从文本文件Python创建字典

Question

I have to create a dictionary based on a csv file that looks like this:

'song, 2000, 184950'
'boom, 2009, 83729'
'boom, 2010, 284500'
'boom, 2011, 203889'
'pow, 2000, 385920'
'pow, 2001, 248930'

from this, I have to create a dictionary that consists of the word as a key and then a list of class objects as a value.

This is what I have so far...

class Counter():
   __slots__ = ('year', 'count')
   _types = (int, int)

def readfile(file):
   d = dict()
   with open(file) as f:
      for line in f:
          element = line.split(,)
          for word in element:
              if word in d:
                  d[word].append([Count(int(element[1]), int(element[2]))])
              else:
                  d[word] = [Count(int(element[1]), int(element[2]))]
   print(d)

the output I'm getting is weird and it is giving me a dictionary similar to what mine should look like but it's using the counts (183930) as the key instead of the name. I also need it to add the class onto the value if it is already listed in the dictionary.

for example, since 'boom' should already be in the dictionary with {'boom' : Count(year = 2009, count = 83729)} I want there to be a list of those Count objects under the one value.

expected output:

{'song' : [Count(year= 2000, count= 184950)], 'boom' : [Count(year=2009, count=83729),
Count(year=2010, count= 284500), Count(year=2011, count=203889)], 'pow' : ...etc..} 

Answer 1

With this loop:

for word in element:
    if word in d:
        d[word].append([Count(int(element[1]), int(element[2]))])
    else:
        d[word] = [Count(int(element[1]), int(element[2]))]

You are iterating trough all words on line, so you call (for first line):

d['song'].append([Count(int('2000'), int('184950'))])
d['2000'].append([Count(int('2000'), int('184950'))])
d['184950'].append([Count(int('2000'), int('184950'))])

Just use:

for line in f:
    element = line.split(,)
    word = element[0]
    if word in d:
        d[word].append(Count(int(element[1]), int(element[2])))
    else:
        d[word] = [Count(int(element[1]), int(element[2]))]

You also can replace your condition if word in d if you use collections.defaultdict :

import collections

def readfile(file):
   d = collections.defaultdict(list)
   with open(file) as f:
      for line in f:
          element = line.split(,)
          word = element[0]
          d[word].append(Count(int(element[1]), int(element[2])))

   print(d)

Answer 2

Here's some simple code that does what you're looking for.

from collections import defaultdict
from pprint import pprint


class Counter(object):
    __slots__ = ('year', 'count')

    def __init__(self, year, count):
        self.year = year
        self.count = count

    def __str__(self):
        return "Counter(year=%d, count=%d)" % (self.year, self.count)

    def __repr__(self):
        return self.__str__()


def import_counts():
    counts = defaultdict(list)
    with file('data.csv') as f:
        for line in f:
            name, year, count = line.split(',')
            name, year, count = name.strip(), int(year.strip()), int(count.strip())
            counts[name].append(Counter(year, count))

    return counts

pprint(import_counts())

However, I changed the data format into a proper CSV as below.

song, 2000, 184950
boom, 2009, 83729
boom, 2010, 284500
boom, 2011, 203889
pow, 2000, 385920
pow, 2001, 248930

The output generated is as below:

{
    'boom': [
        Counter(year=2009, count=83729),
        Counter(year=2010, count=284500),
        Counter(year=2011, count=203889)
    ],
    'pow': [
        Counter(year=2000, count=385920), 
        Counter(year=2001, count=248930)
    ],
    'song': [
        Counter(year=2000, count=184950)
    ]
}

Do note that the above doesn't validate input and would error if given an invalid CSV.