我有一个包含日期范围的元组列表，如下所示：
[('2012-01-01', '2012-01-25'),
('2012-01-26', '2012-02-20'),
('2012-02-21', '2012-03-16'),
('2012-03-17', '2012-04-11'),
('2012-04-12', '2012-05-06')...]
def pull_unsampled(date_list):
df123 = []
for start, end in date_list:
df123.append(ga.read_ga(metrics,
dimensions = "yearMonth",
start_date = start,
end_date = end,
account_id = account_id, profile_id = profile_id))
print start, end
return df123
pull_unsampled(unsampled_date_ranges)
在上面的功能中，我获取元组列表，并通过Pandas将开始日期和结束日期传递给Google AnalyticsAPI请求。我打印了通过for循环的开始和结束日期，并打印出它们在原始列表中的显示方式。当我收到回复数据时，我会收到没有像输入那样格式化的日期，事实上我给了不同的日期。
[ visits transactionsPerVisit \
deviceCategory operatingSystem yearMonth
desktop (not set) 2020-12-01 137 0.000000
Chrome OS 2020-12-01 20 0.000000
Google TV 2020-12-01 5 0.000000
Linux 2020-12-01 363 0.275482
Macintosh 2020-12-01 111467 0.909686
Nintendo Wii 2020-12-01 2 0.000000
Playstation 3 2020-12-01 1 0.000000
SunOS 2020-12-01 3 0.000000
Windows 2020-12-01 98147 0.996464
mobile Android 2020-12-01 2976 0.302419
BlackBerry 2020-12-01 1051 0.000000
Nokia 2020-12-01 2 0.000000
Samsung 2020-12-01 1 0.000000
SymbianOS 2020-12-01 4 0.000000
Windows Phone 2020-12-01 46 0.000000
iPad 2020-12-01 2 0.000000
iPhone 2020-12-01 24754 0.379737
iPod 2020-12-01 331 0.000000
tablet Android 2020-12-01 189 0.529101
iPad 2020-12-01 24973 0.828895
由于某种原因，('2012-01-01', '2012-01-25')的日期范围会返回上面yearMonth列的不正确日期，但后来的响应会正确显示。例如，('2014-03-06', '2014-03-31')的最后一个日期范围提取正确显示yearMonth：
avgTimeOnSite pageviewsPerVisit
deviceCategory operatingSystem yearMonth
desktop (not set) 201403 1.127708 1.013683
Chrome OS 201403 159.150562 3.811236
Google TV 201403 0.000000 1.000000
Linux 201403 431.220165 8.705575
Macintosh 201403 219.625070 5.193008
Nintendo Wii 201403 0.000000 1.000000
Playstation 3 201403 65.000000 3.000000
Windows 201403 186.650353 4.077768
Xbox 201403 0.000000 1.000000
mobile (not set) 201403 51.151515 2.212121
Android 201403 143.776396 3.195142
BlackBerry 201403 132.644037 3.049541
Nokia 201403 0.400000 1.000000
Series40 201403 5.071429 1.214286
SymbianOS 201403 0.000000 1.000000
Windows Phone 201403 172.685446 3.305164
iOS 201403 126.005593 2.954413
tablet Android 201403 183.609062 3.746048
BlackBerry 201403 224.421053 3.789474
iOS 201403 217.915084 4.462462
这是list.append，pandas还是Google AnalyticsAPI中的错误？我的代码中是否有一些东西可以防止这些奇怪的日期代替实际的日期？
编辑：
这是一个简单的调用，它返回相同日期范围内的yearMonth数据：
df = ga.read_ga(metrics,
dimensions = dimensions,
start_date = start_date,
end_date = end_date,
account_id = account_id, profile_id = profile_id)
df
返回为：
visits
yearMonth
201001 270094
201002 254034
201003 305295
201004 311824
201005 281570
201006 249348
201007 247005
201008 245882
201009 252296
201010 266417
201011 274349
201012 272913
201101 259959
201102 261080
201103