从列表中获取两个日期之间的元素

Question

I have a list of list which goes as follows:

A = [['05-22-2013', '0.5553', '0.887', '0.14'], 
     ['05-22-2013', '0.3442', '0.345', '0.0'], 
     ['05-22-2013', '0.3', '0.7', '0.4'], 
     ['05-23-2013', '0.53', '0.87', '0.4'], 
     ['05-23-2013', '0.9', '0.8', '0.1'], 
     ['05-23-2013', '0.0', '0.799', '0.214'],
     ['05-24-2013', '0.053', '0.7', '0.1422'], 
     ['05-25-2013', '0.5', '0.110', '0.200'], 
     ['05-25-2013', '0.311', '0.799', '0.426'], 
     ['05-25-2013', '0.311', '0.091', '0.41']]

I want to select all the elements between dates of '05-22-2013' to '05-24-2013'. What I am trying is manually getting all the elements corresponding to each date in an array and appending them all into a single list of list. Like for one date:

date_1 = []

for u in A:
    if '05-22-2013' in u:
        date_1.append(u)

So, specifically stating, what is the best possible way to get all the items within a given date range (in this sort of list of lists)?

Answer 1

I don't know if this is the "best" way, because let's be honest, "best" is a very subjective term.

from datetime import date

start_date = date(2013, 5, 22)
end_date = date(2013, 5, 24)

data = []

for i in A:
    month, day, year = i[0].split('-')
    record_date = date(year, month, day)
    if record_date >= start_date and record_date <= end_date:
        data.append(i)

Answer 2

To simply get all list entries that match a given date, you can build a list comprehension:

print [entry for entry in A if entry[0] == '05-25-2013']

RETURNS:

[['05-25-2013', '0.5', '0.110', '0.200'], ['05-25-2013', '0.311', '0.799', '0.426'], ['05-25-2013', '0.311', '0.091', '0.41']]

This approach only addresses one date, of course, but could be modified in the if to handle additional dates. The downside is you'd have to explicitly list off each date in the range, as they are all interpreted as mere strings.

If you really need to be able to enter a range, though, you'll have to get more indepth and probably use the datetime module, which would convert it into a datatype capable of using > and < -like operators.

Answer 3

If linear-time preprocessing is an option, then first get the keys (dates) out into a separate list. To make them orderable, convert them to datetime objects first.

>>> from datetime import datetime
>>> def parsedate(s):
...     return datetime.strptime(s, '%M-%d-%Y')
>>> keys = [parsedate(x[0]) for x in A]

Then you can do binary searches to cheaply (in O(lg n) time) do range searches:

>>> from bisect import bisect_left, bisect_right
>>> left = bisect_left(keys, parsedate('05-22-2013'))
>>> right = bisect_right(keys, parsedate('05-24-2013'))

Now A[left:right] is the range you're looking for.