the list in reversed order.
Upvotes: 1
Reputation: 142146
Assuming that count
is your dict of IP->Count, then:
from operator import itemgetter
sorted_counts = sorted(count.iteritems(), key=itemgetter(1))
for ip, cnt in sorted_counts:
print ip, 'had', cnt, 'results'
Upvotes: 4