使用多个拆分选择文本

Question

I've started to learn python and am stuck on an assignment regarding manipulating text data. An example of the text lines I need to manipulate:

From stephen.marquard@uct.ac.za Sat Jan  5 09:14:16 2008

I need to extract the hours from each line (in this case 09) and then find the most common hours the emails were sent.

Basically, what I need to do is build a for loop that splits each text by colon

split(':')

and then splits by space:

split()

I've tried for hours, but can't seem to figure it out. What my code looks like so far:

name = raw_input("Enter file:")
if len(name) < 1 : name = "mbox-short.txt"
handle = open(name)
counts = dict()
lst = list()
temp = list()
for line in handle:
    if not "From " in line: continue
    words = line.split(':')  
    for word in words:
        counts[word] = counts.get(word,0) + 1

for key, val in counts.items():
    lst.append( (val, key) )
lst.sort(reverse = True)

for val, key in lst:
print key, val

The code above only does 1 split, but I've kept trying multiple methods to split the text again. I keep getting a list attribute error, saying "list object has no attribute split". Would appreciate any help on this. Thanks again

Answer 1

First,

import re

Then replace

words = line.split(':')  
for word in words:
    counts[word] = counts.get(word,0) + 1

by

line = re.search("[0-9]{2}:[0-9]{2}:[0-9]{2}", line).group(0)
words = line.split(':')
hour = words[0]
counts[hour] = counts.get(hour, 0) + 1

Input:

From stephen.marquard@uct.ac.za Sat Jan  5 09:14:16 2008
From stephen.marquard@uct.ac.za Sat Jan  5 12:14:16 2008
From stephen.marquard@uct.ac.za Sat Jan  5 09:14:16 2008
From stephen.marquard@uct.ac.za Sat Jan  5 09:14:16 2008
From stephen.marquard@uct.ac.za Sat Jan  5 15:14:16 2008
From stephen.marquard@uct.ac.za Sat Jan  5 12:14:16 2008
From stephen.marquard@uct.ac.za Sat Jan  5 09:14:16 2008
From stephen.marquard@uct.ac.za Sat Jan  5 13:14:16 2008
From stephen.marquard@uct.ac.za Sat Jan  5 12:14:16 2008

Output:

09 4
12 3
15 1
13 1

Answer 2

Using the same test file as Marcel Jacques Machado:

>>> from collections import Counter
>>> Counter(line.split(' ')[-2].split(':')[0] for line in open('input')).items()
[('12', 3), ('09', 4), ('15', 1), ('13', 1)]

This shows that 09 occurs 4 times while 13 occurs only once.

If we want prettier output, we can do some formatting. This shows the hours and their counts sorted from most common to least common:

>>> print('\n'.join('{} {}'.format(hh, n) for hh,n in Counter(line.split(' ')[-2].split(':')[0] for line in open('input')).most_common()))
09 4
12 3
15 1
13 1