去除字典

Question

具体文字及字符有像这样的列表（只有这样，更大）被称为“DATA_LIST”：

2017-04-01, available 
2017-04-02, available 
2017-04-01, available 
2017-04-02, available 
2017-04-02, available 
2017-04-01, available 
2017-04-02, available 
2017-04-01, available 
2017-04-02, available 
2017-04-01, available 
etcetera 

我用下面的代码，

dates = collections.defaultdict(list) 
for date, xyz in data_list: 
    dates[date].append(xyz) 
counts = {date: collections.Counter(xyz) for date, xyz in dates.items()} 

要创建像一本字典这个：

2017-04-01,Counter({'available': 9}) 
2017-04-02,Counter({'available': 12}) 
2017-04-03,Counter({'available': 9}) 
2017-04-04,Counter({'available': 4}) 
2017-04-05,Counter({'available': 9}) 
2017-04-06,Counter({'available': 2}) 

我将如何继续删除“计数器”？ （最终像“（”和“{”字符）

目前，我有这样的代码，但它不会做任何事情。

for x in my_dictionary: 
try: 
    x = x.replace('Counter','') 
except: 
    pass 

主要目标最终是让.csv文件像这样：

date, available 
2017-04-01, 9 
2017-04-02, 12 
2017-04-03, 9 
2017-04-04, 4 
2017-04-05, 9 
2017-04-06, 2 

字典的打印输出部分：

'2018-12-12': Counter({'available': 3}), '2018-04-28': Counter({'available': 4}), '2017-12-16': Counter({'available': 2}), '2017-12-17': Counter({'available': 2}), '2017-12-14': Counter({'available': 2}), '2017-12-15': Counter({'available': 2}), '2017-12-12': Counter({'available': 2}), '2017-12-13': Counter({'available': 2}), '2017-12-10': Counter({'available': 2}), '2017-12-11': Counter({'available': 2}), '2017-12-18': Counter({'available': 2}), '2017-12-19': Counter({'available': 2}), '2018-05-31': Counter({'available': 4}), '2018-05-30': Counter({'available': 4}), 

Answer 1

就快可以得到AV通过使用available键ailable数从柜台得到的，就像这样：

counts = {date: collections.Counter(xyz)['available'] for date, xyz in dates.items()} 


import csv 

def to_row(date, counter): 
    return date, counter['booked'], counter['blocked'], counter['available'] 

counts = [to_row(date, collections.Counter(xyz)) for date, xyz in dates.items()] 

writer = csv.writer(open('<filename>.csv', 'w')) 
writer.writerows(counts) 

Answer 2

你不需要在这种情况下collections.Counter可言的，你甚至可以用collections.defaultdict做了。这将做伎俩：

dates = {} 
for date, value in data_list: 
    if value == "available": 
     dates[date] = dates.get(date, 0) + 1 
# dates contains (date, count) pairs 

而且应该也快得多。然后你可以使用csv.writer或csv.DictWriter（取决于你想要的输出CSV）写出你的最终CSV。例如：

import csv 

data_list = [['2017-04-01', 'available'], 
      ['2017-04-02', 'available'], 
      ['2017-04-01', 'available'], 
      ['2017-04-02', 'available'], 
      ['2017-04-02', 'available'], 
      ['2017-04-01', 'available'], 
      ['2017-04-02', 'available'], 
      ['2017-04-01', 'available'], 
      ['2017-04-02', 'available'], 
      ['2017-04-01', 'available']] 

dates = {} 
for date, value in data_list: 
    if value == "available": 
     dates[date] = dates.get(date, 0) + 1 

with open("output.csv", "wb") as f: # open output.csv for writing 
    writer = csv.writer(f) # create a csv.writer 
    writer.writerow(("date", "available")) # write our header 
    for row in dates.iteritems(): # sorted(dates.iteritems()) instead for date-sorted output 
     writer.writerow(row) # write the row 

让你有效的CSV为：

date,available 
2017-04-02,5 
2017-04-01,5 

，您可以在几乎任何电子表格应用程序打开它。如果您想将其格式化为您的输出，请注意这不是有效的CSV。

UPDATE - 一个版本有多个可用值对于每一日期（在使用collections.Counter变得更方便了这一点，但为了保持与主题）：

import csv 

data_list = [['2017-04-01', 'available'], 
      ['2017-04-02', 'available'], 
      ['2017-04-01', 'booked'], 
      ['2017-04-02', 'available'], 
      ['2017-04-02', 'booked'], 
      ['2017-04-01', 'available'], 
      ['2017-04-02', 'blocked'], 
      ['2017-04-01', 'blocked'], 
      ['2017-04-02', 'blocked'], 
      ['2017-04-01', 'available']] 

dates = {} 
values = set() # just so we know what are possible values for the latter CSV header 
for date, value in data_list: 
    values.add(value) 
    dates.setdefault(date, {})[value] = dates.get(date, {}).get(value, 0) + 1 

with open("output.csv", "wb") as f: # open output.csv for writing 
    header = ["date"] + list(values) # set header to date,<available_values> 
    writer = csv.DictWriter(f, header) 
    writer.writeheader() 
    for k, v in dates.iteritems(): # sorted(dates.iteritems()) instead for date-sorted output 
     v.update({"date": k}) # add the date to our row 
     writer.writerow(v) # write the row 

创建output.csv为：

date,available,booked,blocked 
2017-04-02,2,1,2 
2017-04-01,3,1,1 

您可以拥有任意数量的“价值”字段，因此它不必仅为3。